Talk:Paper - The course of the blood through the heart of the fetal mammal, with a note on the reptilian and amphibian circulations (1909)

From Embryology

The Couese Of The Blood Through The Heart Of The Fetal Mammal, With A Note On The Reptilian And Amphibian Circulations

Pohlman AG. The course of the blood through the heart of the fetal mammal, with a note on the reptilian and amphibian circulations. (1909) Anat. Rec. 3: 75-109.


Augustus Grote Pohlman, Indiana University.

The problem of the course of the blood through the heart of the fetal mammal has been taken up because there are three distinct theories regarding the fate of the blood entering the heart through the superior and inferior caval veins. Each of these theories is based upon anatomical findings — a correlation of function to structure, and while injection experiments have been carried out which seem to substantiate each theory, experimental evidence derived from a study of the fetal circulation in the living embryo is entirely lacking. It is our purpose, therefore, to review the important literature; to state the position of the various theories together with their modifications; to analyze critically the evidence produced by the observers ; to present the results of personal findings in the living mammalian embryo ; and to offer a general summary of the physical, anatomical, and pathological evidence in favor of the one as opposed to all other theories. Finally to give a preliminary account of our observations on the circulation of the reptile and amphibian.

The literature up to the time of Harvey (1628) is drawn from Dalton's excellent work. Extremely comprehensive reviews are to be found in Knabbe's dissertation in Latin, and also in Kilian's article in Gterman. With the exception of the first six references, where direct control is offered by comparison with the original text, and of Wolff's treatise, the articles have been read personally. The reader is referred to literature reviews mentioned for more exhaustive study,


D 76 Augustus Grote Pohlman.

The first definite information on the structure of the fetal heart is found in Galen's work: "In this matter, we have reason to admire the provisions of nature. For so long as the lung has only to be nourished and grow, it is supplied simply with blood; but, when it is ready to take on an active motion, its tissue becomes lighter and capable of expansion and compression by the movements of the chest. On that account the vena cava (right auricle), in the foetus, communicates by an opening with the arteria venalis (left auricle). As this latter vessel thus performs for the lung the office of a vein (that is, supplies it with blood for its nourishment), its companion (the pulmonary artery) must need at this time serve the purpose of an artery, and it is consequently made to communicate with the aorta. As these two vessels (pulmonary artery and aorta) are situated a little distance apart, their communication i? effected by means of a third smaller one (ductus arteriosus), which forms a junction with each. In the case of the other two (auricles), which lie in contact with each other, there is a kind of orifice or fenestra (foramen ovale) common to both. At this orifice there is attached a membrane, like a lid or cover, opening toward the pulmonary vessel (left auricle), so that it will yield to the infiux of blood from the vena cava (right auricle), but will prevent its regurgitation into that vessel."

Galen furthermore mentions the fate of the foramen ovale and the ductus arteriosus after birth. It must be remembered that at this time (third century) practically all of the work on the heart had been done hundreds of years before by Aristotle, Herophilus and Erasistratus, and despite the investigations of the latter two men on the character of the veins and arteries and the valves in the heart, the doctrine of circulation was shrouded in mystery. Galen assumed that the blood passed from the right into the left ventricle through the ventricular septum, and this may in part account for the curious notion that the ductus arteriosus transmitted blood from the aorta into the pulmonary artery — the passage is ambiguous and this is one interpretation of it. His description, however^ of the foramen ovale, its valve and the method of its obliteration, together with the atrophic changes in the ductus arteriosus, are quite accurate.

D Course of the Blood Through the Fetal Heart. 77

Some thirteen centuries later Vesalius questioned the teachings of Galen regarding the passage of blood through the ventricular septum. "And accordingly, notwithstanding what I have said about the pits in this situation, and at the same time not forgetting the absorption by the portal vein from the stomach and intestines, I still do not see how even the smallest quantity of blood can be transfused, through the substance of the septum, from the right ventricle to the left." While this questioning the infallibility of Galen's work was little more than rank heresy at the time, Vesalius paved the way for others, and doubt was again raised in Servetus' discovery of the pulmonary circulation in 1563.^ Servetus says,

'Tliis communication, however, does not take place through the median wall of the heart, as commonly believed; but by a grand device the refined blood is driven from the right ventricle of the heart, in a long course through the lungs. By the lungs it is prepared, assuming a bright color, and from the vena arteriosa is transferred into the arteria venosa" (pulmonary vein). Servetus also recognized the foramen ovale, and but for his unitarian views, which resulted in his untimely death at the stake, might have contributed more than a mere description of the pulmonary circulation.

The contemporaries of Vesalius and Servetus, Coluiabus and Csesalpinus are also to be mentioned in this connection. We fail to see, however, wherein Columbus bettered the description given by Servetus, and Csesalpinus did not add materially to what was already known on the subject. It must not be thought that the newer teachings were eagerly accepted, for Fallopius held to the Galenic views to the time of his death (1563). In 1566 Botallus reported the persistence of the foramen ovale and re-described the ductus arteriosus. We agree with many other writers that the use of Botallus' name in connection with these structures is questionable since Galen first mentioned them.

The gradual relief from religious persecution and the works of Vesalius and others stimulated investigation. Comparative anatomy,

We do not care to enter into a discussion of who actually did dlscoyer the pulmonary circulation and favor Servetus for the reason that his work appeared six years before that of Oolumbus.

D 78 Augustus Grote Pohlman.

particularly the investigations of Harvey, threw new light on the question. In Harvey^s classical work (1628) we note the first scientific description of the course of the blood through the fetal heart. Unfortunately, the terminology is rather vague and the translation leaves much to be desired. For example, Harvey uses the term Vena cava' as Galen used it — to denote the right auricle, while the left auricle is termed arteria venosa or pulmonary vein; the translator supplied plurals to these structures Venae cavse' and 'pulmonary veins' at discretion, and the meaning is far from being the same. We present a corrected excerpt from Harvey's work, which reads as follows: "The first contact and union of the vena cava (right auricle) with the pulmonary vein (left auricle) which occurs before the vena cava opens into the right ventricle, or gives off the coronary sinus, a little above its escape from the liver, exhibits a lateral anastomosis that is a wide open passage-way from the cava (right auricle) into the vessel already mentioned (left auricle) ; in such a manner, that (as if by a single vessel) the blood can flow very freely and copiously through that opening from the vena cava into the left auricle and through the left auricle all the way into the left ventricle."

Note the similarity of this passage to the description given by Galen. We must give Harvey credit that he knew of the two caval veins and of the multiple pulmonary veins, otherwise the passage would mean nothing. It is generally accepted that the interpretatiQn is as follows — the blood contained in the right auricle passes through the; foramen ovale into the left auricle. He continues: "Further, in this foramen ovale, from the part which regards the pulmonary vein, there is a thin tough membrane, larger than the opening, extended like an operculum or cover ; this membrane in the adult blocking up the foramen, and adhering on all sides, finally closes it up, and almost obliterates every trace ©f it. In the foBtus, however, the membrane is so contrived that falling loosely upon itself, it permits a ready access to the lungs and the heart, yielding a passage to the blood which is streaming from the cava (right auricle), and hindering the tide at the same time from flowing back Into that vein. All things, in short, permit us to believe that in the

D Course of the Blood Through the Fetal Heart. 79

embryo the blood must constantly pass by this foramen from the vena cava into the pulmonary vein, and from thence into the left auricle of the heart; and having once entered there, it can never regurgitate."

It was many years before Harvey^s doctrine of the circulation in the adult was generally accepted, while his views concerning the course of the blood through the fetal heart were greatly obscured through the work of Mery, who claimed to demonstrate the passage of blood from left to right. Mery says, "L'hypothese du passage du sang de I'oreillette gauche par le trou ovale dans le ventricule droit du coeur du foetus humain que j'y propose comme une simple conjecture, n'y appuyee que sur le seul rapport que j'ay trouve entre le trou ovale et le canal de communication du coeur de la Tortue, et les memes conduits du foetus."

This position was championed successfully by Mery throughout his life and despite the objections raised by Duvemey and others, it was approved by Winslow, Littre and practically the entire French Academy. Winslow speaks of the anomaly reported by Vieussena where no foramen ovale was present (heart incompletely described — probably ventricular defect), and Steno's case of defect in the ductus (probably incomplete separation of the ventral aortic stem).

Some years after Mery's death, the injection experiments of Trew and Eoederer proved the scheme a faulty one and in the work of von Haller we find no mention of it whatever. Here again are found vague descriptions; the wording is suggestive of Galen and Harvey, while the context appears to be a forerunner of the Sabatier theory. "But yet the septum betwixt the right and left auricle, conjoining them together, is perforated by a broad oval foramen, through which the blood coming from the abdomen, and a little directed or repelled by the valvular sides of the auricle, flows in a full stream into the pulmonary sinus." In 1773 Senac repeated Mery's injection experiments and found that blood passed from the right into the left auricle but not the reverse.

About this time Wolff found that the relation of the opening of the inferior cava to the foramen ovale was somewhat different than hitherto described. He placed the orifice of the inferior cava dorsally at the border between the two auricles and considered the auricular

D 80 Augustus Grote Pohlman.

septum to be defective in this situation (foram^i ovale). The orifice of the cava inferior was, therefore, split on the limbus Vieussens in such. a manner that the left part of the opening transmitted blood directly into the left auricle through the foramen ovale, while the right part of the opening connected with the right auricle. The foramen ovale, in other words, did not afford a communication between the two auricles. This theory differs from the following one in that it was based on anatomical findings rather than inferred physiological necessity.

Some years later Sabatier presented his famous theory on the course of the blood through the fetal part In his article, he deals with the formation of the inferior cava, devotes a few lines to the passage of blood through the liver, and says of the blood entering the heart through the inferior cava, "Le trou ovale le transmet a I'oreillette gauche." Practically v. Haller's statement The theory was accepted by Bichat and incorporated in his text book on descriptive anatomy: "All the blood that the trunk of the inferior cava receives, . . • , instead of stopping in the right auricle, as in the adult, passes entire into the left through the foramen ovale, the superior edge of which is so arranged that nothing can mix with the blood of the superior cava ; so that it is really with the left auricle that the inferior cava is continued." As we read the literature, it appears that this was really the scheme proposed by von Haller a few years before, but it was probably suggested independently by Sabatier. This theory, the prevalent one to-day, was confirmed by Homer, Murray, and Eeid.

Wolff's anatomical findings were substantiated by Kilian: "Die Vena cava inferior offnet sich nicht allein in den rechten Vorhof, sondem sie ergiesst ihr Blut durch zwei vollkommen isolierte Miindungen durch eine rechte und eine linke, sowohl in das rechte, als wie in das linke Atrium." Kilian, however, went still farther and his monumental work probably came into disrepute because of the curious view he held regarding the distribution of the blood from the right and left ventricles. "Es giebt im Foetus noch keine Arteria pulmonalis, sondern die f alschlich mit diesem Namen belegte Arterie, sammt dem sogenannten Ductus arteriosus Botalli, sind

D Course of the Blood Through the Fetal Heart. 81

ein und dasselbe fortlaufende Gebilde und der Ursprimg eines sich in die untere Korperhalfte fortsetzenden Gefasses, welches den Namen Aorta abdominalis zu tragen verdient, im Gegensatze der Aorta cerebralis, welche aus dem linken Ventrikel entspringt." Kilian believed that all of the blood of the left ventricle went to the head and upper extremities, while the blood from the right side was distributed to the lungs and Aorta descendens. An Aorta cerebralis, therefore, in contradistinction to an Aorta abdominalis.

Meckel's case occlusion of the descending aorta at the fourth thoracic vertebra seemed to conform with this scheme, but a careful examination of the drawing shows the constricted area to be well marked above, as well as below, the remains of the ductus. Even granting Kilian's scheme, the case could not represent a persistence of fetal conditions.

Injection experiments were carried on by Reid (1835) in three specimens. He injected a red mass into the cava inferior and a yellow mass into the cava superior, equal quantities under equal pressure ; and found in one, that some of the red passed into the right auricle, none into the ventricle, while it fiUed the left ventricle. In this case a mistake was made of injecting the superior cava the wrong way. In the second attempt, the two masses mixed in the right auricle, with the comment that the injection was not well managed. The third case showed no mixing of the two masses ; all the red went to the head, while all the yellow into the Aorta descendens. In his second article, commenting on the first, Eedd mentions that he agrees with Sabatier but also states that Magendie's Physiology considers the scheme impossible, while Bostock's Physiology alludes to it as fanciful.

The chief objections to the von Haller-Sabatier theory were taken by Williams and by Peaslee. Williams's article, lamentably hidden, is worthy of no little consideration, because it represents the first critical analysis of the scheme. "Prom a careful examination of the anatomical character and dependence of the Eustachian valve, notwithstanding the opposing experiments of Dr. John Reid, I have recently convinced myself that it is mechanically inefficient as a means of preserving the individuality of the two caval currents

D 82 Augustus Qrote Pohlman.

as they traverse the chamber of the right auricle; at the period of its diastole, when the auricle has attained a moderate limit of distension, it may be readily demonstrated, that the two streams must freely intermingle. It is not true, therefore, that the difference in quality is so considerable as that generally taught by the anatomists between the blood distributed to the anterior segment and that circulating the posterior segment of the body of the foetus." Peaslee, who by the way believed in a marked aspirating action of the auricles, arrived at a similar conclusion. He considered "The foramen ovale, a temporary arrangement to allow the rapid conversion of the reptilian to the mammalian heart," the "mixture of blood in the right auricle," and that "no artery in the foetus contains arterial blood." The statement r^arding the conversion of reptilian to mammalian heart is probably a slip in English — ^what he undoubtedly meant was a change from the type of reptilian circulation to that of the mammal. With no article published in its favor since 1835, the von Haller-Sabatier theory came into disrepute in the literature even if it still occupies its original prominence in the text books on anatomy, physiology, embryology and obstetrics, both human and comparative.

With Eiidinger's finding that the orifice of the inferior cava was divided on the limbus Vieussens as described by Wolff, Kilian and others, interest seemed to be reawakened in the course of the blood through the fetal heart. Preyer, although a champion for the Wolff theory, still inclined toward the idea that the head of the embryo received a better arterial supply and this he gained as follows : The circulation through the lungs he granted was free, and inasmuch as the fetal lung occasioned little waste, the return through the pulmonary veins would be of better quality than the return through the superior cava; now inasmuch as the blood of the inferior cava was passed in equal amounts to both ventricles, the left ventricle would contain a more arterial blood. The idea in a way offers a compromise between the Wolff and von Haller-Sabatier theories and is incorporated in the elaborate scheme of the fetal circulation in Preyer's book.

Ziegenspeck, working under Preyer, waived the question as to

D D D Course of the Blood Through the Fetal Heart. 83

the quality of blood and presented a unique scheme for the pladental circulation. The heart was diagrammed as two hearts, in order to render evident the Wolff idea that the foramen ovale did not afford communication between the two auricles, and this in turn necessitated picturing the vena cava inferior as a forked vein. The scheme proved confusing and in his last article he has redrawn his figure which we present later.

In review, we find the following theories arranged in chronological order :

I. The theory of Galen-Harvey (300 [?] and 1628). Foramen ovale affords communication between the two auricles. Passage of mixed blood from right to left.

II. The theory of Mery (1692). Passage of blood from the left auricle to the right ventricle through the foramen ovale.

III. The theory of Wolff (1775). Foramen ovale does not afford communication between the two auricles but connects with the left opening of the vena cava inferior. Eight opening of that vein leads into the right auricle.

IV. The theory of von Haller-Sabatier (1779-91). Blood of the inferior cava passes over to left auricle through foramen ovale. Foramen ovale does not connect the two auricles.

V. The theory of Kilian (1826). Same as that of Wolff with this modification— division of vessels leaving the heart into Aorta cerebralis and Aorta abdominalis. Descending aortic arch conveys no blood during fetal life.

VI. The theory of Ziegenspeck (1881 and 1905). Same as that of Wolff with modification; that return to heart through superior cava equals return through pulmonary veins ; that Pars communicans aortse transmits the same amount of blood as ductus and that each carries one half of the contents of the left and the right ventricles respectively. "Gesetz der Halbierung des Blutes."

Mery's theory was refuted in the eighteenth century. Kilian's theory has not met with approval since 1835 and neither has the von Haller-Sabatier scheme. The latter one, however, demands some attention. We, therefore, consider it first, next the Wolff and Ziegenspeck contentions, and finally the theory of Galen-Harvey.

D 84 Augustus Grote Pohlman.

The Theory of von Halleb-Sabatiee.

This theory may be interpreted in one of two ways: either the orifice of the inferior cava is placed in close relation to the foramen ovale and that all or practically all of its blood passes to the left; or, interpreted frem the usual diagrams, the streams from the two cavsB cross in the chamber of the right auricle without mixing. The only difference between the former interpretation and the Wolff theory is one of degree and we, therefore, consider it later. We present our criticism to the latter reading taken from a preliminary article. "A critical examination of this theory shows it to be physically impossible, morphologically inaccurate, and developmentally unnecessary." We believe it is simple to show that it is physically impossible to preserve the identity of two currents when they cross within a distending chamber. Bom has already pointed out the morphological inaccuracy of the scheme in that the condition is not to be found in the sauropsidian embryo. The scheme is developmentally unnecessary. It attempts to account for the more rapid growth of the head because that segment then obtains the better quality of blood. The head end of all vertebrates develops more rapidly than the tail end whether this alleged better arterial supply is present or not.

It is our opinion that the von Haller-Sabatier scheme of the fetal circulation should be eliminated from the text books except as a matter of historical interest, and we also believe that the investigators favoring the Wolff school will concur in this statement. This represents the neutral greund, and from here on we differ.

The Theory of Wolff, including Zieqenspeck's Modification.

This theory is based on the anatomical findings that the orifice of the inferior cava is placed dorsally on the auricular septum which is deficient at this point. The free edge of the septum constitutes the limbus Vieussens, and the current in the inferior cava is directed against it in such a manner that the blood is split into two streams ; the left part of the current passes to the left of the limbus directly into the foramen ovale, while the right part passes to the right of

D Course of the Blood Through the Fetal Heart 85

the limbus directly into the right auricle. The foramen ovale does not afford communication between the two auricles and inasmuch as the more arterial blood in the inferior cava is distributed to both ventricles, a mixing of bloods, arterial and venous, occurs in both ventricles. Ziegenspeck's work is the most scientific and recent and we, therefore, direct our attention to his article on the subject.

Ziegenspeck criticises both the von Haller-Sabatier and GalenHarvey theories. We agree that the former is incorrect, but, inasmuch as we favor the latter, we present his views on the question.

Ziegenspeck claims that the Galen-Harvey theory is refuted ; first, by the anatomical findings of Wolff which he substantiates, and second, "Wer kann bei Betrachtung der Abbildungen, . . , es fiir warscheinlich oder moglich halten, dass wahrend der Diastole beider Vorhofe, wdhrendem doch beide aspvrieren, ein Blutstrom sich durch die rechte Miindung nach rechts, dann wieder durch dieselbe Miindung nach links ergiessen? 1st es nicht vielmehr ohne weiteres klar, dajss sich jeder Vorhof aus der Vena cava inferior direct so viel Blut aspiriert, als zu seiner Fiillung noch notig ist?" We can answer the question quite frankly — it is not clear.

While we are tempted to agree with Ziegenspeck that the illustrations he presents seem to support the argument, and while these relations appear to hold in pig (the animal used in our experiments), we would hesitate before accepting these conditions to obtain necessarily in the living animal. Practically Bom's criticism. Again we note that particular emphasis is laid upon an aspirating action of the auricles, a point also mentioned by Peaslee in his claim for a mixing of the blood. It must be remembered that the aspirating action of the auricles is by no means a certainty and that even if it be present, it is probably feeble and transient.

The physical data offered by Ziegenspeck are based on the following preanise: if both ventricles expel the same amount of blood under the same pressure, then the vessels transmitting the blood must be of like caliber. Upon this assumption, therefore, if the ductus and Pars communicans aortse (segment of aorta between left subclavian artery and ductus) are of like caliber, they carry ■equal quantities of blood. He tabulates 33 measurements; in 28

D arger, and in 3, For variation, he ities of blood. Jd have a direct ents, for we are ns and compare sns =

where the three md Pars. comm. 2.97 — ; Pars. 3 + mm. Com?his error, while er cent in carryaspeck's figures,


►< •

P ^

« ^


•< B










+ 12.50




+ 12.50






+ 15.50
























— 9.84


— 5.47




— 2.00



D Course of the Blood Through the Fetal Heart.


the Aorta descendens can carry away but 82 per cent of the blood fed to it by the ductus and descending aortic arch — ^unless the resistance in the descending aorta is less than in the carotid-subclavian system and in the right and left pulmonary arteries. If the Aorta descendens actually carries away one half of the contents of both ventricles, as Ziegenspeck maintains, then it is also possible to compute the caliber of the aortic and pulmonary stems — ^measurements which he unfortunately does not give us.

Furthermore, much depends on the accuracy in calibration of small vessels. If we select certain of the measurements, we find that individual variations between the actual and computed value of the Aorta descendens are extreme. In No. 1, for example, the carrying capacity of the descending aorta is 35 per cent less than calculated. In only three cases does he obtain the vessel slightly larger; in ten, it is smaller; and in nine, it is too small by about 31 per cent.

To test the validity of these measurements we selected two pigs at random, taken from the same uterus and hardened by formalin injection. Ring segments were cut out of the vessels named in Table II; the rings split and carefully straightened to avoid stretching. A linear measurement on the intima when divided by 3.1416 ought to give a relatively accurate calibration if this method can be employed, and if the vessel lumina are circular. According to Ziegenspeck, the following equations obtain :

1. Caliber Aorta descendens^ = caliber ductus « + caliber of Pars. comm. aortae*,

2. Caliber Pulmonary stem* = caliber ductus" + caliber right pulmonary* + cali ber left pulmonary*.

3. Caliber Aorta descendens? «= caliber aortic stem* + caliber pulmonary stem*.

Table II.


Pulmonary Rtem

3.8 " 14.44 — 14.44

DuctuB . . 7

2.6 - 7.76)

2.1 - 4.40V - 14.72

1.6 "- 2.66)

Right pulmonary

Left pulmonary

Aortic Bttnx

3.8 — 14.44 — 14.44

3.0 - 9.00

Aorta descendens

4.1 « 16.81 — 16.81

4.0 2.2 1.9 1.2 3.8 3.0 3.8



3 .61 y

1.44) 14.44

9.00 14.44

- 16.00

- 9.89

- 14.44

- 14.44

D 88 Augustus Grote Pohlman.

Equation 1.

In A.

16. 8 : 16.76.


14.44 ;


Equation 2.

In A.

14.44 : 14.72.


16.00 :


Equation 3.

In A.

16.81 : 14.44.


14.44 ;


Equation 1 conforms nicely in pig A ; in B an error of 3 per cent carrying capacity. Equation 2 conforms nicely in pig A; in B an error of 38 per cent carrying capacity. Equation 3. In pig A, the Aorta descendens is 16 per cent larger than necessary (carrying capacity), and in B, it is 5 per cent smaller than necessary. In all of these measurements the coronary circulation has not been taken into account In any event the results will show that measurements of this character are valueless for exact conclusions because we must grant that :

1. The vessel lumina are exactly circular.

2. The vessel elasticity must be equal.

3. The expansion of these vessels must be equal in all directions.

4. The intrinsic vessel resistance must be the same.

5. The capillary resistance in all vessels must be equal or known.

6. The quantity of blood expelled by the two sides of the heart must be the same and the pressure exferted equal.

7. The vessels must undergo no particular change after death and fixation.

When all these points have been established, there are still a number of factors to be considered before we can arrive at definite conclusions. We, therefore, raise the question of reasonable doubt to Ziegenspeck's major premise, and state in opposition that we believe we can show that the Aorta descendens carries away more than half of the contents of both ventricles, and further, that the ductus carries more blood than the Pars communicans aort«e.

Data Obtained feom Injection.

It would be difficult, indeed, to grant that the relations of the blood currents in the living fetal heart could be studied in the dead animal, and this is more true in Ziegenspeck's experiments, because only one vein was injected and no attempt made therefore to repro

D Course of the Blood Through the Fetal Heart. 89

<iuce, as far as it was possible, the life-like conditions. We have seen that Trew, Eoederer, S6nac and others found that injected material did pass through the foramen ovale; Beid found that material injected into the inferior' cava in the human embryo passed over entire to the left, or at least he did in one case, even when a simultaneous injection was carried out in the superior cava; Ziegenspeck finds that material injected into the umbilical vein passes equally to the right and left ventricles. We do not object to this as a finding, but we do not see how of itself it proves anything for the living heart under entirely different conditions, and minus the factor of auricular and ventricular aspiration that Ziegenspeck himself uses as an argument against the theory of Galen-Harvey.

We now come to a critical examination of Ziegenspeck's contention : "Das Blut, welches in einem Herzrhythmus das Herz durchstromt, wird gevierteilt. Die Halfte wird von der V. cava inf. geliefert und gleichmassig auf den linken und rechten Vorhof verteilt. Jedes Viertel der Gesamtmenge mischt sich links mit der gleichen Menge aus den Lungenvenen, rechts mit der gleichen aus der V. cava sup. Diese Mischung: % Cava inf., % Cava sup. geht zu ^ durch die Lungenarterien nach links zu % unverbraucht in die Aorta descendens (durch den Ductus arteriosus). Die Mischung links geht ebenso zu ^/^ unverbraucht durch das Schaltstiick in die Aorta descendens, zu % ^^ den Oberkorper. Auch das Blut der V. cava inf. wird somit gevierteilt."

To state this proposition in our o^vn way, grant that the ventricular capacity, right and left, equals a volume of say 4.0 cc, then the matter can be expressed in the form of an equation :

Lungs — 2 Pulmonary return =■ 2

R. V. - 4/ \l. V. - 4

Ductus « 2 Left auricle « 2

^Aorta descendens ■= 4 = Cava inf.^

Pars. comm. aortse » 2 Ri^t auricle «« 2

L. V. « 4/ \r. V. - 4

Carotid — subclavian =-2 .Superior cava = 2

Correlating this equation with Ziegenspeck's diagram (Fig. 1), we note that several important things have been omitted :

D 90 Augustus Grote Pohlman.

1. The coronary circulation — of the 4r cc. leaving the left ventricle a certain amount is returned to the right auricle but not throu^ either inferior or superior cavse. This, however, might be granted, practically speaking, in the return through the superior cava.

2. The azygos circulation — of the 4 cc passing down the Aorta descendens a certain amount is not returned through the cava in

Fig. 1.

ferior but through the azygos system either to the cava superior or to the coronary sinus (pig). Results in 4 cc. minus returned through the inferior cava.

3, The lymphatic return of the entire region supplied through the Aorta descendens is returned to the heart through the superior cava. Results in 4 cc. minus returned through the inf erioi*' cava.

D Course of the Blood Through the Fetal Heart 91

Leaving out a consideration of the bronchial system, the objections would mean that according to Ziegenspeck's contention and diagram, the return through the superior cava is greater than through the pulmonary veins, and second, that the cava inferior returns less than the 4 cc. necessary to fill both ventricles one half.

The only way to make this scheme a tenable one would be (inasmuch as the foramen ovale does not afford communication between the two auricles — ^Wolff theory), to abandon the exact division of blood and to grant that more than one half of the return through the cava inferior passes to the left. Here, however, we read, "Dass die linke Miindung der Vena cava inf. enger ist als die rechte andert daran nichts", which is far from reassuring; and further, "Das Herz wirkt als Saugpumpe und jeder Ventrikel aspiriert in der Diastole das zu seiner voUigen Fiillung noch notige Quantum aus der V. cava inferior." Now even granting the marked suction action of the ventricles (which we do not believe) inasmuch as the beat of the two sides is synchronous, the only way for the left ventricle to fill itself through the narrower channel would be to aspirate more markedly. But here Ziegenspeck answers the question himself by presenting Table II with thirty-three measurements to show that the right and left ventricular walls are of equal thickness, and by his assumption in his major premise that both ventricles exert an equal pressure during systole.

We are not able to see how this proposition may be made a feasible one, and also for reasons presented as the result of our own investigations must oppose "Das Gresetz der Halbierung des Blutes im Foetalkreislauf". Further we do not see wherein Ziegenspeck is justified in his claim that he has simplified the description of the placental circulation, or wherein the anatomical findings of the Wolff school can lead directly to the assumption that the foramen ovale does not afford communication between the two auricles.

Peesonal Findings.

It became evident from the varying results^ obtained through observation and injection of the dead fetus, that if any further work

D 92 Augustus Grote Pohlman.

was to be done on the course of the blood through the fetal heart, it must be undertaken in the living animal, and with the placental circulation intact. The fetal pig was chosen because of the accessibility and abundance of material and the several propositions demanding an answer through the experimental method were considered as follows :

I. Is the ventricular capacity an equal one in the fetal heart ? II. Is the pressure exerted by each ventricle equal ?

III. What is the course of the blood entering the heart through

the cava inferior?

IV. What is the course of the blood entering the heart through

the cava superior ?

I. The capacity of the two ventricles in the fetal mammal has always been considered equal (note the exception in the Mery theory) because that is the condition in the normal adult heart and because of the necessity of this condition at birth. There appears, however, to be no experimental evidence on the question. Accordingly, the pig embryo was opened (see later) and a ligature slipped around the heart at the auriculo-ventricular sulcus with the idea that if the ligature was tightened at the completion of auricular systole, the aortic-pulmonary and auriculo-ventricular orifices would be occluded and the contents of the ventricles isolated. The experiment proved successful in two out of ten trials. The heart was next removed from the body, washed, the contents of each ventricle bled into separate vials and the volumes compared. Comparison of the two vials showed equal capacity as nearly as this rather primitive method permitted in both cases. There being no valid objection to equal ventricular capacity (generally accepted), the point was considered as settled in the affirmative. The two ventricles in the living fetal pig contain or at least expel equal quantities of blood.

II. The pressure exerted by the right and left ventricles in the fetus has alsd been considered equal, because both ventricles expel blood into the Aorta descendens, and secondly by observation, nicely shown in Ziegenspeck's table, because the right and left ventricular walls are equally well developed until after birth when the left ventricle wall hypertrophies rapidly. Our later experiments required


Course of the Blood Through the Fetal Heart, 93

the simultaneous recovery of blood under identical conditions and to this end the following technique was employed:

Pieces of glass tubing about 10 cm. in length were carefully drawn in the flame to a fine connecting piece of about 1 mm. in diameter; laid aside to cool and then carefully broken at the point indicated (Fig. 2). This procedure resulted in pipettes of the same

1 Fig. 2.

opening and, when fastened together with a small elastic band, permitted sufficient spreading to allow the pipettes to be passed one to either side of the ventricular septum and permitted their use as a single pipette. The opening in the pipettes was small enough to necessitate an actual pumping on the part of the ventricles, while the capillary attraction aided in holding the contained blood in place. This was further assisted by mouth pieces of rubber tubing which were pinched off on the withdrawal of the pipettes from the heart.

The beating heart was laid bare (see later) and the pipettes thrust one into each ventricle simultaneously. In all cases where the pipettes were properly introduced and where the heart continued to beat, the blood mounted progressively and evenly in both ; proving to our satisfaction that the pressure exerted by the right and left sides is an equal one. Further there was little, if any, appreciable oscillation of the blood in the two pipettes which went to show that in the opened chest little aspirating action was manifested by the ventricles. The results, thus far, are in perfect harmony with what has been quite generally accepted and may be said to substantiate these views in an experimental way.

It was found that in the majority of pigs, the heart suffered but little inconvenience through the introduction of the pipettes and in some the heart beat quite rhythmically for many minutes even after they were withdrawn. Inasmuch as it was impossible to estimate how long it would take the blood to reach the heart from a given point, a requirement was set that the heart must beat at least five times after the introduction of the pipettes and that the blood

D 94 Augustus Grote Pohlman.

must mount evenly and pix)gressively in both pipettes. Each pig used therefore directly controlled the point that the ventricles exerted an equal pressure.

Injection Experiments.

We have seen that the legitimate cry of artifact was raised by Born to the anatomical findings of Ziegenspeck^ and that it may also be raised to all injection experiments on the dead animal even if the animal be used directly after death and all precautions taken to avoid imdue pressure. The heart itself is no longer the active agent and there is no way of determining how much the contracture of the heart muscle may influence its normal intrinsic relations.

Technique. The following idea was carried out: to inject a nonirritant granular substance suspended in normal salt solution into a selected vein; to allow the blood current to propel these granules to the heart; to recover some of the blood from both beating ventricles under identical conditions; and to examine the blood recovered for the granules injected.

Stand was taken in the abattoir where the pig uteri were removed and dropped into a tank truck. The larger and uninjured uteri were selected and laid upon a table. Next a small incision was made into the uterine wall at some distance from the markedly vascular area and the incision widened by tearing to allow the escape of the pig. It was found that tearing through the uterine wall practically eliminated all hemorrhage and pigs were rejected if any amount of oozing occurred.

Injection was made only in those pigs in which the cord pulsation was strong. An ordinary hypodermic syringe was filled with cornstarch granules suspended in normal salt solution, the air expelled, and about one half of the syringe contents was injected slowly into the imibilical vein, some 5 cm. from the navel. The needle withdrawn, the pig was rapidly opened with a large blunt scissors by cutting through the length of the sternum and by a lateral cut through the abdominal wall just below the diaphragm. A blunt instrument was selected with the idea of tearing rather than cutting through the tissues and pigs were again rejected if anything further than

D Course of the Blood Through the Fetal Heart. 95

a slight oozing resulted from the incisions. Next the pericardium was incised and the paired pipettes thrust simultaneously into the two ventricles. An arbitrary requirement was established that the heart must beat at least five times after the introduction of the pipettes, giving a better chance of recovering the granules injected. The blood in the pipettes was immediately expelled into paired vials, marked right and left, containing a small quantity of half per cent acetic acid. The vials corked and shaken thoroughly. Later the vials were separated, and the contents diluted to an equal quantity with dilute acetic acid. Each was shaken a given number of times, and a small amoimt of fluid withdrawn by a pipette from the central area of the vials. Samples from the right and left sides were placed on one slide and compared under the microscope for the number of starch granules.


1. The death of the mother. It is well known that pigs will live in the removed uterus for many hours after the death of the mother. In pur experiments the time rarely exceeded half an hour and in some cases about fifteen minutes after the sow's death.

2. The contraction of the uterus. This was not marked but was present in some cases. We shall show later that this is not a serious objection to our results.

3. The artificial factors introduced in opening the chest and manipulating the heart. Granted present. The collapse of the lungs especially offers an abnormal condition which probably limits the pulmonary return.

4. The introduction of the pipettes. This is undoubtedly placing the heart under some disadvantage, but we do not consider the objection a serious one because only a small quantity of blood was withdrawn and because the heart showed no signs of interference for at least five beats.

5. The introduction of a foreign substance in the circulation. Cornstarch granules are non-irritant and non-toxic but are of sufficent size to plug the capillaries, hence the blood was obtained as soon as it reached the heart.

D 96 Augustus Grote Pohlman.

It will be seen from these objections to our method that it is practically impossible to reproduce the normal conditions in all details as they are found in the fetus in utero. All that we claim is that the artificial elements were avoided as far as facilities permitted, and that our procedure is an improvement on experiments made thus far. We at least allowed the blood stream to propel the granules through what appears to be the normal course of the blood in the fetal heart with a minimum of abnormal conditions imposed. The method can, therefore, not be called exact enough for definite proportions, and we make use of the term 'about equal' in this paper as a personal equation set within 10 per cent of difference in comparison of the two blood samples.

III. What is the course of the blood entering the heart through the cava inferior ?

Injection of about one half of the contents of a hypodermic syringe filled with a suspension of cornstarch granules in normal salt solution was made into the umbilical vein about 5 cm. from the naveL The pig was opened immediately, and the blood recovered from the beating heart as in Experiment II. Seventeen paired samples were recovered which registered equal coloration on dilution to equal volumes, and these were examined. Five paired samples proved negative — ^no com starch granules found in either ventricle and twelve paired samples were positive — starch granules present In all twelve paired samples the number of granules proved to be 'about equal' on both sides.

The experiment proves beyond a doubt (as far as pig is concerned) that the von Haller-Sabatier theory is incorrect and that the objections of Williams, Peaslee and Macdonald were well taken. It seems to prove that the blood from the inferior cava is distributed about equally to the two ventricles, and is, therefore, in accordance with the Wolff theory or with the theory of Galen-Harvey. The former states that the blood from the inferior cava is split upon the limbus Vieussens and that the foramen ovale does not afford communication between the auricles; the latter assumes a small pulmonary return, a mixing of the blood of the two cavro in the right auricle, and a passage of mixed blood through the foramen ovale.

D Course of the Blood Through the Fetal Heart. 97

It now remained to inject the superior cava — ^for if the granules did not pass through the foramen ovale, the Wolff theory was sustained; while if the granules were recovered from both ventricles, the theory of Galen-Harvey was substantiated.

IV. What is the course of the blood entering the heart through the cava superior ?

This experiment was found more difficult than the preceding test.

It was found almost impossible to open the chest the full length

and expose the great veins at the root of the neck without injury to

these structures. The insult offered proved to be greater and the

death rate proportionately larger. Next only one or two drops

could be injected into the superior cava to avoid undue pressure, and,

when the needle was withdrawn, the unsupported vein tended to

bleed freely. The time limit was reduced to one to two seconds from

injection to introduction of the pipettes. Seven paired samples

were finally obtained which registered equal coloration on dilution.

One proved negative — ^no starch granules on either side, and six

positive — starch granules present. In four, the number of granules

on the right and left sides proved to be ^about equal ;' in one, more

were found on the left than on the right, and in one, more on one

side than on the other (labels confused). In both of these samples

the excess was easily 50 per cent.

The fact, however, that starch granules injected into the superior cava did pass through the foramen ovale in all cases where they were demonstrated at all showed conclusive evidence in favor of the theory of Gralen-Harvey that the caval currents mix in the right auricle and that mixed blood passes from the right auricle into the left through the foramen ovale. Inasmuch as about ten seconds elapsed from the injection of the umbilical vein to the recovery of blood from the heart in Experiment III, and one to two seconds from the injection of the superior cava, it was thought possible to make a double injection in the same pig using colored granules in the one and colorless granules in the other injection. If both varieties of granules were found on both sides, the experiment would show that the currents mix in one and the same pig. Pigs in this experiment were opened first and about ten seconds

D 98 Augustus Grote Pohlman.

allowed from the injection of the umbilical vein with colored granules (I-KI) to injection of the superior cava with colorless granules. Only six pigs lived through the requirements in two full mornings work, and of the six, three samples were lost owing to the hurry. Two paired samples were obtained and one from the left side (right pipette struck the septum). Superficial examination of these samples revealed the presence of both varieties of granules on both sides in the two and both varieties in the one from the left. There was of necessity a delay in counting, and when it was attempted some hours later, it was found that the iodine had diffused and colored a large proportion of the colorless granules so that a comparative count was impossible. The experiment, however, further substantiated the Galen-Harvey theory and opposed, therefore, all the more the Wolff theory.

The results from our experiment in the living embryo lead to the following statement: that the ventricular capacity and pressure is an equal one; that the foramen ovale does afford communication between the two auricles; that the blood of the two cavsB mixes in the right auricle ; and that mixed blood passes through the foramen ovale. We agree, therefore, with the theory of Galen-Harvey and believe to have established it through experimental evidence.

It now remains to consider what objections may be raised to our results and wherein the evidence supports the Galen-Harvey theory as opposed to all other theories.

It will be seen from our results in the living pig embryo that about one half of the return through the superior and inferior cav» passes through the foramen ovale into the left auricle. This fact might be interpreted in one of two ways if we grant, as we must, that the collapse of the lungs through opening the chest interferes with the pulmonary circulation — ^the passage of blood through other parts of the fetal body not necessarily being affected :

a. The pulmonary return is relatively free, as stated in the Wolff theory, and that the artificial factors (collapse of lungs and manipulation of the heart) are sufficient to practically prevent blood from passing through the lungs ; or,

b. The lung circulation is relatively small in amount and that

D Course of the Blood Through the Fetal Heart. 91>

the pulmonary return is reduced by the artificial conditions so that it might be well within the personal equation set in our experiments (10 per cent).

The first interpretation offers a serious objection to our results because those investigators favoring the Wolff theory will hold that if we interfere with a large pulmonary return (about one half [Wolff], exactly one half [Ziegenspeck] of the contents of the right ventricle), we also destroy the normal balance of return to the auricles through increase in flow through the cava inferior and through decrease in flow through the pulmonary veins. Hence, the normal position of limbus Vieussens to the orifice of the inferior cava and the function of the foramen ovale may be rendered false. This criticism we have foreseen and we, therefore, discuss the position of the Wolff theory rather fully.

The results of our first two experiments have confirmed the major premise of the Wolff theory that both ventricles expel the same amount of blood under the same pressure, and we now come to an examination of the physical laws underlying the flow of the blood through the arteries. Ziegenspeck based his measurements on the arguments that if the caliber of the ductus and Pars comm. aorts& was an equal one, they transmitted equal quantities of blood; that this quantity transmitted by each vessel was equal to one half of the contents of a ventricle; and that the Aorta descendens carried away one half of the contents of both ventricles. We stated in our objections to these propositions that we believed it might be shown that the ductus transmits more than the Pars comm. aortce ; and that both feed into the Aorta descendens more than one half of the contents of both ventricles.

If we can prove that the resistance to flow in the Pars comm. aortffi and in the ductus is less than in the branches of the aortic arch and in the right;^ and left pulmonary arteries respectively, then we also prove that the Aorta descendens carries more blood than the caliber of its lumen would indicate, while the other branches convey relatively less. We believe we can substantiate the generally accepted idea that the placental area is the point of least resistance in the fetal circulation for the following reasons :

D 100 Augustus Grote Pohlman.

1. The two large umbilical arteries feed into one large umbilical vein — a proportion of lumina which would indicate that either the arteries are under lower pressure than usual, or the vein is under relatively higher pressure. In either case, it would present a low resistance in the placental capillaries.

2. The umbilical vein, practically of round lumen, transmits blood directly and indirectly into the intra-thoracic cava inferior without any marked increase in lumen of the latter vessel, showing that by far the larger proportion of blood passing down the Aorta descendens is returned through the umbilical vein.

3. The umbilical arteries and vein have a long and tortuous course through the jelly-like cord which probably offers little support to the vessel walls, and were not the placental resistance lower than the resistance in the vessels of the embryo body, little blood would pass through the umbilical vessels, whereas we know the reverse to be the case. It must be remembered that in human embryos the cord usually averages about 55 cm. at birth, and that the umbilical arteries may be reckoned on an average of 50 cm. longer than any other arteries in the fetal body, and the vein, while not proportionately long, is easily 40 cm. longer than any other vein. This distance of a little less than a metre represents an appreciable amount in terms of intrinsic vessel resistance. •

4. There can be but little doubt that the contraction of the uterus must increase the resistance in the placental site and still the fetal heart is able to force the blood through the long course and quite freely. This is our answer to the objection that opening the uterus rendered false the circulatory condition in the fetal pig.

6. Taking Ziegenspeck's measurements at their face value if a 2.97 — ^mm. ductus and Pars comm. aortse feed into a 3.832 mm. Aorta descendens, then the resistance in the latter vessel must be considerably less than the resistance found in the lungs and carotidsubclavian vessels, for the lumen is too small to carry off the blood. It should read 4.14 + mm.

From these data we are able to assume with reasonable certainty that the resistance to flow in the placental area must be considerably less than in the fetal body, and that, therefore, until this resistance

D Course of the Blood Through the Fetal Heart 101

is a known quantity, the lumen of the Aorta descendens is no criterion of its carrying capacity. We believe that the Aorta descendens carries more than one half of the contents of both ventricles, and believe this is not only a logical deduction but that it is confirmed by observation.

Conversely, if the circulation through the Aorta descendens is relatively free, then the blood flow through the carotid-subclavian and pulmonary arteries is relatively small. In other words, these two systems return less than one half of the contents of both ventricles. While this view is entirely contrary to Ziegenspeck's law of the equal division of blood, it is not entirely contrary to the Wolff theory of the splitting of the current of the inferior cava upon the limbus Vieussens. It will be necessary to substantiate our evidence from the injection experiments by an attempt to show clearly that the lung circulation is relatively smaller in amount than the circulation through the carotid-subclavian systems, or negatively, that the ductus carries more blood than the Pars comm. aortCB. The two points will be argued under separate headings, although they lead to the same result.

Evidence that the circulation through the lungs is relatively small in amount:

1. The histological appearance of the fetal lung, even when hardened in situ, does not substantiate the theory that large quantities of blood pass through the pulmonary circulation. The air sacs are collapsed, the capillaries are compressed and tortuous and possibly more numerous than elsewhere in the fetal body, l^ot only is the blood current interfered with directly in the capillary system, but the expansion of the vessels to the blood stream is limited.

2. The right and left pulmonary arteries are placed practically at right angles to the blood impact, while the blood wave passing aroimd the aortic arch meets the carotid-subclavian vessels with their openings more nearly parallel to the current. Showing if the lumina of these vessels read alike, the carotid-subclavian arteries wiU receive a trifle more blood (the grain of truth in the Sabatier theory).

Evidence that tKe ductus conveys more blood than Pars comm. aortfiB and that it carries more than one half of the contents of the right ventricle :

D 102 Augustus Grote Pohlman.

1. The laws govering the equal flow of fluid through pipes have certain limiting clauses ; not only must be the *head', the direct resistance at the pipe opening and the pipe lumen and character be the same, but the pipes must be of the same length and have the same course. Therefore, the Pars comm. aortse will transmit less blood than the ductus, even if the caliber be circular and of the same diameter, because the course of the blood from the left ventxicle is a longer one ; because the course is curved as opposed to the relatively straight line to the ductus; and because the branches on the arch are more advantageously placed to interfere with the current.

2. If in our experiments we interfered with a large flow of blood through the lungs, then one of two things must have occurred :

(a) We increased the pressure on the right side in order to force the excess of blood through the ductus, or

(6) We decreased the amount of blood expelled by the right ventricle. We have shown in our experiments that the pressure on the right and left side continued to remain the same, for the blood mounted progressively and equally in the pipettes when they were properly introduced and where the heart continued to beat. Inasmuch as no difference was observed in the character of the heart beat, we may infer that lie ventricles continued to expel equal quantities of blood as demonstrated in Experiment I under the artificial conditions mentioned. If we grant that the lung circulation is free, the ductus carried away the excess without any appreciable effort on the part of the right ventricle; according to Ziegenspeck, the ductus would have to carry double the amount, and according to Wolff perhaps not quite double. If we grant that the circulation is relatively small, the ductus carried but little more than normally.

We, therefore, do not see any evidence that we interfered seriously with an alleged large amount of pulmonary return or that we oversupplied the vena cava inferior with blood. Further, waiving all this evidence aside, if we did increase the amount of return through the inferior cava, we also raised the pressure in that vein, and we still do not see why the limbus Vieussens should not divide the current in the manner demanded by the Wolff theory. If we inter

D Course of the Blood Through the Fetal Heart 103

fered markedly with any of the return, it was that througli the cava superior and this should if anything lessen the chances of ihat blood passing to the left.

If we now grant that the pulmonary return is relatively small in amoimt^ then we can group the von Haller-Sabatier and the Wolff theories under one head, for in neither does the foramen ovale afford communication between the two auricles and in both the vena cava inferior must send an excess of blood to the left auricle to make up for the deficient pulmonary return. Against these theories we can present the view of Galen-Harvey that the foramen ovale does afford communication between the two auricles and that mixed blood in the right auricle passes through that opening to make up for the deficient pulmonary return. Which of these theories do the facts support ?

Neither the von Haller-Sabatier nor the Wolff theory can account for the relatively good circulatory conditions found in embryos with bilocular and trilocular hearts; in the cases of incomplete separation of the ventral aortic stem into aorta and pulmonary artery; or in the anomalous cases where the lungs are supplied, in part, from the Aorta descendens. All of which conform to the Galen-Harvey theory.

For the Harvey theory, as opposed to all others, comes our evidence obtained in the living pig that in no case were cornstarch granules injected into the inferior or superior cavse or both — ^not recovered from both sides of the heart. For the Harvey theory comes the simple explanation of a mixing of blood in the right auricle, and the passage of mixed blood through the foramen ovale to the left auricle. In this theory, we gain the point raised in the Wolff theory that no artery in the embryo contains arterial blood, without the extremely complicated and incorrect arrangement presented by that scheme.

We trust that this article carries conviction with it and that the coloring of diagrams of the fetal circulation to render the impossible theory of Sabatier dear to the student will hereafter be omitted. The arrows to indicate the course of the blood through the fetal heart may be replaced by the statement — the blood of the two caval

D 104 Augustus Grote Pohlman.

veins mixes in the right auricle, and mixed blood passes through, the foramen ovale into the left auricle to make up for the deficient pulmonary return (the theory of Galen-Harvey).

Peeliminaby Note on the Reptilian and Amphibian


Every article that has appeared since 1835 has been in favor of the proposition that no arteries in the mammalian embryo contain arterial blood. It would be fitting, therefore, to retrace our steps to the time of Harvey, Mery, Winslow and others. Here we find that the relations of the blood currents in the mammal were based largely on the turtle. In this animal, although the auricles are completely divided, the undivided ventricle lends itself to a similar scheme^

Fig. 3.

of the crossing of currents, and diagrams illustrating this condition of affairs have been piresented. We present *the scheme taken from Parker and Haswell after Huxley (Fig. 3), together with the following description : "From the cavum pulmonale arise the pulmonary artery, and from the cavum venosum, the two aortic arches. When the auricles contract the cavum venosum becomes filled with venous blood from the right auricle, and the cavum arteriosum with arterial blood from the left auricle; the cavum pulmonale becomes filled with venous blood which flows into it past the edges of the incomplete septum. When the ventricle contracts, its walls come in contact with the edges of the septum, and the cavum pulmonale becomes cut off from the rest of the ventricle. The further contraction consequently results in the venous blood of the cavum pulmonale being driven out throu^ the pulmonary artery to the lungs, while the blood

D D D Course of the Blood Through the Fetal Heart 105

that remains in the remainder of the ventricle (arterial and mixed) is compelled to pass out through the aorta (ae)." Here, as in the mammal, we find anatomical observation correlated with inferred physiological necessity and no direct experimental evidence ; and here again comes the question "Does this actually occur ?"

We present a preliminary report of our findings in some twentyfive turtles of three species. We do not offer the data for anything more than its face value, or do we present it as an exact result. The method was somewhat primitive, but the evidence derived is suggestive.

The carapace was removed under precautions to eliminate hemorrhage, the heart laid bare and kept moist. ^ The experiment was divided into three parts: (1) injection of cornstarch granules in normal salt solution into the right auricle; (2) into the left auricle; and (3) double injection of colored and colorless granules into the two auricles.

The blood was recovered under identical conditions in the following way: A double ligature was placed around the three vessels at the transverse pericardial sinus; the cornstarch-salt solution was introduced into the auricle during its diastole ; the auricle allowed to contract, giving time to have the distal ligature ready to tie ; when the ventricular contraction was well under way, the distal ligature was tied and immediately the proximal ligature — ^thereby isolating a segment of the blood in each of the three vessels. The'ligated part was then removed and washed to avoid any granules which might have been in the ventricle, and each vessel bled into separate watch glasses containing a quantity of % per cent acetic acid. Washing between incisions to avoid any mixing. Comparison of the three glasses containing the blood recovered from right and left aortse and from pulmonary artery respectively revealed that, whether cornstarch granules were injected into the right or left auricle or both, they were always recovered from all three vessels. Comparative count proved rather indefinite, although in one case equal quantities were counted in each vessel.

■The writer expresses his obligation to S. C. Murphy for the careful assistance rendered and for the use of his modification of Ringer's solution.

D 106 Augustus Grote Pohlman.

Further experiments on larger animals must be undertaken before the exact proportions can be established. The results show, however, decidedly in favor of the statement that in the turtle the arteries contain mixed blood.

It now remained to investigate the evidence presented in the amphibian, to which can offer as yet no personal experiments. The usual descriptions, however, are as follows : "When the auricles contract, the blood from the left auricle, which has come in from the pulmonary vein and is therefore oxygenated, is forced into the left side of the ventricle, while the impure blood from the right auricle, which comes through the sinus venosus, pours into the right side and middle portion of the ventricle. The blood from these different sources is prevented from becoming mixed by being received into slit-like chambers in the ventricular wall. During the contraction of the ventricle, the impure blood, lying near the opening of the bulbus, naturally passes out first, while the pure pulmonary blood from the left side is forced out only toward the close of the ventricular contraction. When the ventricle first contracts, the wall of the bulbus cordis is relaxed, and the impure blood flows freely over the edge of the spiral valve into the left compartment, when it is free to issue through the pulmo-cutaneous arches through their common opening. ^Now the blood is under less pressure in the pulmo-cutaneous arches than in the others, because its route is shorter and there are no impediments to its flow. The blood first issuing from the heart takes the line of least resistance, namely, the pulmo-cutaneous arches, and is forced through the first two pairs of arches- only when it has no easier avenue of escape." (Holmes.)

Here are two definite statements, to which we may raise the plea of reasonable doubt: "Is the pulmo-cutaneous system really under less pressure and does it actually receive blood first?" With all preparations made to investigate this phase of the problem, we accidentally came across an article by Gompertz, who made simultaneous tracings from the two vessels and found that the curve in the aorta agreed with that in the pulmonary both for synchrony and for pressure. This statement again would sustain the objection made to the identity of currents in the amphibian circulation.

D Course of the Blood Through the Fetal Heart 107


1. The capacity of the right and left fetal ventricles is equal.

2. The pressure exerted by the right and left fetal ventricles is also an equal one.

3. The blood entering the heart through the superior and inferior oavse mixes in the right auricle.

4. The foramen ovale affords communication between the two auricles.

5. Enough mixed blood passes from the right auricle into the left through the foramen ovale to make up for deficient pulmonary return*

6. The pulmonary return during fetal life is relatively small in amount^ and probably does not exceed one fifth of the capacity of the ventricle.

7. The ductus carries more blood than the descending arortic arch ; probably the proportion of 4-3 is not far from accurate.

8. The Aorta descendens, being under lower resistance, transmits more blood than the caliber of its lumen would indicate and the greater part of its blood passes out through the umbilical arteries.

9. No artery in the fetus contains pure arterial blood — all contain mixed blood.

10. We oppose the theory of von Haller-Sabatier and also the theory of Wolff and ZiegenspecL

11. We substantiate the theory of Galen-Harvey.

. 12. We oppose the theory that the pulmonary artery in the turtle transmits only venous blood and hold for a mixture of blood in the common ventricle as opposed to Briicke's view.

13. We believe there is evidence for a mixing of blood in the amphibian ventricle or in the vessels or in both.

14. We believe that the closed circulation of an arterial and a venous blood is first found in the mammal and bird after birth ; in the fetal mammal and bird ; and in the reptile and amphibian we believe that the circulation is one of mixed blood. In the fetal mammal and bird, the mixing takes place in the right auricle ; in the reptile, in the common ventricle or, where the ventricle is more completely divided, in the arterial orifices or in the foramen PanizzfiB or in both ;

D 108 Augustus Grote Pohlman.

in the amphibian, the mixing occurs in the common ventricle, in the vessels or in both.

Received for publication, November 16, 1908.

BIBLIOGRAPHY. BicHAT, X., 1822. General Anatomy. Transl. Geo. Haywood, Vol. I, p. 367. BoBK, G., 1889. Beitr&ge zur EntwicklnngageBcbicbte des Sftugetierberzens.

Arcb. f. mik. Anat, Vol. 83, p. 368-9. BoTAixiJS, L., 1565. Opera Omnia. Observationes anatomic®. Obs. III.

Loc. cit. Dalton, p. 137.

GAEBALpnnTB, A., 1593. Qusestionum peripateticarum. Quaes. lY, L, I. Loc. cit.

Dalton, p. 132. Ck>LUMBUS, M. R., 1569. Be Re Anatomica, p. 223. Loc. cit. Dalton, p. 126. Dalton, J. C, 1884. Doctrines of tbe Circulation. Lea's Son & Co., Phila delpbia. Galen, Claudius. Opera Omnia, Vol. lY, p. 243. KtUm, Leipzig, 1821-33.

Loc. cit Dalton, p. 68. GoMPEBTz, C, 1884. Ueber Herz und Blutkrelslauf bei nackten Ampbibien.

Arch. f. Anat. u. Pbys., Phys. Abt., p. 257. VON Hallfr, a., 1779. First Lines of Physiology. Eng. transl. Wm. Cullen,

Edinburgh. P. 476. Habyky, W., 1628. Anatomical Dissertation on the Movement of the Heart

and Blood in Animals. Frankfurt edit. P. 38. Holmes, S. J., 1906. Biology of the Frog. Macmillan. P. 278. HoRNEB, W. E., 1818. Plan of the Foetal Circulation. Philadelphia. KiLiAN, H. F., 1826. Ueber den Kreislauf des Blutes im Kinde, welches noch

nicht geathmet hat Karlsruhe. P. 200. B^ABBE, J. H., 1834. Disquisitiones historico-critiC8& de circulatione sanguinis

in foetu. Dissertation Bonnse. Macdonald, W., 1867. Objections to the Theory of Foetal Circulation. Med.

Press and Circ, July, Vol. IV. Meckel, A., 1827. Verschliessung der A<»rta am vierten Brustwirbel. Meckel's

Archiv, Vol. II, p. 846. M£by, J., 1692. Nouveau systftme de la circulation du sang par le trou ovale

dans le foetus humatn ; avec les responses aux objections de MM. Duverney,

Verheyen, Silvestre et Buissiere centre cette hypothdse. Paris, Jean

Boudot, 1700, P. 10. MusBAY, R., 1816. De circulatione sanguinis in foetu. Edinburgh. Pabkeb and Haswsll, 1897. Text-book of Zo6Iogy, VoL II, pp. 333^. Peaslbe, E. R., 1864. A Monograph on the Foetal Circulation. Am. Med.

Monthly, May.

D Course of the Blood Through the Fetal Heart- 109

PoHLMAir, A. G., 1907. Tbe Circulation of the Blood through the Fetal Heart

Johns Hopkins Bulletin, August PSKYES, W., 1885. Physiologie des Embryos. Leipzig. RmD, JoHir, 1885. Injection of the Vessels of the Fcetus to show some of

the peculiarities of its circulation. Edinb. Med. and Surg. Jr., Vol. 43,

pp. 11-13. Bsm, John, 1835. Additional Observations. Ibid., Vol. 43, pp. 306-310. BoEDEBBB, J. G., 1750. De Fcetu Perf ecto. Argentorati. RVdinqeb, N., 1871. Ueber die Topographie der beiden VorhSfe und die

EinstrSmung des Blutes in dleselben bei dem Fcetus. Jr. f. Kinderkr.,

Vol. 56-7, p. 402. Sabatisb, R. B., 1791. Traits complet d'anatomie» VoL II, p. 493. StNAd, J. B., 1773. Trait6 de la struct du c<Bur, I, p. 309. Sebvetus, M., 1553. Ghristianlsmi Restitutio, p. 170. Loc. cit, Balton,

p. 115. Tbxw, 1736. Diss, epist dc. different etc. Nov. ViBSAUus, Ain>B£A8, 1543. De Humanl Ck>rpori8. Liber VI, cap. XV. Loc. oit

Dalton, p. 108. Williams, T., 1843. On the Homology of the Foetal Circulation. Lend. Med.

Gaz., Vol. 32, pp. 17-22. WiNSLbw, B., 1725. Beschreibung einer sonderbaren Klappe, etc. Erlftuterung

einer Abb. K. Akad. d. Wiss. in Paris, p. 528. Wolff, G. F., 1778. De foramine ovali ejusque in dirigendo sanguinis motu.

Observ. noyse. Nov. comment sclent Petropolit XX. ZiEOENSPECK, R., 1882. Welche Verftnderungen erf&hrt die foetale Herz th&tigkeit regelm&ssig durch die Geburt Inag. diss. Jena. ZiEGENSPECK, R., 1884. Ueber den Blutkreislauf des Sftugethier- u. Menschen Foetus. Preyer's Physiologie des Embryos, pp. 596-607. ZiEGENSPSCK, R., 1902. Ueber Foetal-Kreislauf. Mtlnchen. Zeboensfeck, R., 1905. Die Lehre von der doppelten ElnmUndung der unteren

Hohlvene in die VorhOfe des Herzens. Samml. kiln. Vortrftge, Ser. XIV,

Heft 11, No. 401.




In order to secure a series of clean cut sections of the adult human head for th« demonstration of some of the anatomical relations of the accessory sinuses of the nose^ the following method was found most satisfactory. It consists in polishing sawn sections^ while stUl frozen hard, on a rapidly revolving wooden wheel wet with water and finely powdered pumice stone. A similar process is employed for beveling glass.

After employing this method, for more than two years, with a great variely of material, the results have been so uniformly successful as to warrant a detailed report. Connective tissue, muscles, bone and teeth polish with equal facility and smoothness. The resulting surfaces have the appearance of having been cut with a sharp knife.

Having secured a good adult head, it should be well frozen in a near-by refrigerating plant or during cold weather in the open. The sections, either frontal or horizontal, should then be sawn in the usual way by hand or by using the band-saw. The latter is much the quicker and easier proceeding. Whichever course is adopted, the section should be polished before it has a chance to thaw on a rapidly revolving wooden wheel (Fig. 1). This wheel must be run by power, horizontally, while a mixture of water and finely powdered pumice stone is playing constantly upon it. Heavy woolen gloves are useful and greasing the hands with vaseline, an advantage if many sections are to be polished at one time. Some little experience will be necessary to familiarize the operator with the proper use of the polishing machine. Plenty of pumice should be on hand, as one section may require a pound or more. It is best, therefore, to purchase the pumice in bulk, as the fine pumice supplied to dentists


D 112 Joseph P. Tunis.

in small cylindrical boxes does not mix well with water, and is not to be compared with the commercial variety.

As soon as the section b^ns to thaw, it should be frozen again, as a soft specimen is quickly spoilt. This is due to the fact that the bones polish much more readily than the softer tissues. If it is desired to polish both sides of a specimen, it is usually necessary to freeze again after one side is finished. This refreezing, however, often detracts from the cleat cut appearance of the finished surface, especially if the section is not immersed in water before it is frozen for the second time. If the section has a tendency to float above the surface in the freezing pan weighting with a strip of sheet lead is an advantage. A series of numbered pans make it easier to keep track of the sections.

Fig. 1. — tVooden wheel for polishing frozen sections.

Fig. 2 shows photographs of the eighth section, made from the occiput forward, of a series from a woman's head (aged about seventy). They were made on a band saw and were each about half an indb thick. The anterior surface (Fig. 2, A) was not polished while the posterior surface (Fig. 2, B) was. A comparison of the surfaces illustrates the advantage of the method. Tracts in the spinal cord, or the distribution of the white and grey matter of the brain

D A Method of Polishing Frozen Sections. 113

can be demonstrated in this way. In order to secure the best results /with such material, however, the brain must be specially prepared.

To illustrate the anatomy of any particular part of the head, a good preliminary section having been secured, it would be an easy matter to photograph that surface, polish a little deeper, make a second photograph, polish again and so continue until a series of photographs, of the r^on desired, would have been made. The only objection td this plan would be that only the last polished surface would remain for preservation.

A. B.

Fig. 2. — ^A section of the human head. A, Anterior surface, not polished. B, Posterior surface, polished. •

The most satisfactory plan to adopt in preserving a series of sections.that I have found is to place them in a large jar horizontally, in a four per cent solution of formaline, with pieces of ordinary window glass between them. In this way any distortion of the section is prevented and they may be kept indefinitely for future reference, Kept in formaline, however, they will blanch more and more, so that the best photographs are obtained as soon as the specimen has

D 114 Joseph P. Tunis,

been polished. The wooden wheel shown in Fig. 1 was twenty-nine and a quarter inches in diameter^ one and three quarters inches thick. A considerably smaller wheel would have answered the purpose just as well.

As far as I can determine from a careful search through the literature of this subject, no report has been published of any similar attempt to polish the frozen surfaces of sections. For the use of all the needed apparatus I am indebted to the Wistar Institute.

Received for publication, October 27, 1908.



H. E. RADASCH. From the Anatomical Laboratory of the Jefferson Medical OoUege.

Although a number of slideholders have been described, nothing seemingly suitable for large 2 by 3 inch slides, used in serial embryologic work, has been suggested. The writer desires to present an apparatus that is not only light and cheap, but that can be readily constructed by any one. The holders are made of thin aluminium, this metal having proved satisfactory from all standpoints.

A strip about l/75th of an inch in thickness and measuring 6^ by 2% inches is utilized for the base and sides. Each end of this strip is then nicked ^th inch deep and %th wide for ten slides, as shown in Fig. 1; it is then bent at aa, bb, and a' a' and b'b' to assume the shape of Fig. 2. The distance bb' should be 31/^ th inches. Another piece of the metal 3l/4th inches long and % inch wide is then nicked in the same way, having a ^/4th inch nick at each end at which place the metal is to be used as a clamp (Fig. 3, beyond cd). This strip is then bent at right angles at cc', then at cd and c'd'

D A Slideholder for Serial Work.


Fig. 2.

Fig. i.

^ nn n nnrmnnnn ^


Fig. 3.


Fig. 4.


Fio. 5.

D 116 H. E. Kadaach.

at an opposite right angle, as illustrated in Fig. 4 ; it is then placed at the base of the uprights of the main part and clamped into place, as seen in Fig. 2. These strips serve to support the lower edges of the slides. The base is then opened in four places, as seen in Fig. 5, and a handle and support of No. 10 aluminium wire is bent as in Fig. 5 ; the end is carried through a hole in the center of the base and a hook formed at its free end. This wire support is clamped into place by means of the strips cut out of the base, the openings left will serve as points of drainage for the space between the slides, when the trays are removed from one solution to another.

This holder is suitable for all ordinary solutions. When iodin is used, the trays must not be left in too long, as prolonged action of iodin tends to corrode the metal, as the writer has found by previous tests; this, however, is a matter of hours. If acid alcohol is used, the same precautions must be observed. With proper care these trays will last indefinitely. The first set made is still in use, although the trays have seen active service for over three years and no corrosion has been noted.

In order to utilize the least possible space and solution, the writer finds that the pint Barclay candy jars are admirably adapted and serve well as stain jars. With half a dozen trays and about eight jars for the various stains and wash solutions it is no difficult matter to carry through fifty slides in a few hours. The excess of stains and other fluids is always washed oflf by the use of a washbottle of alcohol or water, as the case indicates, before placing the trays in the succeeding jars. The step of clearing, however, the writer prefers to carry out on the table and upon fewer slides than ten each time.

Received for publication, January 23, 1909.


A Laboratory Guide fob Histology, Laboratory Outlines for the Study of Histology and Microscopic Anatomy by Irving Hardesty, .A.B., Ph.D., Associate Professor of Anatomy in the University of California. With a Chapter on Laboratory Drawing by Adalbert Watts Lee, M.D., Assistant in Anatomy in the University of California, vi + 193 pages, 30 illustrations, 2 colored. Philadelphia, P. Backiston's Son & Co. Cloth, $1.50 net.

How best to lay before the student in the laboratory the work to be done/ with the greatest economy of time and effort for teacher and the best educational result to the student, is a problem which «very laboratory teacher meets and which is solved differently according to the peculiar ideaa and ideals of each teacher. Perhaps, the most fundamental question involved is to what degree it is best to provide specific directions as a guide to the student, and one of the greatest dangers that will appeal to many is the thoughtless following of too specific directions by the student, in order to obtain the required results without any real* appreciation of the purpose of the experiment or outlined work; the less the student is made to think, ihe less are the results attained his own, and in corresponding degree the chief educational value of laboratory work falls. The more individual and personal the laboratory instruction, therefore, and the less detailed and specific the ^neral directions, the less will this defect in laboratory instruction be apparent. As a rule, however, individual instruction makes greater demands on the instructing staff, requires more of the student's time and is less easily carried out as the nxmiber of students increases. Particularly is this true in such a subject as Histology and some form of printed laboratory directions, bound or as separate sheets, has seemed to several teachers in this field almost a necessity when the number of students is large.


D 118 Book Reviews.

When such directions are put out in printed form as in the case of the book here reviewed, the value of the work is increased if sufficiently specific directions can be presented with a flexibility and adaptability that wiU make them serviceable in other years and other institutions, and this the author has attained with a marked degree of success.

The plan of the work appeals to the reviewer as excellent : to spare the student all unnecessary technique and yet permit him to gain a certain knowledge of general methods; to give him direct contact with the tissues and organs of the body by means of preparations uniformly and carefully made which become the property of the student ; to furnish a correct idea of the natural appearance of tissues by means of their examination in the fresh state ; to attempt to bridge the unnatural gap too often left between the gross and microscopic anatomy by providing for a preliminary examination or dissection of the organ as a whole.

The laboratory directions proper are presented in twelve chapters as outlines or papers (submitted as reports) as they are termed. Each chapter is closed with a bibliography intended to present the more important papers and monographs in the field covered by that particular outline to be made use of in collateral reading. In the various outlines the selection of the preparations to be studied is on the whole excellent and the directions to the student concise. Nearly every preparation is to be drawn or sketched, while numerous questions are introduced which are of distinct value in overcoming the defect inherent in specific laboratory directions. These twelve outlines are designed for a year's laboratory course of three weekly periods of three hours each, the whole being easily divided into three portions: (a) Histology, (b) Microscopic Anatomy, (c) Neurology. These may be given together (as in the institution for which the guide was primarily designed) or as separate courses if necessary To adapt it to shorter courses, the omission of some of the preparations is suggested.

The outlines for the laboratory work are preceded by a chapter on laboratory drawing and are followed by one on technique and the care of the microtome knife. The first of these, by "Dr. A. W.

D Book Reviews. 119

Lee, is, as far as the reviewer knows, unique and of distinct value, not only, or perhaps principally, to the general student but to the advanced student as well, who must in too many laboratories provide for his own illustration and for whom the suggestions will be very helpful As a whole the guide presents a good laboratory course in Histology and Microscopic Anatomy, and it should be quite generally useful. The rather frequent typographic errors will doubtless be corrected in a second edition.

B. F. Kingsbury. Received for publication, January 23, 1909.


In order to encourage the spirit and method of scientific investigation among students of medicine, the University of Chicago offers three scholarships for the session of 1909-10 to he awarded to applicants presenting the best thesis* in any of the sciences fundamental to medicine. It is to be hoped that the students who receive these scholarships, as well as other medical students, will thus be encouraged to continue, in their medical course, independent work in anatomy, physiology or pathology, for we are much in need of specialists in these branches who are investigators.

Students of abilily who have had a taste of scientific work in physics, chemistry or biology, before they begin the study of medicine are best fitted to undertake such work in the medical course, and since the latter is very flexible at the University of Chicago, the authorities there wisely encouraged scientific work early in the medical course. We sincerely hope that their plan will be adopted by the better medical schools elsewhere, for only by a more liberal medical course can we hope to produce a sufficient number of trained anjltomists, as well as of scientific medical men generally.



Vol. III. MARCH, 1909. No. 3




The researches on sensory ganglia before 1875 established the fact that the predominating type of ganglion cell is, in fishes, bipolar, and in higher vertebrates, unipolar.

In 1875 Ranvier demonstrated that the single process of unipolar cells divides by a T-shaped formation into two branches, one of which goes to the central nervous system, and one to the periphery. Later it was shown (Key and Retzius, '76, Retzius, '80, Lenhossek, '86) that the unbranched process of unipolar cells is the equivalent of the two processes of a bipolar cell.

Freud, in 1878, demonstrated that the ganglion cells of Petromyzon are both bipolar and unipolar, and he found a complete series of intermediate forms connecting these two types.

Retzius, in 1880, first pointed out that in many forms, particularly tnanimals, the axone of imipolar cells enters into a more or less complicated spiral formation, the initial glomerulus, immediately upon leaving the cell.

With the introduction of the Golgi and methylene blue methods, began a series of investigations that yielded important results.

Retzius, in 1890, demonstrated by the latter method a complete series of transformation stages from oppositipolar to unipolar cells in Myxine.

HZJontributlon from the ZoSIogical Laboratory of Northwestern University, Evanston, III., under the direction of William A. Locy.


D 122 Martin R Chase.

Von Lenhossek, in 1892, found in his work on the spinal ganglia of Pristiurus embryos by the Golgi method, that, while the great majority of the ganglion cells were bipolar, a few unipolar forms and intermediate stages were present. He makes the point that the peripheral process of unipolar and intermediate forms enter the dorsal ramus of the spinal nerve.

Disse, '93, Cajal, '93, Von Lenhossek, '94, and Spirlas, '95, reported the finding of multipolar cells in the spinal ganglia of chick and mammalian embryos. These cells seem to be transitory in nature.

Dogiel, '96, '97, '98, published the accounts of his studies of the spinal ganglia of mammals by the methylene blue method. He found besides the typical form of unipolar cells (Type I) a new shape (Type II). The single process of this latter type divides at the first node into 2-4 finer fibers, which, dividing repeatedly, diverge over the ganglion, and end by forming pericapsular and pericellular plexuses about cells of Type I. He also traces sympathetic fibers from the anterior roots to their termination in basket-likt plexuses about cells of Type II. Thus sympathetic fibers come directly into relation with cells of Type II, and through them with Tyi^e I. He finds typical sympathetic (multipolar) cells in ganglia.

Other workers (Cajal and Oloriz, '98, Kamkoff, '97) have confirmed Dogiel's description of pericellular plexuses, but not that of cells of Type II.

Our knowledge of sensory ganglia has been much increased by the recent introduction of the reduced silver nitrate method of Ramon y Cajal, who in the Ergebnisse, etc., for 1906 gives a good review of the work done by this method to that time. In general, the work of Eetzius, Dogiel, Cajal and others is confirmed, but many new details are added. Particularly to be mentioned are fenestration of ganglion cells, and the presence of accessory processes terminating in end bulbs.

Diagrammatically considered, the fenestrated type cell is characterized by protoplasmic processes, arranged in a netlike apparatus more or less complicated, by means of which a part or even the entire periphery of the cell is broken up into a system of anastomosing

D D D A Histological Study of Sensory Ganglia. 123

fibers, which for the greater part go into the formation of the axone. Often there are slings which connect different parts of the cell-body and are independent of the axone. It should be noted that some of this work was anticipated by Daae, who in 1888 described the formation of the axone in the spinal ganglia of the horse, by means of the union of a varying number of roots arising from different parts of the cell.

CeUs with bulbed processes. This type of ganglion cells was first described by Huber in 1896 in amphibians, and was re-found by Cajal in 1904 in man and larger mammals. Fine fibers arise from the cellbody or the axone of a regular glomerulated cell and end with rounded or oval bulbs on the cell-body, among the mantle cells, or on the inner surface of the capsule. Usually these processes are entirely within the capsule, but in some cases they pierce the capsule and end among the ganglion cells.

G. Levi, in 1906, found that the cells of the cerebrospinal ganglia of Chelonia are provided with two or three prolongations, varying from large lobes united to the nucleated portion of the cell by broad protoplasmic bridges, to fine fibrils ending with small bulbs. He observed that the latter were often numerous, and often by anastomosis formed a complicated system.

In his study of Selachians, Levi ('06) found the bipolar type of spinal ganglion cell in great majority, with rare unipolar and intermediate forms. Occasional cells showed short thick fibers arising from the cell-body, or from the axones, which were normal, and ending usually Tvithin the capsule.

In Teleosts, his observations show that transition and unipolar cells are more numerous. The accessory processes are sometimes fine non-anastomosing fibers, which form a sort of wreath about the periphery of the cell. In other cases they are coarser, and by anastomosis give rise to a condition similar to fenestration in mammalian ganglion cells. In some cases Levi thought he saw plexuses of sympathetic origin in continuity with accessory fibers from cells, but some doubt as to this observation is expressed by Cajal, ^vho examined the same preparations.

Levi ('07 ) also considers fenestration and bulbed proces^^os in mammals and deals with their development.

D 124 Martin E. Chase.

V. ' Lenhossek's observations by the silver method confirm those of Cajal and Levi.

According to recent developments (Kohn, '07) the mantle cells are developed from the ganglionic anlage and are consequently of ectodermal origin, rather than of connective tissue origin.


The observations which follow were begun at the suggestion of Professor WiUiam A. Locy, and were carried on during 1907-1008 in the Zoological Laboratory of Xorthwestem University.

I wish to thank Dr. Locy for assistance at all stages of the work.

Material, etc. The material used was embrj^os of Squalus acanthias of about 16-20 cm length, and adult brains of Mustelus canis, fixed and preserved in formalin. A short study was also made of the spinal and Gasserian ganglia of certain mammals.

The ganglia of Squalus acanthias embryos were studied by sections, and by maceration and teasing. The best results were obtained by Cajal's reduced silver nitrate method, which was also applied to the mammalian material. (For this method see Cajal's article in the Ergebnisse, etc., for 1906, or v. Lenhossek's article in Arcbiv f. mik. Anat., Bd. 68.)

A. Selachians.

General Comments, The ganglia especially studied in the Selachian material were the ganglia of the Ninth and Tenth cerebral nerves. The latter nerve is composed of several branches which preserve their identity, although bound for some distance in the same sheath of connective tissue. There are four of these branches which supply the gill slits, a branch to the lateral line organs, and a visceral branch, each with its individual ganglion.

As is well known, the ganglia, as a whole, are enveloped in a connective tissue sheath, a continuation of the epineurium of nerves. This sheath continues into the interior of the ganglion as the perineurium, surrounding the bundles of nerve cells and fibers, and as the endoneurium encloses the individual cells (capsule and fibers (filler sheath).

D A Histological Study of Sensory Ganglia. 125

The connective tissue is not strongly developed in the ganglia of specimens of 16-17 cm. length, and the cells and processes are in close proximity to each other. The ganglion cells are quite evenly distributed throughout the ganglion, among the nerve fibers. Owing to the direct manner in which the processes leave the cells, the arrangement is such that in general the fibers and their cells lie in a direct course.

The cells are enclosed in a connective tissue capsule (Figs. 11, 12, Cap.) which takes the form of a thin structureless membrane. Tn sections the cells were contracted away from the capsule by the action of the reagents. (Figs. 11 to 13.)

On the interior of the capsule, and in the fresh condition pressing directly upon the cell body, are a number — 4 to 6 or more — of round or oval nuclei, the nuclei of the mantle cells. (Figs. 3, 11, 12, 13, M. c.) These cells are also known as capsule cells, but the designation mantle cells" seems the better term inasmuch as they are developed independently of the capsule.

On the Interior of the sheath of the nerve fiber are found elongated fiat nuclei, the nuclei of the Sheath of Schwann, S. c. Figs. 1, 2 and 6. They are similar in structure to the nuclei of mantle cells.

The ganglion of the ninth nerve of Sqiudus Embryos, Study of macerated material was instructive as to the nature of the nervous elements of the ganglion. Such preparations demonstrated that by far the most prevalent type of nerve cell is the bipolar form. The cell body is somewhat elongated in the long axis of the ganglion, being oval or spindle-shaped, and from each pole is given off a process. (Figs. 1 to 3, c.p., p.p.) Figs. 2 and 3 are good illustrations of oppositipolar cells, that is, cells whose two processes arise from directly opposite poles.

But the oppositipolar form is not the invariable one. Cells were found which form a series from bipolar to unipolar cells similar to those described by Freud for Petromyzon, Eetzius for Myxine, and Lenhossek for Pristiurus embryos.

The first stage in this transformation series is a cell, one of whose processes leans a little to one side, destroying the symmetrical shape of the cell body. Further approach of the processes toward one

D 126

Martin R. Chase.


^c. Mc


Fig. 2.


Fig. 3.

Fig. 1 .

Fig. 1. Bipolar cell from a teased ganglion of the Ninth Cerebral Nerve or a Sqiialus acanthias embryo of about 16 cm. length. X 260.

Fio. 2. Bipolar cell from a preparation similar to Fig. 1. X 260, Sc, Cells of sheath of Schwann.

Fig. 3. Bipolar cell from a similar preparation, x ^30. M. c, mantle cells, c. p. central process, pp. peripheral process, nucl, nucleolus.

D A Histological Study of Sensory Ganglia. 127

another results in the assumption by the cell-body of a still more asymmetrical position with reference to the origins of the axones. Fig. 4.

In Figs. 5 and 6 the origins are progressively nearer to each other. From this stage to that seen in Figs. 7 and 8 is but a short step.

Fig. 4. Intermediate form from a teased ganglion of Squalus embryo. X530.

Here the processes have a common origin in the short thick axone, (ax.) dividing almost at once into two branches, p. p. and c. p.

Lengthening of the common origin into a more slender, true nerve fiber gives the final step in the process of unipolarization (Fig. 9). Here the single axone (ax.) leaves from the side of the cell body, dividing after a relatively long course into two fibers of almost equal


Fig. 5. More advanced intermediate form from the same preparation as Fig. 4. X 530.

thickness. The unipolar cells were seldom seen, although many intermediate forms and quite a large number similar to Figs. 6 and 7 were found.

The following enumeration, the result of careful count of the cells in one slide, gives an idea of the proportions of the different elements :

D 128

Martin R. Chase.

Fig. 6.



Fi(i. 8.


Fig. 7.

F^iG. 9.

Fig. 0. Intermediate stage from ninth ganglion of the same individual as Fig. 4. X 530.

Fig. 7. Early unipolar cell from same preparation as Fig. 6. x 530. Fig. 8. Early unipolar cell from same preparation as Fig. 6. x 530. Fig. 9. Unipolar cell from same preparation as Fig. G. x 530.

D A Histological Study of Sensory Ganglia. 129

Cells undoubtedly bipolar 208 50.6%

Cells showing one process 124 30.0%

Cells broken or uncertain 77 19.0%

Unipolar cells 1 25%

Of the 208 bipolar cells, 20 were intermediate forms of various stages. This does not include all which departed slightly from oppositipolar condition. Of those classed as having one process, the majority undoubtedly must be interpreted as mangled bipolar cells, but some must be considered unipolar cells whose process has been broken off back of the bifurcation.

-P/rS. -Pns.

Fig. 10. CeU from a reduced silver preparation of the gangiion of the Ninth nerve of Squalus embryo, showing protoplasmic slings, Pr. s. X 530.

In the majority of the cells drawn (Figs. 3 and 4 to 8) it will be noted that one process (p. p.) is of greater thickness than the other (c. p.). This is in conformity with the results of many WTiters who have found the peripheral process much thicker than the central.

The transition series above described is confirmed by the reduced silver nitrate method. Silver preparations show that while unipolar and intermediate stages are mainly located about the periphery of the ganglion, they may be found in any portion of it, and their processes have the distribution normal to bipolar cells. As stated in the review of the literature, v. Lenhossek finds in the spinal ganglia of Pristiurus embryos that these forms send their peripheral processes into the dorsal ramus.

Cells with protopldsmic slings. The silver method shows also the presence of cells with structures similar to those foimd in the adult forms of other animals. These cells exhibit fine protoplasmic processes which after a short curved course return to the cell at a short

D 130 Martin R. Chase.

distance from their origin, describing a sort of half circle. Fig. 10 illustrates such a cell. So far as seen the accessory processes in specimens of 16-20 cm. have formed closed slings, but it is supposed that they correspond to the complicated structures found in the adult.

Nucleus, The nucleus is relatively large and varies much in






Fig. 11. Fig. 12.

Fig. 11. Bipolar cell from a reduced silver preparation of the Tenth ganglion of a Squalus embryo 17 cm. long, x 530.

Fig. 12. Intermediate stage from the same ganglion as Fig. 11. x ^30.

size, shape and position. It is rounded or oval in shape and, as is seen from the figures, rarely centrally located in the cell. The nucleus contains at least one large deeply staining nucleolus (nucl). Quite regular is the presence of two such elements. Figs. 1, 3, 5 and 7.

The Ganglia of the Tenth Cranial Nerve of Squalus acanthtas

D A Histological Study of Sensory Ganglia 13L

Embryos, Of the ganglia composing the vagus group, those supplying the gills resemble very closely the ganglion of the ninth nerve.

There was observed in sections a series from bipolar to unipolar cells similar to that described for the ninth. Figs. 11, 12 and 13. In Fig. 13 the processes arise from a common prolongation of the cell body.




Fig. 13. Intermediate stage showing common origin of the processes Pr. n. From the same ganglion as Fig. 11. X 530.

Figs. 11, 12 and 13 are all from the same ganglion. In all three cells the peripheral process, p. p., is considerably thicker than the central process, c. p., and this occurrence is constant for the ganglion.

Sling arrangements, similar to those on cells of the ninth, were rarely seen, and the accessory processes of the adult was lacking.

The other ganglia of the vagus, so far as observations go, are

D 132 Martin R. Chase.

simpler than the branchial and the cells have kept more completely their oppositipolar condition.

The Ganglion of the Ninth Cranial Nerve of Adult Mustelus. This adult ganglion is more complicated than those of the embryos studied. The interstitial connective tissue is more rich, and the course of the nerve-fibers is not so direct. The capsule is more highly developed, and the mantle cells are more numerous.

The predominating type of cell is bipolar, but intermediate forms are frequent. Cells around the periphery of the ganglion sending

Fig. 14: Unipolar cell from the ninth ganglion of an adult Mustelus tanis. Reduced silver, x 256.

their processes toward the center are common, and in some cases they wen? seen to be unipolar — Fig. 14.

Many cells give off besides the regular processes a variable number of accessory branches of various sizes. (Cells A, B, C, of Fig. 15, A.c.p.)

These accessory branches range from very fine fibers ending with small bulbs, usually within the capsule, (Cell C, ac. p.) to relatively thick processes w^ith large round or oval end-bulbs outside the capsule. (Fig. 15, L. eb, CellB.)

Usually these extra processes are quite evenly distributed about the periphery of the cell, as in Fig. 15. In other cases a short

D A Histological Study of Sensory Ganglia 133

knob-like prolongation of the cell body gives rise to a large number of fine branches which may divide, diverge with a tortuous course, and end near the cell, usually with small thickening.

Fig. 16 shows a cell with an accessory process which divides repeatedly, each terminal fiber ending with a bulb-like enlargement. Often these accessory rami seem to anastomose with one another.


- f

Fig. 15. Small portion of a section of the ninth ganglion of an adult Mnstelas canis, showing accessory processes. Ac. p. and small ganglion cells, a to f. Reduced silver method, x 25C.

The occurrence of this condition is quite general, and in portions of the ganglion nearly every cell is affected.

Many small cells, hardly larger than the largest bulbs above described, are seen (Cells a to f., Fig. 15). These small cells are nucleated, have processes, and give a picture in miniature of the large ganglion cells. Cell c. Fig. 15, represents a geminipolar cell, indicating that in these cells also a transformation series is to be found.

D 134 Martin R. Chase.

The Vagus of Adult MusteAus. In the adult, as in the embryo, the branchial ganglia resemble the ninth very closely, although they are somewhat smaller and simpler perhaps in construction. Bipolar and intermediate cells were demonstrated well in silver preparations, and in addition unipolar forms in small numbers. The accessory processes observed so constantly in cells of the Ninth are present, although not seen in such numbers or in such complicated form.

The ganglia of the lateral line and visceral branches are relatively simple. Here the structure is more diffuse and the cells are not

Fio. 16. Cell showing accessory processes, Ac. p. From the same gauglion as Fig. 15. X 530.

crowded so closely together as in the Ninth. Consequently strictly bipolar, indeed oppositipolar cells, are the rule. These elements may be very readily isolated by teasing.

The spinal ganglia of Squalus embryos were studied by teasing. While much smaller, the spinal ganglion is, so far as observed, similar to the ninth cerebral ganglion in constitution, although relatively more simple.

The ganglia of the Fifth and Seventh cerebral nerves have been very briefly studied. The cells in the maxillo-mandibular and superficial ophthalmic branches were found to be exclusively oppositipolar.

D A Histological Study of Sensory Ganglia 135

The main Gasserian and Facial ganglia are very complex — perhaps the most modified of any of the cranial ganglia.

B. Observations on Mammals.

Spinal and Gasserian ganglia of the cat and dog were treated by the reduced silver nitrate method, which gave excellent results.

..Mc. ..-Ac. p. ...e.b:

Fig. 17. Cell from the spinal ganglion of a 10 days kitten. Ac. p. accessory process, e. b., end-bulb. Silver Method, x ^30.

The nerve fibers stained a deep brown; the other elements of the ganglion colored less heavily.

Spinal Ganglia of the Cat. Ganglia from a ten-day kitten and the adult cat were studied. In these specimens the ganglia differ


/f; "^^'

Fig. 18. Spinal ganglion cell of a 10 days kitten. Ax., axone, ac. p., accessory process. X ^30.

characteristically from those of the Selachians by the predominance of unipolar cells.

The cells in the ganglia of the kitten vary much in size, those in the interior of the ganglia being much smaller as a rule than those about the periphery. The round or oval mantle cells are very

D 136 Martin R. Chase.

numerous (Figs. 17, 19), sometimes appearing heaped up about the origin of the axone. The axone rarely leaves the cell in a direct line, but iisually goes into a glomerulus which is relatively simple in the kitten. (Fig. 19.)

The nerve-fibers show a longitudinal striation due to the neuro

U ^•^


Fig. 19. Typical spinal ganglion cell of a kitten, showing a simple glomerulus of the axone (ax.), x 530.

fibrillse. The cytoplasm exhibits a complicated network, also due to neurofibrillar. (Fig. 22.)

The process often leaves by an implantation cone from a sort of hollow in the body of the cell. Figs. 19 and 22. In some cases the process follows around the side of the cell for a short distance within

He. .Ax.

Fig. 20. Spinal ganglion cell of a kitten. X 530.

the capsule, without convolutions, and then leaves the capsule in a direct line (Fig. 20), a mode of origin described by v. Lenhossek ('06).

In many sections T-formations could be sc^en in large numbers, one branch going centrally and one peripherally.

D A Histological Study of Sensory Ganglia. 137

Processes with end bulbs, as described by Iluber, Ramon y Cajal and others were present in rather limited numbers. They take the form of round or oval bulbs of various sizes, staining very evenly a brown or reddish color. They are connected with the cell or an axone by a fiber which varies greatly in thickness. Figs. 17 and 18, ac. p., e. b., illustrate such processes.

6 I6c.


Fig. 21. Spinal ganglion cell of a kitten, showing a large vacuole or ai)erture (Vac), x ^530.

Some processes were found which suggest the fenestrated cells of Cajal and Levi, although not exactly agreeing with his descriptions of them. Sometimes the ^"aperture" is very large, and bounded by a relatively thick concentrically striated band. The fibrillse of this band extend into the cell body and appear to connect with the

Fig. 22. Spinal ganglion cell of a kitten, showing immediate division of the axone (ax.), x 530.

network in the cytoplasm. (Fig. 21.) The aperture in this cell ran through four sections of 10 microns thickness and appeared to be a closed cavity, suggesting the vacuoles descril)ed by Athias ('05). Other small vacuoles or apertures appeared in cells, but apparently not in close relation with mantle cells, as described by Ramon y Cajal and Levi.

D 138 Martin E. Chase The cells in the ganglia of the adult cat are much larger than in the kitten. The initial glomerulus is immensely complicated, and processes with end-bulbs are more frequent

The Spinal Ganglia of the Dog. The study of the ganglia of the dog was brief, owing to the lack of time. In general appearance the sections are very similar to those of the adult cat The glomerulus is strongly developed. V. Lenhossek ('06) found the glomerulus in the spinal ganglia of the cat and dog the most complicated of all mammals.

Fig. 23 shows a cell whose axone (ax), after a simple glomerulus, divided into two branches of unequal thickness (c. p., p. p.)* -^ ^^^ fiber (c. p.) arises near the axone and ends freely within the capsule.

^- .Acp.

PP Fig. 23. Spinal ganglion cell of an adult dog. ac. p., small accessory process. X 530.

The walls of the nuclei are sharply defined. The nucleolus is large, round usually, not heavily stained, and seems to be granular in composition.

In these ganglia a great many fine fibers in addition to the thick medullated processes were seen — supposedly the nerve-processes of the smaller ganglion cells. As such fibers were sometimes found to branch repeatedly, the impression that they are partly of sympathetic origin is strong, although pericellular plexuses have not been seen.

Summary, In addition to the bipolar cells which are the predominating type, certain of the ganglia of Selachians show a transformation series from bipolar cells to unipolar cells, similar in all respects

D A Histological Study of Sensory Ganglia. 139

to those described by Freud, Retzius and v. Lenhossek as above noted. In the adult, besides the normal axone or axones, a characteristic appearance is the presence of a variable number of accessory branches which are of varying thickness, and terminate by end-bulbs, freely, or by anastomosis with one another or the cell body. Fine fibers, forming closed slings in the embryo of 16 cm. constitute a very simple corresponding structure. As mentioned above, Levi ('06) has described for Selachians "short thick processes which terminate after a brief course for the most part within the capsule." While Levi shows no figures, and has not described end-bulbs in this connection, the structures found in Mustelus probably correspond to the processes described by him. They correspond more closely, however, to his figures of homologous accessory processes in Chelonians, and are similar in nature to those found in amphibians and mammals.

Observation of mammalian ganglia confirmed, so far as the study, progressed, the accounts of Retzius, Cajal, v. Lenhossek, etc. The processes terminating in the end-bulbs offered no new features. Typical fenestration affecting the origin of the axone was not seen, and the apertures in ganglion cells resembled closed vacuoles rather than true perforations of the cell body.

Received for pnbUcation, January 8, 1909.

BIBLIOGRAPHY. Athias, M. La vacuolisatlon des Cellules des ganglions spinaux les animaux

a retat normale. Anat. Anz., Bd. 27, 1907. Cavazzant, B. Sur les Ganglions Spinaux. Archives Italiennes de Biologie,

+ 28, 1897. Daae, H. Zur Kenntniss der Spina Iganglienzellen bei SRugetieren. Arch.

f. Mik. Anat, Bd. 31, 1888. DissE, J. Ueber die Spinalganglien der Amphibien. Anat. Anz., 1893, Jahrg.

VIII, Suppl. II. DoGTEL, A. S. Der Ban dor Spinalganglien boi den Siingetieren. Anat. Anz.,

Bd. 12, 1896. DoGnsL, A. S. Zur Frage fiber den feineren Bau der Spinalganglien und In

deren Zellen bei SS-ugetieren. Internat. Monatss. f. Anat. u. Physio'l.,

Bd. XIV, 1897. DoQiEL, A. S. Zur Frage liber den feineren Bau der Spinalganglien und

Zellen bei Saugetieren. Internat. Monatss. f. Anat. u. Physiol., Bd.

XV ('98).

D 140 Martin R. Chase.

Huber, G. Carl. The Spinal Ganglia of Amphibia. Anat. Anz., Bd. 12, 189C.

Kamkoff, G. Zur Frage iiber den Bau des Ganglion Gasserl bei den S&ugetieren. Intern. Monatss. f. Anat. u. Phys., Bd. xiv, 1897.

KoHN. Ueber die Seheidezelleu (Randzellen) der peripheren Ganglienzellen, Anat. Anz., Bd. 30, 1907.

V. Lenhoss£k, M. Beobachtungen an den Spinalganglien und deiu RUckenmark von Pristiurus-embryonen. Anat. Anz., Vol. VII, *92.

X. Lenhoss^k, M. Nervensystem, Spinalganglien. Ehrgebnlsse der Anat. u. Entwick., Bd. 7. 1897.*

V. LENHossfiK, M. Untersuchungen tiber die Spina Iganglienzellen des Frosches. Archiv f. Mik. Anat., Bd. 26, 1886.

V. LENHOSSfeK, M. Zur Kenntnis der Spinalganglienzellen. Archiv f. Mik. Anat.. Bd. 09.

V. LENHOSsfeK, M. Zur Kenntnis der Spinalganglien, Beitr^ge znr Histologie des Nervensystems und d. SinnesorKaue. 1894.

Levi, G. La struttura del gangli cerebrospinal i del Cheloni. La struttura dei gangli cerebrospinali nei SelachI e nei Teleostei. Monitore Zoologico Italiano Anno 17, 1906. Struttura et istogenese dei gangli cerebrospinali dei niammiferi. Anat. Anz., B<1. 30, 1907.

Ramon y Cajal, S. Die Structur der senslblen Ganglien des Menschen und der Tiere. Ergebnisse der Anat. und Entwickel., Vol. for 1906.

Ranvier, L. Des tubes nerveux en T et de leurs relations avec les cellules ga ngl ion na ires. Comptes rendu de I'acad. d. sc., Vol. 81, 1875.

Rawitz, B. Ueber den Bau der Spinalganglien. Archiv f. Mik. Anat., Bd. 21, 1882.

Retzius, G. Untersuchungen Uber die Nervenzellen, etc. Archiv f. Anat. u. Physiol., Anat. Abt, 1880.

Retzius, G. Ueber die Ganglienzellen der Cerebrospinalganglien und Uber subcutane Ganglienzellen bei Myxlne Glutinosa. Biologische Untersuchungen, Neue Folge 1. 1890.

V. Smirnow, A. E. Beobachtungen Uber den Bau der Spinalganglionzellen bei eineni vierinonatllchen menschllchen Embryo. Arch f. Mik. Anat.. Bd. 59, 1902.

Spirlas, A. Zur Kenntnis der Spinalganglien der Saugetiere. Anat. Anz.. Bd. 11.

TiMOFEEw. I). Beobachtungen Uber den Bau der Nervenzellen der Spinalganglien und des Sympnthlcus beim Vogel. Intern. Monatss. f. Anat. u. Physiol., Bd. XV, 1898.


Thk DEVELOPMBifT OF THE Chick. An introduction to embryology. By Frank R Lillie, Professor in the University of Chicago. Henry Holt k Co., 1908.

Every one Ivho has long wished for all ilp-to-date text book on the developtnent in the egg of the common foTfl, to replace the classic book of Foster and Balfour, and the most excellent book of Marshall, will need to judge by trial whether this is his ideal, or but the voice of one crying in the wilderness. This is a book of some four hundred and fifty pages and two hundred and fifty most excellent illustrations, six of these being full page colored plates. The illustrations are well chosen from many recent sources and very many, both surface views and sections, are original. In fact, it is this aspect of the book which is emphasized ; it is no mere complication but full of the original work of the author, and the- work of others has been very judicially and fairly balanced. It may be said that the vast body of known facts on this subject has been digested by the author, digested to fit his needs ; whether the process has been carried far enough to suit the tender assimilative powers of the "beginners in embryology," for whom it is said to be written, may well be doubted.

Every teacher of embryology must have the book, and the man who really wants to study the embryology of the fowl must use this as the most complete book of reference. The average beginner, however, who actually knows nothing of embryology, will scarcely have either the time or thfe pertinacity of purpose to get an adequate notion of the events in the egg of the fowl from such a technical treatment.

Until the second part, which treats of the organology from the fourth day on, shall have been much shortened and the first part reCdst in presentation, the book can scarcely be judged as a text book for beginners, though doubtless it will be an inspiration to those fortunate, good students whose teachers introduce it to them.


D 142 Book Eeviews.

As the foremost book, in English, that presents the subject of the development of the fowl, as it appears to-day, the book of Professor Lillie deserves only praise. Throughout it is marked by conscientious matter of fact treatment and a precision that shows both in the illustrations and the text. The author's attitude is indicated by the section on the eleventh cranial nerve which is merely, "No observations on the development of this nerve in the chick are known to me." The book is restricted to the description of the visible developments of the common fowl, except for a few necessary illustrations from other birds. It is neither experimental nor comparative, barring a few very brief references to probable recapitulations. It will be xmderstood also that there is no reference to methods and no directions for laboratory work. There is, on the other hand, an excellent bibliography of some four or five hundred recent original contributions to the subject of the embryology of the fowl.

The introductory first fifteen pages take up the subjects of the cell theory, the recapitulation theory, the physiology of development, the law of genetic restriction, the sperm and ovum and the polarity of the ovum. The book being essentially a chronicle of facts, this introduction might well be omitted.

The facts are arranged in fourteen chapters, as follows :

A brief account of the reproductive organs of the hen, the formation of the egg and the structure and chemical make up that it has when laid, makes the first chapter. As nothing is known of the fertilization, and maturation, and little enough of the cleavage, of the hen's egg; these phases are figured and described from the excellent work that has been done upon the pigeon. The second chapter devotes some thirty pages to the details of these events in the bird's egg before it is laid, including the formation of the ectoderm and entoderm. The third chapter contains a discussion of the rate of development, with a folding plate presenting the amount of development of organs in the hen's egg at different stages up to forty-one somites, at 96 hours. Here also is a brief sketch of the orientation of the embryo in the egg and its relations to the yolk sac, but the expected epitome of the entire development within the egg, as far as embryo is related to membranes, is deferred for later chapters, which must prove troublesome to the novice.

D Book Kaviews. 143

In chapter four the events from laying to the formation of the first somite are given twenty pages with excellent views of sections, as well as some surface views taken from Schauinsland and representing the blastoderm of the sparrow. The fifth chapter, though of some forty pages, includes only the events from the head fold to the period of twelve somites, thus advancing only to about the 33d hour of incubation. This indicates the detailed care with which the subject is treated, even though here as throughout, much of the space is taken up with most useful and finely executed illustrations. Chapter six runs over the events from 34 to 72 hours; but this, ending at the lapse of three times twenty-four hours, is about the only remnant of the old method of describing the chick's progress as if done by day labor. As most of the organs are laid down in this important third day, all the eighty, pages of this chapter are necessary.

With the next chapter begins the second part of the book, the events of the fourth day to the time of hatching. This part is chiefly the completion or advancement of organs already begun, and such organology could, from the view point of beginners in Embryology, well be more briefly described than it is in the following two hundred and thirty pages, the more so as many students who are interested in medicine may get such information from the study of. pig or other mammalian embryos.

The seventh chapter deals with the general form and the fate of the embryo with reference to the anmion, allantois and yolk sac, and much of this, as above noted, might have been acceptable earlier in the book. The beautiful colored plates of the vascular area add much to this part of the volume.

The eighth chapter deals with the nervous system, the neuroblasts, neurons, brain, cord, and origin of cranial and spinal nerves.

The ninth chapter takes up the eye and the ear where they were left in the sixth chapter and carries them to their definitive form.

The tenth chapter gives thirty pages to the alimentary tract. The eleventh some fifteen pages to the difficult problems of the body cavities and mesenteries, while the twelfth devotes thirty pages to a very good account of the later stages of the heart and the

D 144 Book Reviews.

vascular system. The remaining two describe, in each about the above number of pages, the urinogenital system and the skeleton. This presentation of the facts of the development of the sexual and the excretory Organs will be most welcome to all who have not ready access to all the recent literature, and it is a good sample of the great value of the book throughout, both for the teacher and the student who is prepared to realize what has been done for him. Throughout, points that have received no attention as yet are noied, and one will not get the idea that the subject is a completed one despite the great mass of fact that has accumulated since our old texts were written. The press work and the execution of the illustrations do credit to the publishers.

The volume well represents the present status of our knowledge of the development of the fowl : a great monument of hard won facts, evidently connected with the embryology of other animals, in fact faintly illuminated still by the principle of recapitulation, but on the one hand shorn of the mysteries of the unfamiliar and on the other awaiting some future more fimdamental interpretation and explanation. With the increase of accuracy and solid achievement which this volume should incite in the study of the chick, we may expect even better progress in the future and reaction for the better in the study of embryology in connection with medical schools. Received for pubUcation, February 19, 1909. E. A. Andrews.

Quain's Elements of Anatomy. Eleventh Edition. Vol. I. Embryology. By T. H. Bryce. Longmans, Green & Co., London and Xew York. 1908.

The appearance of a new edition, the eleventh, of Qiiain's Anatomy is a noteworthy event, since, among English text-books, this has long occupied a foremost place as an exposition of Anatomy from a scientific standpoint. The first volume of the present edition, like that of the tenth, is devoted to Embryology, a subdivision of Anatomy concerning which our knowledge has made remarkable progress since the publication of the earlier edition (1892), not only as a result of the study of greater numbers of embryos and of embryos of earlier stages of development than were formerly available, but also

D Book Eeviews. 145

in consequence of the investigation of embryos of a larger series of mammalian types having yielded results whereby the gaps in our knowledge of human development can be more accurately filled in and a significance attributed to ontogenetic phenomena which formerly were more or less unintelligible.

To cover the wider extent of territory thus made available within the compass which the conditions rendered necessary must have been no easy task, but it may be said at once that Dr. Bryce has fully maintained the high standard of excellence set by Professor Schafer in the earlier volume. But this has been accomplished only by the extension of the present volume to 260 pages, as compared with the 169 of the tenth edition, notwithstanding that considerable space has been gained by the omission of the bibliographic lists at the close of the various sections, which were such characteristic features of the older book. Dr. Bryce has not, however, failed to supply in the foot-notes numerous references to the most recent embryological literature, those papers giving more or less extended bibliographies being especially noted. He has also added a large number of new figures, the volume containing a total of 313 as compared with the 200 of the earlier edition, and it is deserving of mention that over seventy of the figures are reproductions of original draw^ings by the author, many of which are of a high degree of excellence and contribute materially to the value of the book.

A detailed review of the text is hardly feasible; it can truly be said to present a very complete resume of our knowlijdge of human embryology and organogeny. But the effort to condense the mass of material at the author's disposal within the proper limits is evident on every page and has occasionally led to a degree of conciseness which somewhat obscures the descrij)tion ; indeed, in one section, that on the skin and the cutaneous glands, condensation is carried to such an extent that no mention is made of the epitrichium, of the develop^ient of the nails, or even of that of the mammary glands. As a result of this condensation the volume is one that may be read with profit, though not altogether with ease, by one more or less familiar with the results of recent embryological investigation, but to a beginner it is to be feared that the book will prove, in parts at least, exceedingly difficult of comprehension.

D 146 Book Reviews.

For this the author is not altogether to blame, but rather the plan which endeavours to concentrate within the limits of a single text-book a presentation of our information in all its details concerning all departments of human anatomy. This is impossible without the expansion of the volume to an unreasonable size, unless much of the material be condensed to such an extent that it becomes unpalatable from its very condensation. In a text-book of Anatomy, using that term in its narrower sense, it is the histology and especially the embryology that naturally suffer in this respect, and it would be far preferable to include in such a book only so much of these subjects as is absolutely necessary to make the facts of gross structure intelligible, the student obtaining that further knowledge of histology and embryology that he ought to possess from special treatises. In other words, these two subjects, as ordinarily presented in an anatomical text-book, are treated more or less as appendages to the main topic and are, therefore, not given that space and treatment which are necessary for their successful presentation and which they deserve. They are integral parts of the anatomical discipline and also have applications apart from their importance in rendering intelligible the facts of gross anatomy. They should be taught as essential parts of the anatomical course, but they should not be presented in such a way as to give the student the impression that they are merely academic addenda to a more important and "practical" subject.

J. P. McMurrich. Received for publication, February 12, 1909.


On the Peeservation of Data Used in Biometeio


At the present time there is an increase in the number of biological investigations in which long series of measurements on different characters of animals are employed.

On account of the expense, most of our journals hestitate to print the individual measurements, and it thus often happens that valuable data are either lost, or at least fail to remain accessible. In view of these facts, the Wistar Institute wishes to aid in the preservation of such data, so far as they relate to vertebrates, and to that end the following suggestions are made to authors working in this field.

1. That authors communicate with the Institute to learn whether their data fall within the scope of this plan.

2. In case they do fall within the scope of the plan, then that the data, in a complete form be sent to the Institute.

3. The Institute will copy the records on cards, and file the same as a reference archive.

4. The Institute will also make a typewritten copy of the records, to be certified by the author, and with the permission of the author, to be used by other investigators who desire to consult his data.

5. A duplicate typewritten copy will be returned to the author, together with the original manuscript.

M. J. Gbeenman^ Director. The Wistar Institute of Anatomy and Biology Philadelphia, Penna.




Vol. m. APRIL, 1909 No. 4


Papers and Abstracts of Papers Read at the Twen^ty-Fourth Session, - Baltimore, Md., December 29, 30, 31, 1908.




EDWIN G. CONKLIN. Princeton University.

The central aim of all embryological study is to trace to their origins the principal differentiations of organisms ; in addition the aim of experimental embryology is to be able to control these differentiations. In this study experiment and observation can never profitably be separated. More accurate observations will always be needed as a basis for more critical experiments. These two methods of study are, therefore, not antagonistic, but mutually dependent upon each other. Experimental embryology is not a wholly different study from descriptive embryology, but rather a more refined and accurate form of observation, in which emphasis is placed upon physiological processes rather than upon morphological structures. In any large view of the science the results obtained by either of these forms of study cannot be separated easily or profitably.


D 150 Edwin G. Couklin.

All sciences as they become more detailed and accurate pass from the descriptive to the experimental stage, and it cannot be otherwise with the various branches of biology. Where the materials with which a science deals are relatively simple, experiments may profit-* ably begin at a much earlier stage than where those materials are very complex. In the case of embryology the materials are so complex that there is still a large opportunity for studies of a descriptive sort, although the experimental method will here play a larger and larger part as the science develops. But experimental studies to be of much value must always be founded upon a knowledge of the normal condition, and the more thorough this knowledge is the better.

Again to be of real value experiments must be of a detailed and individual character. Much of the early work in experimental embryology has been of a general and explorative sort ; for example, eggs were treated by the thousands and the end results only in the case of a few of them noted. Now every one who has done such work knows that one of the most usual of all results is the great variation in the types and structures produced. This diversity, probably, depends upon different conditions of the organism at the time of the experiment and also upon varying actions of the experimental conditions upon the organism. Thus it has been shown that eggs are much more susceptible to injury during division stages than during resting periods, and in many cases it can be shown that the precise time and manner of modifying the normal conditions are of the greatest importance in determining the end result. All of the variables should be known for each case, and this requires individual rather than mass experiments.

These general remarks apply with especial force to the study of the organization and early differentiations of the egg. As the result of both observation and experiment we know now that e^gs are not unorganized, homogeneous, isotropic, as they were once supposed to be, but that they are morphological and physiological systems, possessing certain differentiations which are correlated with cprresponding differentiations of the adult. Although much valuable work has been done in this field, the results so far gained constitute

D Organization and Early Differentiation of the Egg. '151

only a sort of preliminary program of the work which is yet to be done. Some of the chief points to which attention has been and must still be directed are the following:

1. The Origin and Causes of Polarity and Bilaterality.

It is well known that the chief axis of the ovarian egg, in many animals, coincides with the chief axis of the fertilized egg and gastrula, and that the latter, either directly or by certain bendings, becomes the chief axis of the adult animal. Is it possible to shift this axis in the egg by shifting the positions of the ooplasmic substances? Lillie (1906) and Morgan (1907) have shown that in the eggs of Chsetopterus, Arbacea and Cumingia some of these substances may be shifted into new positions by centrifugal force without shifting this egg axis. Lillie believes that the polarity persists in the 'ground substance' which is not altered by the centrifugal force used. On the other hand, I find that this axis may be shifted in the eggs of Cynthia and Crepidula if they be centrifuged for a considerable time after maturation and before the first cleavage. Development may be normal after this shifting of the chief axis. In this case there is good reason to believe that polarity consists in the heteropolar arrangement of certain ooplasmic substances, though the results obtained by Lillie and Morgan indicate that not all of these substances are concerned.

Bilaterality appears at different times in the development of different animals ; in echinoderms it is first evident in the late gastrula stage; in most annelids and moUusks at the time of the formation of the mesomere, 4d; in some annelids and mollusks at the first cleavage of the egg, in others before cleavage; in the frog and ascidian at the time of fertilization; in cephalopods and insects during the development of the oocyte in the ovary. Is there a common cause of bilaterality, and if so what is it ? According to Roux the path of the entering sperm in the egg determines the plane of the first cleavage in the case of the frog, and this usually coincides with the plane of bilateral symmetry. But it is now knowTi that the first cleavage plane in this animal bears no constant relation to the plane of symmetry and, therefore, the path of the sperm cannot determine

D 152 Edwin G. Conklin.

the plane of bilaterality. The fact that this plane frequently coincides with the first cleavage may indicate that although the sperm may enter at any point on the egg there is a line of least resistance along one plane, the plane of future symmetry, and that, therefore, bilaterality may be present in the egg before fertilization. Among ascidians the first plane of cleavage always coincides with the plane of symmetry, and here also there is evidence that this plane is predetermined in the egg. There is convincing evidence in ascidians that one cause of bilaterality is the bilateral localization of certain egg substances. If this bilateral arrangement of substances is changed, the bilaterality of the embryo is destroyed. The bilateral arrangement of egg substances cannot be changed after the first cleavage is completed, and in general the arrangement of substance cannot be changed after cell walls have been formed. These facts lead to the conclusion that in these animals bilaterality is dependent upon a bilateral arrangement of egg substances.

2. The Potency of Blastomeres and Ooplasmic Substances.

It is well known that many investigators, following the lead of Driesch, have found that individual blastomeres of the early cleavage stages of many animals may give rise to entire larvse. On the other hand, it is now known that in a large number of other animals isolated blastomeres produce only those parts of a larva which they would normally form. In the one case the blastomeres are said to be totipotent, in the other to be specified. In those forms in which partial development of isolated blastomeres takes place, the ooplasmic substances are either segregated into different blastomeres, or they are very definitely and fixedly localized in the blastomeres. This indicates that in these cases the potency of blastomeres depends upon the completeness with which the different substances are represented in them, or upon the power of localized substances to rearrange themselves into a typical whole.

Are these different ooplasmic substances specified, or is each totipotent? In certain molluscan and ascidian eggs, where these substances are very plainly visible and definitely localized, it has been shown that the development is a strict mosaic work, based upon the

D Organization and' Early Differentiation of the Egg. 153

localization of these substances. Here it is possible to speak of organforming or histogenetic substances, since each gives rise only to definite organs or tissues. In the absence of a certain substance from an egg, a corresponding part of the embryo is lacking. On the other hand, when substances have been thrown out of their normal positions by centrifugal force, they still develop, in some cases at least, into their characteristic structures. Here, then, we have both negative and positive evidence that these substances are definitely specified, that they are organ-forming. There is no doubt that different eggs differ greatly in the capacity which parts of eggs show for development. This may be due to varying degrees of differentiation or localization of these substances, or to varying powers of regulation. Further experiments must be depended upon to harmonize the conflicting results already obtained.

3. The Mechanism of Differentiation.

So far as the process can be directly observed, differentiation consists in the origin, localization and progressive transformation of unlike substances in cells. How these substances arise in the first place is unknown, though there is reason for believing that they are formed through the interaction of nucleus and cytoplasm ; further study on this subject is much needed. The localization of these substances in the cell body is accomplished, in large part, through the achromatic portion of the mitotic figure. Lillie has shown that a limited amount of differentiation may take place in the absence of cleavage, and it has been shown repeatedly that there is no necessary relation between planes of cleavage and lines of differentiation. Nevertheless, cleavage is necessary to progressive and orderly differentiation. I have found that cell walls limit and fix the movements and localization of substances, and that the movements of substances within cells take place largely through the instrumentality of the astral systems of the mitotic figure, or of the entering spermatozoon. Differential divisions of the cell body are thus brought about, though there is no evidence that differential divisions of the nucleus ever occur, except in certain maturation divisions of the egg and sperm. Where nuclei become differentiated it is probable.

D 154 Edwin G. Conklin.

as Boveri has suggested, that it is through the influence of the surrounding cytoplasm. It is, therefore, probable that while one of the principal functions of the mitotic figure is the eqyal distribution of the chromosomes to the two poles, another scarcely less important function is the localization and differential distribution of the ooplasmic substances to the daughter cells.

By a series of remarkable observational and experimental, researches, Boveri and Wilson have shown that individual chromosomes may possess different hereditary value. There is here open one of the most promising fields in the whole science of biology for the application of experiment to the solution of fundamental problems of cytology and development. In this connection may be mentioned also the fundamental experiments of Loeb, Garbowski, Herbst and others, on the relative influence of the egg and sperm on differentiation and inheritance.

4. Modifiahility of Organization and Differentiation.

Finally, mention should be made of the need of experimental work to determine to what extent the organization and differentiations of the egg may be permanently modified. Hitherto, such work has been taken up only incidentally, but it is one of the greatest problems of biology, upon the solution of which the artificial production of new types must largely wait.

This brief summary of results and aims of experimental work as applied to the organization and early differentiation of the egg, is, as I indicated at the beginning, in the nature of a preliminary and very incomplete program. Experimental embryology is a new science, and the valuable work so far done is more or less isolated and disconnected. This symposium can serve no more useful purpose than to point out the lines in which work is especially needed, and to stimulate interest in the solution of the fundamental problems of development.



THOMAS H. MORGAN. Columbia University,

In recent years the results of experimental embryology have led to the view that however significant in the physiology of the cell, and even in heredity, the nucleus of the egg may be, the protoplasm plays also a very important role in early development. . It has long been taken for granted by embryologists that the presence of such inclusions as the yolk is an important factor in determining the fate of embryonic cells. The presence of pigment in certain parts of the egg, as in the red ring in Toxopneustes, and the yellow band in Cynthia, has made it possible to follow movements that take place in the cytoplasm, and these movements are definitely associated with the formation of organs. The pigment allows us to follow certain movements that might otherwise escape our attention. The movements show, at least, that there are accumulations of materials in certain regions of the egg preparatory to the develoj>ment of organs in those regions. How far this accumulation means that previously existing differential materials are segregated, and how far it means that diflFerences are arising in the egg materials, is the point that I shall here consider.

We owe to E. P. Lyon the introduction of a new method into embryology by means of which we shift at will many of the substances contained in the egg, and in consequence analyze further their function in embryonic development. The experiments have already yielded most important results, and I venture to prophesy that we are only at the beginning of further discoveries.

If the visible substances of the protoplasm — inclusions I may call them — ^act as the initiators of later diiferentiation, we might expect, if their distribution is altered, to find an extremely abnormal product.


D 156 Thomas H. Morgan.

If, on the other hand, the inclusions are not organ-forming materials, we should expect to find normal development after their displacement. These alternatives put in sharp contrast the extremes of expectation. As I shall point out later, they do not exclude other less radical interpretations.

If the egg, either before or after fertilization, is placed in the centrifuge and rotated at the rate of about 8000 revolutions per minute, a stratification of the materials takes place. The heaviest materials are driven to one end, and the lightest to the other. Even the nucleus will be carried through the protoplasm and pass to one pole of the egg. In most cases, it appears, that the eggs fall without regard to their polarity, hence the nucleus and the inclusions of the protoplasm may come to occupy positions with respect to the unmoved parts of the protoplasm that are entirely different from their normal relations.

If the egg is rotated more slowly but for a longer time, the same result is accomplished, and this method has in some cases a decided advantage. For instance, the egg immediately after its fertilization may be placed in the machine and kept rotating during its early development. In this way we can insure against the possible redistribution of the stratified materials.

I need not point out to you the wonderful delicacy of the centrifugal force, by means of which we can shift at will the contents of the egg without injuring the egg, as the sequel will show. The method, as I have said, opens a new era in experimental cytology.

I wish now to bring before you some of the results already obtained, and, if I speak more especially of those forms that I have studied, it is because I have here a knowledge of the facts at first hand.

The egg of the sea urchin Arbacia has given, I think, the most definite, positive results. The egg after centrifuging is shown in the first figures of the diagram (Fig. 1). Four distinct zones are present — a light whitish cap of oily matter, a middle perfectly clear protoplasm, a band or segment of yolk and a red pigmented base. The nucleus lies just beneath the oil cap.

The normal egg has a definite polarity which, as Boveri showed,

D The EflFects Produced by Centrifuging Eggs.


stands in relation to the point of fixation of the egg to the ovarian wall. This point is marked later by the attachment funnel that runs through the jelly-like outer membrane. It serves, therefore, to determine the original axis of the egg.

The centrifuged eggs show that the stratification stands in no constant relation to the egg axis (see first figure). In other words, the egg is centrifuged as it happens to fall in the tube.

The first cleavage begins, almost invariably, at the white cap where the nucleus lies and cuts through the egg, dividing the stratified materials at right angles (Fig. 2). The second plane is at right angles to the first (Fig. 3), lying near the line of separation of yolk and clear zone, and the third at right angles to both (not shown in figures). The fourth cleavage is a differential cleavage (Figs. 4, 5, 6). Four micromeres appear at one pole; the cells of the opposite hemisphere divide equally. This cleavage is a critical one, for the

D 158 Thomas H. Morgan.

micromes show in the normal egg where the digestive tract will develop. The mieromeres become the mesenchyme.

If the position of the mieromeres in the centrifuged egg is followed, it will be found that in these eggs also the archenteron develops at the micromere pole. Furthermore it has been determined that the mieromeres do not develop with respect to the secondary induced axis, but lie opposite to the attachment funnel. In other words, while the first three cleavages come in with respect to the stratification, the fourth or differential cleavage comes in with respect to the original egg-axis (Figs. 4, 5, 6).

Around the micromere pole as a center the development of the embryo takes place. A perfectly normal pluteus is formed. In some of these plutei the red pigment is contained in one region ; in others, in other regions. Similarly for the yolk.

The results prove triumphantly that the materials that centrifuge in the egg of the sea urchin are not organ-forming.

Let us turn now to another egg, that of the mollusc, Cumingia. The former egg, that of the sea urchin, has at first a less determinative type than the mollusc; hence the importance of taking up a different form.

Three substances appear after centrifuging, the yolk at one pole, the pigment at the other, and a clear zone between.

The first cleavage is shown in the following figures. (Figs. 7, 8, 9.) It will be noted that the first cleavage plane pays not the least attention to the distribution of the materials. All of the yolk may be in the small cell, or the small cell may contain all of the pigment. This difference of results may be due to the fact that in the sea urchin the nucleus is driven to one pole, while in the mollusc's egg the polar spindle has developed when the egg is laid. The centrifuge is unable to change the position of the spindle, although it can change readily the resulting nucleus. I suspect, however, that the difference is more profound and that in Cumingia the first cleavage plane is a differential one, like the fourth plane in Arbacia. But note, it is differential not with respect to the inclusions of the egg but with respect to the polarity of the egg.

For some reason unknown to me the eggs of Cumingia do not

D The Effects Produced by Centrifuging Eggs.


develop well in dishes as far as the bivalve stage, despite the fact that the cleavage and early development is quite normal. I have had great difficulty in obtaining embryos, and this applies equally to normal and to centrifuged eggs. Whenever I have obtained later stages in the one, however, I have also obtained them in the other, so that after a prolonged study I can now state that the centrifuged eggs of Cumingia are also capable of producing normal embryos.

These results show, I think, that the visible material of these eggs, such as yolk, oil, pigment and perhaps other granules, are not necessarily organ-forming or even organ-determining.

I should, however, give you a wrong impression if I left the

7 ' . 8 g

matter here. In other eggs it has been shown that the centrifuge acts injuriously. In the fertilized frog's egg, for example, I found that if centrifuging were carried on for too long a time, abnormal development takes place. Just what happens is not clear. That the disturbance of the materials may interfere with normal movements in the protoplasm essential for development is quite possible. I could give examples of such conditions. It may also be possible that some of the inclusions are necessary for the cells, as food, for instance. Their removal from the cells might, therefore, act injuriously. It seems almost a foregone conclusion that even if the young stages of centrifuged eggs are normal, later stages may become abnormal or fail to be attained, for with all the yolk in the ectoderm

D 160 Thomas H. Morgan/

and none in the digestive tract we should expect abnormal nutrition. A.gain the nucleus may be carried so far from a position essential to it when the differential cleavage occurs that it fails to reach its proper location at the proper time. I could give evidence to show that displaced nuclei do endeavor to reach the proper place for a differential cleavage.

All of these conditions, and probably more, must ultimately receive careful attention. Failure of a centrifuged egg to develop normally may be owing to any one of them, but the positive results show, I think, with all clearness that the visible inclusions in the cytoplasm of the kinds here referred to are not organ-forming substances ; their role in development is of secondary importance. Behind them lies an organization that is the chief director of the series of events that characterize development. The visible materials of the egg follow and do not lead in the development.

Both in the case of the sea urchin and of the mollusc I have spoken of certain cleavages being differential. You may fairly ask for an explanation of my meaning ; whether nuclear or cytoplasmic ; and if the latter how differential if not a separation of different materials.

Here it seems to me is the really vital question that remains to be settled by further study. The evidence we have does not justify us, I think, in supposing such a cleavage to be the result of special nuclear interference; nor do I think it due to the segregation of already existing preformed materials. I am inclined to adopt the view that a differential cleavage is one in which a physical condition is reached that marks a further step in development. The division is differential not in the sense of separating things previously mixed, but it is a creative phase by which something new is produced by the bioplasm. It occurs at the first division in the mollusc, but at the fourth in the sea urchin. The same kind of change may even go on before any division has occurred. The results here described for the egg of two species does not, of course, preclude the possibility that in other eggs a change similar to that which occurs at a differential cleavage may occur in the egg prior even to the first cleavage and after the ripening period. The results that Conklin has obtained

D The Effects Produced by Centrifuging Eggs. 161

in the egg of Ascidian, Cynthia, fall under this class. If, as I suppose, the early differentiation of different regions of the bioplasm is a physical fact of the living substance, such a change might readily take place before the first cleavage. The results of isolating blastomeres show definitely that removal of parts of the egg, after a differential change has taken place, seriously affects in many cases the subsequent development, and if the centrifuge is capable of displacing such differential areas or of interfering with their formation, or of making difficult the subsequent development by filling them with indifferent materials, Abnormalities may take place.

It has been shown, especially by Conklin, that extensive movements of materials take place before the first cleavage and also in subsequent cleavages. Such movements seem to be connected with the formation of organ-forming regions. Whether such movements mean that differential materials already developed are moved to their definitive locations, or whether the movements are themselves the expression of differential changes taking place, or whether such movements are simply the outcome of karyokinetic movements, we do not know positively.

Perhaps I should add, in order to prevent misunderstanding, that, although I dispute the view that the visible substances in the sea urchin's egg acted upon by the* centrifuge are organ-forming, I hold that development is the outcome of physical changes in the egg. The materials that characterize the different structures and organs of the body are the end product of changes that the egg-substance undergoes in its development. Differentiation is a product of the activity of the egg, the egg itself before cleavage is not the sum total of those materials that characterize the organs of the body.





CHARLES RUSSELL BARDEEN. University of Wisconsin.

In experiments with living organisms, it is necessary to take into* account variations in internal conditions and in the environment, as well as the nature of the specific influence brought to bear upon the organisms. Thus toad or frog ova are much more readily injured when over-ripe at the time of fertilization than when fertilized at the normal period, and, as a rule, are hardier in moderately cool weather. than in hot. Owing to variable factors of this kind, the results of exposure of sex-cells to x-rays of a given intensity for a given period of time are not uniform in detail in a series of experiments, although the general results are fairly uniform. There is, moreover, great individual variability in susceptibility in any given lot of sex-cells, so that the percentage of organisms affected as well as the extent of the abnormalities of development of the organisms affected must be taken into account in. estimating the effects of the rays.

Exposure of females, in which the eggs are still ovarian, will prevent the ripening of the eggs. The eggs remain indefinitely in the ovaries and are not capable of artificial fertilization. Exposure of the sperm or of ripe ova to intense x-rays for a considerable period of time seems not markedly to affect the power of fertilization. The earlier stages of cleavage in fertilized ova, when one or both sex-cells have previously been exposed, are apparently nearly normal, but if the exposure has been sufficiently severe, abnormality in development of a considerable percentage of eggs appears at the time of gastrulation. Gastrulation may be markedly interfered with, being either interrupted at an early period or so modified that spina-bifida or similar abnormalities result. If the effect is less


D 1G4 Charles R Bardeen.

severe, the abnormalities begin to appear at the time of the differentiation of the alimentary canal and the neural tube, either or both of which may be abnormally formed. When the action of the rays is less marked, the abnormalities appear in one or more parts of the organism at a still later period. The vascular system may fail to develop, or various abnormalities may appear in any of the newly-forming organs. When the effects are slight, the deformities may not appear until the larvae are well advanced and then merely in a small percentage of the larvae. Thus, for instance, a hind leg may fail to develop in some individuals.

Exposure of a fertile female toad for one hour and the subsequent fertilization of the eggs with normal sperm caused in my experiments most of the eggs to develop marked abnormalities, many of them of the spina-bifida type. Nevertheless, a few individuals developed into normal tadpoles. When a female frog was exposed for an hour and fifteen minutes, and the eggs were subsequently fertilized, at the end of seven days all the ova were abnormal in form. Exposure of spermatozoa for a corresponding period leads to essentially similar results. In one experiment with toads in which the sperm was exposed for one hour, about 88 per cent of normal ova fertilized by the sperm developed into tadpoles abnormal in form, while, when the sperm was exposed one hour and a quarter, all the ova fertilized, in one instance over 250 specimens, were abnormal at the end of five days. In one experiment in which toad sperm was exposed for one-half hour to the x-rays, only 1.2 per cent of the eggs developed normally, but in other instances the percentage was greater than this.

Several experiments were made to test the action of the x-rays on fertilized ova at different stages of development. In one series of experiments, from a single female toad eggs were removed, fertilized with fresh sperm and then exposed at successive intervals for one-half hour to the x-rays. Of those exposed during the first half hour after fertilization, 22.3 per cent were undergoing apparently normal development at the end of two weeks. During this^ period of exposure the second polar body is given off. Of those exposed during the second half hour, 17.5 per cent were apparently normal

D Susceptibility of Amphibian Ova. to X-Eays. 165

in development at the end of two weeks.. Of those exposed during the third half hour, 68 per cent were normal at the end of two weeks. During this period of exposure the male and female pronuclei were approaching one another. Of those exposed during the fourth half hour, but 4.8 per cent were normal in appearance at the end of two weeks. None of those exposed for half an hour during the subsequent stages of cleavage up to the 64th to 128th cell stage developed normally. In later cleavage stages, however, exposure had a less marked eifect. Thus of eggs exposed thirteen hours after fertilization for half an hour, at this time in a state of advanced cleavage, 64.2 per cent were apparently normal in development at the end of two weeks. Eggs exposed between thirteen and twentyfour hours after fertilization, for half an hour at a time, varied in the percentage of those developing normally from 60 to 80 per cent. Eggs exposed for one-half hour periods after the first twenty-five hours, that is after the blastopore was closed, practically all developed normally. Exposure of the eggs after closure of the blastopore and of young larvae for several hours to the x-rays failed to produce any noticeable abnormalities of form.

In other series of experiments a considerably smaller percentage than in the series cited of eggs exposed for one-half hour during the first hour and a half after fertilization developed normally, but otherwise the results corresponded with those given.

These experiments show conclusively that both the male and the female sex-cells may be so altered by the x-rays as to give rise to the formation of monstrous forms. The susceptibility of the male and female sex-cells is approximately equal, although the abnormalities appear earlier in development and are greater when the ova are exposed. After fertilization until cleavage begins, the ova at first appear to be no more susceptible than the sex-cells before fertilization. During the earlier stages of cleavage the susceptibility of the eggs to the x-rays is markedly increased, but during the later stages of cleavage before closure of the blastopore the susceptibility of the eggs becomes much less, and after the blastopore is closed the power of the x-rays to influence development becomes strikingly reduced. The period of greatest susceptibility is the period during which there is the most rapid production of nuclear material.



CHARLES R. STOCKARD. Cornell University Medical College, yew York City,

The differenoes which exist between two animal species are doubtless in some way associated with the different chemical compositions of the eggs from which the species arise. Whenever the chemical complex of an egg is so disturbed that it cannot be readjusted, the animal which arises is abnormal. Theoretically the abnormality should be definitely the same under identical changes of composition. It might be expected that each chemical change induced within a giv6n variety of egg would result in a characteristic embryo. However, development is so difficult to analyze in detail, and the egg composition is so complex that until now we have only succeeded in recognizing a few specific abnormalities as results of definite chemical actions. Among vertebrate embyros only one monstrosity, known to occur in nature, has been consistently produced by the use of a given chemical substance. Other known defects, such as spinabifida, cauda-bifida, etc., often occur in embryos resulting from eggs treated with metallic salts, but the occurrence of these abnormalities is irregular and uncertain so that the experimenter can never predict with confidence his result.

On the other hand, in the action of the metallic ion Mg on the eggs of the fish, Fundulus heteroclitus, the case is very different. Here the experimenter may predict with certainty that a given strength solution of MgClj or Mg (N08)2 will cause a considerable percentage of embryos to present the cyclopean defect. The oneeyed monsters occur in 50 per cent of the eggs when a favorable strength solution is employed. All of the eggs do not give rise to


D 168

Charles R. Stockard.

W 3

Fig. 1. — A normal free-swimmiug Fundulus embryo.

Fig. 2. — An Individual from the magnesium solutions with its two eyes approximated.

Figs. 3, 4 and 5. — Three views of a typical cyclopean monster from the magnesium solutions. The single eye is antero-median in position and causes the mouth to project ventrally as a proboscis-like organ. M, mouth; ys, yolk-sac

D The Artificial Production of One-Eyed Monsters. 169

Cyclopean individuals merely because they are not all affected by the same delicate point of concentration in the solutions. The more hardy eggs would require stronger solutions to produce a response while weak eggs often die in a strength of MgCl2 which causes cyclopia in eggs of the average powers of resistance.

The cydopean fish embryos are in all respects exactly comparable to the human cyclops. One eye exists in the middle of the face and the nasal pits are often represented by a single or double pit in front of the eye. The eye conditions show all steps in a series beginning with two eyes unusually close together, two approximated eyes, a double eye with two lenses, two pupils, etc., a laterally broad eye with a double retinal arrangement, and a single lens and pupil and a typically single eye showing no indications of its double nature. The lattei: condition may be termed typical or perfect cyclopia, from this we pass to extreme cyclopean eyes which are unusually small, sometimes deeply buried in the head, others with small optic cups and illfitting lenses which protrude beyond the eye, and, finally, all retinal or optic cup portions of the eye may be absent, with independent lenses present, or both optic cup and lens may fail to form and eyeless creatures result. Many illustrations of all these stages have been found and studied among almost three hundred cyclopean fish, which were produced during the past summer.

Some of the embryos present perfectly normal bilateral brains and show no abnormality other than the cyclopean eye and characteristic probocis-like mouth. Many of the monsters hatched in the usual manner and swam normally, giving every evidence of being able to see perfectly by means of the single anterior eye. The bodies of others were slightly twisted or bent aftet hatching and these swam rather abnormally, although their vision seemed unimpaired. Figs. 1 to 6 represent the free swimming fish. Fig. 1 shows a normal individual, below this is a cyclopean monster with its eyes approximated and, finally, a dorsal, lateral and ventral view of a typical one-eyed cyclops is shown by Figs. 3, 4 and 5.

The development of the cyclopean eye in human monsters has been difiicult to interpret on account of the scarcity of material and want of early stages in the defect. Such abnormalities are not easy to

D 170 . Charles E. Stockard.

interpret from later stages. In the summer of 1906, when these monstrous fish were first produced, I studied only later stages and on finding all degrees of doubleness in the eyes concluded that the Cyclopean condition resulted from a more or less intimate fusion of the two eye components after they had arisen from the brain. This position has been held by other workers both before and since my study. A more careful investigation, however, of the earliest stages of cyclopia in the living eggs and in sections shows' that the final condition of the eye is foreshadowed in the first appearance of the optic anlage from the brain. The early eye is either perfectly single or double from the start, and the union of the two components does not become more intimate during development, even though the eye may develop partially within the brain itself.

An incomplete diprosopus monster with three eyes and one additional lens extends the series of eye monstrosities in the fish to the opposite side of the usual two-eyed condition. This monster has a pair of eyes arising from one head component, and a single eye from the outer side of the other head, while on the inner side of this head there is entire absence of an optic cup, although a perfectly differentiated lens exists.

A new type of eye monstrosity occurred with regularity in the Mg solutions. These individuals had one perfect eye of the normal pair in its usual position, while the other eye was either abnormally small or represented by a solid mass closely applied to the brain, or still in other cases all evidence of the optic cup parts was absent and a free lens occurred on that side, or, finally, both optic cup and lens were wanting on one side of the head.

Again as in the cyclopean defect, the embryo will hatch with its eyes in dissimilar conditions comparable to the state of things shown when they first become differentiated from the brain. The difference in size of the two eyes is not overcome during development, and, on the other hand, there is no further degeneration of the small eye. I have termed these individuals "Monstrum Monopthalmicum Asymmetricum," the monster with one asymmetrical eye, as distinguished from the cyclopean monster with the single eye in the middle of its face. These unequal eyes may possibly result from an

D The Artificial Production of One-Eyed Monsters. 171

unequal allotment of eye material to one side or the other, or the entire anlage might by chance occur on one side. In a sense this would be lateral cyclopia.

A point of particular interest to experimental anatomists is the frequent independent occurrence of crystalline lenses in these embryos. Experiments on frog embryos in which the optic cup was cut away or transplanted have seemed to show that the lens arises from the ectoderm as a result of a stimulus received from the optic cup ; and further it has been shown that in some amphibians the lens is incapable of self-differentiation after it has arisen unless the optic cup stimulus continues. The cyclops fish is entirely out of accord with such conclusions. Here we find lenses arising from the ectoderm in embryos which lack optic cups entirely, or in others where the optic cups are far removed from direct contact with the lens.The lens, then, arises independently of the optic cup stimulus, and it is also evident that after its formation it continues to self-differentiate and gives rise to lens fibres and, finally, results in a clear transparent body perfectly invisible in the living embryo but clearly demonstrated in sections.

Thus it is shovni that by changing the chemical environment of an egg the experimenter may produce conditions similar to those which have been tested by mechanical operations. He has the advantage of knowing that no tissues have been cut away or mechanically destroyed.

We may now ask what causes these one-eyed monsters? Many theories have been advanced to account for cyclopia. The French observer Dareste attributed the condition to a closed brain or the failure of the anterior vesicule to develop, thus allowing the parties retiniennes to come off in approximation. This idea is opposed by the fact that Spemann has found cyclopia to occur in Triton embryos while the brain tube was hollow. Peculiarly enough I find that the optic-outpushings in Fundulus are normally given off while the brain is yet solid. Thus according to Dareste all of these fish would be cyclopean in nature.

The single brain condition or the failure of the fore-brain to develop is not a definite cause of cyclopia since such conditions are

D 172 Charles R Stockard.

not always accompanied by the defect, and, on the other hand, many of the Cyclopean fish have perfectly developed bilaterally lobed forebrains.

Spemann thinks that in his Triton cyclops the anlagen of certain tissues are lost, consequently these parts never begin development, and organs situated lateral to them develop in contact from the start. The cutting experiments in which tissue between the eyes is destroyed seems to support Spemann's view.

This explanation, however, does not seem to fit all cases. In the "Magnesium Embryos why should tissue between the eyes fail to form and no other tissues ? why are the nasal pits sometimes united and sometimes separate in cyclops? A close microscopical examination of the brain floor in cyclopean and two-eyed embryos shows no absence of recognizable parts in the former. The monstra monopthalmica asymmetrica are to be explained. Are these due to the absence of the early anlage of one eye ?

The various degrees of cyclopia, the imperfect formation or absence of one eye and entire absence of eyes are conditions common to the magnesium solutions and very rare or never occurring in other solutions or in the hundreds of eggs observed developing in seawater. These conditions, probably, have a common cause, and I suggest hypothetically that this cause is an inhibitory or anaesthetic effect of the magnesium on the process of out-pushing and separation of the optic vesicles. Magnesium is known to exert a decidedly anaesthetic effect upon both vertebrate and invertebrate animals and is an inhibitor of muscular activity. It might possibly inhibit the giving off of the optic vesicles or prevent their early separation in the brain, so that both might come off together as in cyclopia. It is, of course, necessary to find a definite point in the strength of the solutions in order to obtain the proper amount of inhibition.

Finally, these experiments show that an egg which had begun development normally and which would have given rise to a twoeyed individual may, at the will of the experimenter, be caused to produce a cyclops monster. Thus the germinal theories of cyclopia are shown to be unnecessary as explanations of its cause ; since here

D The Artificial Production of One-Eyed Monsters. 173

it is undoubtedly due to the influence of external conditions acting upon the egg substance.

The suggestion is evident, though highly speculative, that cyclopia in man and other mammals might be due to a similar chemical cause, an excess of magnesium salts in either the mother's blood or the amniotic fluid surrounding the developing embryo.



WARREN H. LEWIS. Johns Hopkins University,

In a series of experiments on localization and regeneration in the fish embryo it was noted that defects, made in the anterior end of the embryonic shield, often gave rise to cyclopean forms. With the renewed interest in the subject since the publication of the papers by Spemann and Stockard, it seems desirable to indicate this meclianical method by which these forms were produced and to enter somewhat upon the bearing they may have upon the theories of the origin of Cyclopia. Physicians have been interested in Cyclopean monsters for centuries and many theories concerning their origin have been advanced. Two groups of theories are to be recognized, the germinal and the enviromental. Concerning the germinal origin there is no direct or indirect proof such as might be obtained were cyclopean monsters viable and capable of sexual maturity and subsequent reproduction with the possibility of transmitting to their offspring the same peculiarity, through peculiarities in the germplasm. The very fact that this chance of transmission is eliminated would speak against the germinal origin. There is considerable evidence in favor of the second view, that Cyclopia are due to some modification of the embryo during the early stages of development. The experiments of Spemann and Stockard and likewise of the ones recorded below seem to support this view. Spemann's^ experiments consisted in constricting the eggs of triton during the two cell or later stages by a fine thread about the circumference of the first cleavage plane. Double headed monsters, exhibiting^ varying degrees of fusion of

'Ueber experimentell erzeugte Doppelblldungen mlt cyclopischem Defekt. Zool. Jahrbuch., Supp. VII, 1904.


D 176

Warren H. Lewis.

Figs. 1, 2. — Showing embryonic shield at operation stages. The black area shows usual amount of tissue lost by the operation.

Fios. 3, 4, 5. — Dorsal, lateral and ventral views of normal head shortly after hatching.

Fig. 6. — Experiment ha,. Operation on stage 2 (see Fig. 2).

Figs. 7, 8, 9. — Experiment hai«. Operation on stage 2. Right eye completely absent.

Figs. 10, 11, 12. — Experiment hai. Operation on stage 2. Left eye entirely absent.

D The Experimental Production of Cyclopia. 177

two or more eyes were often produced. This fusion of the eyes was probably primary before the eye rudiments were recognizable. In Stockard's^ experiments on fish embryos (Fundulus heteroclitus), the eggs shortly after fertilization were placed in sea water solution of magnesium chloride with resulting production of the cyclopean condition in a large percentage of the eggs. The cyclopean condition is also primary in these experiments, "the earliest indication of an eye is just as truly cyclopean as it will be later," Stockard.^ Neither Spemann's nor Stockard's cyclopean monsters can be looked upon as germinal in origin but are truly due to environmental conditions. The additional data from the following experiments extends the possibility of cyclopean monsters depending upon abnormal influences exerted during early embryonic development. These experiments were done at the Marine Biological Laboratory, Woods Hole, Mass., and a few of the typical ones are given below. Such forms were easily reproduced during two succeeding seasons.

The experiments were made on the eggs of Fundulus heteroclitus during the embryonic shield stage. The egg was held with a small pair of forceps and a very fine needle was thrust through the egg membrane into the anterior end of the shield ; as the needle was withdrawn slight pressure on the forceps caused some of the material of the embryonic shield in the region of the needle prick to be extruded. As the experiments were done under the binocular microscope, it was possible to determine with some degree of accuracy about how much material had escaped. This is indicated in Figs. 1 and 2 by the solid black patch at the anterior end of the embryonic shield. The amount of material extruded varied somewhat in each experiment, the variations becoming more evident during the later stages of development. After the embryonic shield begins to appear, there is very little or no regeneration of the central nervous system and defects caused at this time consequently become more and more apparent as development proceeds.

"The artificial production of a single median cyclopean eye in tlie fish embryo by means of Sea Water Solutions of Magnesium Chloride. Arch. f. Entwlcklungsmechanlk der Organismen, Bd. 23, 1007.

•Science, Vol. XXVIII, p. 455.

D 178

Warren H. Lewis.



Figs. 13, 14, 15.— Experiment lim«. Operation on stage 1 (see Fig. 1). Byes in contact in median plane.

Figs. 16, 17, 18.— Experiment h,^ Operation on stage 1. Eyes fused, with two lenses and two cup cavities.

Fios. 19, 20, 21. — Experiment hai,. Operation on stage 2. Eyes fused with two lenses and one large cup cavity.

Fios. 22, 23, 24. — Experiment ha„. Operation on stage 2. Eyes completely fused with one lens and one cup cavity.

D The Experimental Production of Cyclopia. 179

In a number of instances the material taken out with the needle point was from one side of the anterior end of the embryonic shield with the resulting absence at the time of hatching of the eye on that sida Fig. 6 shows the head of such an embryo, which was operated on at the stage shown in Fig. 2 and killed a few days after hatching, 15 days after the operation and 17 days after fertilization. The right eye consists merely of a small bit of retina remaining in the otherwise almost normal brain wall. The left eye is apparently normal as are also the brain and nasal pits. Figs. 7, 8, 9 are from another embryo operated on at the same time and stage as the one shown in Fig. 6, and killed 15 days after the operation. The sections show complete absence of the right eye, but with an otherwise normal brain and head. Figs. 10, 11, and 12 are from another embryo operated on at the same time and stage and killed 15 days later. In this embryo the left eye is entirely wanting, the forebrain is slightly reduced in size and the nasal pits are quite close together. The right eye lies nearer the median plane than normal. See Figs. 3, 4, and 5.

Figs. 13, 14 and 15 are from an embryo in which the operation defect was about medial and done at a stage such as seen in Fig. 1. Fifteen days after the operation the embryo was pulled out of the membrane and killed. The two eyes are in contact, but each one is surrounded even at the place of contact with its own pigment layer. Two optic nerves are present and two lenses. The two nasal pits are in contact, but the brain is apparently about normal in size. Figs. 16, 17 and 18, from an embryo operated upon as above, show a median cyclopean eye in which the pigment layer is wanting between the two components. The two cup cavities are also slightly reduced in size. This fusion of the eye rudiments in the median line has taken place in such a manner as to separate the cranial from the facial portions at the anterior end of the head, contrast Fig. 17 with a similar view of the normal head, Fig. 4. Figs. 19, 20 and 21 show a somewhat similar cyclopean eye. The operation was done at the stage shown in Fig. 2, and the embryo killed 15 days later. The sections show a common cup cavity, the retinal and pigment layers forming a continuous wall about the cavity. There are two

D 180 Warren H. Lewis.

lenses and two pupils, however. The lenses are in contact. There are also two distinct optic nerves. The brain is reduced in size and the eye separates it from the mouth region.

Figs. 22, 23 and 24 are from an embryo operated upon at a stage shown in Fig. 2 and killed 15 days later. Here is a single median Cyclopean eye with one pupil and one lens and one cup cavity. A slight median notch on the anterior side of the optic cup indicates its origin from portions of two eye rudiments. The large optic cup shows in sections a very beautiful median eye with complete continuity of the layer* of the retinas of the two components about a single large cup cavity and a single lens. The two nasal pits are in contact and lie dorsal to the eye. The brain is somewhat reduced, and its anterior end separated widely from the mouth region by the medially placed eye.

The explanation of the formation of these various abnormalities is in a way a comparatively simple one, if we assume that already in the early embryonic shield stage the various parts of the central nervous system and the eyes are, probably, already predetermined, and secondly that there is very little or no power of regeneration in this tissue. Numerous experiments on regeneration indicate very clearly that there is very little or no regeneration of the tissue (at least that of the central nervous system) extruded during the operation. The repair, taking place after the operation, consists merely of a rapid closing together of the parts left behind, and thus, a healing of the wound occurs without regeneration of lost parts. This closing of the wound is accomplished in a few minutes, and rudiments are thus brought into contact that normally are quite widely separated, those of the two eyes, for example. The subsequent differentiation adjusts itself to the new relations of these rudiments with the resulting abnormal forms. Thus as one examines these developing embryos, from the very first time the eye rudiments are visible in the living specimen imder the binocular microscope, they appear to have the same amount of fusion or loss of an eye that is clearly to be found in the same individual at later stages and at the time of hatching. So we can explain these cyclopean forms through a fusion of the rudiments of the two eyes immediately after the

D The Experimental Production of Cyclopia. 181

operation, even though at this time no rudiments are visible. Differentiation of the eye tissue evidently occurs sometime before it becomes visible by our crude microscopic methods.

Thus cyclopia in man can be explained through the influence of external factors acting during early stages of development in such a manner as to produce a single eye rudiment, and we need not seek for a germinal explanation. In these experiments on fish embryos, the eye rudiments were brought into contact and fused soon after the operation determining at this time the end result There was not the formation of two eyes and then their subsequent fusion into a single median eye. It seems likely that in man similar early fusion of the eye rudiments must take place to produce cyclopia. These experiments throw no light, of course, on the cause of the early defect in man, although Stockard's experiments indicate that chemical factors might be responsible for such defective or altered early development.

The great similarity between these cyclo])ean forms and those produced by Stockard suggests that the MgClg may have in some manner prevented the growth of certain cells at the anterior end of the embryonic shield during the embryonic shield stage. It is possible that the MgCU solution might have the same effect on eggs subjected to its influence during and just preceding the formation of the embryonic shield.





ELIOT R. CLARK. From the Anatatnical Laboratory of the Johns Hopkins University,

Secent studies have led to a distinct advance in our knowledge of the lymphatic system. Yet there are many points in which this knowledge is still incomplete. The observations here recorded seem to have a bearing upon some of these points, and hence seem worthy of presentation to the Association.

The studies were made on living frog larvae ; the species used were Eana sylvatica, E. palustris and E. catesbiana. Two devices were employed, both of which were essential to the success of the observations, an upright chamber and chloretone anesthesia. The former permits observations to be carried on with the larva in its normal upright position. Chloretone serves to keep the larva motionless while under observation. The use of chloretone introduces slightly abnormal conditions. Yet by alternating the periods of anesthesia with return to fresh water, the same larva may be kept under observation for several hours daily, for three or four weeks. During this time growth continues, though somewhat retarded.

In order to preserve accurate records of the various stages, drawings were made both with and without the aid of the camera lucida. The micrometer eye-piece was employed in making measurements. With the assistance of careful records, little difficulty was experienced in finding the same structures — ^blood-vessel, lymphatic, even connective-tissue cell — in successive observations.

The fin expansion of the tail of the frog larva, in early stages, is rather opaque, owing to the j)resence of pigment and yolk pjanules. During this period it is possible to distinguish the course of the blood-vessels only by the moving blood-corpuscles. Gradually this opacity diminishes, until eventually, at lengths which vary for dif (183)

D 184 Eliot E. Clark.

ferent species, the tail becomes beautifully transparent. Now the walls of blood-vessels, individual connective-tissue cells, nerves, wandering cells and lymphatics may be readily distinguished, furnishing a picture of rare simplicity and beauty. Little wonder that it has been selected so often as a field for the study of elementary problems in anatomy ! From this period an accurate record may be kept of the changes which take place, until the picture is again somewhat clouded by the development of pigment cells and by the increase in thickness of the tissues.

The blood-vessels first demand a short description. When the tail becomes transparent, they form anastomosing loops in both dorsal and ventral fins, extending outward from the edge of the axial mass. The limit of the vascular area is roughly parallel to the free edge of the fin and leaves a wide non-vascular area. Sprouts are sent out forming secondary plexuses which gradually approach the edge of the fin, so as to give a festooned appearance. Hand in hand with this new growth there is a general expansion of the tissue with an increase in the distance between neighboring capillaries. Accompanying these processes, adaptive changes take place in the older capillary mesh-work. Here some of the capillaries increase in size to form arterioles and venules, while others disappear. These disappearing vessels may be readily observed in all stages. The sequence of events in a single degenerating vessel is usually a stasis, so that a cell in the lumen remains in equilibrium — a narrowing so that no other cells enter — ^the appearance of a solid portion, which extends until the whole vessel becomes a solid Qord — the breaking of this cord and gradual shortening of the two ends until the only vestige may be a slight swelling on the wall of the blood-vessel.

In no instance during these observations has there been seen a lumen-containing portion of a blood-vessel isolated from the. actively functioning vessels. The relation l)c^twoen blood-vessel and lymphatic will be mentioned later.

The lymphatics in the tail of the frog larva have often been studied, by Kolliker^ in 184(), then independently by Remak* in

'KoHlker, Annal. d. Sc. Natur., 1846.

'Remak: Muller's Arch. f. Aiiat., Phys., u. wissensch. Med., 1850.

D Lymphatics in Tail of Frog Larva.


1850 and subsequently by numerous investigators, in the live animal, in fixed and in injected specimens. In our work, lymphatics were noticed during observations on the blood-vessels, as narrow, irregular, double-contoured structures with free ending toward the margin of the fin and traceable to the edge of the axial mass. They

Fig. 1. — Three successive stages of growing lymphatics and blood-vessels In dorsal fin of Raua catesbiana larva. X 30.

Lymphatics in solid black, blood-vessels in lines. Vessels near tip are omitted. A,A', A", etc., indicate same vessels in different stages. M., muscle edge; N., notochord.

appeared independent of blood-vessels. In order to be quite certain, injection was resorted to. With the aid of Dr. Knower's^ delicate injecting apparatus, a capillary glass tube with glass bulb to furnish pressure by air expansion, it was found possible to fill with India ink the vessels in question, quite independently of the blood-vessels.

•Knower: Anat. Rec, Vol. II, No. 5, Aug., 11X>S.

D 186 Eliot R. Clark.

When the fin first becomes transparent, the only lymphatics to be seen are sprouts, some branched and some unbranched, often anastomosing with one another, extending out from under the cover* of the axial musculature. They are of varying lengths, and in number correspond somewhat irregularly to muscle segments. Their tips stop considerably short of the limit of the blood-vascular area, 80 that we have three areas: a peripheral non-blood-vascular, nonlymphatic area, an intermediate blood-vascular non-lymphatic area, and a proximal area containing both blood-vessels and lymphatics, next the axial mass. As growth proceeds the lymphatics rapidly encroach upon the blood-vascular, non-lymphatic area, while both together grow into the non-blood-vascular area, until eventually the two systems are practically coextensive, reaching in older larvae nearly to the fin border. During growth the primary anastomoses noted above disappear in part. Later there are gradually formed secondary anastomoses between neighboring capillaries. (Fig. 1.)

In a minute study of the growing capillary several questions at once present themselves. Does the lymphatic capillary grow out independently? What is its relation to blood-vessels, connectivetissue cells, wandering cells ? Can we gain any clue as to the factors underlying the periphereal extension of Irayphatics — ^what is the stimulus ?

If a single lymphatic capillary is selected for minute observation, it is found to present a characteristic appearance well described by KoUiker and Remak and well figured in KoUiker's Grewebelehre.* The wall is of irregular thickness; most of it is extremely delicate, while at intervals are nuclear thickenings. In the earliest observable stage there are small globules in this nuclear area (yolk?), which soon disappear leaving a granular appearance. From the walls extend numerous fine pointed projections, at various intervals, and of varying lengths. The diameter of the lumen of the lymphatic is considerably less than that of the blood-capillary. The lumen always extends beyond the last nuclear thickening. The tip ends in one or more fine pointed processes, usually somewhat longer than

Von Ebner, A. Kolllker*s Gewebelehre, III, 1899.

D Lymphatics in Tail of Frog Larva.


the prooesses at the sides. Lito the bases of the larger of these processes, the lumen of the lymphatic may be followed for a short distance. The tip is never bulbous but rather pointed or angular.

When such a living tip is carefully studied, an extraordinary phenomenon is noted, for it is found that the appearance of the capillary is perpetually changing. So complex are these changes that they almost elude description. Most noticeable are the changes in the

Fig. 2. — Successive drawings of same lymphatic tip in tail of Rana catesbiana larva, which had been stained with neutral red intra vitam. X ^'^^• Nuch, nuclear thickening.

contour of the lymphatic. The fine pointed processes already noted are not constant, they are continually appearing and disappearing. They vary much in length, some form mere short blunt projections, while others reach a length of many micra with all intermediate gradations. They may appear at any portion of the wall, including the nuclear area. The longer processes are usually seen at and

D 188

Eliot E. Clark.

near the tip of the lymphatic, but not invariably — for occasionally long processes appear at the sides with those at the tip quite short. Just as they vary in length, so also they vary in the time of persistence. If careful drawings are made at ton or fifteen minute




Fig. 3. — Same larva as in Fig. 2. Note shiftings of protoplasm of (n) nuclear thickenings as indicated by pigment granules.

intervals, in no two tracings will the pattern be identical. In fact, one gains the impression that were it possible to photograph the lymphatic at one minute intervals, careful study would reveal definite changes in successive pictures. Moreover, it is to be noted that these

D Lymphatics in Tail of Frog Larva. 189

observations were made on larvae whose body processes were slowed by the use of chloretone. (Fig. 2.)

In addition to the imceasing change in- the contour there are to be seen changes and shiftings in the wall. The nuclear thickenings are perpetually changing both shape and position. Now they appear crescent-shaped with the convexity encroaching on the lumen, again they become much elongated. Often the nuclear area is accompanied 'by a large black pigment granule. The nuclear thickening may shift its position relative to this pigment granule — being now at its proximal, now at its distal side. If a larva is stained with a weak solution of neutral red, there is a red granular coloration in the area of nuclear thickening. Later, instead of the red, there are to be seen here numerous small black granules. If a nuclear area containing these granules is observed closely, there will be noted a continuous shifting of the granules; sometimes several become grouped together, and again they sej^arate. Two nuclear thickenings may appear in place of one, — -and again two may appear to be moulded into one. (Fig. 3.)

While these changes are taking place, there may be, on the part of the capillary as a whole, a definite increase, or, more rarely, a decrease in length. The successive changes concerned in the increase are as follows: From a main lymphatic may be sent out numerous fine pointed processes. One of these processes persists and grows longer. As it increases in length, the lumen follows farther and farther into its base, until there is formed a short delicate-walled tube with one or two pointed tips, without a nuclear thickening. As this increases, gradually, from the wall of the main lymphatic, there passes into the branch a nuclear thickening. This whole tube now becomes longer and longer, the nuclear area passing further and further from the main lymphatic, but always remains at a distance from the tip. The lumen always extends beyond this nuclear thickening. (Fig. 4.)

After a time there appears in the wall a second nuclear thickening, and again a third and a fourth. These new thickenings seem to arise by division of the pre-existing ones, though definite proof of this has not yet been adduced by staining selected stages. Branches form in aIC


Eliot R Clark.

MdU /6-l(-34d.m. fJldV lb'll:30p.Tn. Nldyn- 11-66 d.W.

Maij 18-1X36 p.m. Af^y i<f^ ;,..^^d.^. Maifn-imp,


Fio. 4. — Successive stages In growth of lymphatic capillary in tall of Rana palustrls larva. X 207. b.-v., blood-vessel ; lym., lymphatic ; n., nuclear thickening.

D Lymphatics in Tail of Frog Larva.


similar way. Anastomoses arise by the growing together of tips from neighboring capillaries or their branches, and appearance of a lumen in the solid connection thus formed. (Fig. 6.)

The shortening of a lymphatic capillary may be quite pronounced, for a branch which has attained the length of .2 mm. may be entirely withdrawn. The process of withdrawal has been studied less minutely than that of increase. It has been noted, however, that it takes


Fig. 5. — Three successive stages superimposed of lymphatics In tip of regenerating tall of Rana catesbiana larva. Camera luclda drawing, x 82. On July 21, about 4 mm. of tail of larva 12 mm. long was cut away. On August 2, regenerated part measured 3 mm.

place more slowly than does the increase ; that two nuclear thickenings may approach one another; that the fine processes are shorter and less numerous; and that the lymphatic undergoing regression often ends in a long solid thread. In one instance such a long process contained a narrow lumen separated from the main lumen by a solid portion.

■Growth processes, especially branch and anastomosis formations, are well seen in a regenerating tail, for here the changes are more rapid than in the normal tail.

D 192 Eliot R. Clark.

If we examine the growing lymphatic over a wide area, we find a continuous balance between neighboring sprouts. Of two or more sprouts which start in a new area, often only one increases while another may remain stationary or may be withdrawn.

Between neighboring sprouts which grow out to the fin margin the distance is fairly constant for any stage. As the larva grows, the tail expands in thickness and in length. The distance between neighboring lymphatic sprouts increases, and coincidently there arise branches and anastomoses. For a time these all lie in the same sagittal plane. Later, in the thicker portions next the axial mass, branches may be seen extending toward either surface of the tail. As the lymphatic grows longer, the lumen of the more central portions increases in size, and thus the capillary becomes converted into a larger duct.

The relation between the connective-tissue cells and the growing lymphatic was carefully studied. These cells, with their numerous branched processes form a richly anastomosing supporting network. Unfortunately their finest processes are invisible in the living animal, so that the relation of these finest fibrillae cannot here be determined. Of the visible processes, the smallest seem sometimes to extend to the lymphatic, yet whether they actually join the lymphatic cannot be decided beyond dispute. The larger of the visible processes appear quite certainly to be independent of the lymphatic capillaries. The main bodies of the cells are distinctly separate from the lymphatic. In a study of the growing lymphatic nothing may be seen which would even remotely suggest the bodily addition of one of these cells to the growing capillary.

Wandering cells are always present in the fin, usually well scattered. Xo connection was noted between them and the growing lymphatic.

The relation between lymphatic and blood-vessel was studied with great interest, for around this point many controversies have arisen. As noted above, when the tail first becomes transparent, the bloodvascular area extends beyond the lymphatic area. Later the two become practically coextensive. The growth on the part of each is an invasive one — as is readily seen by noting their relations to

D Lymphatics in Tail of Frog Larva. 193

selected fixed connective-tissue cells. During this invasion, however, the two remain totally separate.

Occasionally a lymphatic grows into an area in which no bloodcapillary is present and in which none has been present. When the two are invading the same area, no regular relationship is maintained, for the lymphatic now wins parallel to and now crosses at right or oblique angle the blood-capillary. The lymphatic pays no attention to the new blood-capillary sprout, sometimes passing near by, and sometimes at a distance. Nor is the lymphatic influenced by the presence of the degenerating blood-vessels previously described. The entire process of degeneration may take place with no lymphatic near. Even when a lymphatic is near a degenerating capillary, no . transfer of tissue may be detected. As no portion of a blood-vessel cut off from the main blood-vascular system has been observed, evidently there has been no opportunity for the growth of the lymphatic by appropriation of such a portion. Never have we observed an anastomosis between lymi)hatic and blood-capillary; or the direct passage of blood-cells from one .to the other. In brief, all our evidence favors the absolute independence of the two systems in their peripheral extension. We thus differ from S. Mayer,® who finds the lymphatics formed in part from the blood-vessels, ilayer quite certainly confused the true lymphatics with the constricted blood-vessels — a confusion easily possible when the larva is placed on its side beneath a cover-slip, and' the electric current ilsed* to assist the anesthesia. For it has often been shown that both electrical and mechanical stimuli cause constriction of blood-capillaries in the tail of the frog larva.

In quite late stages Dr. Knower has noted that the lymphatics near the axial mass are very close to the veins. Our observations in the living lymphatic have not yet been carried to this stage, but we have seen the same in the relation between the ventral caudal lymph trunk and the caudal vein, in cross-sections of the tail. So. close are these two that they seem to share in part the same wall.

•S. Mayer: SltziiiiKsber, d. kais. Akad. d. Wissoiisch.. Wien. Abt. 3. Bd. 01, 92, 1885.

D 194 Eliot R. Clark.

It is not surprising that such pictures have given rise to difficulties in the interpretation of cross-sections.

Thus far the changes in the lymphatics which seem concerned in growth have been considered, but there is another process which is going on simultaneously, perhaps inseparably connected with growth, perhaps the phenomenon at the basis of the growth, namely, functional activity.

By accident a few red blood cells were extruded from a new forming blood-capillary sprout into the extra-vascular tissue. To our astonishment two days later a lymphatic capillary had grown down to this group of cells and was seen taking them in one after another, without the agency of leucocytes. We had subsequently many opportunities to observe a repetition of this process, which deserves a careful description. The changes undergone by capillary and cell are extremely characteristic. From the lymphatic is sent out a fine pointed process, indistinguishable from the processes previously mentioned. This gradually extends to the blood cell. After coming in contact with the cell, the delicate tip is lost to view against the deeper colored blood cell and for a time the mode of procedure may only be observed by noting the changes in the shape of the blood cell. After the fine tip has been lost to view, there appears opposite the point of contact, a slight blunt projection from the blood cell. This projection gradually becomes longer until the cell is pear-shaped. The narrow portion is always paler than the remainder of the celL Gradually this nose-like projection becomes longer, until the cell assumes an elongated oval shape as if in a narrow passage. Soon there appears beyond the cell, the tip of the lymphatic, slightly dilated, ending in a point. Slowly the cell moves centralwards and as it advances it is usually preceded by a short constricted area of lymphatic, while the portion between the cell and the tip is dilated. Gradually the cell passes along the lymphatic to the main caudal trunk, along which it advances with a steady uninterrupted motion, as if borne on by a definite stream. In one instance in which the cell entered the lymphatic quite rapidly, the tip could be seen for about a half minute distinctly open. (Figs. 6 and 7.)

If we arrange in order of intensity the reactions on the part of the



11=05 dm. uzoA.m uHOd.'m. ii-ss d.^

FiQ. 6. — Successive stages in taking up of red blood cell by lymphatic capillary In tall of larva of Rana catesbiana. x 163. A before, B during, after the process of taking up. r, r', two red blood cells, one of which (r') is removed, the other (r) left, lyin., lymphatic; b.-v., blood-vessel.

D 196

Eliot R. Clark.

lymphatic capillary as a whole to the stimulus caused by the presence of red blood cells in the tissue, we find the following series : If a single red blood cell is extruded, the lymphatic near by stnds out a fine process, which takes in the cell and then disappears. If several

TiucL \ ^

f ce

(30 m.)


(3f m.)



Z:¥8dLJn, on f3Tn.)

Z'SCdM (^m.)

IZ:36 v.m

3;Q7d.m. (in . 3z m.)



Fig. 7. — Successive stages in taking up of red blood cells by lymphatic capillary in tail of larva of Rana cate«biana. X 275. A group of cells was extruded at 9.30 P. M., June 13. Two hours later a lymphatic sprout was observed taking in one of these cells. This process was rei>eated until the extravasated cells were removed. Figs, a-h illustrate the taking up of one of these red blood cells. Between h and i, the larva was left in fresh water for 8 hours, i and k represent same sprout as shown in a-h. Note branch formation and moving down of nucleus.

cells are extruded near a lymphatic, the process takes in first one, then another, until all have been taken in. During this time- it may increase somewhat in size, first without, later with a nucleus. If

D D D Lymphatics in Tail of Frog Larva. 197

several cells are extruded at some distance from a lymphatic, the sprout sent out may reach a considerable size, with a nucleus, before the cells are taken in. If a large number of cells are extruded near a lymphatic, the sprout sent out may branch, the cells being taken in through two or more processes. After all the extruded cells are removed, the lymphatic formed may remain or may gradually be withdrawn.

The function changes in the lymphatic capillary revealed by this series of accidental yet beautifully clear-cut experiments give us a suggestion as to the meaning of the growth changes described above. If we compare the two, we find a striking parallel. In fact, if, in the " description of the taking up of the cells, we should substitute for "red blood cell" a substance microscopically invisible, and make the necessary changes, the two descriptions would be practically identical. Thus the changes concerned with peripheral growth and with function seem inseparably connected. We cannot avoid the suspicion that the fine processes continuously sent out represent a reaction of the lymphatic to ultra-microscopic substances, perhaps products of cell metabolism, that the greater the accumulation of these substances the longer and more persistent the processes, and that it is the varying formation of such substances which regulates the peripheral growth of the lymphatic capillary.

In connection with recent studies of the lymphatic system the tail of the frog larva furnishes an excellent field for testing methods. The two devices principally used have been injections and serial sections. Some confusion has arisen in the results obtained by these two methods. It has been suggested that this confusion has arisen from limitations of methods, yet they have not been subjected to rigid tests. In the fin expansion of the tail of the frog larva, where every lymphatic may be seen, a record may be made of the living non-injected lymphatics to serve as a control. The lymphatics may now be injected to see whether the entire system will be filled with the injection mass, or the tail may be cut into serial sections to see whether a reconstruction may be made which shall correspond with the control drawing.

A bull-frog lar\'a (Rana catesbiana) 16 mm. in length has been used for the latter test. A drawing was made of the lymphatics and blood

D 198 Eliot R Clark.

vessels during life. The tail was then cut into serial sections, ten micra thick and stained in hematoxylin and congo red. In attempting to reconstruct, it was found that, while blood-capillaries could often be fairly well reconstructed, it was impossible to reconstruct the lymphatics beyond the muscle margin. Further tests, however, are needed, with different stains, before definite statements are justifiable.

The injection method has been tested frequently and it is foimd that, while it carries us much farther than do reconstructions, the mass injected does not always fill the entire lymphatic system.

Let us now take a hurried glance over our results in their relation ' to present knowledge of the lymphatic system. Recent studies indicate that the first lymphatics arise from veins at various points. Of this primary origin we have made no test. As to the mode of extension of the lymphatic system into the different organs of the body, present views are divergent. Some observations lead to the conclusion that from several primary centers there is a centrifugal extension into the rest of the body, others that there is a transformation in situ of veins or mesenchymal tissue into lymphatics. The results here recorded are all in favor of the view that the peripheral lymphatics are formed by a process of centrifugal extension ; that this extension, so far as relates to the endothelium, is strictly invasive, with no addition from connective-tissue cell, wandering cell or blood-vessel. Our observations indicate, however, that outside factors may exert a modifying influence on the growing lymphatic, that there is a close relationship between peripheral growth and functional activity. The question as to whether the lymphatic is open or closed is not definitely determined for aU organs and tissues. The evidence here adduced favors a closed system of tubes, without direct openings into tissuespace or blood-vessel capable, however, of taking through its wall solid bodies of the size of red blood cells. But, perhaps most important, it has been demonstrated that the tail of the living frog larva, studied by special methods, furnishes an excellent field for the testing of problems relating to the growing and functioning lymphatic.

It is a pleasure to have this opportunity of expressing gratitude to Dr. Mall for his generous interest and numerous suggestions.



GEORGE L. STREETER, University of Michigan,

The accompanying figure represents a reconstruction of the brain, eye and two ear vesicles of a tadpole about one month old, in which the experiment was made of transplanting the left ear vesicle to the right side in the space between the normal right ear vesicle and the eye. This experiment was made as a supplement to a series of

similar experiments showing the effect of change in environment upon the posture and development of the labyrinth, and which have been previously reported (Jour. Exper. Zool., Vol. IV, 1907).

In the present experiment the effort was made to determine the influence of two adjacent ear vesicles upon each other; to see if on transplanting a very young ear vesicle, while still a simple primitive epithelial cup, and placing it against another similar ear vesicle, whether the two would fuse and develop into a single large labyrinth,as has been supposed to occur in. cyclopia, or whether the transplanted vesicle would retain its individuality and continue to develop as a separate structure.

The experiment was carried out on Rana pipiens larvse during the premotile stage, at a time when the ear vesicle consists of an invaginated epithelial cup just in the process of being pinched off from the deeper layer of the skin. The procedure adopted was similar to that used in the experiments previously mentioned; in this case the left vesicle being loosened from its natural bed and transplanted in a pocket in the loose tissue closely against the front


D 200 Gteorge L. Streeter.

surface of the right ear vesicle. After the operation the specimen was reared and at the end of a month was killed in preserving fluid, embedded in paraffin and prepared in serial sections. A model was then made as shown in the accompanying photograph by means of the wax-plate reconstruction method of Bom.

Examination of the sections and the model immediately shows that ear vesicles under the circumstances of this experiment maintain their identity. There is no trace of fusion or communication between the two. The experiment was repeated on other specimens and the specimens dissected with results to all apparances the same, though the duplicate specimens were not modelled. It may be pointed out that this result is in harmony with Stockard's recent experiments (Science p. 455, 1908), in which he produced cyclopia by the action of magnesium salts, and found that the defect was not due to a subsequent union or fusion of the two eye elements after they had become free and distinct. In all his cases where the cyclopLan defect was present it could be recc^ized at the first appearance of the optic vesicles.

In addition to the. original problem, the result of sftch a modification of environment upon the individual growth of the two vesicles is worthy of note. The effect produced upon the right labyrinth by the presence of the foreign one is limited to an abnormality of the anterior semicircular canal. A protruding pouch, corresponding to this canal, was formed in the normal way, but the central part of its walls failed to approximate and there was consequently no absorption area, such as is necessary for the completion of the closing off of the canal.

As regards the transplanted vesicle it can be seen, in the first place, that it has developed into a characteristic labyrinth. Furthermore the two canals, seen in the figure, possess the characteristics of the lateral and posterior canals respectively, that is, the labyrinth is a left-sided one. It may be pointed out that the distinction between the anterior and posterior canals can be easily made out by their relation to the lateral canal ; the ampulla? of the anterior and lateral canals branch out together from the utricle like the two arms of a ^^Y," while the lateral canal is completely separated from the

D Development of Amphibian Ear Vesicle. 201

posterior canal by a sharp cleft. Thus, in this instance the transplanted vesicle maintained its left-sided characteristics. It is next to be noted that, though in the transplantation it was placed haphazard as regards the planes of space, it has developed, like those described in previous experiments, in nearly a normal posture, with the endolymphatic appendage toward the brain. The tip of the appendage can be seen in the figure. The only serious defect in the transplanted vesicle is found in the region of the ampullse of the anterior and lateral canals, where they press against the other labyrinth. The labyrinth wall here is markedly retarded in growth and there is a very incomplete development of the anterior canal. Otherwise we have two practically normal labyrinths, and both are connected with the brain by well developed separate ganglia and nerves.




ROSS G. HARRISON, Yale University,

No abstract of this paper is given here, as the paper itself in amplified form was published in The Anatomical Record^ Vol. II, No. 9, December, 1908.



The following is a report of progress in experiments with food mixed with Sudan III.

The first step reported was in 1896, when Daddi found that Sudan III fed to animals, colored growing adipose tissue red.

The next was Riddle's report during Convocation Week a year ago that Sudan III fed to laying hens reappeared in the yolk of the developing egg.

Last July we found that in chicks hatched from such colored eggs the adipose tissue was of the characteristic Sudan pink. The fact of transmission of this coloring matter from mother to offspring was presented before the Graduate School of Agriculture in July and printed in Science of October 9, 1908.

As this substance can be transmitted from mother to offspring in a bird, it seemed that there might also be a similar transmission in mammals.

White rats were used, these responded at once to the feeding. The adipose tissue of half grown animals of both sexes showed the pink color when first tested, that is within five and seven days. As our animals were all immature, we sought the aid of The Wistar Institute.

Dr. J. M. Stotsenburg, who has direct charge of the rat colony at the Institute, undertook the necessary experiments for us. He mi:^d Sudan III with the food of pregnant rats, and continued the experiments during a considerable period; but the new-bom rats have not ias yet yielded a trace of the Sudan that we could detect either by visual examination of the minute fat masses present in the body at birth, or by ether extracts of those masses.

Further experiments were carried on for us by feeding the mother rats food mixed with Sudan III during the first eight days after


D 204 The AnatomicaL Record.

the hirth of their young, that is during a time in which only the mother's milk is used by the young as food. At the end of eight days the yoimg rats showed an abundance of pink adipose tissue, and the milk filling the stomach was so pink that it showed clearly through the stomach wall.

This foreign substance then not only colors the adipose tissue in the adult but also colors the fat of the milk, and the young living upon this pink milk has its growing adipose tissue colored.

As Sudan III gives the adipose tissue in the living animal what might be called a mass coloration easily recognized by the unaided eye, or with a simple magnifier, but is not satisfactory for microscopic investigation, we adopted the suggestion of Dr. Stotsenburg and turned from the rat to the guinea-pig in which the fat masses are well developed at birth.

Through the kindness of Dr. Theobald Smith of Harvard University and his assistant, Herbert R. Brown, guinea-pigs were fed the sudanized food during the last half of gestation. One specimen fed 13 days gave birth to offspring in which, as usual, much fat was present. No Sudan color could be seen and on extracting 4.5 grams with ether no color was obtained. Five other new-bom specimens from sudanized mothers were examined, but in no case could the characteristic color be found in the adipose tissue.

While in the hen the Sudan III is transmitted to the young through the yolk, and in the rat through the milk, contrary to our expectations neither the rat nor the guinea-pig fed during gestation showed transmission of this substance through the placenta.

DESCRIPTION OF A 5 MM. HUMAN EMBRYO. By H. E. Jordan, Adjunct Pro feasor of Anatomy, University of Virginia.

The material upon which the following contribution is based is a human embryo 5 mm. in greatest length. The specimen, for which I am indebted to Dr. Stephen H. Watts, Professor of Surgery, came into my hands in normal salt solution two hours after hysterectomy. It was immediately transferred to 95 per cent alcohol. Subsequent measurement showed a shrinkage of 1 mm. The specimen was stained in toto in Delafield's hematoxvlin and sectioned at

D Proceedings of the Association of American Anatomists. 205

10 microns. The tissues are excellently preserved and the degree of development is very similar to that previously described for embryos of aproximately this length. Thus in respect to the vascular and alimentary systems, it appear* similar to embyro No. 148 of the Mall collection (length 4.3 mm., myotomes 28) carefully studied by Mall,^ Bardeen and Levs^is^ and more recently by Mrs. Gage ;* and to embryo G. 31 of O. Hertevig's collection (length 4.9 mm., myotomes 35) studied and described by Ingalls.* It differs markedly, however, from these embryos in respect to the brain and the nephric system, the difference probably representing a slight advance in development. In external form it appears similar to embryo R of the His* collection (length 5 mm., myotomes 35), and is probably of about the same age, i. e., between 22 and 24 days.

I have been able to procure the following history of the case: The woman wa^ 33 years of age and had previously had four children, the youngest of whom is 9 years old, and two miscarriages. Menstruation had been regular. Her last period began September 16th. The operation was performed October 27th.

The above dates leave an interval of 13 days between the first omitted menstruation and the time of operation. The minimum and maximum ages of the embryo are probably about 21 and 25 days, respectively. Fertilization must have occurred at least 12 days after the last menstruation or at least 8 days before the omitted one. These data indicate independence of ovulation and menstruation.

External Form. The following points are important concerning the external appearance: The head turns slightly to the right and the tail to the left (see text figure). The umbilicus appears to be median. The nuchal bend is almost a right angle. There are 35 (perhaps 36)

^MaU, F. P., Johns Hopkins Hospital Bull., xii, 1901. 'Bardeen, C. R., and Lewis, W. H., Amer. Jour. Anat., 1901.

Gage, S. P., Amer. Jour. Anat., IV, 1905. ♦IngaUs, N. W., Archlv f. Mik. Anat., Ixx, 1907.

•His, W., Anatomie menscliliclier Embryouen. Text and Atlas. Leipzig, 1880-1885.

D 206

The Anatomical Record.

somites (lo, plus 8c, plus 12 1, plus 51, plus 5s, plus 4 or 5c). External corrugations simulate segmentation anterior to the single occipital somite. The sections show that these represent remnants of three additional occipital somites. The arm-buds extend from the 4 to 8 cervical somite inclusive ; the leg-buds from the 1 lumbar to the 1 sacral inclusive. The left arm-bud is turned back, thus presenting

its medial face. Four gill-arches appear externally; a fifth lies within the sinus praecervicalis. The maxilla is just beginning to take shape.

Three corrugations appear on the anterior face of the diencephalic region. The extent of the thin roof of the fourth ventricle is plainly visible. In the mid-region there is a small notched ele

D Proceedings of the Association of American Anatomists. 207

vation of folded ectoderm. Eye, ear, auricle, ventricle, bulbus and truncus arteriosus, liver. Wolffian ridge and ganglia (5, 7 and 8, 9, and 10) are also visible externally. Anterior to the maxilla, the ectoderm of the ventral face of the head is very thin and slightly depressed. On changing the level of focus one sees in this region a sucker-like projection with median groove and central indentation. Sections show that this is the region of the optic recess, and the indentation probably marks the site of final closure of the neuropora The ectoderm consists generally of a single layer of ouboidal cells. Ventral to the eye on either side of the head, occur patches of thickened ectoderm extending through 24 sections representing the anlagen of the nostrils. Thickened patches of ectoderm occur also in relation to the cranial nerves forming so-called "placodes." Sections show that the somites have differentiated into sclerotome (with loose anterior and denser posterior segment) and myotome. The otic vesicles lie between the base of the third gill arch and the ectodermal fold of the roof of the fourth ventricle above mentioned.

Internal Stbucture.

(a) Alimentary canal. — The mouth forms a broad transverse slit bounded laterally by the mandibles and maxillae, anteriorly by the fore-brain and posteriorly by the fused mandibles. No remnants of an oral plate can be found. The mouth leads into a broad wide pharynx bounded laterally by the gill arches. Between the latter are the entodermal extensions of the pharynx, or gill pouches, forming with the apposed invaginated ectoderm thin membranes stretching between the arches. A dorsal cephalic extension of the pharynx is the anlage of the hypophysis, posterior to which lies a second extension of the pharynx, SesseFs pocket. Closely applied to the curved anterior border of the hypophysis, rests a projection from the diencephalon, the infundibulum. In this same region the notochord terminates with a sharp ventral curve.

On the floor of the pharynx in the region of the second arch the tuberculum impar is well developed. The alveolo-lingual grooves and the lateral rudiments of the tongue are not well marked in the sections. In the same region the median thyroid, unconnected

D 208 The Anatomical Record.

with the pharyngeal epithelium is present, consisting of a small irregular mass of spheroidal cells. Anlagen of lateral thyroids and the thymus are just recognizable in the sections.

The larynx, represented merely by a slight depression, leads into the trachea, about one half millimeter in length, which bifurcates into two branches each with a terminal expansion. The latter are enclosed by extensive mesoderm forming the lung-buds.

The pharynx leads into a short oesophagus with thick epithelial wall and narrow lumen. On the dorsal border of the liver, the tube enlarges slightly, forming a short spindle-shaped stomach. This in turn leads into the duodenum with wider lumen and finally into the caudal intestines. About the level of the fourth thoracic somite the vitelline duct arises. The gall-bladder and bile-duct lie on the lower border of the liver ventral to the intestine. About 50 microns anterior of this point the anlage of the dorsal pancreas can be recognized as a shallow outpocketing. The intestine can be traced to the cloaca, which is closed by the anal plate. From the cloaca the allantois extends outward into the umbilical cord. Proximally the allantois shows a slight expansion, the anlage of the urinary bladder. Pericardial and pleural coelon are continuous dorsally with the abdominal coelom. The liver has grown out in all directions into the septum transversum. The hepatic tissue is invaded by the omphalomesenteric veins forming sinusoids.

(b) Vascular System, — The heart appears very similar to the His model for a 5 mm. embryo. The two atria are continuous, forming a single large sac with lateral expansion, and opening by the atrial canal into thq left ventricle. The sinus venosus has moved to the right. The left ventricle passes into the bulbus arteriosus which makes a sharp turn dextrally and cephalward, and passes into the trimcus arteriosus which extends along the ventral aspect of the atrial part of the heart. The wall of the heart consists of a loose mesh of undifferentiated muscular tissue. The wall is thickest in the left ventricle. The endothelial tube is loosely applied in the heart proper, but more closely in the bulbus and truncus arteriosus.

The truncus passes into the floor of the pharynx where it expands into a wide sinus. From the latter extend two anterior branches (ventral aortal) each with two lateral twigs. The most anterior of

D Proceedings of the Association of American Anatomists. 209

these breaks up into capillaries in the mandible and the second supplies an aortic arch to the second gill-arch. From the sinus, and projecting backwards arises the larger third aortic arch. Caudally from this point the fourth and sixth aortic arches take origin. Slender backward extension of the ventral aorta behind the sixth arch supply capillaries to the floor of the larynx. The aortic arches on each side unite dorsally to form two dorsal aortas which in turn unite about the level of the fifth cervical myotome, or the cephalic border of the liver, into the single dorsal aorta. Just anterior to this point the subclavian arteries, and immediately beyond it the coeliac artery, are given off."

A large ventral branch, the right omphalomesenteric or superior mesenteric artery leaves the dorsal aorta at about the level of the fifth thoracic myotome. Near its termination it bifurcates into two branches which accompany the vitelline duct for a short distance on either side. The inferior mesenteric artery arises as a delicate ventral branch about the level of the seventh thoracic myotome. The dorsal aorta again divides into two (the hypogastric or umbilical ffrteries) before it passes into the umbilical cord. Frequent renal twigs are supplied to the glomeruli of the Wolffian ridge. From each dorsal aorta there extends cephalad a twig (internal carotid) to a short distance beyond the hypophysis. The vertebral arteries can be traced forward as far as the diencephalon.

The jugular veins (precardinals) arise from the union of numerous twigs lying close to the surface in the anterior head region. Posterior to the eye, where the vein becomes an elongate vessel, it passes mesad of the gasserian ganglion and laterad of the ganglia of the 7th and 8th nerves and the otic vesicle. Ramifying within the ganglia of the 9th and 10th nerves it passes laterad of the former and mesad of the latter and unites with the duct of Cuvier. The umbilical veins arise at the umbilicus from a single large vessel and thence forward through the somatopleure to the Cuverian ducts. The ductus venosus, formed by the union of the vitelline veins in

•I am unable to And any evidence of multiple subclavian arteries such as Evans (Evans, H. M., Anatomical Record. Vol. 2, No. 0, 190S) reiwrts from embryo No. 148 of the Mall collection (two segmental vesicles), and as I have myself observed in very young turtle embryos (three segmental vessels).

D 210 The Anatomical Eecord.

the liver, also connects with the sinus venosus. A single vein, the "primary ulnar" (subclavian) drains each arm-bud. The leg-bud also contains only one vein, the "primary fibular" or "vena.ischiadica" (common iliac). All blood vessels and the heart contain well preserved erythroblasts ; occasionally these are seen in mitosis.

(c) Nephric System. — In the nephric and central nervous systems the greatest variations are found between this embryo and the several above mentioned. The Wolffian ridge extends from the level of the mid-region of the arm-buds to the cloaca. Anlagen of the genital ridge and metanephros have not yet appeared. The Wolffian ducts appear continuous from end to end of the ridge. There are approximately 30 nephric tubules. The anterior 18 or 20 consist of well-developed glomeruli with Bowman's capsule connected by patent S-shaped tubules with the Wolffian duct. Several of the most anterior glomeruli lie very close to the coelomic epithelium; the remainder lie deeply embedded within the ridge. The tubules have comparatively thick walls and narrow lumen. The posterior 12 or 14 end distally in expanded vesicles, but no true glomeruli have formed ; and these tubules also are connected with the Wolffian duct. Several delicate tubules at the cephalic end of the ridge which are difficult to trace in sections may represent the remnant of a pronephros. However, no tubules connect either with the coelom or the myotomes. The nephric system at this stage consists essentially of a mesonephros apparently considerably less generalized than that of the embryos described by Mrs. Gage and by Ingalls.

(d) Central Nervous System. — The eyes at this stage are represented by evaginations from the diencephalic margin of the forebrain and are already slightly cupped. The ectoderm has thickened into a plate of tall cells over the region of the cup forming the anlage of the lens. The posterior roots of the spinal nerves are represented by delicate bundles of neuroblasts. The wall of the neural tube, which consists of an internal ependymal layer of tall cells, a middle layer of neuroblasts and a peripheral marginal velum, appears in the sections to have a perfectly smooth internal contour. Study of a carefully constructed model of the brain (by the blotting paper method described by Mrs. Gage in the Anatomical Record for November 10, 1907) together with the sections disclose the following

D Proceedings of the Association of American Anatomists. 211

conditions: Externally the brain tube gives no clear evidence of folds. As already stated, however, three ectodermal corrugations appear in the region representing the diencephalon and the mesencephalon; and the same number appear anterior to the occipital somite. The fact that these latter corrugations correspond to >occipital somites in process of disappearance shows that metamerism had at an earlier developmental stage extended into the region between the first cervical and vagus nerves. The ganglia of the 5, 7 and 8, 9 and 10 nerves are symmetrical in position and similarly developed except that the right gasserian ganglion is considerably larger than the left. The segmental arrangement of ganglia indicates a still more anterior extension of metamerism. When one regards a slight external bulging just anterior to each gasserian ganglion as the "cerebellar folds," and the ganglia of the 7 and 8 nerves as a fusion of two segments, and the region of the otic vesicles as the fifth "neuromere" the full number considered typical (seven) for the mammalian hind-brain (Bradley^) is accounted for. In six distinct regions, also, there is a decided thinning of the neuroblast layer of the wall of the hind-brain and a reciprocal thickening (in sections conforming to a blunt wedge-shaped area) of the marginal fibre layer corresponding more or less closely with ganglia of the 6, 7 and 8 nerves, the otic vesicle, and the ganglia of the 9 and 10 nerves. Moreover, there is in these regions the merest indication of a bulging of the wall. In the model, however, one seeks in vain for distinct evidence of folds, with the exception of a wide embayment in relation to the gasserian ganglion. This "neuromere" is by various writers on different forms described as the most pronounced and least transitory fold. I can find no evidence of distinct internal folds in the fore and mid-brain. The foregoing facts seem to indicate, then, that my embryo has attained to a slightly later stage of development than that of the embryos described by Mrs. Gage and by Ingalls. It seems probable, accordingly, that folds anterior to the gasserian ganglion have already disappeared, as also those related to nerve roots posterior to this region including the vagus. The only persisting "neuromere" at this stage is the one associated with the trigeminal nerve, or the second of the rhombencephalon.

'Bradley, O. C, Rev. Neurol, and Psychiatry, il, 1904.

D 212 The Anatomical Record.

A STUDY OF PATHOLOGICAL CAT EMBRYOS. By H. E. Jobdan. From the Anatomical Laboratory of the Vmversiiy of Virginia.

The science of descriptive teratology is founded mainly on facts relating to the embryonic pathology of man. Recently, Denison (1) made a study of ten abnormal pig embryos ajid reported results in harmony with the conclusions of Mall (2) regarding the origin of merosomatous human monsters. The results of a microscopic study, which I am making of diseased cat embryos, are thus far also strikingly consonant with recent opinion respecting the etiology of human terata.

Schwalbe (3) regards "amniotic constrictions and bands" as among "the most abundant of anomalies of the amnion" and states that ' "abnormalities thus produced are manifold" (p. 192-193). Mall in his article on "The Origin of Human Monsters" says, "Xo amniotic bands are found in any of the 169 specimens which I have studied." Denison likewise finds no amniotic bands in pig embryos, but the "amnion is often thickened, rough and smaller than normal." Ballantyne (4) in summarizing his chapter on "Amniotic Diseases in Teratogenesis" says that in the case of some of the terata at least, "the amnion would seem to act by pressure, and so delay, or altogether stop the progress of events in ontogenesis."

In one horn of the uterus of a cat I found two embryos measuring 12 mm. and 9 mm., respectively, and in the other horn one embryo measuring 7 mm. The first two embryos with their adnexa appeared normal. The third embryo was enveloped in a closely fitting amnion which was adherent to the uterus over a wide area (Fig. 1). The amnion had formed a band extending across the body between the two limb-buds, and the constriction associated with the band had produced a partial separation of the posterior from the anterior portion of the embryo. Three additional minor constrictions of the amnion produced adhesions with the ectoderm of the head, fore-limb and the body-wall in the region of the heart. The eyes were faintly visible externally. No sign of somites could be recognized. The head appeared considerably swollen. The embryos and a portion of the uterine wall were immediately fixed in Zenker's fluid.

The first striking fact is the great variation in length between em

D D D D D Proceedings of the Association of American Anatomists. 213

bryos from the same uterus and probably of the same age. Professor McClure, who has studied many cat embryos, in his work on the development of the venous system and lymphatics, writes me that from 1 to 1.5 mm. is the greatest diflFerence he has ever noted. He says, moreover, that he has not found many embryos with amniotic bands. In the case under consideration one naturally infers that the amniotic band interfered with the nutrition of the smallest

Fio. 1. — Photograph of 7 mm. cat embryo showing the area of fusion between the closely-fitting amnion and the wall of the uterus; also the deep amniotic constriction between the limb-buds. Magnification about 3 diametei-s. (Made by Prof. Theo. Hough.)

embryo and prevented normal growth. However, the difference of 3 mm. between the larger embryos indicates the influence of a more primary malevolent factor. I undertook a comparative study of the pathological embryo with the amniotic band and the two apparently normal embryos from the same uterus. The sectioned material showed that all three embryos were similarly pathological

D 214 The Anatomical Record.

though in varying degrees, thus indicating a common fundamental inciting cause.

The above facts combined with others even more obvious, make

Chorion ^.^ Amnion

Fio 2. Semidiagrammatic drawing of transverse section through region of fore-brain showing solid cord and brain (the unstippled areas represent an acidophile coagulum) ; also the engorged right anterior cardinal vein (V. A. C).

it improbable that the union between amnion and chorion represents an incomplete original separation. The union is intimate

D Proceedings of the Association of American Anatomists. 215

(Fig. 6), but is probably due to secondary fusion following amnionitis. A better understanding of the early development of the cat would aid in distinguishing a primary from an acquired adhesion. The problem is complex, since, in early stages, there is an allantois-amnion, an allantois-chorion, and a yolk-sac placenta. (O. Schultze) (5).

Both amnion and chorion have become much thickened in places, and both may perhaps be best described by the term "fibro-cystic" adopted by Denison for apparently similar changes. Cells with large fragmenting nuclei lie in the cyst-like interstices of the chorionic mesodermal tissue. The cells of the amnion are for the most part smaller, the fibrous tissue is less compact, and the lacunae are wanting. Sections of the uterine wall show that chorionic villi are present and apparently normal in some regions, while in others they are covered with an exfoliating epithelium or are absent. In short there is evidence of necrosis, but no sign of inflammation.

Xeither umbilical cord nor vesicle is present. The embryo appears attached directly to the amnion and chorion (Figs. 3 and 4). Extreme strangulation evidently obtained, centering about the point of entrance of the blood supply of the embryo. Since there was no endometritis, the pathological condition may be the result of the "faulty implantation" (Mall) or some other elusive cause producing "disorderly ontogenesis". (Ballantyne.)

Sections of the 7 mm. embryo reveal the following points: The embryo is much deformed in the facial region. A portion of the head has fused with the wall of the thorax thus obliterating the mouth and involving the base of the tongue (Fig. 2). However, the mandible, maxilla and two gill-arches can be distinguished. No distinct epidermal ectoderm can be recognized. The left forelimb has turned upward (dorsal ward, Fig. 3) and rests over a thickened area of the chorion. Caudalward from the fore-limb, the bodywall is much deformed (Fig. 5). In the region where the amniotic band has. not cut through the entire body-wall, the internal organs are much compressed and misshapen (Figs. 4 and 5); they have been invaded by blood cells. Still more posteriorly the remains of chorionic villi, appear with exfoliating epithelium and necrotic areas.

D 216

The Anatomical Record.

The brain and spinal cord are enlarged and almost solid. Their cavity is filled with a mass of coagulum and round cells, probably derived from the dissociating nervous elements. No wandering

N —

Fig. 3. — Semidlagrammatic drawing of transverse section through the region of the fore-limbs and heart, showing a necrotic area (N) in the chorion; also the area of fusion between body-wall and amnion (X) and the fusion of heart with body-wall and the amnion with the right forelimb. X 20.

blood cells are present. The nerves are merely masses of pale disintegrating fibres, and the ganglia are in process of dissociation.

D Proceedings of the Association of American Anatomists. 217

The epithelium of the ear has broken down and the cells are disintegrating. The eye appears as a confused mass of broken elongate cells (the prbduct of the dissociating and disintegrating

Lung.— —

Liver— —


Fig. 4. — ^Drawing of region 85 sections posterior to the last, showing enlarged area of fusion (X) between amnion and body-wall, and the strangnlated condition of the blood vessels and intestines in the region of the umbiUcus. X 20.

lens), surrounded by a layer of dissociating retinal cells, and by large pigment granules (the residue of disintegrating choroid cells).

D 218 The Anatomical Kecord.

The epithelium of the trachea, oesophagus, stomach, duodenum and mesonephros is also detached from its basement membrane and

■*• Mesonephros

^V. P. C.

_ Aorta

Sp. C

Fig. 5. — Drawing of section through the mesonephron and the associated posterior cardinal veins (V. P. C), showing also tlie thickened character of the amnion and the malformation due to pressure of the amniotic band. X 20.

dissociating. The liver is represented merely by an amorphous mass of round hepatic cells mixed with blood cells (Fig. 4). The

D Proceedings of the Association of American Anatomists. 219

pharynx is small ; the infundibulum, thyroid gland and thymus are dissociating.

The aortic arches are very small and filled with dissociating endothelial cells. The blood vessels and the heart are filled with blood. The right anterior and posterior cardinal veins are much dilated and engorged with blood.. (Fig. 2.) In a few cases the walls of the blood vessels have disappeared and erythroblasts have wandered

Fio. 6. — Photomicrograph of region of fusion between amnion and chorion. X 300. (Made by Dr. Frank P. Smart.)

into the surrounding tissues. Many of these have fragmented nuclei. The heart appears almost normal, though the atria are small and irregular and there are signs of tissue dissociation.

In the head region the mesenchymal tissue seem oedematous and the nuclei of the cells are fragmented. In other regions the mesenchyme appears generally in healthy condition. At points of fusion between the embryo and the amnion the mesoderm seems

D 220 The Anatomical Record.

continuous from one to the other. In the region of the heart the body-wall has fused with the amnion over a wide area (Fig. 3), and the heart has fused with the mesenchyme of the body-wall.

Myoblasts are sparsely scattered here and there through the dorsal regions of the body. Precartilage and cartilage everywhere appear normal. The notochord is in process of dissociation.

Sections of the 9 mm. embryo show similar pathological changes, though it is plain that the embryo has attained a slightly later stage of development. There is a smaller area of fusion between amnion and chorion. The umbilical cord is very short and compressed, and the yolk-sac has disappeared. The embryo is misshapen and flattened in the region of the umbilicus. In the oral region and the spinal cord the embryo is more nearly normal. But again the brain is solid, enlarged and filled with round cells; the epidermal ectoderm is lacking; ganglia, nerves, eyes and epithelial linings are dissociating.

The liver is a confused mass of large round cells. The walls of the blood vessels have very generally disappeared and the blood cells have invaded the tissues. The aorta and aortic arches are filled with dissociating and probably proliferating endothelial cells. Myoblasts are more numerous, but never aggregated in myotomes. Mesenchyme and cartilage again appear perfectly normal.

The 12 mm. embryo has attained a considerably later stage of development, but similar diseased conditions prevail in the tissues, apparently with great severity. There is no blood in the heart or vessels. The walls of the vessels have disappeared. The tissues including the brain are invaded with nucleated blood cells. Here also the brain and cord are solid. The area of adhesion of amnion to chorion is very small; but there is undoubted strangulation and consequent interference with nutrition. The face in the region of the mouth is much misshapen. All the tissues except cartilage and mesenchyme have dissociated as in the other embryos. Myoblasts are very numerous and individually they appear in good condition. There are regions where mucoid degeneration has taken place. All the organs belonging to this stage of development are present, but dissociating, the liver, brain and blood vessels being most seriously affected.

D Proceedings of the Association of American Anatomists. 221

Summary. The stage of development is inversely proportional to the area of adhesion between amnion and chorion or to the degree of strangulation. Barring the deformity due to the pressure of the amnion and the presence of an amniotic band in the 7 mm. embryo, the degree of abnormality varies slightly but directly as the development, as indicated more especially by the character of the blood vessels. Since the three embryos of the same uterus are similarly diseased and since only one has an amniotic band, the latter can only have been secondary to some underlying more primary cause. This was not endometritis, but some other pathologic agent producing necrotic areas in the placenta, destruction of chorionic villi, fusion of amnion and chorion, strangulation of the cord and interference with nutrition. Another interesting fact is the selective influence of the disease-producing factor on the various tissues. Liver, brain, cord, nerves and blood vessels show progressively less susceptibility to the morbid agent in the order named. Mesenchyme and cartilage seem most resistant. The pathological cat embryos agree among themselves and with certain human and pig embryos in being hydrocephalic and oedematous, with tissue dissociation and local histolysis. The results of this study support the position of Mall respecting human embryos that amniotic bands are secondary factors in the production of merosomatous monsters.


1. H. S. Denison, Anat. Record, Vol. 2, No. 7, 1908.

2. F. P. Maix, Jour. Morph., Vol. 19, No. 1, 1908.

3. E. ScHWALBE, Die Morph ologie der Missbildungen des Menschen und der Thiere, Jena, Pt. 1, 1906.

4. J. W. Ballantyne, Manual of Antenatal Pathology and Hygiene* 2 Vols., Edinburgh, 1904.

5. O. ScHULTZE, Grundriss der Entwickelungsgeschichte des Menschen und der Sttugerthiere, 2 Vols., Leipzig, 189G-*97.

D 222 The Anatomical Kecord.

REMARKS ON THE DYES USED IN THE HISTOLOGICAL LABORATORY. By Robert Retzer, Anatomical Laboratory, John Hopkins University ,

An effort to arrange the dyes used in this laboratory, so that not only the members of the staff, but also the errand boy should have no difficulty in finding them and putting them back to their proper places, was found to be a more difficult task than it seems at first sight. The dyes are known by English, French and German names and translating them into the terms of one language is but the first step. Most dyes have compound names and the question then arises, "Under which name shall they be classified?" The problem was solved by arranging them alphabetically according to the colors indicated on the bottle. Where the color was not mentioned (dahlia, fuchsin, etc.) the bottles were placed in an alphabetical order interspersed between the other bottles. With this method of classification, the dyes were not misplaced and a great deal of annoyance ob\"iated.

Before coming to the decision of adopting this manner of arrangement, it was considered whether it would not be advisable to place synonymous dyes together, with a view of thus utilizing the old stock. To my surprise, this turned out to be an impossible labor.

What according to one author is a synonymous dye means according to another an entirely different one. This, in turn, led me to look into the literature more thoroughly and study the origin of the names and the constitution of the dyes we use. For reference, I used the Encvklopadie der mikroskopischen Technik, Mann's Physiological Histology, Bernthsen's Lehrbuoh der organ ischen Chemie, Xietzke's Organische Farbstoffe, Beilstein, the chemical dictionaries of Watts and Thorpe, and the catalogues of Griibler and of Merck.

Even the cursory examination of a few of these books will reveal the lamentable state of our knowledge of the dyes. No two authors seem to agree, a fact readily understood when we consider that one is a chemist, another a histologist, another a manufacturer or dealer. The confusion seems to arise with the manufacturer who caters to the demand of the dyer and calls the product he sells by the name of the dye from which it is derived. For instance, Echtgriin is dinitroresorcin to the chemist, while one manufacturer

D Proceedings of the Association of American Anatomists. 223

places the sodium salt and another the potassium salt on the market as Echtgriin, Frequently it is still more confusing and complicated. The chemist applies the name to the base while the histologist buys a salt of its sulphonic acid under the same name. I might have stated that the histologist not alone suffers from the impositions of the manufacturer. If the chemist wants pure methyl alcohol in his laboratory he calls for Columbian spirits, if he asks for methylalcohol he is not sure to get a pure product.

In tabulating^ the synonyms of the dyes used by histologists, and there are about three hundred names to be found in biological literature, one is beset by a great many obstacles due to the inaccuracy of the authors. Each one stretches the synonyms a little further, until by comparing the first with the fifth author we find two entirely separate and distinct dyes called by the same name.

So, for instance, helianthin becomes crocein 3 B, dahlia (sol. in water) becomes anilin violet (insol. in water) and alcohol blue becomes primula. The confusion of terms is partly due to the histologist who will dissolve a dye in water that is according to all the authors insoluble. Evidently he used a different dye from the one he mentioned. In this paper, I intend to present but a few of the most commonly used dyes and point out some of the errors we fall into.

Hsematoxylin, the dye obtained from the wood of Haematoxylon campechianum (logwood), is spoken of by the older German authors as Blauholzextrakt or Campeschaholzextrakt. The crystals are colorless and are but little soluble in cold water. They oxidize very rapidly by exposure and become haematein and some other oxidization products, that are more soluble. It follows from this that the solubility depends upon the age and exposure of the hjematoxylin

'The tables contain : First, tlie name of the dye ; second, all of its synonyms and pseudonyms; third, the chemical formulae of each of these; fourth, the manufacturer; and fifth, the process of manufacture or Its derivation. Rows 1, 2, and 3 of the tables are fairly complete, but 4 and 5 are necessarily incomplete, because the manufacturers* catalogues were inaccessible. The size and incompleteness of the tables prevent me from having them published. They form, however, the basis of this study, which has extended over a period of six months.

D 224 The Anatomical Eecord.

and when making up a saturated solution it must be taken into consideration.

Eosin is probably the next most commonly used dye and is manufactured by a great many German, English and American factories. Whenever this is the case the products differ in their physical and chemical properties, because these dyes are not made for the chemical laboratories, but for dyeing and staining. Eosin is derived from fluorescein, the sodium salt of which is occasionally used in histology under the name of uranin. When fluorescein is treated with bromium it forms among the bromium compounds tetra-bromfluorescein, called by chemists eosin. We buy the sodium or potassium salt under the name of eosin, Eosin gelblich, Eosin w. gelb, or water-soluble eosin. The pure tetra-bromfluorescein is practically insoluble in water, however.

The alcohol soluble eosins are the ethylethers or methylethers of tetra-bromfluorescein. The former are but little used in histology, while the latter we buy under the name of methyl eosin, primerose and Spriteosin. Kaiserroth, saffrosin, phloxin, Bengal rose, Eosinscharlach and Lutetienne have been called synonymous, but wrongly. They are iodine, chlorine or other compounds of tetra-bromfluorescein.

By erythorosin is meant an alkali salt of either tetra- or di-iodide fluorescein. We may get a sodium salt, potassium salt, or ammonium salt of tetra-iodide fluorescein, and the same salts of di-iodide fluorescein, and mixtures of all. The difference in results we notice by solubility and color reaction. Unless we get the stain from the same bottle we are never sure to get the same result the second time, because we deal with arbitrary mixtures and not chemical compounds.

1. BaBic Fuchsin. G. RosaniUn. 11. Anilin Red.

2. Neutral Fuchsin. 7. Rosanllln hydrochloride. 12. Azalelne.

3. Fuchslclenne. 8. Solferlno. 13. Harmallne.

4. Magenta. 9. Rubin. 14. Rublanlte.

5. Magentaroth. 10. Erythrobenzlne.

These are the names which are given as synonyms of fuchsin. The one most commonly used in this country and England is

D Proceedings of the Association of American Anatomists. 225

frequently spoken of as a separate and distinct dye. Histologists recommend magenta for staining the inner substance of elastic fibres, but fuchsin for a nuclear stain. The dealers naturally make stock of this and we find the two names in their catalogue. The original magenta was an English manufacture, while fuchsin is made in Berlin and Elberfeld. We find frequent errors, especially in every-day parlance, made in the opposite direction. Histologists will speak of sudan, when they mean sudan III, pyronin when pyronin G' or pyronin B is meant, and methyl violet when they mean any one of six or seven dyes that are put on the market.

A word about the methyl violets and the use of numerals and letters. Methyl violets are oxidization products and are the mixtures of hexa-methyl-para-rosanilin with penta, tetra, or even di and nonomethylpararosanilin. The salts are called methyl violet 6B, 5B, 4B, etc., the numerals indicating respectively the preponderance of 6, 5, 4, etc., methyl groups. Methyl violet 6B, therefore, is frequently called hexamethyl violet. Just as the methyl violets are the methylated rosanilins (or fuchsins) so the anilinblues are the phenyl ated rosanilins. With one or two phenyl groups the color is violet, with three or more blue. There are more than twenty anilinblues mentioned in histological literature. Most authors do not state which anilinblue they have used, and it is, therefore, little wonder that other histologists fail to get the same results.

Every person who has worked with methylene blue knows of the inconstancy of the results. The main cause of this lies in the impurities. There are various methods of manufacture and each method brings in a diflFerent impurity. Knowing what chemists can do, I am convinced they can put a pure methylene blue on the market for the. use of histologists. As soon as we get a pure stain we must discard all the others, because the results obtained from them are but empirical. I have frequently heard the term methyl blue used synonymously for methylene blue, more from carelessness than from ignorance, but as it is just such carelessness that brings the vast number of pseudonyms into literature, it is worth while calling attention to it.

What conclusions must we draw from these observations ? What can be done to give us more uniformity in the matter of dyes?

D 226 The Anatomical Record.

It seems a hopeless state of affairs and even an optimist can find little pleasure in the contemplation of the problem. First of all, we must be more careful in the use of names. When we use eosin we must state what kind of eosin, that is, we must be explicit and not use class names such as methylene blue, anilin blue, etc. Secondly, we must state whether it is the sodium, potassium or ammonium salt. If we do not know, then stating the manufacturer's or dealer's name will be of some use. To-day we know the process of manufacture of most of our dye-stuffs, as the patents have long expired, and when we find out the manufacturers' name we can also conclude what the most likely impurities are. I may here say that it is immaterial whether we consider staining a chemical . or a physical action. We well know that the solubility of a salt is not dependent on the chromofore radical, but upon a number of undetermined factors, and where some stains are done with a watch in the hand, the concentration of a solution is very important.

To-day nearly every histologist is a dyer, as soon as he uses stains he belongs to the same class as the silk manufacturer and dyer. Histological staining is to-day unscientific, or if we dignify it by the name, let us call it an empirical science. As long as we do not know the nature of the reagents we are dealing with, histological staining is to-day where medicine was a hundred years ago. We must do away with empiricism if we want to learn something about the principles underlying all structures.

Yet, staining has once and for all time become a necessary adjunct to our histological course and our own researches, and it is not my intention to suggest a sweeping reform, but to advocate the elimination of a great many evils connected with it. Every teacher knows that it is difficult to demonstrate to a student the form of it when the nucleus or cell is unstained, yet it can be done and it trains the student's power of observation. We must not forget that there are so many artefacts connected with staining that we do not always get a true picture of the structure when we study it stained. Staining has very little to do with morphology, it deals with far more complicated phenomena, which will never be approached until the histologist learns that he must not only know that a given dye will stain a nucleus blue, and another protoplasm

D Proceedings of the Association of American Anatomists. 227

red, but that he must also know what its chemical constitution is. There are a great many dyes that have formed the basis of chemical work, of whose physical properties we know something. The histologist can have more, if he demands more. How can he learn anything about the physical properties of a gland or of a cell if he is" ignorant of the essential properties of the reagents he uses 'i

To repeat, staining has very little to do with morphology. Sufficient evidence is given by the progress of histology. We can use Strieker's Handbuch to-day, and learn all of the essentials; indeed, a number of facts we find in it have been forgotten and rediscovered. And this book was written before the introduction of anilin dyes. It is hard to recall any great discovery with the exception of Ehrlich's and his followers, that has been based on the reaction of cells to dyes. The discoveries were made by the physiologist, who pointed out certain physiological actions of cells, glands or tissues, and these discoveries were only verified by the histologist. With all the complicated methods for staining the nervous system we have only managed to verify the work or suggestions of the physiologist. The dispute about the structure of ganglion cells and nerve-endings will not be settled until we put staining on a sound physical basis.

This seeming digression was necessary in order to make clear the conclusions. It is every evident that the histologist is being imposed upon by the manufacturer of dyes, and this imposition is due mostly to his ignorance and stands in the way of progress. The study of the synonyms has led us to get at the root of the evil. There are, as far as I have been able to find out, about ten factories from which we get our dyes. Each one of these factories turns out a product which is intended to stain blue, green, violet, etc., but not to have definite chemical reactions, and the name which the manufacturer gives, means something to him but nothing to us. We have, chemically speaking, a number of synonyms for a great many dyes, but practically — practically to the histologist — there are no two dyes alike. This evil must be remedied.

We owe to Weigert and Ehrlich the science of histological staining, which is still in its infancy. Because it is a difficult task it needs the comhined efforts of all histologists to ])lace this science


228 The Anatomical Eecord.

in the same rank with the others. The efforts must at first be directed towards the obtaining of pure products. As soon as we have these, and not before then, can further discussions be legitimately introduced. Isot so many years ago the chemists were in the same predicament and had to purify practically every reagent with which they worked. They could purify them because they had the appliances and the knowledge, but the histologist cannot do so to-day. It may mean that eventually the histological laboratory will have its chemical division, but in all events that lies in the far future. We must consider the present, and so the histologist should use only those dyes of which the chemist has found the constitution, and of which the most likely impurities and their influence upon the reaction are known. Even those histologists who do not believe in the importance of chemistry must admit that to-day it is difficult, sometimes impossible, for a man to obtain the same results in this country as another gets in France or Germany. This applies especially to results obtained from the newer stains, with which the market is being flooded. It may be unscientific to consider the protection of American industry, but we certainly could get products fresher and quicker by patronizing American dye manufacturers. Competition is great enough to make the insistence upon purity tenable. To-day we get impure products because there is no demand for other. We must therefore —

1. Exercise more care in the use of names, because carelessness is the prime cause of misleading pseudonyms.

2. Insist upon the name of the manufacturer, if not his, then the dealer's, because each manufacturer means a definite dye when he puts it on the market, and does not consider that other dyes carry similar names.

3. Give the chemical formula, so that there be no misunderstanding and our work can be repeated and verified by other observers.

4. Require pure products, made for the use of the chemist and histologist and not the dyer, and following this, require a standardization of dyes.

D D D Proceedings of the Association of American Anatomists. 229

THE ORIGIN OF THE SEX-CELLS OF AMIA AND LEPIDOSTEUS. Bt Bennet M. Allen, University of Wisconsin,

It has been shown in a number of papers that have appeared during the last fifteen years, that sex-cells of representative fishes, amphibians, birds and reptiles undergo a more or less extensive migration from the position in which they are first distinguishable, in order to finally come to rest in the sex-gland anlage.

The papers of Wheeler, '99, on Petromyzon, Woods, '02, on Acanthias, King, '08, on Bufo, Jarvis, '08, on Phrynosoma, and Allen, '06-'07, on Chrysemys and Kana have each shown with greater or less certainty that the sex-cells arise in the endoderm, and that they migrate from it up through the mesentery to the sex-gland analgen.

Striking differences in these processes, as seen in Amia and Lepidosteus, are extremely interesting. In Amia the sex-cells arise from the endoderm a short distance lateral to the junction of that portion forming the roof of the subgerminal cavity, with the vitelline mass. Further work may show a more extensive anlage. From this zone they pass up into the lateral plates of mesoderm, in which they migrate toward the axis of the embryo, passing along between the splanchnic and somatic layers. On account of their large size and the large size of the yolk particles with which they are filled, they are clearly distinguishable from the mesoderm cells. Furthermore, the nuclei are larger, paler and more rounded than are those of the mesoderm. These characters also serve in a lesser degree to differentiate the sex-cells from the endodermal cells of the alimentary tract, which they approach as they near the goal of their migration. Most of the sex-cells undergoing this migration reach the medial ends of the lateral plates of mesoderm before the splanchnic and somatic layers separate to form the body cavity. When this does occur, the sex-cells remain attached to the peritoneum in such a way as to occupy a position a short distance on each side of the root of the developing mesentery. As the two mesodermal layers separate, the sex-cells are very clearly seen to lie between them, only later sinking into the proximal portion of the somatopleure — ^the sex-gland anlage. The sex-cells are found in the mesoderm opposite the hind-gut, in a region beginning with its anterior end, and extending backward about five-sixths of the distance to the cloacal opening.

D 230 The Anatomical Record.

It will be seen from the above account, that this process, as it occurs in Amia, shows conditions almost identical with those described by Woods and Beard in the Elasmobranchs and less fully by Hoffmann in several species of birds.

In Lepidosteus, on the other hand, the sex-cell migration is strikingly different in many important features. The sex-cells are distinguishable only at a much later period of development than in Amia. They are first to be seen in the endoderm of the hind-gut, becoming clearly distinguishable only when the yolk material of the surrounding cells has become so far absorbed as to more sharply reveal the cell boundaries and at the same time render the yolkfiUed sex-cells more distinct by contrast with their neighbors. The sex-cells at this time (embryo of 7.5 mm. length) are found in the <^ndoderm of the hind-gut, from its proximal almost to its caudal end, — chiefly in its lateral and dorsal walls. They differ from the ordinary endoderm cells surrounding them, not only in their much greater yolk content, but also in their greater size and in their very distinct and well rounded cell outlines. There are at this time no striking nuclear differences between the two classes of cells.

\Mien the embryo attains a length of about 8.5 mm., the sexcells begin to migrate from the gut endoderm up into the loose mesenchyme, dorsal to it. This migration continues for a considerable period — embryos from 8.5 to 9.5 mm. — until all but relatively few of the sex-cells have left the endoderm. During this period, the lateral plates of mesoderm gradually split to form the body cavity, and by the slow approximation of the splanchnopleuric plates, the mesentery is being formed. As these splanchnopleuric layers approach one another from both sides, they gradually compress between them the loose mesenchyme in which the sex-cells lie, until finally when the larva has reached a length of about 12 mm., the tissues have become so condensed and the mesentery so narrow that it would seem no longer possible for the sex-cells, remaining behind in the endoderm, to migrate through it While the mesentery is slowly forming, the sex-cells are migrating up through it to finally come to rest in the sex-gland anlap:en on each side of its dorsal end (radix mesenterii). Some very clear instances were found, in which these

D Proceedings of the Association of American Anatomists. 231

sex-cells showed definite amoeboid form. This seems to indicate their mode of progression. In Lepidosteus, as in other forms studied, some sex-cells go astray, never reaching the sex-gland anlage. In the oldest stage studied, sex-cells were seen lying in the endoderm and mesoderm of the intestine.

In a recent paper upon Bufo by Miss King, '08, and a slightly earlier paper of the author's upon Kana, the sex-cells were shown to migrate en masse, instead of singly as in Lepidosteus. Aside from this, the sex-cell migration of these two amphibians shows striking points of similarity with that process as observed in Lepidosteus, in that the sex-cells are first recognizable in the endoderm, in which they undoubtedly have their origin; furthermore, they migrate at the time when the mesentery is forming. In Kana and Bufo on the one hand, and in Lepidosteus on the other, the sex-cells are quite similar in appearance; this applies to relative size, yolk content and sharp boundaries. The absence of indications of cell division, during the migration process, is characteristic of the sexcells of aU vertebrates studied.

In a previous paper, '06, I showed that the sex-cells of the turtle (Chrysemys) arise on each side of the caudal half of the embryo in the extra-embryonic endoderm at the junction of the area opaca with the area pellucida, and that they migrate medially from these anlagen, continuing in the endoderm as they do so — ^pushing their way among the surrounding endoderm cells to a position immediately beneath the axis of the embryo. From this point they migrate dorsally through the developing mesentery to the sex-gland anlagen.

This process is essentially like that observed in Lepidosteus and Kana, with the difference that the sex-cells in Chrysemys can be easily traced back into the early stages, when they are seen to lie in the extra-emiryonic endoderm. It is quite probable that further investigation in Lepidosteus and Kana may show a more or less extensive migration within the endoderm from more lateral anlagen to the median line beneath the mesentery.

The sex-cell migration in Acanthias, as set forth by Woods, differs from the corresponding process in Chrysemys only in the fact that in that form the sex-cells migrate from their anlagen into the meso

D 232 The Anatomical Record.

derm immediately above, in which their further migration is accomplished. In Wheeler's, '99, very suggestive work upon Petromyzon, he shows that the sex-cells arise at some distance on each side of the axial plane of the embryo in a region where the lateral plates are not yet split off from the endoderm. When the splitting is finally carried to such a point as to involve the sex-cell anlagen, the sex-cells adhere to the mesoderm, through which they finally migrate to the sex-gland anlagen. While this is true, they closely resemble the endodermal cells, their large size and great yolk content being in sharp contrast with the small size and small yolk content of the mesodermal cells, among which they lie. I think, as Wheeler suggests, that we are justified in considering this to represent a very precocious migration from the endoderm into tissues potentially mesodermal.

However striking these differences may be, they are not of fundamental importance. The really important generalizations, toward which these facts point, are (1) That the sexKjells are formed in the endoderm in the forms mentioned in this paper, in some forms certainly, and in others possibly, at some distance on each side of the median line. (2) They migrate from these anlagen to the sexgland anlagen. This path may carry them either through the endoderm or through the mesoderm. (3) This migration is so timed that the sex-cells pass through the anlage of the mesentery at the time when it is forming. However interesting from a morphological standpoint it may be to ally the sex-cells with the endoderm, I do not wish to have it understood that I deny the specific character of the sex-cells in contradistinction to the somatic cells. Experimental work may ultimately help us in distinguishing between them, even though they show no visible morphological differences in early stages.

D Proceedings of the Association of American Anatomists. 233

ON THE GROWTH OF THE ALBINO RAT (MUS NORVEGICUS VAR. ALBUS) AFTER CASTRATION. By J. M. Stotsenburg, M.D., Curator and Junior Associate in Anatomy at The Wistar Institute.

No systematic study of the growth of any mammal after castration has yet been reported. The existing literature on castration deals mainly with its application to domestic animals for economic purposes. There are, however, some general descriptions and measurements made during life on human castrates; Matignon ('96), Pelikan ('76) and Jameson ('77), also several reports of dissections of eunuchs; Ecker ('64r'65), Gruber ('47), Lortet ('96), Becker ('99), Tandler and Grosz ('09), together with a considerable number of investigations on animals showing the dependence of the secondary sexual characters on the integrity of the testes; Eibbert ('98), Eorig ('99, '99A, '01), Sellheim ('98) and Foges ('02).

Further we have some literature, based on animal experiments, touching the interdependence of the hypophysis and of the thyroid gland on the testes: Fichera ('05, '05 A), Richon and Jeandelize ('05, '05 A).

In chickens, rabbits and dogs studies have been made on the growth of parts of the skeleton, especially the growth of the limb bones which become longer than normal: Poucet ('78), Eichon and Jeandelize ('05B, '05C), Sellheim ('99).

Finally, McCrudden ('08) has studied the metabolism of castrated dogs, using the excretion of salts as an index. These experiments show the operation to be without any marked influence on metabolism in this animal.

The chief result of the experimental work is, therefore, to show that many secondary characters in mammals and birds are modified in their development by injury or removal of the testes, and that the latter probably produce their effect through some form of internal secretion, as shown by the observations of Walker ('08). The growth of parts of the skeleton and some of the ductless glands are also in a measure and in some animals affected by castration.

In the present conmiunication, however, we shall consider the influence of castration only in relation to the increase in body

D 234 The Anatomical Record.

weight with age, making incidentally one application of the results to the phenomenon of prepubertal growth in^an.

The observations to be presented were made on the albino rat (mus norvegicus var. albus) and are arranged in three series; one made for Dr. Donaldson in the Neurological Laboratory of the University of Chicago by Dr. S. W. Eanson in 1905-6, and the others by myself at The Wistar Institute in 1907 and 1908.

The comparison in growth was made always between members of the same litter, some of which were castrated, while the others were left intact to serve as controls. All the members of one litter were reared together in the same cage and fed similarly. The diet was ample and varied and included milk, except in series two.

The operation proved to be simple, and was performed on the fourteenth or fifteenth day after birth, at which time the sexes can be easily distinguished.

The males of the litter were removed from the nest and weighed to determine whether they were of normal weight. Each one was then marked upon the pinna of one or both ears. Those selected for controls were returned to the nest, while those assigned for operation were placed in warm cotton. For operation the animal was anaesthetized and the operation conducted under antiseptic precautions : —

The incision was made in the mid-line of the perineum and each testis drawn forward and its connections severed. The wound was washed with bichloride and dressed without stitches with thin celloidin. No case of infection of the wound occurred, and all the operations were successful.

All traces of blood or its odor must be removed before the rat is returned to the nest, otherwise the mother is apt to kill it.

In returning the operated rats to the nest, the mother was first removed and kept away until the operated animals had become satisfied to remain with the balance of the litter and had acquired the odor and warmth of the nest.

The mother was then allowed to enter and was not disturbed for twenty-four hours, after which time it was found there was not much danger that the young would be destroyed by her.

The weight record was taken at regular intervals, increasing

D Proceedings of the Association of American Anatomists. 235

from daily records to those taken once a week. The rats were disturbed as little as possible in the process, the cage always being taken to the balance table, the rat gently placed in a perforated tin box (balanced by a counterweight) and weighed as quickly as possible, and the record set down opposite the record of the distinguishing ear-mark.

It was found best not to attempt to weigh the rats immediately after any unusual excitement in the colony-room, as under such conditions they show a temporary loss of weight; even the presence of strangers may cause them to become unusually restless and easily frightened.

The weighing was continued as long as the animals remained in a healthy condition, and the weights of a litter maintained a comparative uniformity.

During the period between 150 and 200 days, the albino rat is subject to numerous affections which disturb its growth, so it was found impracticable to follow more than a few litters beyond 200 days.

The data for these records are based on 99 animals, of which 52 were castrated and 47 were controls. These fall into three series.

While the observation of no one series was continued during an entire year, the combined records include the 12 months, so that any pronounced seasonal influence, if present, could be noted. No indication of such influence has thus far been observed.

In series No. 1, the records for which were made by Dr. S. W. Hanson at the University of Chicago during the summer and fall of 1905, continuing into the spring of 1906, there were ten litters numbering 40 animals, of which 21 were castrated and 19 were controls. The constitution of the series was the following:



2 3 4 5 6 7 8 9 10

Nnmber of castrated.

Number of controls.





















D Numli^r of castrated.




















236 The Anatomical Eecord.

In series Xo. 2, the records for which were made at The Wistar Institute, Philadelphia, during the summer and fall of 1907, there were 8 litters numbering 27 animals, of which 14 were castrated and 13 were controls. The constitution of the series was the following:



2 3 4 5 i) 7 8

In series Xo. 3, the records for which were made at The Wistar Institute during the winter, spring and summer of 1908, /there were litters numbering 32 animals, of which 17 were castrated and 15 were controls. The constitution of the series was as follows:


9 10 11 12 13 14 15 10 17

It will be impracticable to* give the records for all the litters in each series, but to show how the two groups in each litter change during growth, three examples will be given as rejwesented by litters 2, 4 and 5 of Series 2. The tabulated results are not given, but the curves based on them are shown in Figure 1.

These three records serve to illustrate what took place in all the series, i, e., in some litters the castrated grew faster, in some the con

of castrated.

Number of eor



















D Proceedings of the Association of American Anatomists. 237



> 2

h' i/o

Fig. 1

D 238

The Anatomical Record.

trols, and in others the two curves were nearly identical. The immediate effect of the operation on growth was not detectable. A review of all the litters shows that the castrated surpass the controls about as often as they fall below them.

Moreover in a given litter the incidental variations in the two groups tend to coincide, showing that the castrated rats are just as

TABLE 1. Showing for Series 1 tlie body weiglit based on the average of the litters at different stages. In the fifth column are given the numbers of the litters on which the averages are based, and In the sixth column the number of the litters permanently removed.

Box>T Weight in Gus.


Averaice Aj?e.

Number of



Cafltrate». 17.6

Controls. 17.8


LVed for Averages, ^^j^^^:^}










































1-3. 5-10





1-3. 5-10




1 9

1-3. 5-10





1-4. 6-8. 10










1-7, 9. 10.










1-7. 9.















1-7, 10




1 12






1-8, 10





2-7. 9.















1,4,5. 10.





1, 4. 6-8.





1. 2. 4. 7. 10.





1, 4. 8. ,5, 7. 10.

susceptible as are the controls, to the minor influences modifying growth. See Figure 1.

To determine the general relations of the two curves when all of the litters of a given series are taken together, the following method was used:

The average weights of the individuals in each group of each litter was determined at the time of each weighing.

D Proceedings of the Association of American Anatomists. 239

These averages were tabulated according to age in days.

The results were then averaged for- all of the litters in the series and again tabulated according to age in days.

Since different litters vary widely in their growth, it seemed best in the final averages just mentioned to take the average for each


Showing for Series 2 the body weight based on the averages of the litters, at different ages. In the fifth column are given the numbers of the litters on which the averages are based, and in the sixth column the number of the litters permanently removed.


BoDT Weight


Average Age.

Number of



Castrates. >




15.2 ,



















25.0 ;








28.8 1




30.6 1




31.4 '












43.2 1








54.2 1




44.9 ,




48.6 1




60.6 ;
























106.7 1




110.8 ;




119.0 '




122.4 1








131.9 1




140.4 1












156.6 1












182.6 {



Used for Averages.

' Permanently Removed.

1-8. 1-8. 1-8. 1-8. 1-8. 1-8. 1-8. 1-8. 1-8. 1-6.

1-5, 7. 8.

1, 2, 5-8.

1-4, 6.

1, 3-5. 7, 8.

1, 2, 5-7.

2-4. 6.

1, 3-5, 8.


1, 2, 6, 6.


2, 3. 4, 7.


1. 2. 5, 6.

1-4. 8.


2. 3, 5-8.








1 1


2, 4-7.


2. 4-7.

3. 8

2, 4-7.

2, 4-7.

2. 4, 6. 7.

2. 4, 6.


2. 4.


litter as a unit, and not to weight it by the number of individuals in the litter. The averages of the observations are made at short intervals for the first fifty or sixty days, and then at longer intervals to the end of the series.

D 240

The Anatomical Record.





"^^^ *5 50 -70 — * — rte — ik ijo ii6 — rt

Fio. 2.

D Proceedings of the Association of American Anatomists. 241

The results thus obtained are given in tables 1, 2 and 3 and in Figure 2. In explanation of tables 1, 2 and 3, the following comments are in .place. ^

In the age groupings used, not all the litters are represented every time. This of course tends to alter the direction of the curves, but does not modify the value of the comparison between the castrates

TABLE 3. Showing for Series 3 the body weight based on the averages of the Utters at different ages. In the fifth column are given the numbers of the litters on which the averages are based, and in the sixth column the numbers of the litters permanently removed.

Body Weight in Gmb.


Average Age.

Number of

- . _






Used for Averages.

Permanently Removed.






21 23





16, 17.












16, 17.













16, 17.






16, 17.








































































































9 11,



















9 11.

15, 16.







and controls. A notable instance of this occurs at 46 days in Series 2. See Table 2 and Figure 2. In column 5 of the tables, the litters involved in each average are indicated by their numbers.

Moreover, towards the end of the series, observations on some litters ceased earlier than on others, and the effect on the curve is simi

D 242 The Anatomical Record.

lar to that described above, with the additional effect of permanently reducing the number of cases and hence the general significance of the averages.

Observation in the case of any litter was usually brought to a close by some illness which interfered with the normal growth of one or more of the animals.

In such a case, observations on that litter were discontinued. When this occurred, the fact is noted in the last column of each table.


1. In the case of albino rats, the growth curve for the castrates is similar to that for the normals.

2. Castrates are as susceptible as normals to the incidental influences modifying growth.

3. Castrates are as susceptible as normals to the forms of disease and digestive disturbances which hinder normal growth.

Although these observations show that in the albino rat the normal growth curve is not modified by castration, yet it is not uncouMnonly assumed that in man prepubertal growth is casually related to the maturing of the reproductive system at puberty.

Against this assumption, in addition to the direct evidence furnished by the foregoing observations, the following facts may be adduced :

In man castration is not usually practised before the ninth year (Mobius, '96). Castrates are never described as dwarfed, and are often stated to be heavier (?". e., fatter) or to have longer limb bones than normal (Eeker, 'G-4, '05), (Lortet, '96), (Tandler and Grosz, '98).

The amount of groAvth is then certainly not diminished, and it seems probable therefore that prepubertal growth it not retarded by castration. Indeed there are positive stat<'ments in the literature to the effect that it is increased.

Further in support of the idea that the relation between puberty and prepubertal growth in man is merely incidental, we have the fact that in the rat the corresponding point in the growth curve.

D Proceedings of the Association of American Anatomists. 243

where the females are heavier than the males, comes to end at 50 days, while puberty does not occur until about 70 days (Donaldson, '00) and in the guinea-pig the corresponding period comes to an end at 19 to 31 days and puberty is not attained until 150 days (Minot, '91).

It seems highly probable therefore that puberty in man is not a factor in stimulating prepubertal growth.

It is desirable to emphasize in closing that these conclusions are not to be interpreted as invalidating the view, which rests on good experimental evidence, that the full development of secondary sexual characters in some birds and mammals at least, is dependent on the integrity of the testes which act directly or indirectly through some form of internal secretion.

BIBLIOGRAPHY. Beckeb, 1809. Ueber das Kuochensj'steni eines Castraten. Arch. f. Anat. und Physiol., 1S99, p. 83-112.

Donaldson, H. H., 1906. A comparison of the white rat with man In respect to the growth of the entire body. Boas Memorial Volume, p. 5-26.

EcKEB, A., 1865. Zur Keniitniss des Korperbaues schwarzer Euniichen. Eln Beitrag zur Ethnographie Afrlkas. Abhandl. Senckenb. Xaturforsch. Gesellsch., Fraukf. a. M., 1864-5, vol. 5, 101-112.

FicHEBA, G., 1905. Sur I'hypertrophie de la glande pituitaire consecutive A la castrazione. Policlinlco, vol. 12, 1905.

FiCHERA, G., 1905. Sur I'hypertrophie de la glande consecutive A la castration. Archives ital. de Biol., vol. 43, p. 405.

FoGES, A., 1902. Zur Lehre von den secundiiren Geschlechtscharakteren. Archiv f. d. ges. Physiol., vol. 93, p. 39-58.

Gbubeb. W., 1847. Untersuch. einiger Organe eines Castraten. Arch, ftir Anat. und Physiol., 1S47, p. 4(13.

Jamieson, R. a., 1877. Chinese eunuchs. Lancet, vol. 112, July 28th.

LoRTET, 1896. Allongenient des nienibres inferieurs dft h la castration. Arch. d'Anthropol. criminelle, vol. 11, p. 301-304.

Matigxon, 1896. La castration industrielle en Chine. Gaz. hebd. d. sc. m6d. de Bordeaux, vol. 17, p. 403.

McCbudden, E. H., 1908. The effect of castration on metabolism. Am. Soc. of Biological Chemists, Science, X. S., vol. 27, No. 091, March 27th.

Minot, C. S., 1891. Senescence and rejuvenation. J. of Physiol., vol. 12. Xo. 2, p. 97-153.

D 244 The Anatomical Record.

MoBius, P. J., 1906. Beitriige zur Lehre von den Geschlechtsunterecliieden.

I'eber die Wirkimg der Castration. Halle, C. Marhold, 119 pp. I'ELiKAN, E., 1876. Gerichtlich-medicinisdie Untersuchungen aber das Skop zenthum in Russland. Nebst historischen Notizen. Mit Gen^migung

des Verfassers aus dem Russlschen in*8 Deutsche tibersetzt von Nicolaus

Iwanolf. J. Ricker, Giessen. PoNCET, 1878. De Tinfluence de la castration sur le developpement du

squelette. Assoc. Franc, pour Tavanc, d- Sc. C. r., 1877. Paris, 1878,

vol 6, p. 893. RiBBEBT, 1898. Ueber Transplantation von Ovarium, Hoden and Mamma.

Arch. f. Entwicklungsmecli. d. Organ., 1898. RicHON, L., AND Jeandelize, P., 1905. Action de la th^roldectomie et de

cette operation combing avec la castration sur les os longs des membres.

Comparaison avec les effets de la castration. C. R. Soc. Biol., Paris, T.

58, p. 1084-1085. RicHON, L., AND Jeandelize, P., 1905. (A) Remarques sur la t^te osseuse

d'animax tbyroidectomis^s dans le jeune flge. Comparaison avec les effets

de la castration. C. R. Soc. Biol., Paris, vOl. 58, p. 1087-1088. Richon, L., and Jeandelize, P., 1905. (B) Remarques sur la t^te osseuse de

lapins adultes castr^s dans le jeune age. C. R. Soc. Biol., Paris, vol. 58,

p. 1086-1087. Richon, L., and Jeandelize, P., 1905. (C) Castration pratique chez le lapln

jeune. Etat du squelette chez Tadulte. Examen radiographique. C. R.

Soc. Biol., Paris, vol. 58, p. 555-557. RoRiG, A., 1899. Welche Beziehungen bestehen zwischen den Reproduktlons organen der Cervlden und der Geweihbilduhg derselben. Arch. f. Ent wicklungsmech. d. Organ., vol. 8, p. 382-447. RoBiQ, A., 1899. Ueber die Wirkung der Kastration von Cervus (Cariacus)

mexicanus auf die Schftdelbildung. Arch. f. Entwlcklungsmech. d. Organ.,

vol. 18, p. e33-641. RoRiG. A., 1901. Abnoriiie Geweihbildungen und Ihre Ursachen. Arch. f. Entwlcklungsmech. d. Organ., vol. 11, p. 225. Sellheim, H., 1898. Zur Lehre von den secundilren Geschlechtscharakteren.

Beitr. zur Geburtsh. und GynJlkol., vol. 1, p. 229. Sellheim. H., 1899. Castration und Knochenwachsthum. Beltr. zur Geburtsh.

und Gynakol., vol. 2. Tandler. Julius, and Gbosz, S., 1909. Uober den Einfluss der Kastration auf

den Organismus. I. Besclireibung eincs Eunuchonskelets. Archiv f. Ent.

der Organismen, vol. 27, p. 35-61. Walker, C. E., 1908. Enlargement of hen's comb produced by Injections of

testicle extract. Lancet, vol. 174, p. 934-935.

D Proceedings of the Association of American Anatomists. 245


A comparison of the brain and spinal cord weights in the grey rat with those in the albino rat shows that the former has a much heavier central nervous system than the latter. The diflference is considerably greater in the brain, than in the spinal cord weight. In the case of the brain weight, the difference appears at an earlier period of life (little over ten days after birth), while in the case of the spinal cord weight it does not appear distinctly until the body weight becomes about 180 grams, that is at the time when the body length in the grey rat becomes greater than in the albino rat So far as the weight of the brain is concerned, the present investigation confirms the observations of Darwin and of Lapicque that the brain weight is proportionately smaller in the domesticated than in the wild race from which it is derived. The explanation suggested by Darwin, that this may be due to the effect of disuse, seems inadequate.


It was the purpose of this investigation to obtain primarily a linear measure of growth in the albino rat. This being obtained, it would be possible first to get for this form the ratio obtained by dividing the body length by the body weight; second; to relate the weight of the brain and of the spinal cord to the body length and, finally, to compare the growth in body length in the rat with the corresponding growth in sitting height in man. For the mathematical treatment of the results, I am indebted to my colleague Dr. Hatai.

The body length was measured from the tip of the nose to the root of the tail (on the ventral side), the animal lying relaxed on

D 246 The Anatomical Record.

its side. When the data are arranged in groups diflfering by 10 grams in body weight and 10 mm. in body length, the coefficient of correlation between body weight and body length is found to be .90.

For a given body weight the male has a slightly greater body length. This explains the slightly greater weight of the central nervous system in the male.

When the data are arranged in groups differing by 10 mm. in body length and 0.1 grams in brain weight, the coefficient of correlation is found to be .86.

When the data are arranged in groups differing by 10 mm. in body weight and 0.04 grams in spinal cord weight, the coefficient of correlation is very high, being .90.

The curve for the increase in the weight of the spinal cord, according to stature, is a straight line. This departure from the usual form of the growth curve is largely due to the passive elongation of the cord in response to the lengthening of the vertebral column. Body length is the best datum thus far found from which to infer the weight of the brain or of the spinal cord.

The body length in the rat corresponds with the sitting height in man and during active growth both increase in nearly the same proportion, indicating that in both forms the spinal cord is subject to a similar amount of passive lengthening.

D D D Proceedings of the Association of American Anatomists. 247

A MODERN METHOD OF TEACHING THE ANATOMY OF THE BRAIN. By H. J. H. HoEVE, M.D., Professor of Anatomy in Drake University, Des Moines^ Iowa.

In a recent number of the Anatomical Record (Vol. 2, No. 8) Professor Johnston described a method of brain dissection showing the architecture of that organ more effectively than is possible with the methods ordinarily used. My method (which was in part demonstrated at the last meeting of the Association of American Anatomists) is designed for a similar purpose, but differs considerably in the manner and order of procedure. It is here presented, as fully as the limited space will permit, in the hope that it may be of service to those interested in the teaching of this difficult subject.

The different steps in the technique of the removal of the brain 'are well known, but there are a few points which seem to me of great value. After the tentorium cerebelli is cut and the lower part of the medulla also, and the brain ready to be delivered, I cut transversely across the sinus occipitalis, that is just inferior to the confluens sinuum, in order that the falx cerebri and the tentorium cerebelli may remain attached to the brain without being loosened. I ligate the anterior end of the sinus sagittalis superior, the distal end of the sinus occipitalis and the cut ends of the sinus laterales. The falx and the tentorium are left in place on account of the desirability of preventing the rupture of the Vv. maghae (galeni) and the Vv. cerebri superiores; and the ligation of the cut extremities of the sinuses is to prevent the outflow of venous blood. The importance of these two steps is readily appreciated if we remember that the gray matter of the brain is of a much darker color in those brains where the venous blood is retained. The color of the gray matter depends first upon its own pigment and second upon the amount of blood contained.

After having experimented with brains macerated in dilute nitric, hydrochloric and acetic acid, alcohol, frozen, boiled in oil and some slowly hardened in alcohol, etc., I find no chemicals which will harden the brain in such a way that its fibres can easily be dissected. Fixing the brain in formalin and then hardening it in alcohol has its good points. Formalin and glycerine give pretty fair results.

D 248 The Anatomical Record.

but it takes too much time. For these reasons I believe that for general purposes the fresh brain should be hardened in formalin. Put the brain in a 1 per cent, solution and change the second day to a 2 per cent, solution. Change this every day until the brain is hardened, and then increase the strength gradually up to 6 per cent, in which the specimen can be kept indefinitely. The brains of bodies injected with zinc chloride are not as good for our purposes as those injected with formalin, carbolic acid and glycerin. The brain must be hardened . slowly, and that can readily be accomplished by using the weaker solutions of formalin. If the brain is put at the beginning in a 6 per cent, solution of formalin, then frequently the outside hardens to the depth of about an inch and the inside will remain soft, at least too soft to be fit for the fibre dissection.

For Instruments Used for Fibre Dissection, I have never found anything as handy as a small stick of orangewood (which comes in manicure sets), the point of which is flattened, and the heavy extremity left as it is. With this instrument the fibres of the brain can be elevated nicely and rolled out of their respective positions, which is not possible with the handle of the scalpel, forceps or the toothpick. The heavy end of the stick can be used to elevate larger parts of tissue, especially for the removal of small association fibres, which extend from one gyrus into another by curving around the bottom of the sulci. A pair of forceps'is used for the removal of the pi a and of small pieces of tissue.

The method of dissection which I follow consists of breaking, teasing and cutting the brain tissue. As soon as fibres are exposed, the orangewood stick is used, which seems to be just hard enough to handle the fibres nicely without cutting or tearing them. (Metal will break every fibre which it touches.) After a small bundle of fibres is loosened carefully it is taken hold of with the fingers if possible and the point of the stick put behind it in such a way as not to touch the fibres beneath it, and then the fibres are carefully and slowly removed parallel to the direction of the bundle to which they belong. All the association bundles can readily be dissected in this fashion.

Space will not permit to give a complete outline, but I will attempt to give a general idea of the order of dissection. After having dia

D Proceedings of the Association of American Anatomists. 249

sected a good many brains according to the method of F. J. Gall and J. G. Spurzheim (1834) following the fibres from below upward into the brain, I came to the conclusion that the association bundles were destroyed in every case, and that a good many other structures were also damaged by that method. So after making several attempts to save them, I concluded that it would be better to dissect them first, and that is my reason for starting the dissection at the upper part of the cerebrum.

The students dissect first the dura mater encephali, the processus durae matris, the sinus durae matris, the emissaria and the cavum subdurale encephali. Then comes the arachnoidea encephali and the granulationes arachnoideales, the cavum subarachnoideale and the pia mater encephali. Next all the arteries of the brain are exposed. The venous systems of the brain are next followed out in detail. Study and remove the pia, falx cerebri and tentorium. Remove the cerebrum by a cross section just above the midbrain. 'Then separate the cerebral hemispheres by a sagittal section through the corpus callosum. Study the external surface of the cerebral hemisphere, including successively the lobes, gyri and sulci, the Bulbus olfactorius, Tractus olfactorius, Trigonum olfactorium. Area parolfactoria. Substantia perforata anterior, Chiasma opticum, Tractus opticus, Lamina cineria, Tuber cinerium, Hypophysis cerebri. Corpora mammillaria, Substantia perforata posterior.

Make a horizontal incision one-half inch deep on the mesial surface of the hemisphere one-half inch above the dorsum corporis callosi. Insert the fingertips into the incision and tear the cortex above it upward and outward. Look for fibres of the fasc. occipito-frontalis running in an antero-posterior direction. Gently remove the cortex from the gyrus cinguli, the gyrus hippocampi and the uncus, and find the fibres forming the cingulum.

Fasc. Occipito-Frontalis (Forcli). — The horizontal fibres whioh became exposed when the upper part of the hemisphere was removed are found most easily at a point one-half inch external to the junction of the middle with the posterior one-third of the corpus callosum and can be followed backward, downward and inward (Tapetum).

Fasc. Perpendicularis (Wcj-niclce). — Break the gyri of the external surface of the lobus occipitalis and find perpendicular fibres,

D 250 The Anatomical Kecord.

three-fourths inch internal to its external surface and one inch anterior to the polus occiptalis.

Fasc. Longitudinalis Inferior. — Develop and expose the fibres of the Fasc. longitudinalis inferior by boldly following them forward and backward to where it interlaces with the posterior end of the fasc. long, sup., and the lower end of the Fasc. perpendicularis.

Fasc. Uncinatus. — Scrape carefully through the gray matter of the insula and find fibres which extend from before backward and downward external to the claustrum. The lower ones form an arch (Fasc. uncinatus). •

Xext in order come the Corpus Callosum, Ventriculus Lateralis, Septum Pellucidum, Cavum Septi Pellucidi and Fornix. Make an anterio-posterior incision through the corpus callosum one-fourth inch lateral to its cut margin and, inserting the fingertips, lift the white brain substance external to the incision upward and outward, in order to expose the ventriculus lateralis. Remove the outer wall of the cornu inferius et posterius ventriculi lateralis. Cut the corpus callosum transversely and remove it. Follow the colunma fornicis downward and find its fasciculus olfactorius as it passes over the commissura anterior. Cut transversely through the corpus fornicis, one-fourth inch posterior to the foramen interventriculare (Monroi) and dissect it backward. Study the Commissura Hippocampi, Verga's Ventricle, Tela Chorioidea Ventriculi Tertii, Ventriculus Tertius. Find the commissura anterior and then follow the columna fornicis from below upward until the commissura is reached. Scrape the commissural fibres and follow them clear back into the lobus occipitalis, where they interlace with the fibres of the fasc. long. inf. and the cingulum.

Study the following structures: Thalamus, Nucleus Hypothalamicus (Luysi) Commissura Anterior, Claustrum, Nucleus Amygdalae, Capsula Interna, Corpus Striatum, Nucleus Lentiformis, Nucleus Caudatus. Break the cortical and white matter above the insula, from without upward and inward and expose the entire insula. Kemove the insula, the upper part of the fasc. uncinatus, the claustrum and the capsula externa, and expose by gentle scraping the external surface of the nucleus lentiformis. Find the anteroinferior fibres of the pars frontalis laminae superioris capsulae in

D Proceedings of the Association of American Anatomists. 251

ternae projecting forward, just between the antero-superior parts of the nucleus lentiformis and the caput nuclei caudati. By scraping through them, find the two bodies communicating just inferior' to them. Carefully remove the fibres of the capsula interna from between the two bodies and expose the outer surface of the thalamus, but do not injure the cauda nuclei caudati.

Study the Stria Terminalis, Massa Intermedia, Commissura Posterior, Corpus Pineale, Corpus Geniculatum Laterale, Corpus Geniculatum Mediate. Find the taenia terminalis by removing the vena terminalis and follow it downward toward the foramen interventriculare (Monroi). Flap the tractus opticus outward after detaching it. Push the corresponding columna fornicis over to the opposite side and follow the taenia downward to just above the commissura anterior. Find that just external to the columna, above the commissura anterior, and inferior to the foramen interventriculare, it divides into two fasciculi, which should be carefully followed according to the direction of their fibres.

Next proceed to the Mesencephalon-, Pedunculi Cereiri, Substantia Nigra, Tegmentum, Pons, Medulla and Cerebellum, Scrape the fibres away from the midbrain, and expose the substantia nigra. Find that the latter is surrounded by fillet fibres. Expose the nucleus ruber tegmenti and its peculiar relations. Remove the pia and vessels carefully from the entire cerebellum. Make a median sagittal section through the vermis from behind forward and bend the two lobes carefully outward without detaching them. Study the sulci and laminae of the vermis. Examine the ventriculus quartus. Identify the decussatio pyramidum. Separate the lateral halves of the pons and medulla, by a midsagittal section. Remove all the cortical matter from the hemisphaerium cerebelli.

Brachium Pontis,- — Remove the laminae of white matter from the superior surface of the brachium pontis. Remove the tonsilla and the rest of the small leaflets from the inferior surface of the brachium pontis and find the fibres of the Fasc. obliquus separated from the corpus restiforme by the Fasc. transversus pontis. Remove about one-fourth of the Fasc. obliquus pontis, just external to the superficial origin of the nerv. trigeminus, in order to expose the Fasc. transversus fully. Find that the fibres of the Faso. transversus pontis

D 252 The Anatomical Eecord.

foria the greater gart of the vermis superior. Find that by communicating with its fellow of the opposite side it forms a complete ring around the upper two-thirds of the ventriculus quartus. Within this ring are found from behind forward, the fibres of the corpus restiforme, the corpus dentatum, and the fibres of the brachium conjunctivum.

Brachium Conjunctivum. — Kemove the ventricular lining from the internal surface of the brachium conjunctivum and follow its fibres carefully downward and backward, exposing at the same time the corpus dentatum.

Corpus Restiforme, — ^Follow the -fibres of the corpus restiforme by cutting through the brachium conj. just anterior to the hilus corporis dentati. Expose the entire corpus dentatum. Cut through the corpus, just anterior to the tail of the corpus dentatum, and also sever the Fasc. transv. pontis and the brachium conj. at the same level, in order to proceed with the dissection of the pons and medulla.

Xervus Trigeminus. — Eemove carefully the part of the Fasc. obliq. pontis above the nerv. trigeminus and follow the latter horizontally backward and inward. Bend the brachium conj. carefully outward and upward and follow the radix descendens nervi trigemini upward.

The Fasciculi Longitudinales Pontis, — Remove some of the fibrae pontis superficiale and notice that longitudinal fibres (Fasciculi longitudinales pyramidales pontis) pass through the fibrae pontis profundae. Trace the fasciculi longitudinales downward and find that they form the pyramis medullae. Cut the Fasc. long, at the lower border of the pons, and carefully reflect the pyramis medullae forward and downward. This will expose the fibres of the fasc. cerebro-spinalis lateralis and those of the fasc. cerebro-spinalis anterior; and at the same time the decussatio pyraraidum. Find the small masses of gray matter (Nuclei pontis). Cut the fasc. cerebrospinalis lateralis close to the median line (decussatio pyramidum) at right angles to the direction of its fibres.

Lemniscus System, — Trace the lemniscus lateralis downward and forward by removing the fibrae pontis profundae. After the fasciculi longitudinales are removed, a thin layer of transverse fibres (Cor

D Proceedings of the Association of American Anatomists. 253

pus trapezoideum), from which the longitudinal fibres readily separate presents itself. Find that the lemniscus lateralis is continuous with a broad sheet of fibres (Lemniscus medialis) which extends upward in the same plane. Remove all the fibrae pontis profundae and find the lemniscus medialis. A thin strand of fibres (Lemniscus superior) lies anterior to the lemniscus lateralis, at the upper border of the pons. Find that the lower end of the lemniscus can easily be followed downward to a point where it assists in forming the lemniscus interolivaris. Scrape carefully just anterior to the lemniscus lateralis, on the outer side of the brachium conjunctivum, and expose the fibres which correspond to the lemniscus superior. Find that the lemniscus interolivaris consists partly of the fibres extending between the two olivae and find that it consists of a crossing of fibres which can be traced backward, downward and a little outward to the gracile and cuneate nuclei. Find that the fibres which arise on one side in the gracile and cuneate nuclei are continued upward on the opposite side in the lemnisci medialis, lateralis et superior. Follow the anterior external arciform fibres backward over the oliva into the corpus resiforme. Find that some longitudinal fibres (Lamina superficialis fasciculi proprii lateralis) cover the oliva. Find the longitudinal fibres (Lamina profunda fasciculf proprii lateralis) postero-intcrnal to the oliva. Find longitudinal fibres (Fasc. antero-lateralis superficialis descendens) internal to the sulcus lateralis anterior. Find longitudinal fibres externalto the oliva (Lamina superficialis fasciculi proprii antero-lateralis) and follow them downward to just below the oliva. There they join the longitudinal fibres which lie internal to the oliva (Lamina profunda fasciculi proprii antero-lateralis) and form the fasciculus proprius anterolateralis. Follow the fasc. cerebello-spinalis upward from the point where it crosses the sulcus lateralis posterior to the posterior surface of the medulla.

The advantages of this method may be given as follows: The students can accomplish more in less time. They obtain better ideas of the macroscopical structures of the brain. All that can be seen in slices can be worked out by fibre dissection. In reality there is not one structure in the brain which can be displayed to its fullest advantage by any other method.

D 254 The Anatomical Kecord.


The study of the embryology of the Corpus Ponto-bulbare .throws considerable light on the adult body which I described for the human hind-brain.^ In the adult was found a mass of cells and fibres normally present in all brains but showing considerable variations in its size and development. It was described as arising more or less indefinitely from the transverse fibres of the pons mesial to the fifth nerve and gathering into a well-defined bundle which curved into the long axis of the brain behind this nerve and then passed between the seventh and eighth cranial nerves. It continued backward with a dorsal curve, encircling the dorsal cochlear nucleus, and usually ending as a free tip projecting into the roof of the fourth ventricle. The study of its origin shows it to begin at the caudal end and extend cephalad, hence the embryonic description must be reversed. Whereas the mature body varies so greatly in development, often differing on the two sides of the same brain, the embryonic counterpart is remarkably constant. In addition it must be kept in mind that the ponto-bulbar body is present before any pontine formation occurs, appearing first as a thickening in the secondary "Rautenlippe" of His, just behind the dorsal cochlear nucleus. Later the cells migrate over the restiform body and advance to the pontine flexure between the facial and acoustic nerves.

The fact that in the adult this structure varies in amount of development; the fact that in the embryo it is so constant; and finally the fact that at first it is so large when compared to the pons, at once suggests that we are dealing with something which possesses more function in the embryo than in the adult. This study demonstrates that its function is to furnish a path by which tells, arisiiic: at the lateral margins of the medulla, wander to the ventral surface of the brain and form the main mass of pontine nuclei.

In the 20 mm. embryo of the Mall Collection (Xo. 22) immediately

C. R. Essick, The Corpus Poiito-biiU)are, a hitherto undescribed Nucleus In the Human Illnd-hrain. Anier. Jour. Anat., vol. vU, p. 119.

D Proceedings of the Association of American Anatomists. 265

caudad to the dorsal cochlear nucleus, appears a circumscrihed thickening of the secondary "Kautenlippe" in which an active division/ of cells can be made out. This is the point of origin for the pontine cells which travel down the path described for the corpus pontobulbare.

In the 23 mm. embryo (No. 382), the ventricular margin just behind the dorsal cochlear nucleus is rich in mitotic figures, and here due to the active proliferation of cells, the "Kautenlippe" is markedly thickened. From this point as a starting place deeplystaining elongated nuclei extend laterally aroimd the restiform body and pass forward between the facial and acoustic nerves as far as the trigeminal nerve. The sheet of nuclei has become much thinned out in its cerebral portion, so that the layer is only one to two cells thick at the fifth nerve, and the same condition obtains for its mesial border, which is rapidly lost after the sagittal plane of the emergent facial is passed. The distribution of mitotic figures is striking. Numerous karyokinetic figures appear in every section around the central canal, but as soon as this is left, only an occasional dividing cell is made out and the diflBculty in finding them increases the farther cephalad one passes in sections. Clearly then we are dealing not with a growth by extension but an actual migration of cells. This embryo has no pons.

In the 28 mm. embryo (No. 75), the active formation of cells still continues in the domain of the "Rautenlippe" caudal to the dorsal cochlear nucleus and the entire body is much thickened. Cephalad it can now be traced almost to the mid-line on both sides, arching over to the pontine flexure from the lateral portions and presenting an advancing edge of only one or two cells in thickness.

In the 30 mm. embryo (No. 45), the cells of both sides have met and fused across the mid-line directly over the pontine flexure and the primitive anlage of the pons is formed. In the older embryos the addition of cells continues so that there is a heaping up of cells over the mid-line forming the well known crescentic shape which the pons gives in cross-section.

The course of the cells, in sections, is not difficult to follow because of their peculiar affinity for stains, a characteristic of all

D 256 The Anatomical Eecord.

young nuclei, and this property gives a brilliant differentiation of the ponto-bulbar body from the structures with which it oomes into close contact. With the exception of the deeply colored lining which surrounds the ventricular cavity, the porito-bulbar body is the most striking part of sections through this portion of the embryonic medulla and many authors have given good illustrations of the corpus ponto-bulbare, noting its connection with the "Rautenlippe" as well as its extension to the ventral surface of the brain. There has been a failure to connect it with the development of the pons.

This study was greatly aided by sections and dissections of other mammalian embryos where absolutely fresh material is available. This comparative work has simply confirmed the above findings.

A migration of cells from the dorso-lateral margin of the medulla to the ventral portion of the brain has been described by His* for the complex which he designates "die zerrissenen Kerne." To these belong the olives and their neighboring structures. "All of their cells abandon the place where they were originally formed and press through to the mesial lying regions of the medulla. * * * As far as the cells are concerned these structures are, therefore, descendants of the alar plate and morphologically arise from the same longitudinal zone of the medullary tube from which come the higher-lying parts of the brain, i. e., the cerebellum, quadrigeminal bodies, geniculate bodies and the cerebral hemispheres." In the concluding sentence of his classical treatise on the development of the Rhomboid Brain, he mentions his intention to take up the development of the pons and cerebellum and believes that the real key to the development of the former, the pons, will be furnished by the principle of lamina formation. It seems very plausible that this very principle has been observed in the sheet of nuclei arising in the "Rautenlippe" behind the dorsal cochlear nucleus and after encircling the brain for almost 180°, the cells come to rest on the ventral surface of the Rhomboid Brain and later send their processes laterally to form the middle peduncle of the cerebellum. The corpus ponto-bulbare in the fully developed brain represents those

W. His, Die Entwick. des Mensch. Raut., 1891.

D Proceedings of the Association of American Anatomists. 257

cells with their processes which have not descendend to the pons, but lie scattered along the channel where in embryonic life an active migration of cells took place from the ^^Kautenlippe" of the fourth ventricle to the pontine flexure.

THE NERVUS TERMINALIS IN TELEOSTS. By R. E. Sheldon, The University of Chicago, and Chas. Brookoveb, Buchtel College.

Eecently, in some preparations of the olfactory apparatus of the carp (Cyprinus carpio), prepared at the Ohio State University Lake Laboratory at Sandusky by Mr. T. S. Jackson, some three hundred ganglionic cells were observed along a separate and distinct strand of the olfactory nerve. This condition was noted in an adult individual about twenty-five centimeters long. The cells are somewhat larger than the sheath cells of the olfactory fibers, with the Nissl. granules appearing rather indistinct. They are situated on the ventro-median side of the nerve about half way between the olfactory bulb and the olfactory capsule. Cells can be trafeed, however, caudad to the glomerular region of the bulb, the f ormatio bulbaris, and rostrad nearly to the capsules.. The cells diminish in number rapidly as one passes caudad or rostrad from the main group of ganglionic cells. It should be noted that these cells correspond in position and appearance to those described in the ganglion of the nervus terminalis (nerve of Pinkus) in Amia by Brookover.^ Coarse fibers similar to those in Amia can be traced from the ventro-median side of the bulb rostrad to the olfactory mucous membrane where they are distributed to all parts of the nasal capsules with the main rami of the olfactory nerve. In Cajal preparations these fibers impregnate to the exclusion of the olfactory so that they are easily followed. They are most numerous in the mid-rib between the two series of secondary folds or lamelte. The main bundle of fibers is closely associated with an artery.

Later the ganglion has been found in a young carp, two centimeters in length. Here it appears as a compact elliptical mass of

'Brookover, Chas., 1908. Plnkus's Nerve in Amia and Lepidosteus. Science, N. S.. Vol. 27, No. 702, June 12, 1908, pp. 913-914.

D 258 The Anatomical Record.

large cells on the ventro-median side of the olfactory nerve just before it leaves the cranial cavity. This differs decidedly from the condition in the adult, where the cells are scattered, intermingling with the fibers of the olfactory nerve. Toluidin blue and thionin preparations bring out the cells with especial distinctness.

In the young of a species of Ameiurus about twenty-five millimeters long the ganglion has likewise been found.

In Weigert, vom Rath and Cajal preparations of the adult carp brain an unmeduUated tract can be traced caudad from a region closely associated with that in which these cells and fibers are found. At a point midway between the caudal and cephalic ends of the olfactory bulb it can easily be distinguished as it lies in the midst of a mass of medullated fibers near the ventro-median margin of the bulb. Farther rostrad, however, these myelinated fibers end and the tract is lost among the unmedullated fibers of the olfactory nerve in the region where the ganglionic cells first appear. Throughout the long olfactory crus the tract is plainly evident on the ventromedian side partly enclosed by medullated tracts. On reaching the hemispheres it turns dorso-laterad, still closely associated with one of the medullated secondary olfactory bundles, the tractus olfactolobaris medialis. At the level of the anterior commissure the tracts from the two sides turn abruptly mesad and largely decussate in the mid-line, apparently ending in a dense mass of small cells at the meson. Probably part of the fibers do not cross the mid-line but end on the same side.

It should be borne in mind that the connection between this tract and the peripheral ganglionic cells and fibers has not been established and can not be except by fortunate Qolgi or Cajal preparations. There is very little doubt, however, that this is the nervus terminalis, or nerve of Pinkus, for the following reasons. The peripheral ganglion and fibers are practically identical with those found by Brookover^ in Ami a in connection with this nerve. The condition is likewise very similar to that found by Pinkus, ^94,* '95,' in

»Loc. cit.

•Pinkus, Felix, 1894. Ueber elnen noch nicht beschrlebenen Hirnneryen des Protopterus annectens. Anat. Anz., Bd. 9, Nr. 18, 1894, pp. 562-566.

•Pinkus, F., 1895. Die Hlrnnerven des Protopterus annectens. Morph. Arh., Bd. 4, lift. 2, pp. 275-346, Taf. XIII-XIX.

D Proceedings of the Association of American Anatomists. 259

Protopterus, by Sewertzoff, '02,* in Ceratodus and by Locy, '99,*^ '03,® '05,^ '05,® in Selachians. The central tract follows a course similar to that described by Locy for the nervus terminalis in Selachians where the central connections have been established in some detail. Especially strong support comes from the findings of Herrick, '09,® in the frog where the central course of the tract is almost identical with that in the carp and where the nerve leaves the brain rostrad to run in the meninges so that there can be no doubt as to its character.


A nerve is here described in the frog which corresponds in its intra-cerebral course very closely to the new nerve found by Pinkus in Protopterus and by Locy in Selachians and termed by Locy the nervus terminalis. Its relations are essentially similar in both larval and adult frogs (Rana pipiens and R. catesbiana). Its fibers enter the cranium mingled with those of the nervus olfactoriua. In the tadpole they enter the rostral end of the olfactory bulb as a compact bundle among other fascicles composed of fila olf actoria destined to terminate in the glomeruli of the olfactory bulb. The nervus terminalis, however, passes caudad through the bulb, making no demonstrable connections with the bulb, to terminate in free arborizations in the lamina terminalis among the cells of the nucleus medianus septi. In the adult frog the nervus terminalis separates from the

^Sewertzoflf, A. N., 1902. Zur Entwlckelungsgeschlchte des Ceratodus forsterl. Anat, Anz., Bd. 21,. Nr. 21, Aug., 1902, pp. 593-608.

•Locy, W. A., 1899. New Facts Regarding the Development of the Olfactory Nerve. 14 flgs. Anat Am., Bd. 16, Nr. 12, 1899, pp. 273-290.

•Locy, W. A., 1903. A New Cranial Nerve In Selachians. Mark Anniversary Vol,, Art III, pp. 39-55, pis. V-VI, 1903.

Locy, W. A., 1905. A footnote to the ancestral history of the vertebrate brain. 5 flgs. Science, N. S., Vol. 22, No. 554, Aug. 11, 1905. pp. 180-183.

•Locy, W. A., 1905. On a newly recognized Nerve connected with the Forebrain of Selachians. 32 figs. Anai. Anz., Bd. 26, pp. 33-63, 111-123, 1905.

•Herrick, C. J., 1909. The Nervus Terminalis (nerve of Pinkus) in the Frog. Report at Assoc. Amer. Anat., Baltimore meeting, 1909.

D 260 The Anatomical Record.

nervus olf actorius ventrally of the olfactory bulbs and passes caudad in the meninges to a point behind all of the glomeruli of the bulb. Here it turns dorsally and medially to enter the ventro-medial wall of the hemisphere, within which it continues caudad as far as the lamina terminalis, where it rises up and clearly decussates among other fibers of the anterior commissure. Its exact terminus was not demonstrated in the adult^ but is probably in the adjacent nucleus medianus septi, as in the larva. The peripheral relations of this nerve of the frog are still unknown.


The paper contained a discussion of the general morphology of the telencephalon and diencephalon from the genetic point of view and suggestions for some revisions of nomenclature. The new facts brought forward concerned the identification of the velum transversum and pafaphysis in mammals and certain changes in the relations of structures in the pars optica hypothalami (His) in the ontogeny of amphibians and mammals. The neural folds in the early embryo meet in a transverse fold, the terminal or limiting ridge, bounding the neural plate in front. Behind this a transverse groove connects the two optic pits on the open neural plate. When the neural plate rolls up into a tube, the limiting ridge forms the lower boundary of the neuropore, and the groove . behind it continues to connect the optic vesicles. This is the primitive optic groove. When the optic tract grows from the retina into the brain, the fibers cross in the limiting ridge to form the optic chiasma. In the side walls of the brain, ridges are formed, which run from the chiasma caudolaterad obliquely across the primitive optic groove and separate the optic vesicles from the pit behind the chiasma. The optic vesicles are, then, left in connection with a pit in front of the chiasma occupying the lower part of the neuropore space. This is the pit which in later embryos and adults has been called, since His, the optic recess. It is such only secondarily. It should be called the preoptic recess, while the primitive optic groove behind the chiasma

D Proceedings of fhe Association of American Anatomists. 261

should be called the postoptic recess. The preoptic recess must not be confused with the neuroporic recess, which occupies the upper part of the neuropore. The postoptic recess has often been confused with the infundibular recess, which, in embryos of all vertebrates, is situated farther caudad and has connected with it the neural part of the hypophysis.

The velum transversum in mammalian embryos lies immediately behind the interventricular foramina, and in later embryos comes to be involved in the plexus chorioideus and lost from view. In all vertebrates it marks the boundary between diencephalon and telencephalon dorsally. Since the optic chiasma is found in the limiting ridge of the neural plate, it occupies the extreme anterior portion of the floor plate of His. If the telencephalon is a complete segment or ring of the brain, as His defined the term, there is no alternative but to include the optic chiasma in it. On the other hand, since all the structures of the telencephalon are formed from the portion of the neural tube in front of the optic vesicles, the telencephalon should not be made to include in the adult anything behind the primitive optic groove. In mammals, when both velum transversum and postoptic recess disappear, the boundary between diencephalon and telencephalon may be described as passing immediately behind the interventricular foramen and the optic chiasma.

The paraphysis is present in pig embryos just in front of the velum transversum as in lower vertebrates.


In Amblystoma punctatum no mouth-pit or stomodeum is found. Instead, when the hypophysial invagination begins, the ectoderm of the mouth plate commences to degenerate and eventually disappears. About the borders of the mouth plate the ectoderm turns in, forming a sort of collar around the entoderm which projects to the free surface. The tucked-in ectoderm constitutes dental ridges which eventually give rise to the maxillary, vomerine and mandi

D 262 The Anatomical Record.

bular teeth. For a long period the cavity of the f oregut is obliterated by the coalescence of its walls and by the time the mouth opening appears the teeth are well formed, and the taste buds are forming. The formation of the mouth opening takes place by a cleaving of the entoderm, which reaches to the free surface as above described, and the mouth is lined by entoderm to the very lips. The teeth,then, pierce the entoderm to enter the mouth cavity. Some time before the mouth opening is formed, taste buds make their appearance on the roof and floor of the oro-pharyngeal cavity and on the inner surface of the gill arches. At the moment that the mouth cleft is forming, entodermal cells begin to arrange themselves into taste buds in the region of the vomerine teeth, on the totigue and close behind the maxillary teeth. All the taste buds of Amblystoma are of entodermal origin. This is an exception to the supposed law that all nervous structures are derived from ectoderm, and the writer believes that other structures, such as the palatal and intestinal plexuses, require to be investigated with regard to their possible origin from entoderm.

Other facts brought out were: the continuity of hypophysial and neuroporic thickenings in early stages; indications of a connection of hypophysis with archenteron; the presence of preoral entoderm and premandibular somite as in selachians; the union of the nasal sacs with the mouth cavity takes place by way of the preoral entoderm.

THE NERVES OF THE ATRIOVENTRICULAR BUNDLE. By J. Gobdon W1L8ON, M.A., M.B. Hull Laboratory of Anatomy, University of Chicago.

As a result of the examination of the bundle in the calf, sheep and pig, the following conclusions were arrived at :

I. Anatomically the atrio-ventricular bundle contains not only a special form of muscle fiber distinct from the ordinary muscle of the atrium or the ventricle but is an important and intricate nerve pathway in which we find :

1. Numerous ganglion cells — monopolar, bipolar, and multipolar — whose processes may pass

a. to adjacent ganglion cells in the bimdle

D D D Proceedings of the Association of American Anatomists. 263

b. to the muscle fibers in the bundle,

c. through the muscle bundle so far as it was examined.

2. Abundant nerve fibers running through it in strands, the processes of which may end

a. in ganglion cells in the bundle,

b. in the muscle plexus,

or may pass through the part examined.

3. An intricate plexus of varicose fibrils around and in dose relation to the muscle fibers of the bundle.

4. An abundant vascular supply with well marked vasomotor nerves and sensory endings.

11. Physiologically it has been shown that the atrio-ventricular band constitutes the pathway which assures the communication of the atrio-ventricular rhythm. When the bundle is sectioned or crushed, the ventricles cease momentarily to beat though they soon regain pulsation but with a rhythm much more slow than that of the atrium. Pathological anatomy supports this view ; the allorrhythmia of StokesAdams disease can be explained satisfactorily by lesions involving this pathway. As a result of these physiological experiments and from these pathological conditions, it has been asserted that the contraction wave must be myogenic. To such a deduction my anatomical findings are opposed. They demonstrate that in these experiments and pathological conditions an important nerve pathway is equally involved with the muscle bundle. Considering the neurogenetic as opposed to the myogenic hypothesis from the anatomical standpoint, one must acknowledge that the very complex nerve constituents of the bundle indicate an important nerve pathway and are very suggestive of an intricate nerve mechanism.

IS THE ATRIO-VENTRICULAR BUNDLE TO BE REGARDED AS A NEURO-MUSCULAR SPINDLE? By J. Gordon Wilson, M.A., M.B. Hull Laboratory of Anatomy, University of Chicago.

The essential anatomical points in the structure of the neuromuscular spindle, namely, its shape, its lymph spaces, its lamellar capsule, its arrangement of muscle fibers, have nothing similar in the atrio-ventricular bundle. To these must be added that the nerve

D 264 The Anatomical Record.

endings of Ruffini so distinctive of the spindle have nothing comparable in the bundle, and that ganglion cells are present in the bundle and absent from the spindle. From this it must be concluded that whatever the physiological significance of the bundle may be, it has anatomically nothing in common with the neuromuscular spindle.

THE INTERSTITIAL CELLS OF THE TESTIS OF AN HERMAPHRODITE HORSE. By Richard H. Whitehead, Anatomical Department^ University of Virginia.

For the material and photographs of this case I am indebted to the kindness of Professor S. H. Gage, of Cornell University. The horse was a pseudo-hermaphrodite colt two or three years old. The general type of the animal was distinctly masculine, while the external genitals were those of a mare. Operation revealed normal female genital passages including a normal uterus, but the essential organs proved to be testes. After the operation of removing the testes, the horse became a useful animal. Histological examination showed that the epithelium of the seminiferous tubules consisted entirely of Sertoli cells, whereas the interstitial cells were (vpically formed and exceedingly numerous. The structure of the organ was thus quite similar to the ordinary abdominal testis of cryptorchids. Granules such as I have described in the interstitial cells of various mammalian testes were not present, but this can be explained by the fixative used — a solution of picric acid in alcohol. The essential features of the case are the coexistence of male char,ieters with female genital passages, and the presence of undescended testes. It furnishes additional evidence in favor of the view that the male characters of mammals are correlated with the interstitial cells of the testis.

D Proceedings of the Association of American Anatomists. 265


In the course of physical examinations made on 2561 students at

Fig. 1. — Case No. X 3172. An example illustrative of the cases of supernumerary nipples as found in the male — in this case occurring on the right side and below the normal nipple.

the Cornell University Gymnasium there were observed 21 cases or 0.82 per cent in which supernumerary nipples occurred. Only three cases were observed in the examination of the first 1082 students and

D 266 The Anatomical Record.

the number of occurrences was probably greater than this as the nipples were not looked for especially at this time. Ten cases, or 0.87 per cent were observed in the examination of the next 1152 students; and

Fig. 2. — Case No. X 3172. An enlargement from the negative In Fig. 1 showing the supernumerary nipple more in detail.

in the remaining 327 students, 8, or 2.45 per cent, were found having supernumerary nipples. These last 327 men were all farmers attending the Agricultural College and it is interesting to note the

D Proceedings of the Association of American Anatomists. 267

fact of such a high percentage in this class of men as compared with men coming from various conditions of life in general.

In two cases the nipples were paired and located between the normal nipples and the umbilicus. In one of these men there were distinct and well developed papillary elevations surrounded by the pigmented areola. The extra pair of nipples in the other man consisted of small lightly pigmented spots about 4 mm. in diameter, in the nipple line and just above the umbilicus. This case is of particular interest as three other cases were recorded, where there was a pigmented spot on one side only and in the nipple line but which the man was positive had always been present, and these were undoubtedly traces of extra nipples, although from their small size and lack of papillary elevation they might readily be overlooked or mistaken for a slight pigmented skin lesion.

Besides the two cases in which the nipples were paired, there were seven of the cases in which they occurred on the right side below the right nipple and eight cases occurring below the left nipple. In two more cases the extra nipple occurred above the normal right nipple just outside the margin of its ai^eola. These two cases had distinct papillary elevations; in one of these cases there was also a trace of an extra nipple in the left iliac region.

The one case remaining had a large deeply pigmented areola about 15 nmi. in diameter with a large distinct papilla. This nipple was situated below and to the left of the umbilicus.

KEIBEL'S NOTE ON INTESTINAL DIVERTICULA. By Fbedemc T. Lewis, Harvard Medical School^ Boston.

In the paper on intestinal diverticula which Professor Terry kindly presented for me at the last meeting of this Association, and in the subsequent publication by Dr. Thyng and myself, no mention was made of a previous note upon the same subject by Professor Keibel. This note, which forms a concluding paragraph in a paper entitled "Zur Embryologie des Menschen, der Affen und der Halbaffen" (Verh. d. anat. GeseUschaft, 1905, p. 39-50), is of such interest and so brief that it may be quoted in full, as follows :

D 268 The Anatomical Record.

"I come now to a peculiar observation, which I first made on the small intestine of ape embryos, between the orifice of the ductus choledochus and caecum. I found here, in the epithelium of the mucosa, peculiar buds similar to taste-buds or perhaps to hairs in their earliest stages. Later these developed into little diverticula. Then I found similar structures in man, Tarsius, the pig and the deer. The further development of the buds or diverticula I have not yet been able to follow. The accompanying figures show a number of such epithelial buds from human and Tarsius embryos. It is strange that these buds and diverticula were not mentioned by Voigt (1899) and Berry (1902). It is important that the development proceeds throughout from the epithelium and that the mesoderm is only secondarily involved. Moreover, the further development of the buds differs in different animals."

The early stages of the diverticula, which Professor Keibel describes as buds, we have called epithelial pearls; it is clear that his note and our paper deal with the same structures.

In regard to the possibility that diverticula in the adult arise from these pockets in the embryo, a publication by E. Hedinger is of interest (Arch. /. path. AmL, 1904, Vol. 178, p. 25-43). After describing several diverticula in the vermiform process of a child at birth, he states: "Our case represents not only the first observation of congenital diverticula of the processus vermiformis, but also, so far as the literature shows, the first certain proof of such a congenital formation in the entire extent of the intestinal tract." However, Hedinger cites several cas(vs of diverticula found after birth which were believed to be congenital, among them a case of diverticulum of the oesophagus described by Glockner. In a human embryo of 22.8 mm. I have seen several diverticula of the oesophagus, and the possibility of their pathological persistence and enlargement must be considered. The structures found by Hedinger in the vermiform process have at least a superficial resemblance to the diverticula of the embryo.

I have been informed that the embryonic diverticula have, for some time, been studied in Professor Keibel's laboratory, and a further publication concerning them may be expected.

D Proceedings of the Association of American Anatomists. 269

FURTHER OBSERVATIONS ON SUBCUTANEOUS AND SUBPANICULAR HiEMOLYMPH NODES. ,By Abthub VV. Meyer, Professor of Anatomy in the Northwestern University Medical School.

Since the only observations which I was able to find in the English language on this subject, is a short summary published in the Proceedings of the Association last year, it seems justifiable to quote from that note :

"On the examination of carcasses of beeves in abattoirs a number of hsemolymph nodes can usually be found lying in the subcutaneous fat over the neck and shoulder and, more particularly, in the region directly anterior to the hip. These nodes vary in number from one to a dozen on each half of the carcass and have the same appearance as those in the lumbar pre-vertebral region. They vary in size from a half to one and a half centimeters, are oval or circular in outline and usually flattened laterally. In color they vary from a bluishblack to a bright red or pale pink. They are usually firm, the blood cannot be express.ed from them by pressure, and injections of India ink fail to reveal any lymphatic vessels. They are most numerous in young cattle and were found in foetuses of twenty-two or more centimeters in length. In old cattle they are generally small or absent altogether.

"Their structure is very similar to that of the haemolymph nodes of sheep, and as wide variations in structure were found to exist. Such differences as exist are minor even in case of developing nodes. In the latter the occurrence of giant cells is particularly noticeable, and, as in the case of developing haemolymph nodes of the sheep, they arise from mesenchyme."

In order to test further the existence of lymphatic vessels in the mature nodes, a number of carcasses which contained large and easily-accessible nodes were selected from a lot of eighty, condemned because of tuberculosis, pyemia, bruises, etc. Various suspensions and solutions including Prussian and methylene blue and India ink were injected into the nodes by means of puncture with a syringe holding twenty-five cubic centimeters. Since the nodes are so large there is no difficulty in avoiding transfixation of the node or extravasation of the fluid. By detaching the barrel of the syringe when

D 270 The Anatomical Record.

re-filling was necessary repeated puncture of the node was avoided and the injection of large quantities of fluid into the same node made possible. In this manner it was determined that all nodes lying over the lateral thoracic region communicated with the intercostal veins and thence with the main thoracic trunks. In a number of cases enough fluid was injected so that it trickled out of the azygos veins at their cut thoracic ends and ran to the floor. Although a pressure of several pounds could occasionally be used after clamping of the efierent vein, it was never possible to inject lymphatic vessels. Furthermore to avoid error pieces of the injected vessels draining the nodes were excised and examined microscopically.

A striking peculiarity of many of these nodes is their firmnessThere is great resistance to the needle, and penetration is generally accompanied by a crunching sound as if cartilage were pierced. This peculiarity, I take it, is due to the extraordinary large trabeculae. Indeed, the most characteristic thing about the microscopic appearance of these nodes is the thickness of the capsule and the extraordinary size of the trabeculse. It is very rare to find hsemolymph nodes in the sheep or goat with trabeculffl of corresponding size.

Lymphatic spaces or vessels were never seen within the nodes; and the widest variations in the quantity of blood and lymphatic tissue exist. In some nodes only small isolated masses of lymphoid tissues were left containing many large full blood spaces which communicated directly with the much larger surrounding mass of red cells. These large areas of red cells, among which almost no white cells were found, were frequently bounded by thick strands of connective tissue in which very numerous, irregular, anastomosing, empty blood-spaces and vessels were found. This was also the case in the thick layers of connective tissue found beneath the capsule of some nodes. The picture in these nodes, in short, was one of depletion of lymphatic tissue and marked sclerosis.

In order to determine at what age these nodes first make their appearance in bovine foetuses a series of dissections were made in the abattoir immediately after evisceration. In all young foetuses it is very easy to strip off the skin after cutaneous incisions. Sinoe the

D Proceedings of the Association of American Anatomists. 271

small embryonic nodes resemble punctate hemorrhages or minute blood clots very closely, a jet of warm water or gentle stroking with the bare hand was made use of to keep the field of observation free from blood. By this means it was not very difficult to distinguish between haemolymph nodes and small clots or the ends of bleeding vessels even when they were mere specks.

Out of several dozen foetuses examined in this manner nodes were detected in those, 22, 28, 30, 32.5, 38 and 40 cm. long and almost always in those at term, although because of economic reasons, only a few of the latter were examined. The number of nodes found varied from one to six and the size from mere specks to one. and a half to two millimeters. Most of them were found in the pre-crural region and all looked like blood clots to the unaided eye. On section they varied much in appearance however. The earliest contained but little blood and this was almost wholly in vessels, while the older specimens contained but little lymphoid tissue but much blood, mostly all of which was scattered among the lymphoid tissue. The trabeculse were larger in older nodes, there was a better-defined capsule of varying thickness and a peripheral sinus containing blood. All the nodes contained a reticulum composed of branching cells, the processes of which were often of considerable length and great tenuity. The reticulum bounding the peripheral sinus or the central spaces often seemed to form a definite endotheliallike layer.

Besides erythrocytes and lymphocytes a cell the size of the polymorphonuclear leucoyte, the cytoplasms of which took an acid stain, was commonly seen. These cells had a round, vesicular nucleus or several irregularly-shaped nuclei, but seldom contained sufficiently definite granules to justify one in classing them as eosinophiles. Much larger cells with like staining reactions containing a single composite nucleus apparently composed of from six to twelve nuclei or several separate nuclei were foimd in almost all sections. Pigmented cells or lymphatic vessels were not found in the developing node.

D 272 The Anatomical Record.

THE OCCURRENCE OF INTRATHORACIC PARATHYROID GLANDS. By Abthub W. Meyer, Professor of Anatomy in the yorthicestem University, Medical School,

It will be recalled that the thymus in the sheep extends from the angle of the jaw to the heart. That a large pyramidal portion lies in the angle formed by the parotid and submaxillary salivary glands and the dorso-cervical muscles, directly caudad from the ventral border of the pinna, and medial to the jugular and maxillary veins. This portion is united by a very slender cord which lies directly ventral to the carotid artery and medial to the jugular vein, with the voluminous inferior cervical portion. The latter fills all the space between the thyroid gland and the thorax. It is bifid at its upper pole, tapers gradually to a point at its lower pole, to adjust itself between the jugular veins at their junction. Directly ventral to this junction, a very short thin cylindrical portion pierces the thorax, thus uniting the extra and intra-thoracic portions. The latter lies to the left of the supra-pericardial lobe of the lung and is about one-half its size. The weight of the thymus at birth is about 7-10 grams, and it often reaches a weight of 85 grams in lambs four months old.

In a report made two years ago it was stated that the parathymus glands of the sheep "may be anywhere in the region occupied by the thymus itself." At that time no intra-thoracic parathymus gland had been found, but, from embryological facts, it was clear that the finding of such a specimen or specimens was merely a matter of careful search. In order to determine the correctness of this opinion, a series of 63 sheep foetuses from 4.9 to 39 cm. were carefully dissected.

In sixty per cent of these cases the parathymus gland was found on the lateral or median surface of the dorso-cephalic portion of the superior part of the cervical thymus. In twenty per cent of the cases it was found on that portion of the thymus which lies between the stylo-maxillary muscles laterally, and the pharynx medially. In this position, it always lay high up, near the base of the skull and frequently remained behind when that portion of the thymus was withdrawn. In eleven per cent of the cases, it lay in a position in

D Proceedings of the Association of American Anatomists. 273

termediate between these ; in six per cent, somewhere in the remaining part of the thymus — extra or intra-thoracic ; and in three per cent of the cases, the gland was not found.

It follows, then, that the parathymus gland in the sheep is in relation with the cephalic surface of the superior cervical portion of the thymus in ninety-one per cent of the cases. Aberrant glands have been found in all parts of the thymus below this region, and it is unlikely that the finding of an intra-thoracic para-thymus will long remain imique.

Accessory parathymus glands were found in three per cent of the cases. One of these was microscopical in size, one barely visible, and the rest about the same size as those in the usual position. The occurrence of accessory parathymus glands is not at all unusual. In a series of several hundred foetuses from 7.8 to 38 cm. dissected in 1906, four were incidentally found to possess accessory parathymus glands. Had a more careful search been made with this object in view, the proportion would undoubtedly have been greater. The largest number of such glands found in a single foetus was two. Since, however, nothing but a lens was used to assist in distinguishing these glands, and since, moreover, the slices of the thymus gland were about three millimeters thick, it is evident that some must have escaped notice. This manner of examination probably also accounts for occasional failures to find a parathymus gland in a given foetus.

The relation of the parathymus to the thymus gland is a very variable one. Usually, it is imbedded partly or wholly in the substance of the thymus and rarely it is found so deeply buried that it can only be found by microscopical examination. In many young foetuses it is connected with the substance of the thymus by one or two stalks composed partly of parathymus and partly of thymic tissue. In the older foetuses, this connection becomes less intimate as a rule, in spite of the increasing size of the thymus.

Among the sixty-three foetuses, one 25 cm. long was found in which an accessory parathymus gland about one millimeter in size was imbedded in the center of the free surface of the intra-thoracic lobe of the thymus. Although a microscopical examination was

D 274 The Anatomical Record.

made later, the color and size of this gland were so typical that there was no doubt about its identity from gross appearances alone. This specimen had a definite, though loose, connective tissue capsule, delimiting it from the thymus, and was typically parathyroid in structure, except that it contained well-defined alveoli apparently containing colloid.

The occurrence of vesicles containing colloid is not common in embryonic parathymus glands. Rarely, however, one or more of these vesicles may have become cystic in structure and sufliciently large to be visible to the naked eye. Such a cyst was found in a parathymus gland taken from a foetus 26 cm. long. This cyst, which was multi-locular, included one-fifth the area of the parathymus and was bounded by a layer of flattened epithelium for a part of its extent.

THE GLANDS OF THE FRONTAL SINUS OF THE SHEEP. By Elbert Glabk, Assistant in Anatomy, University of Chicago,

The frontal sinuses of the, sheep are relatively large and nearly always contain a great amount of mucus. The mucus is derived from the goblet cells of the epithelial layer of the lining membrane and from mucous glands. Glands were found in all the sinuses examined. They are of two types — intra-epithelial glands and glands in the tunica propia. The intra-epithelial glands are small cup-like depressions in the epithelial layer. They are always numerous and are quite generally distributed over the entire lining membrane on both posterior and anterior walls. The glands in the tunica propia are of two kinds — the smaller tubular glands, usually somewhat coiled, and the larger alveolar glands with most often only a single alveolus.

The tubular glands are found in that part of the membrjjtae lining the inferior lateral portion of the sinus — the region of the ostium. The alveolar glands occur for the most part in the lining membrane of the anterior wall. They are most numerous in the upper and upper lateral parts.

Glandular elements also occur in crypts and furrows of the lining membrane. Small lymph nodules are foimd here and there. It

D Proceedings of the Association of American Anatomists. 275

is not uncommon to find in the tunica propia small and large mucous cysts lined with cubical epithelium.

It is possible that there is represented in these grandular structures different stages in the development of the larger mucous glands. A crypt may be considered as an intra-epithelial gland, which extends down below the surface epithelium into the tunica propia. An alveolar gland of the tunica propia may be thought of as a dilated crypt whose walls have become mucus-secreting.

ON THE VARIATIONS OF THE PALMARIS LONGUS MUSCLE. (AN ABSTRACT.) By J. Pabsons Schaeffer, M.D, Instructor in Anatomy, Cornell University Medical College.

The palmaris longus muscle varies in form, origin and insertion. It may also present the interesting condition in which the muscle has undergone cleavage. It is frequently absent on one or both sides.

The muscle may be entirely replaced by a fibrous strand, or be fleshy throughout. It may have its tendon placed proximally and the fleshy part distally, or be fleshy at both extremities with an intervening tendon. It may also have its fleshy part located centrally with proximal and distal tendons.

It at times has partial or complete insertion into the fascia of the forearm. It is also reported to be occasionally attached either to the pisiform bone or to the scaphoid bone.

I have found a duplicity of the muscle, on the left side, in a icadaver, and at another time, on the right side, in a living individual. A triplicity of the muscles has also be<».n reported by different observers.

In looking over the literature on the palmaris longus muscle, I find that much work has been done, on the determination of the presence and absence of this muscle, on the cadaver. I have, however, been unable to find any reference to work done along similar lines on the living individual. To see how closely data, so derived, would agree with data derived from a study of the cadaver, I undertook the examination of 800 living arms.

Out of 2462 cadavers reported in the literature on the palmaris

D 27G The Anatomical Record.

longus muscle, I find that 440 of them had the muscle absent on one or both sides, or a percentage of 17.8.

LeDouble examined 260 cadavers composed of an equal number of males and females, with results as follows :


Cadavers examined. Muscle absent. Hight. Left. Both. cent.

130 (male) (» cadavers ♦ 4.6

• 8 cadavers • 6.1

10 cadavers ♦ 7.0

Total 24 cadavers 18.4

130 (female) 9 cadavers ♦ 0.0

14 cadavers • 10.7

17 cadavers • 13.0

Total 40 cadavers 30.7

200 (male and female) i'A cadavers '^4.6

Fig. 1.


Arms examined. Muscle absent. cent.

200 (male) 34 muscles 13.0

200 (female) 57 muscles 21.9

520 (male and female) 01 muscles 17.5

I now wish to tabulate the results based on a study of 800 arms in the living individual. In each case the writer was reasonably

D Proceedings of the Association of American Anatomists. 277

sure that the conclusions drawn were accurate ; however, this method of determining the presence or absence of the palmaris longus muscle is not entirely trustworthy, because in cases where the muscle was very feebly developed, or markedly altered in its insertion, it may have inadvertently been classed as absent.

Number persons Times muscle was absent Per cent, of

examined. on one or both sides. individuals.

400 (males and females) 120 30.0

375 (males) 112 29.8

25 (females) 8 32.0

Fig. 2.

Number examined. Times mui^cle absent. Per cent of


375 (males) 20 right side 5.3

36 left side 9.6

56 both sides 14.9

25 (females) 1 right side 4.0

3 left side 12.0

4 both sides 16.0

400 (males and females) 20 right side 5.0

40 left side 10.0

60 both sides 15.0

D 278 The Anatomical Record.

Number of anna examined. Times absent. Per cent of


800 (male and female) 82 right side 10.2

102 left side 12.7

750 (male) 77 right side 10.2

95 left Bide 12.6

50 (female) 5 right side 10.0

7 left side 14.0

Total number of arms Total number of Per cent, of

examined. absences. muscles absent.

800 (male and fem ale) 184 (on both sides) 23.0

By a comparison of the preceding tables, it will be seen that data derived from a study of the living individual agree fairly closely with those obtained from a study on the cadaver. The arrangement of results will be noticed to be similar in both cases, t. e,, the greater number of absences occurring in the female, and on the left side in both sexes. The number of times the muscle is absent on both sides exceeds the number of times the muscle is absent on either the right or left side.


The following report is made on material sent over to me by Dr. W. L. Williams, of the New York State Veterinary College, and at his suggestion :

The specimen, from a nine months pig, was received from an Inspector of the Bureau of Animal Industry at Milwaukee. The external genitals were those of a male. The penis was normal. The perineum, however, presented a well developed ridge suggesting the vulva of the female animal. The internal genital organs indicated strongly that it was a case of true hermaphroditism.

As shown in the photograph of the specimen (Fig. 1), there is the usual small corpus uteri with two relatively long and convoluted uterine horns. Dissection revealed a typical cervix and a vagina. Comua, corpus, cervix and vagina possessed a lumen. As implied above, there was no orificium vaginse. The left comu is

D D D Proceedings of the Association of American Anatomists. 279

prolonged into a Fallopian tube, which terminates in a small fimbria which is attached to a body about the size of a small bean shown by microscopical examination to be an ovary. No trace was found of a testis, epididymis or vas deferens on the left side. Upon the

Fig. 1.

fight side, however, there is an evident testis about two and one-half centimeters in length with an apparently typical epididymis and a vas deferens. The right uterine horn resembles that on the left

D 280 The Anatomical Kecord.

side. Its Fallopian tube ends, however, in a diminutive blind sac, which closely applied to the caput epididymidis with which on superficial examination it appeared to be continuous; this, however, was not found to be the case. No ovary was found on this side.

The microscopical examination of the suspected ovary on the left aide was confirmatory of the diagnosis. The middle third or more of the organ was cut out and serial sections made. In the portion so examined, five follicles containing ova were found, one of them being well developed, with a cavity, cumulus, theca and stratum

Fig. 2.

granulosum. A photograph is submitted showing a young follicle (Fig. 2). Numerous small nests of cells resembling follicle cells were also found, but definite recognizable ova were absent. The stroma ovarii was of normal structure. There was no indication of testicular tissue in the sections.

A longitudinal segment of the testis was also cut out and sectioned, revealing the typical structure of a cryptorchid testis. The tubules were composed of a single layer of cells whose inner ends

D Proceedings of the Association of American Anatomists. 281

were vacuolated. The interstitial cells were numerous and typical, as shown in the photograph (Fig. 3).

Adopting the classification of Klebs, the anomaly would be a case of Hermaphroditismus verus lateralis and as such seemed to me worthy of being put on record. As far as I have been able to ascertain, there are five other cases of true hermaphroditism in the

Fig. 3.

pig in which the presence of ovary and testis were determined by microscopic examination; three cases in which ovary and testis were present on both sides, described by Garth^ and by Kopsch and Szymonowski ;^ one case recorded by Piitz* of ovary and testis

Garth, W., *94. Zwei Fftlle von Hermaphroditismus verus beim Schwein. 59 pp. Glesen, C. v. V. Mttnchow, 1894.

"Kopsch u. Szymonowski, '96. Ein Fall von Hermaphroditismus verus bilateralis beim Schweine nebst Bemerkungen fiber die Erstehung der Geschlechtsdrtisen aus dem Keimepithel. Anatomischer Anzeiger, Bd. XII, p. 129, 1896.

•Pflt^, *89. Ein Fall von Hermaphroditismus verus unllateralis beim Schweine. Deutsch. Zeitschr. f. Tiermed., Bd. XV, 1889.

D 282 The Anatomical Record.

on one side only; a fifth case^ described by Duchanek^ in 1894, was inaccessible to me. More instances of hermaphroditismns verus are reported in the pig than in any other mammal save man in which form six cases are on record, the diagnosis being based on microscopic examination and apparently authentic. Three of these are H. nnilateralis (ovary and testis both present on one side only) ; one is H. bilateralis (Heppner's) (ovary and testis on both sides) ; while three are H. lateralis (ovary on one side, testis on the other).

It hardly need be said that no case in mammals is known of the presence of both ovarian and testicular tissue capable of functional activity other than possibly elaboration of internal secretions.


By most earlier writers the disappearance of the embryonal villi from the large intestine of mammals has been associated with the formation of the crypts of Lieberkuhn, the one process being looked upon as closely dependent upon the other. Brand ('77) believed the crypts to be formed by the gradual upgrowth of ridges connecting the bases of the villi and thus leaving between them pits, the initial crypts, which, as the ridges rise, become deeper and deeper. In the large intestine he supposed the ridges ultimately to reach the tops of the villi, thus completely obliterating the latter, whereas in the small intestine the ridges grew but part way up the villi and the latter hence persisted. Patzelt ('82) and KoUiker in his later views ('84) express agreement with Brand. In an earlier view ('61) KoUiker had regarded the crypts as independent tubular dovmgrowths of the epithelium.

Schulze ('97), while agreeing in the main with this explanation, thought that from the bottoms of the pits left between the upgrowing ridges there were additional hollow downgrowths, the crypts being formed, he supposed, thus partly through upgrowth of the ridges and partly through independent downgrowth. Kollmann ('98),

Duchanek, J. O., '94. Hermaphrodltlsmus belm Schwelne. TlerftrztL Centrablatt, Bd. XVII, p. 1, 1894.

D Proceedings of the Association of American Anatomists. 283

going still farther, concluded that the crypts were formed exclusively as downgrowths between the villi and wholly independently of the latter. His results as to this point were confirmed by Voigt ('99) and by Hilton ('02).

Seeking to determine the manner of disappearance of villi from the colon in accord with his findings as to the formation of the crypts, Voigt, without any stated evidence, advanced the view that because of rapid growth there was, it is to be inferred, a stretching out of the epithelium such as to result in the complete retraction of the villi. The same view has since then been stated by others.

From a detailed study made in Prof. Gage's laboratory in 1902 upon the alimentary canals in a complete series of pig embryos, I was able to reach the following conclusions :

The earlier stages of development in large and small intestine are essentially the same, the latter, however, preceding the greater portion of the former considerably. The villi are formed mainly in the way first correctly described by Berry ('00), who worked on Homo, through the progressive breaking up of distinct longitudinal ridges. In the mid-colon, where the villi are latest to form, their maximum relative length is attained in embryos about 17 cm. long. They are then from cylindrical to clavate in form. In embryos successively older than this it was found that the villi soon lose the clavate or cylindrical form and become apically drawn out to a point, there being clearly reduction or shifting downward of the connective tissue core. This process results soon in the collapse of the epithelium covering the apical portion of each villus. The cells of the portion of epithelium thus no longer in contact with the connective tissue are sloughed oflf singly or in groups, the remaining epithelial sheath constantly adjusting itself anew over the reduced villus. This process continues until the general level of the mouths of the crypts of Lieberkuhn is reached, the villi as such thus becoming wholly obliterated.

It was found from a study of ample material placed at my disposal through the courtesy of Dr. Hilton that the villi in the colon of the white rat disappear in essentially the same way as in the pig.

Schirman's statement ('98) that the villi in the colon of the

D 284 The Anatomical Kecord.

guinea-pig consist for the upper portion of their length entirely of epithelium is doubtless to be interpreted in accordance with these results as applying to the stage immediately before disappearance. In order to compare in some degree the relative rates of growth in large and small intestines, measurements of the diameters of definite portions of the lower ileum and of the mid-oolon were made in such specimens as were available. A plotting of curves from the measurements obtained seemed to indicate comparative uniformity in the rate of increase in diameter of the ileum. The curve for the colon lies below that for the ileum at the outset; but, as embryos of the stage in which the villi have attained a maximum development, as before indicated, are reached, there begins an acceleration, it appears, in the rate of growth of the colon with the result that its curve soon crosses over above that of the ileum and diverges from it until the embryos are from 30 to 34 cm. in length. It is during this period of accelerated growth in the colon, that is when the embryos are from 20 to 30 cm. long, that its villi disappear.

THE DEVELOPMENT OF THE VEINS IN THE BODY WALL OF THE PIG. By Helen W. Smith, Anatomical Laboratory, Johns Hopkins University,

This communication was presented by way of a demonstration of specimens showing the development of the veins in the body wall of the pig.

These specimens show in the earliest stages (7 mm.) the capillaries of the limb bud draining partly into the posterior cardinal and partly into the umbilical vein. Following the development of these blood vessels we find that, at different stages, the superficial body wall drains in great part (1) into the posterior cardinal (2) into the umbilical vein, (3) by way of the thoraco-epigastric into the axilla, (4) finally into the internal mammary.

These changes are effected gradually as follows: The umbilical vein shifts forward and the membrana reuniens is seen filled with blood vessels draining into the umbilical vein from the limb bud and the myotomes (as described by Coste, etc.). As the muscle layer invades the membranous lateral body wall the vessels of the

D Proceedings of the Association of American Anatomists. 285

membrana gradually atrophy and disappear in consequence of the formation of a secondary system which carries the blood from the body wall to the axillary region. This is accomplished by the formation of a chain of capillaries along the body wall that anastomose beneath the arm bud with the primitive ulnar. The primitive ulnar is connected anteriorly with the general capillary mesh which surrounds the artery to the limb bud and empties into the anterior cardinal at its junction with the posterior cardinal. These capillaries enlarge to form a good-sized vessel, running from the body wall beneath the limb bud, receiving the primitive ulnar and emptying into the anterior cardinal. In earlier stages it forms a loop around the artery, but later only the central part of the loop persists. This is the thoraco-epigastric vein and is identical with the "external mammary" described in the rabbit embryo by Dr. F. T. Lewis.

This vein increases in size until it, together with the superficial epigastric, which has formed at the same time and with which it is practically continuous, drains almost all the superficial body wall. The inner body wall is drained by the internal mammary vein and deep epigastric, which lie on the mesial edge of the muscle layers and have been carried ventrally with it. They are connected with the dorsal vessels by the intercostals, and have numerous anastomoses with the thoraco-epigastric. These anastomoses enlarge at a point just below the tip of the sternum, with the result that a channel is formed which deflects the course of the blood from the whole lower part of the thoraco-epigastric into the internal mammary, leaving only the stump of the vessel draining into the axilla.

A CASE OF CYCLOPIA.! By R. H. Whitehead. From the Anatomical Laboratory, University of Virginia,

Deified by the ancients, regarded with holy horror in the middle ages, exhibited as curiosities on the shelves of museums in later times, human monsters have always been subjects of great speculative interest. It is only in comparatively recent times, however,

'Preparation demonstrated at the twenty-fourth session of the Association of American Anatomists, Baltimore, Maryland, December 20-31, 1908.

D 286 The Anatomical Eecord.

that they have been treated as objects of serious scientific study. Professor Wilder's (H. H. Wilder. The Morphology of Cosmobia, Amer. Jour. Anat., Vol. VIII, No. 4, 1908) laudable attempt to bring some order into the chaos of our knowledge concerning the genesis of monsters has encouraged me to put on record a case of cyclopia which I have been holding back for various reasons, the principal one being the hope of obtaining more material for comparative study. Wilder believes that he can establish a fairly complete series extending by successive gradations from monsters which are less than one complete individual, like the cydops, through the normal individual to duplicate twins at the other extreme ; and puts forward as a suggestion, rather than a conclusion, the theory that monsters which exhibit bilateral symmetry owe their development to causes inherent in the germ, and not to any pathological agency. While granting the abnormal character of monsters, he applies to the whole series, including the normal individual, the rather startling term of cosmobia, i. e., "orderly living beings." He thus aligns himself on the side of those who hold to a germinal origin of monsters.

The history of the specimen which I am about to describe is entirely unknown to me. It was given to me by Dr. H. B. Stone, of the Surgical Department, who found it stored in a jar in the University Dispensary while that building was undergoing repairs. The entire specimen had been placed in a solution of formaldehyde without any preliminary dissection or embalming, and consequently was not in good state of preservation. The body is that of a female foetus at or near full term, which on examination of the exterior presents nothing abnormal until the head is reached. Here, as the illustration shows (Fig. 1), there is only one eye, and that is median in position occupying the usual site of the nose ; there is no evidence of an external nose or proboscis. Except for this absence of a rudimentary nose, the specimen is a typical cyclops, and could be placed between I and II of Wilder's series (Fig. 1 of Wilder's article). The cranium, it will be noted, is distinctly microcephalic, and the forehead low and rapidly receding.

The palpebral fissure measures 21 mm. in length, which is practically identical with a similar measurement of the fissure taken in

D Proceedings of the Association of American Anatomists. 287

two newborn children. Close inspection of the margin of the lower lid reveals a small semilunar notch in the median line, on each side of which is a punctum lachrymale. Immediately behind the notch there is a small papilla in the fornix of the conjunctiva, doubtless a lachrymal caruncle. The dissection of the orbit discloses a cavity which is quite symmetrical bilaterally. Its floor is formed by an

orbital plate furnished by the two maxillie, in which there is a median antero-posterior ridge indicating, possibly, a line of fusion; an infraorbital canal is present on each side. The roof is formed by an orbital plate furnished by the frontal, which bone lacks a median suture. The outer walls are formed in the normal way. In the median line of the floor just behind the infraorbital margin is a

D 288 The Anatomical Kecord.

small semi-spherical pit, which may be homologized with the lachrymal fossa. While the frontal bone lacks a median suture, there is in the median line a well-marked notch, which doubtless represents the nasal notch. There is complete absence, however, of the elements of the bridge of the nose, as well as of the lachrymals, ethmoids, and nasal fossae. The transverse diameters of the cornea and eyeball were found to be practically identical with those of the two newborn children. There is one lens, one iris; and the optic nerve is also single, entering the orbit through a median foramen at the apex. Of the other cranial nerves which enter the orbit it was quite easy to identify the stumps of the oculomotor, abducens, and ophthalmic division of the trigeminal left on both sides after removal of the brain; only the fourth pair could not be found. Within the orbit these nerves, unfortunately, were so badly preserved that I was not able to follow them to their destinations.

The muscles of the orbit, however, were in better condition, and I think that I was able to detect all that were present. These consisted of three pairs, as follows: 1. A pair of external recti, one muscle on each side of the ball, was easily identified. 2. A pair of inferior recti, below the optic nerve, one muscle on each side of the median line. 3. A superior pair, both of which were to the left of the median line and inserted into the upper aspect of the ball a short distance behind the cornea. In the absence of the nerve supply it did not seem possible to decide whether these should be considered superior recti, superior obliques, or one of each; if it were certain that the trochlear nerves were wanting, I should regard them as superior recti. No vestige could be found of an internal rectus, inferior oblique, or levator palpebrae.

The brain was in such poor condition that it crumbled to bits during removal. A fairly satisfactory inspection, however, could be had before removal, which showed plainly that the forebrain was badly crippled. The cerebral hemispheres were unpaired, and so rudimentary that the midbrain was uncovered; and the olfactory bulbs and tracts were entirely absent. The dura mater was thickened, measuring as much as 5 mm. in some situations.

In speculating upon the conditions underlying the production of

D Proceedings of the Association of American Anatomists. 289

such a monster — and our imperfect knowledge of teratology forces us to speculate, in large measure — the condition of the brain seems to indicate the path to be taken in the present case. We may suppose that the course of events was somewhat as follows : At a very early stage of the embryo, before . even the optic vesicles had begun to grow out from the brain, the embryonic forebrain was subjected to the action of some pathological agency which brought about an atrophy or destruction of the median portion of that vesicle of the brain, as a result of which the optic vesicles, instead of growing out as two anlagen, appeared as a single one median in position, unpaired yet containing potentially, at least, elements of both vesicles. In like manner the cerebral hemispheres made their appearance as a single, unpaired structure. The subsequent changes are susceptible of easy explanation. The median eye growing forward would come to preoccupy the place on the face normally taken by the nose, and the strong tendency of the embryo to bilateral symmetry would account for the other conditions. It does not seem necessary to invoke the fusion of two already formed optic vesicles here, though that process has occurred in some of the experimental work on cyclopia. Thus the genesis of the cyclops described above would be undoubtedly pathological in nature ; and the cyclopia would be, not a primary condition, but secondary to a defect of the brain. Consequently, according to Dr. Wilder's theory, it would not be entitled to be classed as a true teras (cosmobion). This brings up the question, Are we justified in thus excluding from the class of true monsters those bilaterally symmetrical ones which owe their production to pathological conditions? The fact that human cyclopic monsters always have crippled forebrains (Mall, Study of Causes Underlying the Origin of Human Monsters. Jour. Morph., Vol. XIX, No. 1, 1908) seems to negative the question. But, if it be answered in the affirmative, immediately another question suggests itself: What criteria, in fact, have we for the separation — ^where shall we draw the line of demarcation? The experiments of W. H. Lewis (Symposium on Experimental Embryology, Ass. of Amer. An at., Baltimore, 1908) on fish embryos show that cyclopic monsters quite identical with those produced with magnesium chloride by Stockard

D 290 The Anatomical Eecord.

(Stockard, Science, Vol. XXVIII, No. 718, 1908) can be produced by simply removing a small area at the anterior end of the embryonic shield with a needle. It is quite conceivable that a pathological agent might thus act upon a minute area of the forebrain of a human embryo, and leave no trace behind save the cyclopia.

Note. — Since the above was written Dr. Stockard has published the complete account of his experimental work on the production of cyclopic fish monsters. (The Development of Artificially Produced Cyclopean Fish — "The Magnesium Embryo," Jour. Exp. Zoology, Vol. VI, No. 2, 1909.) His experiments show clearly, as do those of Lewis, that cyclopia can be induced readily by pathological agencies. There is, therefore, no necessity for assuming the hypothesis of germinal variation in the casual genesis of cyclopic monsters.



Vol. III. MAY, 1909. No. 5



S. WALTER RANSON, From the Anatomical Laboratory, Northtoestem University Medical School.

It is an accepted dictum of neurology that the cerebro-spinal nerves (with the exception of the olfactory nerves and the nervus intermedins^) consist almost entirely of meduUated fibers, although it is admitted that a few non-meduUated fibers, arising in the sympathetic system, pass into the spinal nerves along the gray rami communicantes. Experimental studies published during the last three years have cast doubt on the correctness of this conception of the spinal nerves.

In a paper on "Retrograde Degeneration in the Spinal Nerves'" it was shown that after division of a nerve containing 1500 medullated afferent fibers there occurred complete degeneration of 4500 spinal ganglion cells. Thus, three times as many cells disappeared as can be accounted for in terms of medullated axons divided at the time of the operation. But this extensive degeneration in the ganglion was accompanied by little or no degeneration in the dorsal roots. In order to find an explanation for these results a careful study of the literature was made and the results presented

'Weififner. Anatoni. Hefte, 1905.

'Ranson. Jour. Comp. Neurol, and Psychol., vol. 16.


D 292 S. Walter Hanson.

in another paper,* which deals with the "Architectural Relations of the Afferent Elements Entering into the Formation of the Spinal Nerves." It was shown that the spinal ganglion consists chiefly of two kinds of cells, large and small. Although both types of cells are unipolar with T-shaped processes, they present a striking contrast in that the processes of the large cells are medullated while

Fig. 1. — Spencer Objective 2 mm. Eye-piece 8 X. A single non-medullated fiber from a longitudinal section of one of tlie lower cervical nerves of a rabbit Note the slightly irregular contour of the fiber and the small nucleus closely applied to it.

those of the small cells are destitute of myelin. The small cells are more numerous than the large ones, so that it becomes of interest to know what becomes of their non-medullated axons. Dogiel saw them divide in the same way as the medullated fibers and traced the central branches into the dorsal root and the peripheral branches to the point where the nerve is formed by the junction

Fig. 2. — Spencer Objective 2 min. Eye-piece 8 X. A bundle of nonmedullated fibers from the sciatic nerve of the rabbit as seen in longitudinal section. Nuclei are associated with three of the fibers.

of the ventral and dorsal roots. As far as he was able to follow them they remained without a myelin sheath. That the peripheral branches do not end at the point of formation of the nerve from its roots, but pass on through the nerve, is indicated by another series of observations* on the "Alterations in the spinal gang 'Ranson. Jour. Comp. Neurol, and Psychol., vol. 17. ^Ranson. Jour. Comp. Neurol, and Psychol., vol. 19.

D Non-MeduUated Nerve Fibers in Spinal Nerves. 293

lion cells following neurotomy." After division of a spinal nerve nearly all of the cells in the associated gaiiglion show axonal reaction, although only the large ones are associated with injured meduUated fibers. Moreover, the small cells are the first to react, show typical axonal reaction, and undergo complete degeneration. The natural inference is that the non-meduUated axons of the small cells extend into the nerv^e.

Fig. 3. — Spencer Objective 2 mm. Eye-piece 4 x. An area of an oblique section through one of the lower cervical nerves of the rabbit. Note the light granular medullated fibers surrounded by a clear sheath and the bundles of dark nonmedullated fibers. The latter can be followed for some distance as they intertwine with each other.

These fibers have now been demonstrated in large numbers in the spinal nerves of the rabbit. In this work recourse was had to a modification of Cajal's method.* . Pieces of fresh nerve 5 mm. in thickness are placed for two days in absolute alcohol containing 1 per cent, of concentrated ammonia; washed 1 to 3 minutes in distilled water; placed for 3 to 5 days in a 11^ per cent. aqueous solution of a silver nitrate in the dark at 37 degrees C;

"Azonlay. La Presse Medicale, vol. 13, p. 75.

D 294

S. Walter Ranson.

washed 3 to 5 minutes in distilled water ; placed for 1 to 2 days in a 1 per cent, solution of hydroquinone in 10 per cent, formalin The tissue is then imbedded hi paraffin and cut into sections which after mounting are ready for examination.

W 11 I ifc>

Fio. 4. — Spencer Objective 2 mm. Eye-piece 8 X- An area of a cross section through the tibial nerve of a rabbit. The nonmedullated fibers are imbedded in the endoneurium in the interspaces between the larger medullated ones. They are much darker than the axons of the medullated fibers and are devoid of any trace of a myelin sheath.

The nerves of the white rat were found to be too small for satisfactory impregnation, but those of the rabbit gave good results. The sciatic, tibial, and lower cervical nerves were examined.

D D D Non-Medullated Nerve Fibers in Spinal Nerves. 295

The preparations secured by this procedure show both mediillated and non-medullated fibers. The former have a light yellow granular axon surrounded by a colorless medullary sheath. The Inon-medullated fibers are dark brown or black and do not present so granular an appearance. They vary considerably in size — some are as large as the smallest medullated axons, others much smaller. They lie in the interspaces between the medullated fibers directly imbedded in the endoneurium and not surrounded by anything resembling a medullary sheath. They show a pronounced tendency to group themselves into bundles, the individual fibers intertwining with each other. Sometimes a fiber can be seen to leave one bundle and pass into another. Because of their small size and compact arrangement into bundles very large numbers of them can lie in the relatively small interspaces between the medullated fibers. They are in fact more numerous than the medullated fibers themselves. Their distribution throughout the nerve, however, does not seem to be uniform. In cross sections their cut ends seem much more numerous in certain parts of the section than in others. It is often possible to demonstrate a nucleus lying along the fiber in immediate contact with it, apparently belonging to a thin neurilema.

Received for publication, March 3, 1909.



RICHARD W. HARVEY. From the Anatomical Laboratory of the University of California,

With 5 Figures and 1 Table.

For some years during the usual class work in histology in this laboratory, comparisons of the distended bladder epithelium with that in the contracted state have given rise to the impression that in the thinning consequent to distension, the cells of the epithelium were not only flattened, but also were to some extent actually displaced from their relative positions. This impression has suggested this paper, the purpose of which is to describe some variations in the wall, and in the epithelium alone, of the bladder and ureter occurring with their distension.

Material and Methods. — The bladder and the attached ends of the ureters were removed from a freshly killed dog. A portion of the bladder was tied off so that the fluid used in distending the remainder could not enter it, and ligatures were applied to the ureters close to the bladder. Then, by means of a large syringe, Zenker's fluid was forced into the bladder, by way of the urethra, until it attained about the condition of maximum normal distension with urine and a ligature was applied to retain the fluid. The distended bladder, together with the contracted portion which had been tied off, were then immersed in Zenker's fluid, and, of course, became fixed in the condition in which they were when immersed. A piece of one of the ureters about four inches long was removed, and a ligature was applied about its middle. Into one end, the syringe, filled with Zenker's fluid, was inserted, a ligature loosely applied and then a pressure of the syringe was applied until no further extension was possible. The syringe was then removed at the same time that the ligature was tightened. The thus distended portion


Variations in Bladder and Ureter. 297

of the ureter, together with the contracted portion, was then immersed with the bladder in Zenker's fluid for complete fixation.

Pieces for study were cut from the contracted and distended bladder and ureter, washed, dehydrated carefully and embedded in paraffine. In cutting the sections, care was taken to cut perpendicular to the surface of the epithelium. The sections were stained in haemotoxylin and congo red, the latter being chosen as the counterstain because of its eflSciency in bringing out cell boundaries.

Variations in the Wall of the Bladder. — ^Measurements were taken of the entire wall of the bladder in the contracted and distended conditions, the averages in micra being recorded in Table 1. A comparison of these results shows that the thickness of the entire wall in a distension about equal to the condition of maximum normal distension with urine, is about one-tenth that in contraction, or decreases 90.7 per cent, with distension.

The thickness of the muscularis in distension is about onetwelfth that in contraction, or it decreases 92 per cent, with distension. The muscle fibres of the bladder are very much extended during distension. In extreme distension the amount of thinning is aided by the power of distension of the connective tissue to which the muscle fibres are attached. The bladder wall, therefore, with regard to its connective tissue component, at least, is in part like a true elastic membrane in extreme distension, in that, to a certain extent, it relaxes of its own elasticity.

Distension of the tunica propria of the bladder, under the same pressure, of course, as the muscularis, reduces its thickness to onefourth that in contraction, or it decreases 74 per cent, with distension. The difference between the variation of the tunica propria and that of the muscularis is probably accounted for by the fact that during the later stages of contraction of the bladder, tunica propria is thrown into folds along with the epithelium.

Variations in the Wall of the Ureter. — In one respect is the distension of the wall of the ureter quite different from the distension of the bladder wall. The distensibility of the bladder has no definite limit. Pressure exerted will gradually and continuously distend

D 298 Richard W. Harvey.

the bladder until its walls become so thin as to burst, so great is the elasticity of the connective tissue. The distensibility of the ureter is more limited. In distending this, a point is reached at which considerably greater pressure than that used with the bladder fails to further distend it, a definitely fixed limit of distension becoming very apparent. This is indicated in Fig 1. A, of this figure, shows the relaxed ureter with the characteristic folds in its epithelium and tunica propria; B, shows the ureter fixed under extreme distension from the application of pressure probably suf tp


A ^^ B

Fio. 1. — ^DrawiDgs of A, contracted and B, distended ureter showing limits of distensibility. The two are drawn to scale; ^i., epithelium; tp.» tunica propria ; i.l.m., inner longitudinal muscle layer ; m.c.m., middle circular muscle layer; o.l.m., outer longitudinal muscle layer; t.a., tunica adventitia.

ficient to burst the bladder wall. The arrangement and abundance of the supporting tissue, and the more organized arrangement of the muscular investment of the ureter, undoubtedly account for the difference between it and the bladder.

Measurements were taken of the entire wall of the ureter, as in the case of the bladder, in the contracted and distended conditions, the average in micra being recorded in Table 1. Comparing these results it is shown that the thickness of the entire wall of the


Variations in Bladder and Ureter.


ureter in contraction is two and three-quarter times that in distension, or the thickness of the wall is reduced 63.7 per cent, by distension, instead of the corresponding decrease of 90.7 per cent, in the case of the bladder wall.

The distended muscularis of the ureter is about one-half as thick as in contraction, or the muscularis is reduced 56 per cent, in thickness by distension, instead of 92 per cent., as was that of the bladder. The distended tunica propria is likewise about one-half


Thickness of i Entire Wall

Thiokne Muscu


SB of


Per Cent. Deer.

Thickness of Tunica Propria

Thickness of Epithelium

Layers of Nuolei




Per 1 Cent.l Deer.


Per j

Cent. Deer.

i Per Micra Cent. Deer.




Cent. Deer.






90.7 , 486.4 I




172.0 27.6

i 84.0

4.37 2.25



1 76.0




1 751.3

1 243.4 :












51.2 1



1 4.00


Table 1. — ^Recording, in micra and percentages of decrease, yarlatlons in thickness of the wall, and yarlatlons In number and percentages of decrease of the layers in the epithelium of the dog's bladder and ureter in the contracted and distended conditions.

as thick as in contraction, undergoing a reduction of 51.2 per cent, in distension. The close approximation of these variations in the muscularis and tunica propria of the ureter is probably accounted for by the fact that contraction of the ureter does not result in such extctnsive folds in its tunica propria and epithelium as occur in the contracted bladder, where the diflFerence in the variation of muscularis and tunica propria is about four times as great as that of the ureter. The average thicknesses in micra of the muscularis and tunica propria are recorded in Table 1.


]00 Richard W. Harvey. Variations in the Epithelium of the Bladder and Ureter. — The contracted bladder epithelium has the appearance illustrated in Fig. 2. The arrangement consists of a basal layer (a) of cubical cells resting on the tunica propria; a middle layer (b) composed of approximately three layers of irregular, polygonal, or elongated cells; a superficial layer (c) of large ovoid cells from whose under surfaces processes fit into the interstices between the cells of layer (b). The nuclei are full and spherical and the cell boundaries fairly distinct.

Variations in Thickness. — Measurements taken of the epithelium of the bladder in the contracted and distended conditions show that




Fio. 2. — Contracted bladder epitheUum, drawn with camera luclda. x '<^50.

the thickness in distension is about one-sixth that in contraction, or that it decreases 84.0 per cent, with distension. Measurements of the ureter show that the thickness of the epithelium in distension is between one-fifth and one-sixth that in contraction, or that it decreases 82.2 per cent, with distension (see Table 1). Comparing these measurements, there is a difference between the two in percentage decrease with distension of only 1.8 per cent. Thus while the distension results in approximately the same degree of distension in the epithelium of the two organs, the effects are noticeably greater on the niuscularis and tunica propria of the bladder than on the correspond

D Variations in Bladder and Ureter. 301

ing parts of the ureter. This difference between the two is explained as due to the fact that the bladder is capable of, and customarily undergoes, greater extension than the ureter and, when fully contracted, its epithelium is thrown into larger and more extensive folds than that of the ureter. The similar effect upon the epithelium of the two is of interest as indicating a probable proportional arrange

tun/cei ^roj^r/d Fio. 3. — Distended bladder epithelium, drawn witli camera lucida. X 750.

ment of the different coats protective against stretching the epithelium to an injurious extent.

In the distended condition, the bladder epithelium appears as in Fig. 3. The basal layer is so flattened that it is somewhat confused with the connective tissue cells of the tunica propria. The middle layer is represented by a few straggling nuclei; and the superficial

D 302

Richard W. Harvey.

layer is greatly elongated and flattened, the interstitial processes being nowhere in evidence. The nuclei are greatly elongated and flattened, and the cell boundaries of only the superficial cells are found distinctly evident throughout Elsewhere the boundaries appear discontinuous and in fragments, as though the cytoplasm of adjacent cells has fused in places, or distension has rendered the membranes so thin as to be invisible.

An observation of the arrangement of the nuclei was found useful




Fig. 4. — Contracted ureter epithelium, drawn with camera lucida. X 750.

in the study of the structure of the epithelium. This observation was made by counting the nuclei lying in the same focal plane from the basal to the superficial layer. The results of counts taken in different regions of the bladder under the same conditions of contraction and distension are recorded in Table 1. It will be seen that distension reduces the number of rows of nuclei 48.5 per cent, or approximately one-half.

The epithelium of the ureter is of the transitional type, but

D Variations in Bladder and Ureter. 303

relatively thicker in the contracted state than that of the bladder, and disposed in more constant folds, Fig. 1, A. In the contracted state, three layers, comparable with those of the contracted bladder epithelium, are defined, Fig. 4. However, the middle layer is composed of at least five rows of cells as compared with three rows in the middle layer of the bladder epithelium.

The distended ureter has the appearance of Fig. 5. In this condition the folds are lacking, the epithelium is greatly compressed, and the cells and cell-nuclei greatly elongated.

As in the case of the bladder, a comparison of the two conditions shows the number of layers of the epithelium to have been diminished through distension 47.8 per cent, or approximately one-half. The average results of the counts of nuclei lying in the same focal plane between and including the basal and superficial layers in differ

Fig. 5. — Distended ureter epitheUum, drawn with camera lucida. X 750.

ent regions of the epithelium of the ureter, under similar conditions of contraction and distension, are recorded in Table 1.

Comparing the percentage decrease of the layers of nuclei in the epithelium of the distended bladder and ureter shows a diiference of only .7 per cent. The approximation of this result to the 1.8 per cent difference in decrease of thickness of the epithelium of the bladder and ureter due to distension, indicates a relatively equal distension of the two organs.

By comparing the variations in the epithelium of the bladder and ureter, it is seen that the thickness of that of the bladder decreases 84 per cent, with distension or about one-sixth of its thickness in contraction, while the layers of nuclei in vertical section decrease only 48.5 per cent, or to about one-half the layers in contraction, and that

D 304 Richard W. Harvey.

the thickness of the epithelium of the ureter decreases 82.2 per cent with distension, or to about one-fifth the thickness in contraction, while the layers of its nuclei decrease 47.8 per cent, or likewise to about one-half the layers in contraction. If cells are only flattened or spread out by distension, then, to be decreased in vertical thickness to one-sixth or one-fifth of its thickness in the contracted condition, as is the entire epithelium, each cell, on the average, would be flattened out to extend over an area five or six times as great as in the contracted state. This would result in a large number of non-nucleated, thin edges of cells in the sections. Such were described by London, but were not observed in this investigation.

An examination of the literature shows that Paneth ('76) followed a series of changes in the bladder epithelium of the dog resulting from different degrees of distension. He described in the contracted bladder the usual three layers of the epithelium, viz., a superficial layer of broad, thick cells, each overlying several of the cells beneath; a middle layer composed of polygonal or trumpet or tooth-like cells in several layers, and a basal layer composed of more or less cubical or cylindrical cells ; and he observed that as the bladder was distended, the cells became flattened, until in the fully distended bladder the epithelium resembled the stratified pavement type. However, he recorded no measurements of the variations nor observed them in much detail.

London ('81) investigated the bladder epithelium of the dog under different conditions of extension, and reached the obvious conclusion that the volume of a single cell, whether contracted or distended, remains the same. He remarked, that, in contraction, the bladder epithelium consisted of five layers of almost cylindrical cells, while in extreme distension it consisted of apparently only one, a layer of flat cells. He accounts for this variation by the supposition that the apparent diminution in layers relates only to the layers of nuclei, that these latter are pushed widely apart by distension, and the flattened cells really exist in their same relative arrangement, but appear as fine lines in section. He concluded, then, that there is no loss of coherence such as would result from a slipping of the cells of one layer between those of the neighboring layers in distension,.

D Variations in Bladder and Ureter. 305

but merely a flattening of the cells of each layer into thin plates. He disregarded the l§iyers of nuclei as indicating the layers of cells. He further reached a rather peculiar conclusion that, while the return to the normal after distension is devoid of any alteration in previous relation and arrangement of parts, just as in an elastic membrane, in one particular the behavior of the bladder epithelium is different, namely, that the epithelium of the emptying bladder possesses a greater elasticity than that of the filling bladder.

Dogiel (^90) claimed that protoplasmic processes bind the cells of one layer to those of another, and Eggeling ('01) described the superficial cells closely applied to the underlying cells with processes extending between the latter.

Herzog (^04), contrary to London, made use of the occurrence and arrangement of nuclei for a study of the layers of the epithelium of the urethra, when under different conditions.

In the preparations used here, the lines interpreted by London as cell boundaries and as representing the thinned edges of flattened cells were not observed, and congo red is considered one of the best counterstains for bringing out cell boundaries.

It is here suggested as possible that in distension, besides the flattening and elongation of the cells due to compression, there is an actual displacement, not only of the cell nuclei, as concluded by London, but also of the cells themselves. An examination of the sections of distended bladder and ureter shows numerous cases of the apparent displacement of the cells of the superficial layer. In Figure 3, for example, there is an indication that adjacent superficial cells have been separated and so that the immediately underlying cells of the middle layer are exposed to the surface. The epithelium, by being thrown into folds during contraction, is evidently allowed to relax more gradually than the muscular wall, and, if such is necessary, a gradual resumption of the former relations of the cells to one another is possible.

Dogiel ('90) pictures a bit of dissociated epthelium from the bladder of the white mouse, stained with picro-carmine, showing filaments establishing actual protoplasmic continuity between the larger superficial cells and the cells of the subjacent layer, and claims

D 306 Eichard W. Harvey.

the continuity thus indicated is a normal relation between the cells. Numerous areas of both partially and totally dissociated cells were to be observed in the folds of the epithelium of the contracted bladder used here, but no protoplasmic filaments as described by Dogiel could be discerned, other than such as could be more safely explained as products of maceration. When maceration is less extensive, though the superficial cells are completely separated (and numerous cases of such could be seen in the preparations), the cells of both layers show smooth boundaries, the superficial layer having only the processes mentioned above as extending between the cells of the subjacent layer. Even these latter processes were absent in the distended condition, the cells having a smooth boundary throughout. Only in manifestly extreme maceration could irregular and finely ragged borders be seen. Zenker's fluid was used for fixing here, directly applied, while Dogiel used Miiller's fluid and stained with picrocarmine. Congo red was used here and this, as a cytoplasmic stain, is far more efficient for fine processes than is picro-carmine.

Conclusions. i. Comparative measurements of the entire wall of the dog's bladder and ureter in approximately the same degree of contraction and in distension by approximately the same pressure show that the thickness of the bladder wall is decreased considerably more than that of the ureter.

2. The extent of distensibility of the ureter is more definitely limited than that of the bladder, undoubtedly due to the more organized arrangement of the supporting tissue and muscular investment of the ureter.

3. The relative effects of distension are much more evident in the muscularis and tunica propria of the bladder than in these tunics of the ureter, while the thinning effect of the distension upon the epithelium of the two is approximately the same, due to the fact that, in the later stages of the customary contraction of the bladder wall, the epithelium is thrown into more extensive folds than that of the ureter, for, during distension, these folds must first be obliterated before the thickness of the epithelium is much affected by distension.

D Variations in Bladder and Ureter. 307

4. The epithelium of the ureter is thicker in the contracted state than that of the bladder, but, approximately normal maximum distension produces in the epithelium of both approximately the same percentage of reduction in thickness (about 80 per cent) arid approximately the same percentage reduction in the layers of nuclei (about 50 per cent) in both.

5. It is suggested that the diminution in the layers of nuclei may possibly indicate, not only the spreading of the epithelial cells resulting from distension, but, that to some extent at least, in the process of compression, the cells may glide upon one another or be slightly displaced as the epithelium decreases in thickness.

In conclusion acknowledgments are due to Professor Hardesty, at whose suggestion and with whose guidance and advice this investigation was undertaken.

Received for publication, March 15, 1009.

REFERENCES. DoQiEL, A. S. Zur Frage Uber das Epithel der Hamblase. Arcli. f. Mik.

Anat., Bd. 35, 1890, p. 389. EoGELiNo, H. Ueber Deckzellen Im Epithel von Ureter und Hamblase.

Anat. Anz., Bd. XX, 1901, p. 116. Hebzoo, F. BeitrUge zur Eutwickiungsgeschichte und Histologle der MM,nn lichen Harnrohre. Arch, fttr Mik. Anat, Bd. 63, H. 4, 1904, p. 710. London, B. Das Blasenepithel bei verschiedenen Ftlllungszustilnden der

Blase. Arch, fttr Physiol., 1881, p. 317. Paneth, J. Ueber das Epithel der Hamblase. Sltzber. der K. Akademie

der Wlssenschaften. Bd. LXXIV, III, Jul! Heft. Wien, 1876.


Quain's Elements of Anatomy; Eleventh Edition; VoL III, Neubology; Pakt I, The General Structure of the Nervous System and the Structure of the Brain and Spinal Cord. Edited and Revised by E. A. Schafer and J". Symington, with Index by T. W. P. Lawrence. 421 pages and 361 illustrations. Royal 8vo. Longmans, Green & Co., 1908. $4.50.

The many friends of Quain's Anatomy are delighted that the work is being again revised. Volimie I, The Embryology, the first of this, the Eleventh Edition, was issued by the publishers earlier in 1908 and has met much of the approval deserved by the editors for the current information it incorporates and for the retaining of that concise comprehensiveness of treatment characteristic of the previous edition. The second text issued of this revised edition is Part I of Volume III, Neurology, and in this part the preceding standard of excellence is maintained.

Part I, of Volume III, embraces the general structure and briefly considered processes of development of the nervous tissue elements, and is chiefly devoted to the gross and microscopic structure of the central nervous system. While it necessarily treats of their central connections in considerable detail, this part does not include the Anatomy of the Organs of Special Sense. The promise is made that it will be soon followed by Part II, which will embrace the descriptive and special anatomy of the sense organs and of the peripheral nerves in general. The two parts together will constitute Volume III, a complete Text-book of Neurology.

In the present revision of the Neurology, Professor Schafer, the chief editor of the previous edition, has associated with him Prof. Johnson Symington of Queen's College, Belfast. Professor Symington was editor with Professor Schafer of the Splanchnology, Part IV of Volume III of the Tenth Edition, and his name now appears as well in the title page of Embryology, Vol. I, of the present edition.

The part is edited and in considerable measure rewritten and


D Book Eeviews. 809

there is added a large number of illustrations not contained in the previous edition. The result is a book exceeding the previous ^^ Spinal Cord and Brain'* by some two hundred pages. Acknowledgment is made to authors who gave permission for the use of their illustrations and especially to Professor Cajal who, in addition, loaned many original drawings for reproduction in the work.

One of the most striking features of tlie book is the large proportion of microscopic anatomy it contains. The first 57 pages, after giving an introductory survey of the general structure and mode of development of the entire nervous system, are especially devoted to the histology or detailed consideration of the structural elements of the system, including the forms of nerve terminations, the process of degeneration and regeneration, and a couple of pages upon the methods by which nerve pathways are traced. And, throughout the remainder of the text, a large proportion is devoted to microscopic details of the architecture of the system. This excess of the microscopic is one of the features which has contributed toward the peculiar favor in which Quain's Anatomy has been held by many teachers and students. Were the microscopic eliminated to the extent it is in any of our large, one-volume, text-books of Anatomy, the neurology of even this edition of Quain would scarcely exceed, in pages, the text generally devoted to the subject in any of the better of these other works. On the other hand, the part does not attempt to be as exhaustive a treatment of the neurology as is given in the special works on the subject. It is not so full as the latest edition of Edinger^s Lectures on the central nervous system, but, as a textbook for general class use rather than a reference book, its construction will lead many to prefer it to Edinger.

In treating the cerebro-spinal axis, it follows the usual and most advisable plan of beginning with the spinal cord. Thence it passes to the medulla oblongata and pons, taken together, followed by the study of the cerebellum. Then the mesencephalon, diencephalon and telencephalon are taken up in their order, followed by discussions of the development of the cerebral gyri, sulci and fissures, the meninges, the blood supply, measurements of the brain, and craniocerebral topography. The final 47 pages are devoted to an excellent

D 310 The Anatomical Record.

treatment of the intimate structure of the hemispheres and especially that of the Ehinencephalon, to which latter alone about 17 pages are devoted. The sequence is excellent for teaching purposes as well as the generally followed plan of proceeding from the gross to the microscopic structures of the diflferent divisions.

The mesencephalon is classed as not a division of the cerebrum^ being described as ^^a short and constricted part of the brain uniting the pons and cerebellum below with the cerebrum above."

At the end of the discussion of the development and causation of the gyri and sulci of the hemispheres, Professor Symington has inserted 10 pages comparing the size and gross features of the cerebral hemispheres of the Primates. Brains of the Marmoset, Capuchin monkey, Bonnet monkey. Baboon, Gibbon, Orang, Chimpanzee, and Qt)rilla are described in the order named, compared among themselves and with the human cerebrum. This is a commendable and most interesting addition to a treatment of the human brain and made by one especially qualified for it. His illustrations of the different hemispheres are most excellent.

The treatment of the cerebellum is enlarged by several pages, is comprehensive and sufficiently detailed as to gyri, sulci and microscopic structure, and includes a number of new and excellent illustrations by the editors and from Cajal.

The text upon measurements and weights of the brain is altered much less than might be expected, considering the information accumulated upon this subject since the appearance of the tenth edition.

The section on Cranio-cerebral Topography comprises nine pages with three new, full-page figures, and is a brief, up-to-date treatment of the subject. A separate discussion of this subject was not included in this part of the tenth edition at all.

The general description of the neuroglia is not all that might be desired. The neuroglia fibers are spoken of as numerous offsets of and in continuity with neuroglia cells, when, in fact, a large proportion of the neuroglia fibers of the adult nervous system have ceased to be in contact with neuroglia cells and none can be considered as offsets of them any more than the fibers of other forms

D Book Eeviews. 311

of fibrous connective tissue can be considered offsets of the connective tissue corpuscles. Further^ the custom is retained of considering and classifying as neuroglia cells the various pictures of neuroglia obtained by the Gk)lgi method, when it is more than probable that many of these accumulations of the silver deposit contain neither cytoplasm of neuroglia cells nor even a nucleus.

The editors, Professor Symington especially, have added quite a number of new illustrations of the macroscopic features, all of which possess excellent teaching qualities. Most of the drawings selected from the work of other authors are taken from one author alone. Of the total of 361 illustrations in the book, about one-third are from Cajal and most of these are from Golgi preparations. From the fact of the very limited accuracy and completeness of the results of the Gk)lgi method, it might be urged that so great a proportion of Golgi pictures, though taken from an author so pre-eminent in authority and in the use of the method, is not advisable for a text-book. The diagram, Fig. 150, shows no ascending portion of the sensory root of the Trigeminus, though Fig. 171, which is from Cajal, shows such to exist. Fig. 137 is a little misleading in its claim to be of the natural size of the floor of the fourth ventricle.

The BNA is used, translated into English and untranslated, but by no means consistently throughout. This will please some, but many will consider the limited use of the nomenclature as unnecessarily conservative at this stage of its adoption. A bibliography is given, but, unlike the plan followed in the previous edition, the citations are placed at the foot of the pages referring to them and thus where they will be more usually noted and more frequently used by the student.

The press work of the book is good and the size of the paper

and the type, as well as the general style in which it is written, are

the same as of the tenth edition. On the whole, one is impressed

with the very copious illustrations and the construction of the text,

but, after looking it over carefully, one feels a bit disappointed with

it as a part of a newly revised edition of Quains Anatomy.

Irving Hardesiy. Received for publication, January 28, 1909.

D 312 The Anatomical Becord.

DoQiBjj, A. S. Der Bau der Spinalganglien des Menschen und der Saugetiere. Jena, G. Fischer, 1908. 151 pp., 14 plates.

The results of more than ten years of careful research upon the spinal ganglia are embodied in this beautiful monograph. The author, after successful experience with the silver nitrate methods of Cajal, decided that the methylene blue method with which he has already had so much experience gives more valuable pictures of the structure of the ganglia. Accordingly, most of the paper is devoted to the description of such preparations. Most of the anatomical findings of Cajal, Levi, Lenhossek, Nageotte and others are confirmed and supplemented, vnth, however, some differences of interpretation.

The preparations described and figured by Dogiel are of remarkable beauty and the complexity of ganglionic structure which he brings to light is bewildering. No attempt can be made in this review to summarize the characteristics of the eleven types of spinal ganglion cells described; the original should be consulted by all who are interested in ganglionic structure, either from the anatomical physiological or clinical points of view. These findings have important applications in all of these fields, as well as to the general theories centering about the neurone concept.

Each spinal ganglion cell is surrounded by a capsule which is a fine homogeneous structureless membrane usually stained in methylene blue preparations. External to the capsule is a loosely fibrous connective tissue sheath, containing many nuclei, some of which are often closely attached to the outer surface of the capsule. These are all ordinary connective tissue nuclei. The sheath is more extensively developed in those types of cells in which the chief process or dendrites of the ganglion cell arborize in the immediate vicinity of the parent cell. Enclosed within the meshes of the connective tissue sheath are often found some of the complex termini of processes of the ganglion cells ; the glomerulus of the chief processes may lie within it, and also the elaborate sympathetic plexus described in the later sections of the paper.

Amongst the numerous types and subtypes of ganglion cells de

D Book He views. 313

scribed, the familiar unipolar cells with T-shaped processes occupy a quite subordinate position. Usually the neurone is far more complex than this. Cells giving rise to these simple T-fibers (type I of both the older and the present classification of Dogiel) are of all sizes from the largest to the smallest, and seem to be less numerous than some of the more complicated types.

The variations of the remaining ten types are infinitely diverse. Sometimes the chief process divides into a skein or network of branchlets, within or without the connective tissue sheath, all of which reunite to form a single process which later divides T-form in the usual way (types V, VI, VII). Or collaterals may be given off from the chief process with the most diverse sorts of eadings. Some of these have swollen tips which end either within the connective tissue sheath or outside of it, sometimes at great distances from the parent cell. In the latter case the mode of ending is very diverse (types II, III, IV). There are also bipolar cells of the embryonic type, with two simple processes, one of which is central, the other peripheral (type IX). Or the chief process may divide T-form into central and peripheral branches, the latter breaking up into an extensive arborization which ends within the ganglion or its dorsal root, but does not reach the periphery (type VIII). Again, in addition to a chief process of typical T-form, as in type I, there may be one or more thick, short processes (dendrites) which break up within the connective tissue sheath of the cell (type X). These are cells of small size with the chief process unmeduUated and the author regards them as inmiature cells in process of differentiation into functional neurones of other types.

Finally, there are multipolar cells of very peculiar form, not hitherto described (type XI). There are several processes, one of which is a typical chief process like that of type I. The others leave the connective tissue sheath, generally become medullated, divide within the ganglion and end in special free or capsulated endings variously distributed within the ganglion, chiefly about its periphery. These dendrite-like branches are regarded as the peripheral processes ; but they end within the ganglion or in its immediate neighborhood. They seem to be sensory nerves for the ganglia and their

D 314 The Anatomical Record.

connective tissue envelopes. The chief process does not divide T-form, but is thought to pass directly into the spinal cord as a central process.

The spinal ganglion is filled with medullated and unmedullated fibers which arise in large part from the cells of the ganglion itself and which end free or in expanded tips or encapsulated endings within the ganglion and the dorsal root and particularly in the connective tissue envelop of the ganglion and the septa within it and the connective tissue capsules of its cells. These endings and fibers are of very diverse forms. Some of these fibers before, ending wind spirally around the chief process of a type I cell. Sometimes several fine fibers enter into the formation of such a spiral.

Dogiel is of the opinion that the free and encapsulated endings in the spinal ganglia are for the most part sensory organs, thus differing from Cajal and Nageotte, who regard them, especially the expanded endings, as growing nerves analogous with the expanded tips of regenerating nerves. Dogiel gives many reasons against this ; for example, these endings are similar to many familiar types of peripheral sensory endings. If Cajal is right, these, too, would have to be regarded as regenerating or growing terminals.

In general, then, the spinal ganglia of all of the mammals investigated show intra-ganglionic sensory endings which in some cases seem to be termini of ordinary peripheral branches of the T-form division of the chief processes of the ganglion cells; in other cases they are termini of special dendrite-like processes of the ganglion cells. These endings are sometimes capsulated, constituting varieties of the Vater Pacinian (Golgi-Mazzoni) corpuscles; others are free, forming varieties of free end-skeins similar to those found in the nerve termini at the periphery.

The preceding types of fibers all arise from cells of the spinal ganglion. Other fibers enter the ganglion from the outside. The source of some of these fibers is obscure; but some of them undoubtedly come from the sympathetic system. The latter come in by way of the rami communicantes ; they are partly medullated and partly unmedullated, but are all small fibers. Every spinal ganglion cell (including all of the types) is probably enveloped by

D D D Book Reviews. 315

a network of fine sympathetic fibers. These pericellular nets are all connected and bound up in a common network which pervades the whole ganglion. This network lies in the connective tissue sheath of each cell, not under the capsule.

Dogiel's monograph is accompanied by eighty-nine exquisite figures drawn by the author and based, with one exception, on preparation from the horse, which, together with the lucid and wellarranged text, make the work unusually clear and easily read, in spite of the complex nature of the material.

C, Judson Herrick. Received for publication, February 27, 1909.

Economic Zoology, an introductory text-book on zoology, with special reference to its applications in agriculture, commerce and medicine. By Herbert Osborn, The Macmillan Co. 1908.

The scope of this work is sufficiently indicated by its title and by a statement, in the preface, of the author's hope that beyond being a mere text-book, "it may be of service to that ever-increasing body of citizens who wish to familiarize themselves with the general principles and the present status of knowledge concerning the animal kingdom."

The introduction, of nine pages, is, in the opinion of the writer of this review, open to certain criticisms from the pedagogical standpoint ; first because it does not present an adequate setting for such a discussion of the Protozoa as is given in Chapter II ; and second, because of the examples. of a kind of pedagogy of which there are many instances at other places throughout the book. For instance, on page 7 the following sentence occurs, "Locomotion, however, is connected with a great variety of structures, among which may be mentioned pseudopodia, cilia and flagella of protozoa ; the parapodia and setae in worms; the swimmerets and legs in Crustacea; legs and wings in insects, and fins, legs, paddles, wings, etc., in vertebrates." Over half of the terms used in the foregoing sentence are technical ones and indicate things which the elementary student cannot reasonably be expected ever to have seen or to have any conception of, and thus the sentence does not convey any clear picture of what it is in

D 316 The Anatomical Becord.

tended to call before the student's mind, namely, the diversity of the organs by which locomotion is effected in the animal kingdom as a whole. The writer is aware that this kind of description is common in zoological text-books, but its commonness certainly does not justify such a method and it is quite contrary to the method which his own experience tells him is likely to fasten such matters in the mind of a student. Taking this example as a text and the further instances of the use of the terms "cell," "protoplasm," etc, upon page 6, without any previous explanation of what these terms mean, we may say that this way of presenting zoology, or any other subject, is distinctly putting the cart before the horse. In our experience, students are not impressed by words which convey either no meaning at all or only the hazy conceptions of the public at large. If such a thing as locomotion is to be discussed, the way to present it, and have the presentation take root, is to let the students first study along with other things, the structure and function of the locomotor organs in a series of animals and then to set before them such statements as are desirable regarding locomotion in general, using as illustrations the cases studied. If one wishes to talk about the cell or protoplasm to a body of students, having no previous training in biological science, the method of procedure should be likewise by a description of the thing itself, followed by the name and last of all the generalizations. This way of presenting things is, in the opinion of the writer, a habit of thought which is as necessary for the more advanced as for the elementary teacher, but which with more mature students can, of course, be somewhat compensated for by the better trained perceptions of those to whom the instruction is being given. The writer realizes how difficult a thing is an introduction and he would not have so much fault with this one, provided the thing introduced did not begin in such a manner as to make the preceding nine pages a wholly incomplete setting for the subject as it is handled in the chapters on the protozoa and the groups which follow.

In the second chapter, which deals with the phylum Protozoa, we find again examples of the use of technical terms such as "cells," "protoplasm," "germ layers," "chromatin granules," "gastrula," "ectoderm," "endoderm" and "cilia," with no previous explanation

D Book Reviews. 317

of their precise significance. As examples of a careless use of words and facts there may be cited the statement at the top of page 13, "the Amoeba possesses locomotion, nutrition," etc., and on page 20 the statement, regarding the Sporozoa, that "this group includes the parasitic protozoa," as though no other protozoa except the sporozoa were parasites. Or again in this sentence, on page 29, "Calkins has shown experimentally that without this process of conjugation there is in time a deterioration of the Paramoecium, as indicated by its rapidity of growth, ability to multiply by fission and other activities ;" and by the statement that the malaria parasite is an "amoebalike species." Taken as a whole, there is little to commend this chapter on the protozoa and it is particularly weak as the first concrete chapter of a course in the fundamentals of zoology. The third chapter, which is entitled "Sponges, Hydroids and Polyps," is again characterless and shows an apparent ignorance of many important facts in sponge embryology, most of which were well established ten years ago, when the statement is made upon page 37 that the cell layers of sponges "are formed in the same manner and occupy the same position as in all the higher groups of animals," and further on that "perhaps the most striking fact is found in the uniform invagination and formation of the gastrula in sponges and all other metazoa." The chapter upon the flat-worms deals for the most part with the parasitic members of this phylum. The accounts of the structure and life history of Taenia solium and of the liver-fluke are good, though there is nothing in them which is not common to a number of text-books now in use and they thus lack distinctive merit. The few pages, with figures taken from U. S. Department of Agriculture Reports and dealing with forms less frequently described, are an addition to this subject, not commonly found in our texts, and hence of greater value than what precedes them. The chapter upon the round-worms might have been enlivened by some really good illustrations and the brief discussion of "the effects of parasitism," is not nearly so extensive as might well have been given this general topic after an account of the parasitic worms. Two chapters dealing with the Annelids and Molluscoida are again without any degree of freshness or obvious merit and the chapter upon Echinoderms is, in

D 318 The Anatomical Record.

the writer's opinion, distinctly bad, because of its descriptions of structures with no accompanying references to figures and the discontinuity of the style and thought. Indeed; some of the sentences on page 134 read like the blue-books of the elementary students for whom this text was intended. As an example of the inverted order of pedagogy previously mentioned, we may cite the paragraph upon page 139 regarding the evolutionary history of this group, and, as a whole, the chapter shows no great familiarity with the facts of echinoderm anatomy and embryology. The chapter upon the Mollusca has some merits and a larger proportion of new figures than most of the others, but these, however admirable they may be as the drawings of undergraduate students, are in the case of those dealing with Lampsilis luteolus of scant value for a text-book. The Crustacea are again not dealt with in any unusual manner and the chapter devoted to this group is open to the kind of criticism already made.

Chapters XI, XII and XIII dealing with the Arachnida, Prototracheata and Insecta and comprising 106 pages are however distinctly valuable and to be commended in a book of this nature and if the remainder of the book were of the same quality it would have a far greater merit as a whole. The numerous illustrations from Department of Agi'iculture publications are a valuable source of reference and the author's special preparation in this line is at once evident. With the Chordata, however, the same things which constitute the defects of the earlier chapters reappear, though with the birds and mammals the treatment is less open to criticism, and the concluding chapter of "General Considerations," the only one, by the way, which deals with the facts of zoology in any other manner than from the standpoint of the single group, brings nothing to relieve the monotony. "Distribution," "Dispersal," "Adaptation," "Variation," "Heredity," "Evolution," "Animal Industries," "Organization for Research," and the "Geological Succession in Animal Life," make up the 17 pages of this final chapter which must obviously be a wholly inadequate treatment of such general topics, unless they had been elaborated in specific cases throughout the work, which is not the case. One lays down the volume with the feeling that, save for the special part which deals with entomology, it is distinctly corn

D Book Beviews. 310

moiiplace and that it cannot find any wide field of usefulness, despite the urgent need of the right kind of a text-book for the students of zoology in our agricultural colleges and in pre-medical courses. It is too much like a condensed edition of Parker and Haswell, with the merits of that admirable work left out, and its illustrations suggest this even more strongly; for out of a total of 269 illustrations for the entire book, 35 are indicated as taken from Parker and Haswell and 110 others are identical with illustrations appearing in Parker and Haswell, though quoted as from their original instead of their apparently more immediate source. They are moreover in many cases printed from badly worn plates which should have long ago gone to the scrap heap and for this feature the publishers deserve little credit. Several of the figures taken from photographs by the author, are of very little value, the one of a polyzoon (Fig. 78) showing nothing but blotches of black and white' and looking more like the "mountains of the moon" than anything the writer can recall having seen figured. Moreover, the writer would sharply take issue with the entire plan of the course here presented, for any students whether with economic interests or not. This is the old course in comparative anatomy from amoeba to man and one which may be said, with no belittling of the value in its proper place of vertebrate and invertebrate comparative anatomy, to be in no way representative of the kind of course in General or Introductory Zoology which is acceptable to most teachers of this subject to-day. From what the writer knows of the present feeling among zoologists, such a course, like one in General Physics or General Chemistry, should aim at the general, foundational principles of the science, illustrating these by such animal forms as are best suited to the subject in hand. It should represent intensive rather than extensive study and should approach from the viewpoint of each form as an animal rather than as an echinoderm, a mollusc or a vertebrate. It should represent in an adequate way the essentials of general topics such as animal ecology and physiology and the problems which are naturally grouped about the theory of evolution. The trend of zoology to-day is toward the investigation of the fundamental activities common to all living things and this must find expression in our text-books for elementary

D 320 The Anatomical Becord.

college courses by the presentation of the general conclusions which have already been reached and by the subordination of details to their proper illustration of such topics. That a course of this kind is not a difficult one for the student to properly grasp, the writer knows from his own experience, and he has had abundant reason to believe that it is exactly the kind of course which agricultural and pre-medical students need as a foundation for subsequent work in their anatomy, physiology and the like of man and the higher vertebrates. We have no text-book which expresses this point of view in such a way as to include the many important results obtained in the past decade and when such books come they will fill an urgent need, and if they are well done will go far toward adding to the dignity of zoology and toward setting its problems before our students as well worthy the best efforts of men.

W. C. CuHis. Received for publication, March 1, 1909.


The Second International Anatomical Congress is to be held at Brussels during August, 1910. By a recent decision of the International Committee the exact date has been fixed, August sixth to tenth. It is hoped by the Committee that a large number of the members of the Association of American Anatomists will attend the Congress. Those who expect to join the Congress are requested to inform the undersigned, who is serving as the American member of the Committee. A circular giving full information will be issued later and communicated to all members of our Association.

Chables S. Minot.IC



Vol. III. JUNE, 1909. No. 6.





THOMAS DWIGHT, From the Department of Anatomy, Harvard Medical School,


The simultaneous occurrence of fusion of the atlas and occiput, with the presence of a more or less distinct occipital vertebra, is of much importance in the discussion concerning the significance of numerical variations of the vertebral column. The observations by Zoja^ and Swjetschnikow* both deal with the manifestation of an occipital vertebra that is most developed in its anterior and lateral portions. Although we find vague allusions in literature to the simultaneous occurrence of these two conditions, yet, so far as I know, these two anatomists are the only ones who have observed them. In point of fact it was the Russian who recognized the occipital ver

Zoja, G. Intorno aH' atlante stude anthroix>-zootomiei. Con una tavola. Letture fatte nell adunanze 4 niarzo, 1° aprile, 13 inagglo, 17 glugno 1880 e 19 magglo 1881. Pp. 269-296.

"Swjetschnikow. T^eber die Assimilation des Atlas und die Manifestation des Occipltalwirbels beim Menschen. Arch, fiir Anat. u. Physiol., Anat. Abtli. 190(J, pp. 155-193.



322 Thomas Dwight.

tebra in Zoja's specimen. The case to be described in this paper differs from them in that the posterior arch of the occipital vertebra has a free point. Similar appearances have been found on isolated skulls, but I believe this is the first time it has been seen when the atlas is fused. It makes the fact doubly sure to have the free points of the imperfect posterior arches of the atlas and occipital yer tebra appear in series one above the other (Fig. 1). This body presented other vertebral irregularities: a supra-sternal bone and (perhaps) a hypochordal brace of the axis (Fig. 5). The last feature is considered in the second division of this communication.

These observations were made on the body of a white man, aet. 60, dissected at the Harvard Medical School, in the season of 1907-08. The skeleton was distinctly pathological, as is often the case with very exceptional vertebral variations. Thus, the two cases which I have reported of 26 prae-sacral vertebrae on one side and 25 on the other were both extremely pathological.'

The calvaria is a fine specimen of hyperostosis manifested on the inner table and especially in the frontal region, where it presents a hummocky surface with a maximum thickness of nearly 18 mm. Towards the posterior inferior parietal angle the thickness is about 12 mm. The buccal surface of the hard palate presents a similar condition (Fig. 4).

The vertebral formula is as follows, omitting any mention of the occipital vertebra?: C. 7, T. 13, L. 5, S. 5, C. 4. The atlas is fused with the occiput, one might say absorbed into it, so far as some parts are concerned. The axis and third vertebra are fused, the fusion dating undoubtedly from early embryonic life (Figs. 2 and 5). On the left the lateral portions of the two vertebraB occupy their normal relations. On the right they are very close together. The laminae, though fused on the left, retain their distinctness, and the spinous process ends in two^knobs, one above the other, each representing the lateral termination of the bifurcated spines of the two vertebra}, while on the right there is but one knob for the

'Dwight, Thomas. Description of the Human Spines Showing Numerical Variation In the Warren Museum of the Harvard Medical School. Memoirs of the Boston Soc. Nat. History, V, 237-312. 1901.

D Occipital Vertebra and Hypochordal Brace. 323

two. The spines of the fourth, fifth and sixth cervical vertebrae are bifid. The laminae of the sixth vertebra present on each side a sharp point (Fig. 2, L.) rather less than 1 cm. from the spinous process. That on the left is the larger. The fifth vertebra presents a very small similar projection on the left side only. The bodies of the sixth and seventh cervical and first thoracic vertebrae are fused through prominent exostoses. This is evidently the result of a pathological process, and is accompanied with distortion of this region, the details of which do not seem to the purpose. There are several projections on the ventral aspect of the bodies of the vertebrae in the thoracic region which would ultimately have led to fusion, if they have not already done so, and one on the second lumbar vertebra. Apart from this last feature, the lumbar vertebrae are normal, though the relative spread of their transverse processes is not that of a normal lumbar region. This is owing to the transitional character of the twentieth vertebra, which has been reckoned a thirteenth thoracic. The change in direction of the articular processes occurs below the twelfth thoracic. The costal elements of the twentieth vertebra are free, but insignificant. The right one measures 2.5 cm., the left one 3.3 cm. It is really a matter of taste whether we say that there are thirteen thoracic vertebrae and five lumbar, or twelve thoracic and six lumbar. The twelfth rib measures 13.5 cm. on the right and 14 cm. on the left. There are seven sternal ribs on both sides. On the left the cartilage of the eleventh joins that of the tenth — it probably did not do so on the right. The twenty-sixth vertebra (first sacral) is the fulcralis, i, e., the most important one in supporting the sacrum. The third sacral vertebra presents a sudden change in the curve (Hermann von Meyer's co7ijugata vera). The first coccygeal is fused with the sacrum. Thus, encept in the last detail, the sacrum is perfectly normal, in spite of the increased number of vertebrae above it. The ensiforra cartilage, ossified and fused with the body, is bifid below. The most remarkable feature of the sternum is a knob at the top of the posterior surface of the left half, at the median end of the clavicular notch. It is 6 mm. in height, tolerably clearly marked off, and suggests very strongly a snpra'sfenml bone fused with the manubrium. On the

D 324 Thomas Dwight.

right the sternum is prolonged upward perhaps a little more than usual, but presents no corresponding structure.

The Occipital Region, The anterior arch and lateral masses of the atlas are well developed, but it is so closely fused with the occiput that there is no sign of any occipital condyles. There is a small interval between the anterior arch and the occiput and on either side of this a deep groove above the atlas. The height through the region of the articular processes is about 2 cm. on the right and 1.5 cm. on the left. Seen from the intracranial side the fusion is very complete. The left posterior arch is, in the main, well developed, quite free, ending in a point some 5 mm. from the median line (Fig. 4). This piece, which is now separate, was in life connected by cartilage to the atlas just external to the inferior articular process. The left vertebral artery grooved its superior surface. The representative of the right posterior arch of the atlas is a thin sliver of bone, the point of which had been broken off (Figs. 1, 3, 4). It was less than 15 mm. in length when measured and probably never was more than 2 or 3 mm. longer. This was separated by a mere crack from the base of the skull at the border of the foramen magnum. The vertebral artery, smaller than the left one, judging from the size of the foramen in the axis, must have passed below this rudimentary arch. The lateral parts of the atlas differ very much on the two sides (Fig. 3). On the left, the costal element of the atlas is wanting, so that the transverse foramen is completely open in front, except for a very slight hook projecting from the median side. The transverse process is strong, and ends in a knob which rests against the vaginal process of the temporal. On the right the lateral mass is inextricably mixed with the occiput. The transverse process is short and indistinct till near its end, which projects strongly backward; but a projection from near the end runs somewhat forward and is assimilated into an ill-marked paramastoid process. The groove already mentioned, anterior to the foramen magnum at the junction of atlas and occiput, has on the right a deeper part, which extends from the anterior condyloid foramen laterally to the entrance to a canal l)etween the transverse process and the skull, which runs antero-posteriorly. (It is unlikely that there was an suboccipital nerve on the right.)

D Occipital Vertebra and Hypochordal Brace. 325

There is an unquestionable manifestation of the posterior arch of an occipital vertebra on the rightj and a less distinct one on the left. The right arch forms the boundary of the foramen magnum, but some 3 or 4 mm. before reaching the median line it ends in a point, which is separated from the bone above it by a sharp cleft some 3 mm. deep. This is seen most strikingly on the inner aspect (Fig. 1), but is clear also on the outside (Figs. 3 and 4).

A P.A.A. O.V. O.V.?

Fig. 1.

A., Atlas. O.V., Manifestation of occipital verteura. I'.A.A., Posterior i:rch of atlas.

On the right, close to the middle line, and nearer to it than the end of the occipital vertebra above described, there is a small knob which might be held to indicate the termination of another occipital vertebra above it (O. V. (?))• ^^^ ^^^ lower surface of the left posterior border of the foramen magnum there is a very fair manifestation of the arch of an occipital vertebra, marked off by a groove. It ends rather vaguely some 2 or 3 mm. from the median line (Figs. 3 and 4, O. V.).

D 326 Thomas Dwight.

The right anterior condyloid foramen is small. On the left it is subdivided into a larger lower and a smaller upper division by a tongue of bone projecting forward from behind and reaching the anterior border. This foramen opens into a groove, already alluded to, between the front of the atlas and the occiput. From the region of the condyle for some distance forward the borders of the occiptal bone grow downward so as to shut out the atlas from forming a part of the wall of the spinal canal, and placing it in the main in front of the occiput instead of below it. This is particularly marked on the left. Fig. 3 shows a deep vertical cleft between these bones on the interior surface.

Certain peculiarities are to be noted in the region of the odontoid, which is very long, measuring 19 mm. on the posterior surface, from the lower border of the facet for the transverse ligament to the top. The superior part of the odontoid (nearly one-half) is in the same vertical plane as the anterior surface, but does not reach back to the posterior one. Seen from behind this upper part is a roughened irregular pi( cc of bone, giving a suggestion of an additional element (Fig. 2). A noteworthy point is the difference of position of the anterior and the posterior articular facets of the odontoid. In fact, the highest part of the posterior one does not extend above the level of the middle of the anterior one, which latter reaches to the very top of the odontoid. This is shown in Figs. 2 and 5. The superior articular facets of the axis slant perhaps a little more steeply than usual downward and outward, and the pedicles of the axis decline very steeply behind them. All this would imply that the head must have been carried with the chin high. When we consider the fusion, in some cases congenital and in others pathological, of so many of the cervical vertebrae, it would seem that the ])ower of nodding must have been nearly abolished. Presumably the joints above the axis allowed a great deal of irregular motion.

Besides these manifestations of an occipital vertebra in the region of the foramen magnum, there is on the left side an apparent manifestation of one seen from the front, made by modifications of the anterior border of the occipital bone, which is developed into a horizontal shelf where it bounds the jugular foramen, and, continued

D D D Occipital Vertebra and Hypochordal Brace. 327

forward, forms the line marking off the groove above the anterior arch of the atlas. This manifestation is mnch less clear on the right.

82 S3- 34

Fig. 2.

L., Point on left lamina of Otli vertebra.

P., Posterior articular facet. S2, Left lateral knob of spine of 2d vertebra. S3, Left lateral knob of spine of 3d vertebra. S4, Right knob resulting from the fusion of the right lateral knobs of the 2d and 3d vertebrae.

On the latter side a delicate process from the border of the occipital divides the exit of the venous canal from the nervous one (Fig. 3, J. F.).

D 328 Thomas Dwight.

This spine presents also a possible hypoohordal brace which is considered in the second part of this paper.

The following facts are to be noted in this case. The presacral vertebrae are increased bv a transitional one between the thorax and

Fig. 3.

(\C\, Carotid canal. J.F., Jugular foramen. — under the leader is the spicule dividing it. O.V.. Manifestation of o<vipital vertebra. P.A.A.. Posterior arch of atlas. P.M., I*arama8toid processes. H., Ilook by transverse foramen of atlas. T.P., Left transverse process of atlas resting on vaginal prwess of temporal.

loins, bearing short movable costal elemnits. The ribs on the vertebra above it are longer than is usual for last ribs. The second and third vertebrie are fused, the fusion of the arches being more intimate on the right. The atlas is fused much more intimately with the occiput on the right, and the right arch is much less devel

Occipital Vertebra and Hypochordal Brace. 329

oped; yet the manifestation of the occipital vertebra, as shown by its free point, is more advanced on the same side. On the right, also, there is the hint of the presence of the point of the arch of still another occipital vertebra. The suggestion of the antero-lateral part of an occipital vertebra is more distinct on the left.

It is much to be regretted that in so few cases of variations about the foramen magnum do we have an account of the entire column. Smith* remarks that "true assimilation of the atlas is rarely, if ever, an isolated anomaly of the cranio-vertebral axis." There can be no doubt as to the correctness of this statement. In support of it Smith ix)ints out that the cases of ^forgagni, Schiffner and Lambl had also fusion of the axis and the third vertebra; that in two other cases there were cervical ribs; and that a case of his had beside fusion of the axis and third vertebra, a cervical rib and other striking anomalies. To these I would add that among the spines in the Warren Museum showing numerical variation, there are three complete ones with fusion of the atlas and occii)ut, of which one must be discarded because the fusion is to be considered pathological, and one specimen in which eight vertebrie are preservcnl. The first (spine 561) has an extra vertebra at the junction of the back and loins, with small free costal elements, while the ribs of the twelfth thoracic are very long. The twenty-fifth vertebra is more or less sacral ized on both sides. The next (spine 24) is very normal, only the arch of the last lumbar is distinct and there are certain distinct epiphyses ( ?) on the caudal side of some of the lumbar articular processes. I have excluded spine D-7 for the reason given. It has only four lumbar vertebrae. The specimen which consists of the neck and the top of the thorax has suffered the loss of the left transverse process of the atlas, probably by accident. On the right the costal element is wanting. Presumably it was free and was lost. Just the same may be said of the right costal element of the seventh. The label states that the vertebral formula is believed to have been normal. It may be repeated that the spine which is described in this paper has an

\Sinith, G. Elliot. The Significance of Fusion of the Atlas to the Occipital Bone, and Manifestation of Occipital Vertebrae. British Med. Journ., 1908, 594-596.

D 330 Thomas Dwight.

extra vertebra at the junction of back and loins and fusion of the axis and third, which last feature seems remarkablv common in these cases.

We have known for centuries that the atlas may be assimilated more or less completely with the occii)ut, and for a few years we have accepted the manifestation of an occipital vertebra, usually an inseparable part of the occipital bone. The most perfect observation is that recorded by v. Schumacher,* who described several pieces of a fairly developed vertebra between the atlas and occiput, without bony connection with either. Recently we have had discussions as to whether certain "manifestations'*' on the bases of isolated skulls were to be considered as belonging to one region or to the other. Even without the light that is thrown on such peculiarities when the spine is present, we must recognize that the case is the same as at other transitional parts of the spinal column. A twenty-fifth vertebra may be a sacral vertebra or a lumbar vertebra, or a cross between the two. It may ev(>n Im^ both at once in its two halves. Similar observations may be made at the two ends of the thorax. It is idle to discuss which vertebra a particular vertebra is. All we can say is which oiw it is like. 1 agree fully with Bateson that we must not treat the members of such a series as individuals. The details of their structure vary according to circumstances. Many years ago Topinard discussed at length which vertebra was wanting in a spine in which there were only eleven thoracic ones. We now know better; but it would seem that in discussing the occipital region we are slow to apply the principles we follow elsewhere.

The great importance of the present specimen is that it is one of the few undoubted cavSes of assimilation of the atlas and occiput, with manifestation of an occipital vertebra. Perhaps it is the only one. What is most remarkable is that both these processes have made the greatest progress on the same side. The cases of Zoja and Swjetschnikow each show connection between the skull and the spine by means of paramastoid processes from the skull and

'Schumacher, Siegmiind v. Eiii Beltrag zur Frage der Manifestation des Occipital wlrbels. Anat. Aiiz., XXXI, 1907, 145-159.

D Occipital Vertebra and Hypochordal Brace. 331

anomalous upgrowths to meet them from the transverse processes of the atlas. In the latter's case there is a very slight ossification between the bones at one condyle, which probably is pathological. Xow in both these cases the posterior arch of the atlas is apparently both free and well developed. Hence they appear to me to belong to a different class from those in which the body and posterior

P.A.A. O.V. O.V.?. O.V P.A.A.

! ) !

Fig. 4. P.A.A., l*osterior arcli of atlas. O.V.. Manifestation of occipital vertebra.

arch of the atlas are more or less assimilated with the skull. I regret that Swjetschnikow does not give a figure of his specimen, but he tells us that it is demjenigen von Zoja ilusserst cihnlich/' Although the use of color by the Russian anatomist on Zoja's figure is hardly justifiable, as appealing too much to the imagination, it is not to be denied that the tubercle on the anterior border of the

D 332 Thomas Dwiglit.

foraiiieu magniini points distinctly to the manifestation of an occipital vertebra of which other parts are indicated. The occipital vertebra must l)e admitted in these cases; but there is no real assimilation of the atlas and occiput.


Fig. T). II.B., IlypcK'hordal hrjice.

The present s])ecimen shows far more conclusively the presence in the same body and most markedly on the same side) of two distinctly antagonistic processes, according to the popular theory which would have the one a return to the past and the other a step toward

D Occipital Vertebra and Hypochordal Brace. 333

the future. It is therefore one more piece of evidence for the conservative theory of variation around a mean.


It remains to speak of the hypochordal brace (Fig. 5), if indeed that be its morphological significance. It is a stout ossification close on 15 mm. long, with the greatest breadth of some 7 mm. situated free on the front of the axis, its upper part somewhat overlapping the lower border of the anterior articular surface of the odontoid. The anterior arch of the atlas presents a kind of a facet looking downward and forward, which presumably locked with its upper part.

As to the significance of this element, it may be said in favor of its being nothing but an accidental ossification that the body was a distinctly pathological one, showing in many places a tendency to the proliferation of bone. This, in fact, is not to be denied. Nevertheless, the anterior surface of this ossicle has what one may call a "finished" appearance. Whether the fact of its upper dorsal surface being so shaped as to enable it to "lock" with the anterior arch of the atlas is of importance, is a question more easily asked than answered. The fact, however, that an ossification is occasionally found at this precise point speaks strongly for its having a morphological significance. Yet if it be an hypochordal brace, it can be only that of the axis, and embryology leaves one in doubt whether this explanation is legitimate. There is indeed such a cartilaginous ardage, but according to Bardeen it is very transient. This spine is so abnormal, showing so many irregularities that must have occurred at a very early period, that it seems favorable soil for such an unexpected growth; but I have recently seen another instance of this ossification, though a smaller one, in a spine which, if not quite normal, was not remarkable.

Just before revising the proof of this paper I received the Anatoraischer Anzeiger of May 5, 1900, containing Smith's case of fusion of atlas and occiput with the manifestation of an occipital vertebra.IC




JOHN LEWIS BREMER. Harvard Medical School.

In 1902 I published a paper on this subject,^ a resume of which is here given. The pulmonary arteries in man, rabbit, cat, and dog, appear as symmetrical vessels, one rising from each fifth, or pulmonary arch. With the growth of the truncus pulmonalis, and its torsion about the bulbus aortae, the two pulmonary arches are wound, as it were, around the bulbus, and their walls thus brought into contact are absorbed, so that the truncus pulmonalis grows longer at their expense, the point of bifurcation moving continually farther from the heart. The left arch, being the outside one in this rolling "p process, receives the most pull, becomes the straighter and therefore tho larger vess(^l, and is shortened more rapidly. As a result, the point of bifurcation of the truncus pulmonalis reaches the left pulmonary artery while the right pulmonary artery is still seen arising from tlie right arch some distance dorsal to this point. (See diagram, page 3»i8). The portion of the right pulmonary arch between the origin of the pulmonary artery and the dorsal aorta becomes obliterated, the anterior portion of the arch remains continuous with the artery, and we then have the condition described by Rathke, — the two pulmonary arteries apparently arising together from the left pulmonary arch. It should be noted, however, that the right pulmonary artery of the fetus includes, beside the homologue of the left pulmonary artery, the proximal portion of the right pulmonary arch.

In the pig, although the pulmonary arteries first appear, as usual, as symmetrical offshoots, one from each pulmonary arch ; and although the fetal condition is practically the same, the intermediate steps

»Am. Jour. Anat, Vol. I, No. 2, p. 137, 1902.


D Pulmonarj^ Arteries in Mammals. 335

are different. The two arteries, while their points of origin are still far apart, bend toward each other lower down, and soon anastomose to form a long vessel, connected at its upper end with both the right and the left pulmonary arches, and forking at its lower end to send a branch to either lung. Soon the upper, or proximal, part of the right pulmonary artery becomes obliterated, leaving the common stem in communication with the left arch only, thus forcing the blood to both lungs to pass through the left pulmonary arch.

Since 1902 I have been able, through new acquisitions to the Harvard Embryological Collection, to trace the development of the pulmonary arteries in other mammals, — opossum, sheep, and guinea

Li. PA ^*;i?^.

Fig. 1.— Guinea-pig, 7.7 mm. (II. E. C, Series 1512, sections 190-233.) Dorsal view. P, A,y pulmonary arches, left and right ; p. a., pulmonary artery ; T. P., truncus pulmonalis. x ^25 diam.

pig, and to make a few obs(^rvations on the cow and deer. In the opossum and sheep the picture is essentially the same as in man, rabbit, cat, and dog, though in the sheep the two pulmonary arteries are brought to the bifurcation at almost the same time, so that very little of the right arch plays a permanent role in the right pulmonary artery. In the guinea-pig, on the other hand, the development of these arteries follows very closely that described in the pig, but with one important difference. In both animals the arteries originate as symmetrically placed vessels from the right and left pulmonary arches, in both they bend toward each other and anastomose, and in both the upper end of one pulmonary artery, from the arch to the anas

D 336

John Lewis Broraer.

tomosis, becomes obliterated, leaving the anastomosis and the lower ends of both arteries connected with only one arch. In the pig the left arch remains in communication with the combined pulmonary arteries, in the guinea-pig the right ; in the pig the entire right pulmonary arch from the bifurcation of the truncus pulmonalis becomes



FiQ. 2. — Guiuea-pig, 8.0 mm. X 125 diam.

(H. E. O. series 1513, sections 277-315.)

FiQ. 3.— Guinea-pig, 8.2 mm. (11. E. C. Heries 770, sections 230-256.) c. p. a., conjoined pulmonary arteries. Tlie lower portion of the pulmonary arteries not shown, x 125 diam.

obliterated, in the guinea-pig the anterior pii-rt of the arch, as far as the origin of the right pulmonary artery, Incomes incorporated in the adult pulmonary artery, and only the posterior part is lost.

Minor differences of development occur in the two animals, as may be seen by comparing the accompanying drawings with the figures of pig embryos in the former paper. The pulmonary arteries in the guinea-pig are seen to form a meshwork of capillaries and to preserve

D Pulmonary Arteries in Mammals. 337

their irregular course even after the upper part of the left artery has become obliterated. From the beautiful injection of the blood vessels of embryos *made by Dr. 11. M. Evans of the Johns Hopkins Medical School, it is probable that in all embryos the pulmonary arteries, in common with all other small arteries, arise at first by a capillary network, and that only later the main channels become larger and free from the surrounding capillaries. Renmants of this capillary origin of the pulmonary arteries are not infrequently seen in embryos, as for instance the short vessel from the right arch in Figure 2, loop formations near the pulmonary arch, side twigs from the arteries, even (in one instance in a sheep embryo of 10.0 mm., H. E. C. series 1340, sections 398-409) an artery which is double throughout most of its course, making a very long loop. In the guinea-pig this early condition lasts longer than in the pig or the other mammals studied, — the pulmonary arteries are later in straightening out and becoming distinct channels.

Another minor difference lies in the fact that, although in both pig and guinea-pig the two pulmonary arches are wound about the bulbus aortae as described above, in the guinea-pig there seems to be no fusion (or at least a much delayed fusion) between the two, so that the truncus pulmonalis is not lengthened, as in other mammals, at the expense of the two arches ; the two arches merely lie one below the other, side by side. This is shown in Figure 3, in which the left arch is seen to overlap the right for a considerable distance ; if fusion had taken place, as in the pig, the pulmonary artery would already seem to spring from the bifurcation instead of distinctly from the right arch as in the drawing.

In 1904, two years after n^y first article, Sakurai published a paper in which he describes the growth of the pulmonary arteries in the deer.^ The original starting point is the same, two symmetrical buds, one from each pulmonary arch ; but the left pulmonary artery, according to this author, moves toward the bifurcation of the truncus pulmonalis, and then continues farther to the right until it arises distinctly from the right arch, near to the origin of the right artery.

•Anat. Anzeiger, Band XXV, No. 14. p. 321, 1904.

D 338

John T^wis Bremer.

Diagram li b

Otafram li a

Diagram I. — Shows the original symmetry of the pulmonary arteries, and. In the second figure, the result of the torsion about tlie bulbus aortae. A, truncus pulmonalis, at the point of the original bifurcation ; B, point on left pulmonary arch where the left pulmonary artery rises; C, «ame for right side.

Diagram II. — (a) In the pig; shows the original symmetry, the pulmonary arches less wide spreading, the arteries nearer together. In the second figure, the anastomosis oF the arteries, and in the third figure, the result of torsion, (b) Same for the guinea-pig.

D Pulmonary Arteries in ^lammals. 339

I feel obliged to doubt, not the figures in Sakurai's paper, but the interpretation of them. Certainly in the deer^ in this laboratory I find nothing that would lead one to suspect that the deer differed from man, rabbit, sheep, cat, or dog in the development of its pulmonary arteries. In embryos up to 9.8 mm. in length the picture is the usual one, the two pulmonary arteries approaching each other as the bifurcation of tlie truncus pulmonalis is brought farther dorsal; and in an embryo of 18.6 mm. (IL E. C, series 1230), whose general characteristics show it to be younger than the oldest figured by Sakurai, the left pulmonary artery is seen arising from a short stem common to it and the right pulmonary artery. The posterior part of the right pulmonary arch no longer exists. The arteries are well established, with thick walls, so that any migration would seem impossible. A short common stem for the two pulmonary arteries in the fetus is not uncommon, and I should prefer to interpret Sakurai's last figure as an unusual lengthening of this common stem rather than as a migration of the left artery along the right arch, especially as the landmark, the posterior part of the right pulmonary arch, is lacking*

If we accept this interpretation of Sakurai's figures, the different methods of the development of the pulmonary arteries so far reported fall into two main gi^oups, one of which may be subdivided. (1) In man, cat, dog, rabbit, sheep, cow, deer ( ?), and opossum the development may be described by Diagram I. (2) In the pig and guineapig the development differs from that of the other mammals mentioned, and may be shown roughly in Diagram II, (a) representing the pig, (b) the guinea-pig.

In this curious grouping of the animals studied, generic lines seem to have no influence. In my former paper it was suggested that the large size of the auricles in the pig embryo caused the crowding together of the pulmonary arteries and their consequent anastomosis, and I again offer this explanation. In the guinea-pig also the auricles are very large at the time when the pulmonary arteries

Cervu8 cavreolus. Tlie laboratory is indebted to Professor Franz Keibel for the embryos.

D 340 John Lewis Bremer.

are growing, but there seems to be no crowding of the tissue surrounding the trachea from the sides. The mechanism seems to be slightly more complicated.. The large auricles and large sinus venosus separate the trachea posteriorly from the bulbus aortae and the truncus pulmonalis anteriorly more, it seems to me, than is usual in animals without the lai^e auricles. The aortic arches are straightened out more, the figure they present with the bulbus or truncus becomes more like a Y than like a tuning fork, and hence the pulmjonary arteries, starting out at right angles to the pulmonary arches, point toward each other instead of backward, as in other animals. This purely mechanical result of large auricles seems to me to account for the difference of development between the pig and the guinea-pig and all other mammals studied. The cause of the larger auricles I do not know; nor can I explain why, after the anastomosis, the left artery in one case, and the right in the other, should remain permanently.IC



FREDEKIC T. LEWIS. Assii/tant I'rofctssor of Emhryology at the Harvard Medical tScliooL

lu 181)6 Sapcer stated that the lymph glands in sheep and cow embryos arise from a plexus of lymphatic vessels.^ "The connectivetissue between the lymphatic vessels of the plexus has at first a trabecular arrangement, but later one or more compact masses or islands are formed within it. From the beginning, the connective tissue which makes the trabeculae, or masses, is narrower meshed than that which surrounds it, and contains many blood vessels. However, he adds : "There can be no doubt that there are many plexus formations in embryonic tissue, having exactly the appearance of those from which lymph glands arise, which simply degenerate."

Kling, in 1904, emphasized the importance of the plexus stage and modelled the lymphoid trabeculae.^ Although they connect with one another so as to form a continuous mass, his model has "an extremely irregular appearance." It shows that these structures have little resemblance to the future glands. Kling stated that from such a general mass portions were separated by constriction to form the basis for individual glands. But "lymph glands which have an isolated position appear from the first as solitary formations; each one arises independently."

A year later Miss Sabin wrote:' "All of the nodes of the early

'Saxer, F., I'eher die Entwickelung iiiid deii Bau der noriiialeii Lyiiii)hdrliseii. Anat. Ilefte, 1800, vol. G, pp. .'U9-.5.32.

'KHng, C. A., Studien ilber die Entwicklung der LympbdrilRen beiiii Menschen. Arch. f. mikr. Anat, 1904. vol. e.3, pp. 575-010.

•Sabin, F. R., The development of the lymphatic nodes in the pig and their relation to the lymph hearts. Amer. .Tourn. of Anat., 1905, vol. 4, pp. 855-389.


D 342 Frederic T. Txnvis.

embryos, the primary nodes in the sense of GuUand, pass through this (plexus) stage. Lymphatic nodes which develop later in the life of the embry'o, after lymphocytes occur, hurry through the primary process and show a considerable modification of it." Recently* Dr. Sabin published the figure of a section through the jugular lymph sac in a human embryo of 30 mm., "to show the simple bridging of the sac which is the anlage of the first lymph node." In the pig she found that "the first node to appear develops from the lymph heart, which is in the supra-clavicular triangle behind the sternocleidomastoid muscle."

Thus, Saxer, Kliug and ^liss Sabin agree that the first lymph glands arise from trabeculac in a plexus of lymphatic vessels.

The plexus of lymphatics in relation with the internal jugular vein is a conspicuous feature in human embryos measuring from 30 to 40 mm. It is shown in Figs. 1 and 2 from an embryo of 42 mm. A portion of the vein is seen in the lower right corner of each photograph, in places the connective tissue trabeculae arc broad and pale, as shown in Fig. 1. Elsewhere they are more slender and deeply staining, as in the left part of Fig. 1 and in Fig. 2. The latter is a section through the structure which Miss Sabin has described as the primary lymph gland.

The cells in the similar trabeculae of a 31 mm. human embryo are described by Kling as having "chiefly, if not exclusively, the char

Sablii, F. R., The lymphatic system In human embryos, with a consideration of the morphology of the system as a whole. Amer. Journ. of Anat., 1{K)9, vol. 9, pp. 43-91.


Figs. 1 and 2. — Plexus of lymphatic vessels in relation with tlie internal jugular vein. From a human embryo of 42 mm. X 45 diams. (Harvard Fmbryological Colleetio;i, Series 841, Section 432, and Series 838, Section 153, respectively).

Figs. 3, 4, and 5. — Lymi)h glands from a human embryo of 42 mm. X 60 diams. Fig. 3 shows the submental ("submaxillary*') gland (Series 841, Section 5S9) ; Fig. 4 shows the external jugular gland (Series 841, Section 524) ; and Fig. 5. the circumflex scapular gland (Series 8J58, Section 321).

Fig. G. — Sul)scai)ular lymph gland from a rabbit of 20 days, 29 mm. X 60 diams. (II. K. C, Series 170, Section lOSO.)

D Fir.. 1.

Fig. 2.

Fio. 3.

Fig. 4.

Fig. 5,

Fig. 6.

D .344 Frederic T. Ix^wis.

acter of fixed connective tissue cells." At 70 mm. *Ve find among pale oval nuclei, others of rounder form and darker stain which already suggest adenoid tissue." Similarly, in pigs of 80 mm. Miss Sabin found large, faintly staining, oval nuclei belonging to connective tissue, and small, round, deeply staining nuclei with coarser chromatin granules and a more distinct membrane, which belong to lymphocytes. '^Between the connective tissue cell, especially the young forms, and the lymphocyte one can see every possible transition" (1905, p. 371). Saxer likewise found that '^the lymphocytes, which later form the bulk of the lymph glands, arise in loco."

The examination of the bridges in the 42 mm. embryo shows the pale oval cells and the darker round ones apparently derived from them, and indicates that these trabeculae contain lymphoid tissue. They do not, however, constitute a lymph gland, but represent the material from which the chain of deep cervical lymph glands is to be derived. A sufficiently detailed study of the later stages of the plexus has not yet been made. Bonnot' believes that it produces the "interscapular gland" of Ilatai, which seems to be a collective term for the cervical fat and lymph glands.

Almost simultaneously with the lymphoid transformation of trabeculae among the jugular lymphatics, distinct lymph glands appear in the superficial tissues. These do not pass through a plexus stage, but from the first they resemble the glands of the adult. The striking diiference in the arrangement of the deep and the superficial lymphoid tissue seems due to the fact that the deep tissue is molded about an involved pre-existing plexus; but the superficial glands develop freely in the loose subcutaneous tissue. The plexus stage may therefore be regarded as a complication in the development of the glands, rather than a fundamental condition which is sometimes hurried through, modified, or omitted.

In the human embryo of 42 mm. two superficial glands were found on either side of the head. Their position is indicated in Fig. 7. The smaller gland is in intimate relation with the submental branch of the anterior facial vein. A section through it is shown in Fig. 8.

•Bonnot, E.. Tlie interscapular gland. Journ. of Anat. and Phys., 1908, vol. 43, pp. 43-58.

D Lymph Glands in Kabbit and Human Embryos.


At the upper border of the photograph a part of Meckel's cartilage is seen on the left, and the bone of the lower jaw on the right ; the lower border of the photograph passes through the submaxillary gland. Between the submaxillary gland and the mandible the submental vein appears, surrounded by dense tissue. This dense lymphoid tissue is chiefly on the upper side of the vein, and it is bounded by a lymphatic

Fig. 7

Fig. 7. — The head of a human embryo of 42 mm., to show the position of the submental ("submaxiUary") and external jugular glands. X 2-2/5 diams. (H. E. C, 841.)

Fig. 8. — The head of a rabbit embryo of 29 mm. to show the position of the IKisterlor facial gland. X 4 diams. (II. E. C, 170.)

The veins shown are the anterior and posterior facial, the linguo-facial, the external and internal jugular, and the jugulo-cephallc. (The external jugular of man corresponds with the jugulo-cephallc of the rabbit and not with the linguo-facial ; the latter in the rabbit is, however, usually called the external jugular. Cf. I^ewis, Amer. Journ. of Anat., vol. 9, p. 33.)

vessel, crescentic in section. The submental vein sends branches into and through the lymph gland.

The other lymph gland in the head is in relation with the external jugular vein. It is shown in section in Fig. 4. Lymphoid tissue, enclosing small blood vessels, forms a rounded mass attached to the lower part of the vein. Its free surface is in relation with a crescentic lymph sinus. No other lymph glands were found in the head of this embryo.

D 346

Frederic T. Lewis.

In a human embryo of 30 mm. the submental® and external jugular glands were not found. They are not mentioned in four embryos of 46-50 mm. described by Miss Sabin, but she has recorded that in an

Fig. 9. — Keconstructlon of the arteries In the axilla of the human embryo of 42 mm., to show the position of the first axillary lymph gland. X 10 dianis. (H. E. C, 838). The subscapular branch of the axillary artery is seen to divide into the circumflex scapular and thoraco-dorsal arteries. The lymph gland is along the latter. The brachial and lateral thoracic arteries are also shown:

80 mm. embryo "there are secondary lymph nodes along the veins of the neck; for example, along the external jugular vein next the parotid gland and along the facial vein at the angle of the jaw."

•It seems desirable to name the early lymph glands for the veins which they accompany and this has been done. It is to be noted, however, that in the adult there are several glands along the submental vessels, the anterior ones forming the submental group, and the posterior ones the submaxUlary group. The submental gland of the 42 mm. embryo belongs evidently with the submaxillary group of the adult.

D Lymph Glands in Rabbit and Human Embryos. 34:7

Since the early lymph glands develop with such regularity in the rabbit, it seems quite possible that these glands noted in human embryos of 80 mm. are the ones appearing at 42 mm.

The Harvard collection includes three rabbits of 29 mm. (20 days) cut in the transverse, sagittal and frontal planes respectively.

Fio. 10. — A, reconstruction of the arteries in the axilla of the 20 mm. rabbit, to show the position of the first axillary lymph gland. X 10 diams. (11. E. C, Series 170). B, reconstruction of the arteries in the pelvis of the same embryo, to show the first pelvic gland. X 10 diams.

The arteries labelled are the axillary, brachial, subscapular, aorta, iliolumbar and hypogastric.

These embryos all show a lymph gland near the junction of the anterior and posterior facial veins (Fig. 8). Except at this point, no lymph glands were found in the head.

The most distinct lymph gland in the body, in these rabbits and in the human embryo of 42 mm., is in the axillarv rcffion. In the human embryo it is an accumulation of lymphoid tissue surrounding

D 348

Frederic T. Lewis.

the circumflex scapular artery and vein, and forming a lenticular mass bulging into the accompanying lymphatic vessel. Its position is shown in Fig. 9, and a section through it is photographed in Fig. 5. It lies next the muscle in the deep subcutaneous tissue. This gland was not found in a 30 mm. embryo, although at that stage the circumflex scapular vessels are accompanied by lymphatics. It is not specifically mentioned by Miss Sabin, and if it occurred in the embryos studied by Kling it was overlooked. At 70 mm. he found all of the axillary groups represented except the subscapular group

(p. r,88).





Fig. 11. — Wax reconstruction of the bmnan axillary gland shown In Figs. 5 and 9. X 40 dianis. A, dr. s<\ V. cit\ sc, circumflex scapular artery and vein; x. y. z., small bIo(Kl vessels, of which j? is so surrounded by lymphatic vessels, V. Jym., that it seems to perforate them; L.-gh, L.-gV,, nodules of lymphoid tissue.

A corresponding gland occurs in rabbit- embryos. It can be identified in a specimen measuring 25 mm. (18 days), and it is well defined in all three of the 20 mm. embryos. It is in relation with the subscapular vessels, which are relatively large in the rabbit (Fig. 10.4). A section through the gland is shown in Fig. 6.

Since the axillary glands seem to be the largest and most clearly defined, they were reconstructed in wax. The gland in the human embryo is shown in Fig. 11. Along the toj) of the model the circumflex scapular artery and vein pursue a parallel course, accompanied by the lymphatic vessels, V. lym. As the blood vessels approach the gland the mesenchyma around them becomes condensed and forms an intensely staining mass of lymphoid tissue, Ij-gV. Both artery and

D Lymph Glands in Rabbit and Human Embryos. 349

vein are surrounded by this tissue, but the vein seems more deeply embedded. The lymphoid tissue extends for some distance along these vessels and forms a second nodular swelling, L.-gL The position of these swellings may be determined by the small branches of the blood vessels, y and 2;, which they accompany. The main mass of lymphoid tissue, L.-gL, forms a lenticular nodule bulging into the perivascular lymphatic; it has been exposed by removing a part of the wall of the lymphatic vessel. In the photograph, Fig. 5, the dark

Fio. 12. — Wax reconstruction of the axillary gland of the rabbit shown in Figs. 6 and 10 A. X 56 diams. A. subsc, V. subsc, subscapular artery and vein; V. lym., perivascular lymphatics; L. gl., lymph gland; a, &, c, d, the blood vessels correspondingly lettered in Fig. 10 A.

oval area is L,-gL of the model, and the somewhat triangular mass above it is L-gt; in the midst of the latter the vessel z may be seen. Fig. 5 is therefore a horizontal section of the model. The fact that there are two nodular masses of lymphoid tissue connected with one another suggests the twin glands (Zwillingsdriisen) which Kling regarded as malformations due to incomplete subdivision. The bulging of the lymph gland L,-gl. into the lymphatic vessel recalls the following obser\ation by Ranvier :^ "Whenever I have observed a vascular nodule on a lymphatic, the latter has appeared to be interrupted.

^Ranvier, L., Morphologie et d^veloppement du syst^me lymphatlque. ArHi. d'anat. mlc, 1807, vol. 1, pp. 137-152.

D 350 Frederic T. Ixjwis.

. . . . Thus, the lymphatic, divided at the level of the nodide, forms two trunks, of which, the inferior becomes an afferent and the superior an efferent. If a new gland forms along the course of the efferent the latter will become the afferent for the second gland. The efferent for one gland may be the afferent for another."

The subscapular gland of the rabbit is shown in the model, Fig. 12. The subscapular vein and artery are spun about with perivascular lymphatics, which extend along the branches of the blood vessels, a, b, c and d. (Compare with Fig. lOA.) The lymph gland L,-gl. is seen through a window cut in the lymphatic vessel. It rests upon the subscapular vein and bulges into the lymphatic, pushing the endothelium before it. That the gland is more intimately related to the vein than to the artery is shown in Fig. G. The upper portion of the gland is irregularly subdivided, so that in one or two sections there is a suggestion of the plexus formation ; lower down it forms a single rounded mass.

In addition to the well-defined gland in each axilla, other glands were found in the thoracic region of both the human and rabbit embryos. In the human embryo there is indication of lymphoid tissue along the dorsalis scapulae vessel and a somewhat diffuse gland near the anterior end of the internal mammary vein. Where the pleuropericardial septum joins the diaphragm a branch of the internal mammary vein passes inward, accompanied by a large lymphatic. !N^ear the junction of the septum and diaphragm lymphoid tissue is found in relation with these vessels. The left pleuro-pericardial septum is thinner and farther from the median line than the right and has no corresponding lymphoid tissue. In the rabbit there is a developing gland along the thoraco-epigastric, or external mammary vein, nearly opposite the elbow.

The glands of the head and thorax have now been described; the abdominal and pelvic regions remain to be considered. In four rabbit embryos of 29 mm. a gland was found along the ilio-lumbar vein on either side of the body. It appears to be smaller than the axillary gland, but has essentially the same features. It is more deeply placed than the other glands. The ilio-lumbar vessels (Fig. 105) have extensive subcutaneous branches, /, /, f, and a branch, e, to the abdominid

D Lymph Glands in Eabbit and Human Embryos. 351

musculature. As noted by Krause, the ilio-lumbar vessels are highly developed in the rabbit. The lymph gland is found, as shown in Fig. IOjB^ where the subcutaneous branches join the main stem.

In the abdominal part of the human embryo of 42 mm. no distinct glands were found, but along the femoral vessels, in the inguinal region, there is a suggestion of lymphoid tissue. At 50 mm. Miss Sabin describes the posterior lymph sacs as lying in the side of the pelvis opposite the first three sacral vertebrae, and states that "the entire dorsal wall of the sac is occupied by a lymph node" (1909, p. 87). A gland which extends over three sacral vertebrae is clearly unlike any gland in the adult. The structure referred to seems to be the plexus of deep lymphatics, among which lymphoid tissue has appeared, but has not yet formed glands. At this stage Miss Sabin speaks of "secondary nodes" developing near the sacs along the femoral and sciatic groups of veins. In an 80 mm. embryo she describes a true lymph gland which, from its structure and position, as shown in a figure, strikingly suggests the ilio-lumbar gland of the rabbit ; it is not stated along what vessels it occurs. The description of the gland is as follows : "In Fig. 19 is a tiny lymph node . . . which illustrates well the simplest form of a lymph node, a central mass of lymphocytes with a plexus of lymph ducts around it. This plexus of ducts is so close that it may already be termed a sinus, so that the node consists of a single follicle with its peripheral sinus." It may be noted that Miss Sabin has figured such a simple gland in the lung of an adult pig (1905, p. 885), and Kling has described them in the axilla of an adult man.

From the preceding study the conclusion may be drawn that the first definite lymph glands are superficial. They appear with surprising regularity, as shown by comparing the three rabbit embryos of 29 mm. They are situated along the large cutaneous veins, and there is a well-developed pair for the head, thorax, and abdomen respectively. In addition to these, the rabbit embryo of 29 mm. gives evidence of gland formation along the thoraco-epigastric vein. The human embryo of 42 mm. differs from the rabbit of 29 mm. by the absence of the ilio-lumbar gland and the presence of the submental gland, together with indications of glands along the internal mam

D 352 Frederic T. Lewis.

mary and femoral veins. Doubtless, both in man and the rabbit the development of additional glands proceeds rapidly.

At the time when the superficial glands are distinct the deep ones are^ represented by lymphoid trabeculae, which are said to be transformed into chains of glands by the accumulation of the lymphoid tissue in nodules. Something of this sort must occur, but models showing the development of such a chain have not yet been made. It seems undesirable to speak of an extensive plexus of lymphatic vessels, even when associated with diffuse lymphoid tissue, as a lymph gland.

At the time when the lymph glands and trabeculae arise — that is, in the embryos which have now been described — there is apparently no lymphoid tissue elsewhere in the body. The spleen is well developed, but the compact tissue of which it is composed does not appear like that of the lymph glands. The thymus at this stage, in the rabbit at least, is clearly an epithelial organ. This is contrary to the statement of GuUand,^ that ^^the thymus in mammalian embryos is the first place where true adenoid tissue is formed, and it is an active center for the production of leucocytes long before lymphatic glands are formed at all.

The question of the origin of lymphocytes can be answered only by examining thin and specially stained sections. The embryos here described were prepared for general study, and the sections are 10 microns or more in thickness. They suggest, however, that the lymphocytes are forming in the glands and that they are absent from the blood. Maximow, who has studied the embr^^onic development of the blood with faultless technique, has unfortunately not examined the earliest lymph glands.® He considers that "the first leucocytes, the lymphocytes, arise at the same time and from the same source as the primitive erythroblasts. The latter represeat a specially differentiated form of cell, but the lymphocytes always remain undifferentiated.

"GuHaud, G. L., The development of lymphatic glands. Journ. of Path, and Bact., 1894, vol. 2, pp. 447-485.

•Maxinow, A., Untersuchung Uber Blut und Bhidegewebe. I. Die frflhesten EntwickluDgsstadlen dcr Blut, etc. Arch. f. mik. Anat, 1909, vol. 7H, pp. 444-561.

Digitized, by

Lymph Glands in Rabbit and Human Embryos. 363

Therefore, like the primitive blood cells from which they directly proceed, they are undifferentiated rounded amoeboid mesenchymal cells." He states that these lymphocytes of the embryo "have nothing to do with lymphoid tissue" — ^they develop in the yolk sac.

The lymphocytes of the lymph glands are, indeed, round mesenchymal cells, but, except for an occasional cell in the lymph sinus, apparently detached from the gland, they are unlike the forms of corpuscles in the adjacent vessels. It seems probable that the lymph glands, arising in rabbit embryos of 25-30 mm. and in human embryos of 30-45 mm., are the source of a special form of round mesenchymal cell, which is the true lymphocyte. This opinion can be established or disproved only by a cytological study of the early lymph glands, the position of which has been indicated.



An Atlas of Skiaobams. Illustrating the Development of the Teeth, by Johnson Symington, M.D., F. R. S., Belfast Longmans, Green & Co., London, Publishers.

An entirely new method of presenting the development of the temporary and permanent seta of teeth has been brought out by Dr. Johnson Symington, of Belfast, in a series of Rontgen ray pictures taken of the skulls of children of various ages. The skiagrams, twelve in number, have been made from eighteen children, ranging in ages from birth to sixteen years, and one adult.

In the preparation of the charts, the skulls were split in half through the median line, one-half being laid on a photographic plate, and the Rontgen rays applied from above. The plates thus made were developed full size, without any retouching or correction. In addition to these five drawings were made of dissections of certain of the skulls, in relation to the position and size of the maxillary sinus, the drawing being life size. Accompanying all of the charts and drawings there is a short description of findings appended to each.

Upon examination of the work the most striking characteristic is the excellence of the skiagrams and the remarkable clearness of their detail. The developing teeth are clearly shown, lying in crypts in the bone, and their relation at each stage can be determined. In the early cases, the beginning of enamel formation is very picturesquely shown, the first formation being in the tips of each cusp, and in the multicuspid tooth a union of the enamel from the several cusps occurs, forming the whole crown.

The growth of each tooth may be traced through its various stages, from beginning to completion. This is especially true in the lower jaw, in which the teeth are more distinctly shown than in the upper. In the upper jaw, the bones of the face more or less obscure the outline of the teeth. In this respect, the extent to which the tooth is formed at the time of its eruption may be clearly seen — so also, the time of complete calcification of the roots, with the


D Book Reviews. 355

closure of the apical ends of the root canals, can be easily determined.

The relation of the forming permanent tooth to the corresponding temporary one, which it is to supplant, is every nicely brought out, showing the amounts of calcification of the permanent tooth at various ages, and at the same time the amounts of resorption of the temporary tooth. In the case of the permanent molars, which do not supplant any of the temporary teeth, their relation to the jaw and to the teeth present is very interesting, inasmuch as the direction of growth and exact position in jaw can be very clearly made out. An excellent opportunity is offered in these charts for the study of the relations of the teeth, both temporary and permanent, in regard to the lateral pressure which they exert in their growth, and its effect upon the widening and elongation of the dental arch.

In several instances, in these figures, there has been a loss of the temporary tooth before the corresponding permanent one was sufficiently developed for eruption, and the subsequent dipping of the adjacent teeth' and tendency to close the space is clearly shown. In one individual, both lower first molars were lost, and the roots, which were remaining in the jaw after the destruction of the tooth by caries, have become exfoliated to the surface, and below them is shown a marked condensation of the bony structures.

There is little of corollary work, with which this treatise may be compared, outside of the dissections made of the skulls of children, with a removal of the outer plate of the jaws, uncovering the developing teeth in position, or photographs of the same. Although this atlas is scarcely as efficient in its presentation of the subject as the actual dissections, it certainly excels any manner of photograph or text illustration which we have. The views are more comprehensive than any single plane photograph can be. So that the figures made are very commendable for text illustration, and the atlas in itself will be of great service in presenting the subject both to student and to practitioner.

liusspU W. Bunting, D.D,Sc.,

TJniv. of Mich. Received for publication, March 15, 1909.

D 356 The Anatomical Record.

A Study of the Causes Undeblyino the Obioin of Humah Monsters. Franklin P. Mall, Jour, of MorpL, Vol. XIX, Feb., 1908.

The recent publication by MaU on the causes underlying the origin of human monsters marks an egpch in the study of teratology in this country, for he has treated the subject with a breadth of view and a wealth of illustration rarely found in the handling of this complex question. Mall has brought to the task a profound knowledge of the older literature of the subject, an appreciation of the most modern results in experimental teratology, and a thorough familiarity at first hand with the subject of human monsters. The physician and anatomist are brought into close touch with the work generally supposed to be outside of their proper field ; and on the other hand the student of malformations in the lower animals will be made to appreciate the inexhaustible supply of human materials with which the anatomist and physician are familiar. MalFs material consisted of 163 pathological human embryos that include nearly all of the commoner forms of human abnormalities. He points out, that from the earliest ages the study of monsters and the causes that produce them have been two of the capital problems of anatomy, medicine and natural history; that the ancient belief of their supernatural causes has been replaced in part by the theory of maternal impressions ; that this belief also has gradually been replaced in turn by the theory that monsters are germinal or else produced by external causes adverse to the normal development of a normal germ cell. Recognizing that some of the monstrous forms may be germinal, Mall argues with great ability that the great majority of monsters such as anencephaly, spina-bifida and cyclopia are due to external agencies affecting the germ, chief amongst which he recognizes faulty implantation of the embryo in the uterus. Faulty implantation by interfering with the nutrition of the embryo is recognized as the chief mechanical factor in producing the result. When it is recalled that the human ovum is extremely small at its beginning, that it is without a store of yolk and must depend for its growth on materials absorbed from the mother through the placenta, there can be little

D Book Eeviews.

doubt that Mall has indicated the chief sources of abnorm ment. On the other hand, nutrition per se is probably n< cause, for an abnormal condition of the uterus implies the of injurious substances, and these, aside from nutrition, the normal sequence of changes. Mall recognizes in fact ditions as another source of imperfection. He is less fa^ clined to the supposed influences of adhesions, strangulatic like, and suggests that such fusion and restraints are of rather than of primary importance. To the reviewer it too little weight is attached to these factors, although it is too great weight has often been given to them, and on the -^ has done good service in drawing attention away from these towards the other more prolific sources of malformations. Mall's observations that "in very early stages the amnions and embryos are equally susceptible, and the umbilical vesicle and chorion are the most resistant" while "later it is the embryo alone and still later the head, central nervous systems and extremities" that are most affected is a generalization of great importance.

It is pointed out that, while the earlier of the modern teratologists were first inclined to the view that polyspermy is the cause of double monsters, later researches have rendered this view improbable. The early separation of embryonic cells has been shown to produce directly double structures. Almost identical results have been obtained, however, by artificial constriction of later stages. In passing it is also interesting to note that results, externally indentical, may be also produced by chemical reagents.

From details in Mall's monograph concerning pathological ova in relation to the conditions in the uterus; for the cases in which successive births have given rise to pathological embryos; for the relation between tubal pregnancies and abnormalities, etc., the reader must be referred to the able treatment of the problems involved. There is open here a wide field for experimental study on lower mammals where the conditions might be artificially induced.

Following the custom of human embryologists Mall speaks of the earlier embryos and their membranes as ova — a term that will not recommend itseK to the student of general embryology, for here the ovum has a distinct and definite meaning.IC

358 The Anatomical Record.

The embryos that Mall has studied are grouped according to "weeks" from the second to eighth, week. It is impossible in a brief review to do justice to the interesting facts here considered. In all, 138 pages are devoted to this general treatment of the problem; the remaining 223 pages give the data for the individual cases considered.

This great monograph on human teratology should excite widespread interest in this important field of anatomy. The standard here set in the treatment of the questions involved is certain to have a beneficial result on the study of normal and pathological anatomy in this country.

T. H. Morgan, Received for pubUcation, March 22, 1909.


Dr. S. Walter Eanson will have charge of the work in anatomy at the Northwestern University Medical School.

In the department of Anatomy in the University of Virginia, Professor H. E. Jordan, formerly Adjunct Professor, has been advanced to the rank of dissociate Professor. Mr. • J. A. Waddell, A. B., has been appointed Instructor in Anatomy.

Dr. Arthur W. Meyer has been appointed Professor of Human Anatomy in the Leland Stanford Junior University of California. He is to have charge of the gross anatomy, physical anthropology and human embryology, the work being given at Palo Alto. The histology is given by Professor McFarland, the neurology by Assistant Professor Stoltenberg, while the general embryology is given in the department of zoology.

In the University of Missouri Dr. Caroline McGill is to have a year's leave of absence to accept the fellowship in anatomy awarded her by the Naples Table Association for Scientific Research. She will spend the summer working in the laboratory at Woods Hole and in the fall will go abroad, where she plans to work at the universities of Berlin and Tiibingen.

Mr. L. G. Lowrey has been appointed assistant in anatomy.




Vol. III. JULY, 1909. No. 7



C. M. JACKSON. From the Anatomical Lahoi-atory, University of Missouri, Columbia.

With 2G Figures and 1 Table.

The purpose of the present paper is to describe the principal changes- in position which the viscera undergo during the process of development, and to discuss some mechanical principles involved in these changes. The materials used include a considerable number of human embryos and fetuses in my collection. Several of these are cut in serial sections, and from four of them models were reconstructed by Bom's wax-plate method. The embryos modeled are No. 60 (5th week, 11 mm. crown-rump length), No. 58 (6th week, 17 mm. crown-rump length), No. 57 (8th week, 31 mm. crown-rump length), and No. 55 (12th week, 65 mm. crown-rump length). I am greatly indebted to several of my students for assistance in reconstructing these models. That of No. 60 has already been described by Bonnot and Seevers.* The model of No. 57 was in large part reconstructed by C. B. Rodes, Jr., and that of No. 55 by A. W. Kampschmidt and F. O. Kunz. The volumes of the various organs and parts of these embryos (excepting No. 55) have been given by nie in another paper. ^

\Vnatoini.seher Anzeiger, Bd. 21), 1000, S. 4r)2-4r)0. •American Journal of Anatomy, Vol. VIII, No. 1, 1009.


D 3G2 C. M. Jackson.

The models arv especially designed to show the topographic relations of the thoracic and abdominal viscera to the skeleton, the projections of the sternum, ribs and vertebral centra being indicated upon the underlying organs in the models as shown in the various figures. In order to show the general jwsition and relations of the organs repnsented in the models, with reference to the body as a whole, graphic reconstructions were made (Figs. 1 to 4) showing from the left lateral view the outlines of the body, ribs, sternum, vertebral column, etc. The parts corresponding to the models are indicated in the figures by the shaded (stippled) areas.

In the following paper, the gc^neral relations of the viscera to tlie body wall, ])articularly to the vertebral column, will be discussed first, followed by a brief consideration individually of the principal viscera.

Relations to the Vertebral Axis.

The most striking and important feature in the relations of the viscera to the vertebral column in the embryo is the so-called migration of the various organs along the body axis during the course of development. This phenomenon, which is of general occurrence among the higher vertebrates, has long been knowni, but its cause has never been fully and satisfactorily explained.

In general, we recognize that the migration of an embryonic organ may be either active or passive in character. An active migration is due to the (amoeboid?) movement of the component cells, e. g., of the early anlages of the sympathetic ganglia. It therefore does not require actual growth (increase in volume) of the parts concerned. Passive migration, on the other hand, is produced by pressure or tension due to unequal growth, relatively more or less rapid, either in the organ displaced or in neighboring organs or parts. It is evident that the migration of the viscera along the body axis belongs to the latter class. This principle was recognized by His,' who was the first to investigate in a systematic Avay the subject of developmental mechanics.

"Unsere KiJrperforni, Leipzig, 1874; also in Archiv f. Anat. u. Entw., 1894, S. Iflf.

D The Thoracic and Abdominal Viscera. 363

His recognized that the heart and other organs which lie far forward on the ventral side of the embryo at a very early period are at first pushed backward (eaudad) by the ventral flexure of the body axis in the head region. Up to the third week in the human embryo, while the body axis remains comparatively straight, the heart lies chiefly ventral to the head region. During the third and fourth weeks, owing to the flexures in the head region, — the primary cephalic flexure opposite the midbrain, and the cervical flexure at the junction of the head and neck — the heart and adjacent organs are pushed backward farther and farther. In the fifth week (Fig. 1), at the close of the period of greatest flexure of the body axis, the heart lies entirely opposite the cervical region of the vertebral column. It is difiicult to determine with certainty the exact vertebral level of the various organs at this period, on account of the extreme flexure of the body.

The cause of this flexure ventralward of the body was originally explained by His as due to the restraint of the amnion upon the elongating embryo ; but it is now generally believed to be due to the early overgrowth of the central nervous system^ b'^^^S dorsal to the body axis. Later, when the structures lying ventral to the body axis begin to expand more rapidly, the head becomes more erect and the cervical flexure is. largely obliterated.

The influence of the brain and spinal cord in producing the flexure of the body is well sho^vn in embryos with anencephaly and spina bifida. In the absence of brain and spinal cord, the ventral flexure of the body axis is reduced or absent. An anencephalic embryo of about two months (27 mm. head-trunk length) in my collection shows no ventral flexure of the body ajcis. The axis is nearly straight, the face being directed ventralward. In older fetuses of this character, as is well known, the face looks upward as well as forward, giving the so-called "frog-face" expression. In a mid-sagittal section of such a fetus, I find the base of the cranium making an angle dorsalward at the junction with the vertebral column, instead of ventralward as in the normal condition. This is evidently due to the unopposed expansion in the facial region, not meeting the normal resistance of the cranial contents. A study of younger embryos (first

D 364 C. M. Jackson.

and second months) of this class ought to throw more light upon the relation of the growth of the central nervous system to the production of the body flexure. This much is certain, however, that the absence or rudimentary condition of the brain and spinal cord (at least in the later stages) does not prevent the descent of the thoracic and abdominal viscera; for I find these organs at approximately their normal vertebral levels in fetuses with this deformity.

A study of the relations in the normal embrj'o leads to a similar conclusion. However important the ventral flexure of the body axis may be in displacing the heart and other organs in the early embryo, it is certain that this is not the only factor involved. At the end of the period of greatest flexure, as seen in the 11 mm. embryo (Fig. 1), the viscera still lie far above (i. e., cephalad to) their permanent levels. The continued descent of the viscera along the vertebral column is clearly apparent when Figs. 1, 2, 3, and 4 are compared. His* also recognized that the displacement of the viscera is only in part accounted for by the ventral body flexure. The final displacement was explained as an apparent migration of the viscera downward along the body axis. He states that in reality the cervical portion of the spinal cord (and column) undergoes a relative elongation about tlie sixth week, so that it actually pushes upward into the head region, thus moving apast the ventral viscera, which are left lying opposite a lower level on the vertebral column. The neck prominence (Nackenhocker) characteristic of this period he explained as due to this pushing upward of the cervical portion of the spinal cord.

On closer examination, however, there appears no valid evidence to support this view. In fact, His's own figures show no relative elongation of the cervical portion of the spinal cord or vertebral column at this time. On the contrary, there is during this period a constant decrease in the relative length of the cervical region, at least in the vertebral column. In order to show more exactly the growth in relative length in the various regions of the vertebral column during

Anatoiiiie mensohl. Enibryonen, I, S. 78; III, S. 120 flf. Cf. also Archiv f. Anat., 1881. S. 319.

D The Thoracic and Abdominal Viscera. 365

prenatal life, I have arranged Table I, based upon my own observations and those of Aeby,® His,® Merkel,^ Bardeen,® and Ingalls.®

In Table I, the cervical region in embryos under 10 ram. (before the vertebra) are well differentiated) includes the eight cervical somites. There is evidently a constant decrease in the relative length of the cervical portion of the column, with a corresponding increase in the lumbar region, as was first shown by Aeby (Z. c). The change, as may be seen in the table, is most rapid in the first six weeks. During this time the cervical region decreases from about 35 per cent to about 30 per cent of the length of the corvicothoraco-lumbar division of the column. The lumbar region increases at the same time from about 17 per cent to about 22 per cent. During the remainder of the entire fetal period, the cervical region decreases from about 30 per cent to about 25 per cent; while the lumbar region increases from about 22 per cent to about 27 per cent. Aside from individual variations, the relative length of the thoracic region remains fairly constant (average about 48 per cent). The sacral region at the beginning increases in relative length, reaching approximately its permanent relation (individual variations excepted) about the end of the first month. The coccygeal region apparently reaches its maximum relative length in the second month, though it is exceedingly variable.

But even if there were, as His supposed, a relative elongation of the cervical region of the vertebral column in the second month, this would not account for the migration of the viscera downward along the thoracic as well as the cervical vertebne. So it is evident that some other cause must be sought. And that cause, it seems to me, is in gc^neral the relatively more rapid growth beginning at this time in the structures lying ventral to the body axis. This may be noted particularly in the cervico-facial (pharyngeal) region, as may be seen by comparing Figs. 1, 2, 3, and 4. The relatively rapid ex •Archlv f. Anat. u. Entw., 1879.

'Anatomle menschl. Embryonen, I, S. 97.

'Menschliche Embryonen auf Medlanschnitteii uiitersueht. Gottiiigen, 1894.

■American Journal of Anatomy, Vol. IV, p. 277, 1905.

•Archlv f. mikr. Anatomie, Bd. 70, 1907.

D 1


C. M. Jackson.


Relative Length of the Various Regions of the Vertebral Column in the Human Embryo and Fetus, expressed in Percentage of the Cervico-Thoraco-Lumbar Division. (Percentage of the Entire Column in Parentheses.)


Oown- I Total Rump I>enirth Lenfftn. ofCol'mn Mm. Mm.

His 4.

Ingalls 4.9




7.5 10. 11. 11. 13. 14. 16.






Barieen ..



Bardoen. . Jackson . .


Bardeon.. Bardeon..


Jackson. .



Jackson 23.

Bardeon \ 3(».

Jackson 31.

Bardeon 33.

Jackson I 35.

Merkel ; 43.

Bardeon 46.

Bardeen 50.


Jackson ' 65.

Morkol 1 73.

Merkel 89.

Aeby Aeby Aeby Aebv

Morkcl I 140.

His , 143.

Merkel | 170.

Jackson i 190.

Jackson I 205.

Jackson 1 210.

Aeby 1

Jackson 265.

Morkcl I 310.

Jackson 320.

Jackson 340.

His Morkol 365.

Aeby (avorage of 8


Aeby (average of 13



227. 247.





Per Cent.

35.1(28.5) 36. (31. ) 30.7

34.5(28.1) 30.8(22.9) 35. (26.5) 30.5C25.6) 31. (23.1) 31.9(23.8) 30.5(25. ) 29.9(22.7) 31.7(24. ) 30.6(23.9) 31.1(24.1) 29.3('23.4) 29.2(23. ) 27.9(23.2) 29.3(23.7) 20 8(24.3) 28.9(22.3) 26.1(21. ) I 27.7(22. ) 27.9(21.3) 29.2(23.1) 28.2(22.2) 25.4(19.7) 25.5(20.1) 28.1(22.5) 25.6(20.1) 26.1(20.6) 26.4(20.9) 25.5(20.9) 24.5(19.3) I 28.9(21.8) 25.9(20. ) 24.5(19.5) 23.9(19. ) 25. (20.4) 22.1(20.9) 24.

22.9(17. ) 24.5 23.9

23.6(18.5) 20.6(15.3)


Thoracic Reieion. Per Cent.

Lumbar Region. Per Cent.

Sacral Rejpon. Per Cent.

Coceyseal Refiion. Per Cent.

47.7(38.6) , 46. (40. ) , 50.8 '

45. (36.7) 1 49.2(36.6) I 44.4(33.6) I 46.7(30.2) '

46. (34.2)

48. (35.7) , 46.7(38.3) 1 47.1(36.5) 47.5(35.9) 46.9(37.1) 50. (38.7) 47.4(37.7) 47.2(37.2) 51.5(42.7)

49. (39.5) 47.9(37.8) 48.5(37.3) 49.1(31.5) 48.2(38.3) 47 5(36.1)

1 47.5(37.5) , 47.9(37.8) 48.3(37.4) I 50. (39.6) I 46.6(37.2) i 48.7(38.1) I 48.2(37.9) . 48.6(38.4) ■ 49. (40.1) , 50. (39.3) , 47.4(35.7) ^ 48.2(37.2) I 50. (39.9)

48.6(38.9) I 48.7(40.1) I 49.6(38.4)


50.5(37.6) I 48.5 , 48.9



17.2(13.9) 18. (15. ) 18.5

20.5(16.7) 20. (14.9) I 20.6(15.5) ! 22.8(19.2) 23. (17.1) , 20.1(15. ) ' 22.8(18.7) I 22.1(16.8) 20.8(15.7) 22.5(17.9) 18.9(14.6) 23.3(18.5) 23.6(18.6) ' 20.6(17.1) I 21.7(17.5) 21.3(16.9) 1 22.7(17.5) ' 24.8(20. ) 24.1(19.1) 24.6(18:7) 23.2(18.3)

24. (18.9) 26.3(20.4) I 24.5(19.4) I 25.3(20.1) 25.6(20.1)

25.6(20.1) 25. (19.8) 25.5(20.9) 25.5(20. ) 23.7(17.8) 25.9(20. ) 25.5C20.3) 27.5(21.9) 26.5(21.9) 28.3(19.8) 27.4

26.7(19.9) 27.


25.8(20.3) , 26.5(19.8)

12.5(10.1) 10.9(89) 12. (11. ) 4. (3. )


[22.8(18.6)] 19.2(14.3) 15.4(11.4) 20.6(15.5) ' 11.1(8.8)

[18.3(16.0)K?) 19.5(14.5) 14.9(11.1)

18.7(14. ) 13.7(11.3)


8.1( 6.7)

17.4(13.2) I 14.3(10.8) 17. (12.8) 15.4(11.6)

17.8(13.9) 17. (13.1) 17.3(13.8) 18.9(15. ) 14.0(11.6) 16.3(13.2) 18.8(14.9) 19.8(15.3) 18.2(14.6) 18.8(14.9) 18.6(14.2) 17. (13.4)


17.8(14.2) 18.4(14.7)


9.2( 7.2) 12. 2( 9.5) 8.3( 6.6) 8.1( 6.4) 6.6( 5.5) 7.5( 6.1) 7.8( 6.1) 9.9( 7.6) 6.1( 4.9) 7.1(5.7) 12.7(9.7) 9.7( 7.6)

17.9(14.1) 21.1(15.8) 19.6(15.2) 20.6(16.4) 18.3(14.6) 16.8(13.9)

10. 5( 8.2) 8.5( 6.7) 6.8( 5.4)





9.4( 7.4) 11. 8( 8.9) 9.8( 7.6) 5. ( 4) 7.3( 5.8) 4.4( 3.7)


21.9(16.3) 12. 4( 9.2)

19.7(15.4) 22.5(16.8)

7.4( 6.2) 11.8(8.8)

47.5(37.9) ' 26.8(21.4)


21.5(17.0) 1 46. (36.4) 32.3(25.4) I 20. (15.8) , 6.75(5.4)

D The Thoracic and Abdominal Viscera. 367

pansion in this region exerts pressure in all directions. The pressure upward against the massive head, on account of the great resistance, produces comparatively little displacement. The effect is merely to straighten out to a large extent the cephalic and cervical flexures. (The exception in Fig. 4 is only apparent, the cendco-cephalic angle being in reality less marked than in the preceding stages.)

The effect of this expansion in the cervico-facial region is much more marked in the caudal direction, however, where there is relatively more room and the resistance is less. As I have showm in a previous paper (1. c), the alimentary tract, respiratory tract, and in fact most of the organs lying ventral to the body axis (excepting the heart) are also expanding with relatively great rapidity at this time. Thus in the profile views of His,^^ the oesophagus in the 5.5 mm., 7 mm., 10 mm., and 12.5 mm. embryos is found to equal in length about 25 per cent of the body length. In the 13.8 mm. embryo it has suddenly increased to about 40 per cent of the body length. At the same time the respiratory tract (larynx, trachea and lung) increases from about 20 per cent to about 33 per cent of the body length. The stomach similarly increases in relative size and length. The pyloric end becomes fixed at about the 12.5 mm. stage, hence further relative elongation of the oesophagus and stomach increases the obliquity of the stomach in later stages.

During the second month, this period of relatively rapid growth of the viscera is nearly ended. We find accordingly that in the 17 mm. embryo (Fig. 2), and especially in the 31 mm. embryo (Fig. 3), the viscera have reached nearly their permanent vertebral levels, there being relatively little descent after that time. We shall find similarly that the other topographic changes are also most rapid during the first two months. For purposes of developmental to])ography therefore the prenatal term may be divided into two periods, corresponding to those already generally recognized. The first or embryonic period includes the greater part of the first two months, during which the changes are most rapid. During the second or fetal period, from the third month onward, the changes in position are relatively slight.

"Auatomie menschlicher Embryonen, III, S. lG-19.

D 368 C. M. Jackson.

The visceral branches of the aorta, particularly the coeliac and mesenteric, follow their corresponding viscera downward by acquiring successively new roots of origin from the aorta. This was shown first by Mall,^^ later by Tandler^^ and Broman,*^ according to whom these arteries reach their permanent levels of origin during the second month. The nerves, on the other hand, are unable to shift their attachments to the spinal cord, hence their origin indicates approximately the level of the organ at the time it receives its innervation (e. g,, the diaphragm from the 4th cervical).

In addition to the migration caudad along the vertebral axis, the viscera undergo another change in their relation to the vertebral coluum. In cross sections of the trunk in earlier embryos, the body cavity ap|)ears somewhat elliptical in outline (long axis dorso-ventral), and is placed entirely ventral to the vertebral column. Later, as the trunk becomes flattened dorso-ventrally (cf. Rodes^^), the body cavity also appears more roundtd. At the same time, the vertebral column appears to move forward, forming a median dorsal projection into the body cavity. More accurately stated, the body cavity and viscera extend backward (dorsad) on each side of the vertebral axis. The groove thus formed in the viscera by the vertebral column may be called the vertebral groove.

This change is shown in lateral view in Figs. 1 to 4. In Fig. 1 (11 mm.) the vertebral column lies entirely behind the visa^ra. In Fig. 2 (17 mm.) the ribs have appeared, but the body cavity and thoracic viscera still lie entirely ventral to the vertebral column, and the dorsal aorta is visible in lateral view. In the abdominal region, on the other hand, the kidneys and suprarenal glands have begim to move backward on each side of the vertebral column, which projects forward into the shallow vertebral groove b(»tween them. In the 31 mm. embryo, this groove is much dee])er, involving also the lungs, and including about half of the width of the vertebral column as seen in lateral view (Fig. 3). The l)ody cavity and ribs extend still

"Journal of Morphology, Vol. V, 1801.

"Anatomlsche Ilefte, Bd.- 23, S. 180, 1903.

"Anatonilsohe Ilefte, Bd. 37, 1908.

"Zeitschrlft f. Morphol. u. AnthropoL, Bd. 9, 1905, S. 113 ft.

D The Thoracic and Abfloininal Viscera. 369

further backward, the angles of the latter now reaching nearly to the middle of the spinal cord. At 65 mm. (Fig. 4), the groove has become so deep as to include nearly the entire centra of the vertebral column. Corresponding to the width of the vertebral column (centra), the vertebral groove is narrower from side to side in the thoracic region and wider below in the lumbar region (cf. Fig. 20). The pleura at the angles of the ribs extends backward so that in lateral view it covers a large part of the spinal cord (Fig. 4). There is thus formed here (likewise in the 31 mm. and 17 mm. stages) behind the lungs in the pleural cavity a considerable space filled with fluid. In later stages, as the lungs expand the fluid is apparently absorbed, and this space largely disappears. So far as the body cavity is concerned, however, this groove for the vertebral column has nearly reached its I)ermanent relative (fetal) depth at the end of the third month.

The cause of the formation of this vertebral groove in the viscera is in many resjiects difficult to understand. As previously stated, it cannot be explained as due to a pushing forward of the column, which appears to remain relatively fixed in size and position. The change is rather due to movement in the viscera lying ventral to the vertebral axis. The change in the general form of the l)ody wall (which is much elongated dorso-vent rally in the earlier stages, becoming nearly circular in cross-section about the third month) is readily explained as due to the pressure of the rapidly expanding viscera. Since more resistance is offered by the vertebral column, this pressure may be more effective laterally, thus producing grooves on either side of the vertebral column, so that it projects forward from the posterior body wall. Ilenke^^ has proposed an ingenious theory explaining the deepening of the vertebral groove in the thorax during childhood as due to a characteristic growth process in the ribs. This theory, however, will hardly apply to the origin of this groove in the embryo, and fails to account for a similar change in the form of the body wall in the lumbar region. Yet it is quite possible that the formation of the vertebral groove is due to growth processes in the body wall not explainable by simple mechanical principles.

"ADHtomie des Kiiulesa Iters, 2. Aiifl., Tilbingeii, 1881, S. 101 ff.

D 370 C. M. Jackson.

Ribs and Body Wall.

As is apparent in lateral views, the ribs, which extend nearly horizontally forward when first formed (17 mm., Fig. 2, and 31 mm., Fig. 3), soon extend obliquely downward and forward. This obliquity is unusually marked in the 65 mm. specimen (Fig. 4). The obliquity of the ribs is due to the fact that they elongate relatively more rapidly than the corresponding body wall. This may also account for the formation of the angles of the ribs and costal cartilages. There is also a depression of the anterior body wall as a whole, however, as is shown by the oblique course of the abdominal nerves, myotomes, etc., at G5 mm. and later stages. The umbilicus, which from the G5 mm. stage onward is foimd opposite the 4th lumbar vertebra, is located at a higher level in the earlier stages. This is in part, at least, due to a more rapid growth in the anterior abdominal wall, resulting in its relative elongation.


As already mentioned, the heart migrates downward (caudad) from its primitive position in the ventral part of the head region. In the 11 mm. embryo (Fig, 1) it lies opposite the 2d to the 7tli cervical segment; in the 17 mm. (Fig. 2), opposite the 7th cervical to the 5th thoracic; in the 31 mm. (Fig. 3), opposite the 1st to the 6th thoracic; and in the 65 mm. (Fig. 4), opposite the 4th to the 9th thoracic. In later fetal stages, it usually e^xtends approximately from the 3d to the 8th thoracic vertebra.

At the beginning of the second month, the plane of the diaphragm, between the heart and the liver, extends from above and behind in a direction downward and only slightly forward, nearly parallel with the body axis. The downward displacement of the thoracic viscera takes place more rapidly near the dorsal wall than near the ventral wall, however, so that the heart and diaphragm are apparently rotated on a transverse axis. At 11 mm. (Fig. 1), the diaphragm is rapidly approaching the horizontal position, and at 17 rain. (Fig. 2) has pavssed it, now extending from behind upward and forward. The direction is similar at 31 mm. (Fig. 3) ; but at 65 mm. (Fig.

D The Thoracic and Abdominal Viscera. 371

4) the anterior portion of the heart and diaphragm have descended (due to the relative decrease in the size of the liver?) so that the diaphragm has reached a position approximately horizontal.

The long axis of the heart is at first nearly in the mid-sagittal plane, the apex being nearly in the midline at 11 mm. (Fig. 13). It is slightly inclined to the left at 17 mm. (Fig. 14), more so at 31 mm. (Fig. 15), and very decidedly at 65 mm. (Fig. 16). The primitive right side of the heart is thus turned to the front, and the right auriculo-ventricular groove comes to lie behind the sternum near the midline. The asymmetrical position of the heart (which in turn increases the asymmetry of the lungs?) seems to be correlated with the progressive dorso-ventral flattening of the thorax, to which reference has already been made.

The hings in the 11 mm. embryo lie entirely behind and below the heart, being scarcely in contact with it. (Figs. 1, 5, 9, 1). They gradually extend upward and forward upon the auricles at 17 mm. and 31 mm., and upon the left ventricle at 65 mm. (Figs. 1 to 16).

Although the thymus appears in the 11 mm. embrv^o, it does not extend down in front of the heart until in the 65 mm. stage (Figs. 4j 16, th), where it is in contact with the upper part of the right auricle and ventricle. In later fetal stages it continues to expand in the anterior mediastinal space, so that it covers to a large but variable extent the anterior surface of the heart in the new-born.

Lungs. The lungs arise during the first month from the oesophagus in the mid-cervical region, in the angle behind and between the heart and liver (as seen in reconstructions by His, Mall and others). Until the middle of the second month, the lungs remain comparatively small and retain this primitive position. In the model of the 11 mm. embryo (Figs. 1, 5, 9, 17, 1), the lungs are separated by the cpsophagus. They lie on each side in a pocket between the liver and upper end of the Wolffian body below, and arched over by the posterior cardinal vein and ductus Cuvieri above. This pocket is deeper on the right side, which perhaps accounts for the occurrence of an accessory lobe of the lung in this position in the adult more often

D 372 C. M. Jackson.

on the right side. So far as the vertebral level is concerned, the lungs of the 11 mm. embryo correspond approximately to the position later of the lung apex, lying opposite the last oen'ical and the first two thoracic segments, and under cover of the anlage of the upiK»r limb. The migration continues downward along the vertebral column, however, so that the lung in the 17 mm. embryo (Fig. 2) lies opposite the 4th-9th thoracic vertebrae, entirely below the upper limb. In the 31 mm. embryo (Fig. 3), the lung has expanded so as to extend from the 1st to the 10th and in the 65 mm. (Fig. 4) lies opposite the 2d to the 11th thoracic vertebra.

Similarly, the bifurcation of the trachea, which arose in the midcervical region, has in the 11 mm. embryo descended to the level of the 1st thoracic segment; in the 17 mm. to the 3d thoracic; in the 31 mm. to the disk between the 3d and 4th ; and in the 65 mm. to the disk between the* 4th and 5th.

The pulmonary arteries in the 11 mm. embryo arise from the ductus arteriosus and descend along the trachea for a considerable distance Ix^fore reaching the lungs. The distance is relatively less in the 17 mm. embryo, but it is not until the 31 mm. stage that approximately the permanent relation is reached. The change is apparently due, chiefly not to the ascent of the hilum of the lung, but rather to the descent of the heart with the accompanying vessels.

The lungs are relatively small in the 11 mm. embryo, but expand rapidly upward and forward over the auricles in the 17 mm. and 31 mm. stages, reaching the ventricle on the left side at 65 mm. There is little expansion (relatively) in the fetal limg after this time, the anterior borders of the lungs being separated by a wide anterior mediastinal space occupied largely by the thymus. The fetal lungs are relatively largest about the 4th month, when they average 3.3 per cent of the total body volume.

The lobes of the limgs are not distinct in the 11 mm. embryo, but are well marked at 17 mm. The fissures are carried upward and forward by the expansion of the lungs, as may be seen by comparing the relations to the ribs in the various stages (Figs. 1 to 12, I, Is, Im, li). Owang to the large size of the thymus, the cardiac notch is usually not well marked in the fetal lung.

D The Thoracic and Abdominal Viscera. 373

The pleural cavity is from the beginning 'more extensive than the lung, which at no time fills it completely. In the embryonic stages the pleural cavity always contains a relatively large amount of fluid, which surrounds the lung and distends the cavity. The pressure of this fluid may assist in extending the boundaries of the pleural cavity during the process of development.

In the 11 mm. embryo, the pleural cavities are not yet separated oS from the general body cavity. At 17 mm. (Fig. 2, pi) the pleura extends from the 2d to the 9th ribs, and from the plane of the anterior surface of the vertebral column posteriorly nearly to the extremities of the ribs anteriorly. It still communicates with the peritoneal cavity (Fig. 2, pp). In a 24 mm. embryo, and in all later stages, I find the pleura extending from the 1st to the 12th ribs. Anteriorly, however, the pleurse of the twq sides are widely separated, extending forward only to the neighborhood of the internal mammary vessels. They do not approach each other in the anterior mediastinal space until after the atrophy of the thymus in postnatal life. Posteriorly, the pleural cavity in the 31 mm. embryo reaches the plane of the anterior surface of the spinal cord (Fig. 3), and at 65 mm. almost covers the spinal cord in lateral view (Fig. 4).

There is considerable individual variation in the development of the lungs and pleurae, as well as of other organs. For purposes of comparison, the reconstructions by MalP^ wnll be found of especial value.

Alimentary Canal.

Stomach. The descent of the stomach along the vertebral column has been mentioned previously. In the 11 mm. cmbrj^o the cardia lies opposite the 3d or 4th thoracic segment, and the pylorus opposite the 7th or 8th. In the 17 mm. embryo the two ends of the stomach seem to have reached approximately their permanent positions, the cardia opposite the 10th thoracic and the pylorus opposite the 1st or 2d lumbar vertebra. The greater curvature may later extend

"Archiv f. Anatomie u. Entw., 1897, Suppl. Bd. S. 403-434; also in .Tolins Hopkins Hospital Bulletin, Vol. XII, 1901, Nos. 121, 122, 123.

D 374 C. M. Jackson.

much lower, however, especially in cases where the stomach is distended or the liver unusually enlarged.

Superiorly the fundus region is related to the base of the left lung in the 11 mm. embryo (Fig. 1). The liver extends between them in the 17 mm. and 31 mm. stages (Figs. 2, 3, 21, 22), but has retracted at 05 mm. (Fig. 23) leaving this relation constant in all later stages.

Externally the stomach is separated from the hody wall in the 11 mm. embryo (Figs. 1, 5, 17, x) by the thick-wallcd great omentum, in which, in the 17 mm. stage, the spleen appears (Figs. 2, 6, 21, sp). Both are overlapped by the enormous expansion of the liver at 31 mm. (Figs. 3, 22, sp). The stomach is not yet again visible at 65 mm. (Fig. 8), though it is often visible to the left of and below the liver in the later fetal stages.

Anteriorly and to the right, the stomach is at all stages in contact with the posterior surface of the liver (Figs. 17, 21, 22, 23). Posteriorly it is related in the 11 mm. embryo to the left suprarenal gland, sexual anlage, and Wolffian body (Fig. 17). At 17 ram., the pancreas extends across behind the stomach (Fig. 21). Beginning with the 31 nmi. stage, the stomach becomes separated from the sex gland and Wolffian body, but comes into more or less intimate relation with the kidneys and intestines (Figs. 22 and 23).

Intestines. The developmental topography of the intestines has been thoroughly worked out by Mall,^^ so that it is unnecessary to describe in detail the relations shown in my models, and indicated in the various figures.


The developmental topography of the pancreas has l)een described by me in a previous pa]x»r,^^ to which the reader is referred. Figs. 20, 21, 22, and 24 of the present paper exhibit well many of the relations of the pancreas.


From the beginning, the liver is in intimate relation with the diaphragm, and through this with the lieart and lungs. In connec "Arohlv f. Anatoinie ii. Entw., 1897, Suppl. Bd, "Anatomiseher Anzelger, Bd. 27. 1905, S. 488-510.

D The Thoracic and Abdominal Viscera. 375

tion with these organs, it undergoes the migration previously described. The upper surface of the liver apparently reaches its permanent level in the fetus at some time between the 31 mm. and 65 mm. stages.

In size, the liver is at first relatively small, but it rapidly enlarges. In the 11 mm. embryo it measured 4.85 per cent of the total body volume (or approximately the same as at birth) ; in the 17 mm. embryo it has increased to 6.9 per cent; while in the 31 mm. embryo it has reached 10.56 per cent, which represents its maximum relative size. In the 65 mm. specimen the volume of the liver was not measured, but it has apparently decreased to about 5 or 6 per cent, which is the average for the remainder of the fetal period. The importance of the relatively enormous expansion of the liver in producing the characteris^tic embryonal umbilical hernia has been emphasized by Mall.

The relation of the liver to the body wall and ribs anteriorly and laterally is evident in the various figures, and calls for no special discussion.

The posterior or visceral surface of the liver in the 11 mm. embryo (Figs. 5, 9, 17) is related to the suprarenal gland, Wolffian body and sexual anlage on the right side ; to the stomach on the left side ; and to the duodenum and head of the pancreas below. At 17 mm., the relations are similar (Figs. 18, 21). At 31 mm. (Figs. 19, 22), the kidneys and spleen come into contact with the liver. The visceral surface at this stage forms a relatively small depression on the posterior surface of the liver. At 65 mm. (Figs. 20, 23) the relations approach those found throughout the remainder of the fetal period. The lower part of the visceral surface is related to the transverse colon and coils of the small intestine. The relations of the liver to the pancreas are described in detail in my paper on the topography of the pancreas (1. c).


The spleen is not well differentiated in the 11 mm. embryo, but the indistinct anlage lies in the thick-walled great omentum in the region of the window shown in the model (Figs. 1, 5, 17, x).

D 376 C. M. Jackson.

In the 17 mm. stage, the spleen is relatively small but distinct (Figs. 2, 6, 18, 21, sp). It is an elongated prismatic structure, extending from above downward, slightly outward and forward. It has three distinct surfaces (best seen in Fig. 21) corresponding to those of the adult organ. These are: (1) an external surface^ corresponding to the (later) diaphragmatic surface, but here in contact with the lateral abdominal wall below the ribs; (2) an antero-internal gastric surface, in contact with the stomach; and (3) a posterior (later renal) surface, here in contact wath the left suprarenal, sex gland, and Wolffian duct.

In the 31 mm. stage, the liver has expanded so as to separate the external surface of the spleen from the body wall (Figs. 3, 7, 22). The spleen is here nearly vertical, and in contact antero-internally with the tail of the pancreas, as well as the stomach (Fig. 22, sp). Postero-internally, it is in contact with the left suprarenal.

In the 05 mm. stage, the spleen is relatively larger, but still relatively smaller than in the later fetal stages. It here lies entirely under cover of the ribs. The liver has partly retracted, so that it here covers only the anterior portion of tlie external splenic surface ; the posterior portion being in contact with the diaphragm between the 0th and 10th ribs (Figs. 4, 8, 20, 23, sp). The upper extremity of the diaphragmatic surface is now related to the base of the left lung. Antero-internally and postero-internally the relations are similar to those at 31 mm. The lower extremity of the spleen comes in contact with the splenic flexure of the colon (Figs. 20, 23), a relation which is constant throughout all later stages.

In the later fetal stages, the spleen becomes relatively larger, expanding farther upward, inward and downward. The liver usually retracts somewhat, but is often slightly in contact Avith the spleen, even at birth. As the suprarenal gland becomes relatively smaller, the posterior surface of the spleen comes into relation more and more with the left kidney. The fetal spleen is constantly related to the hase of the lung above, and is usually entirely pre-pleural (as in Fig. 4). The spleen is occasionally enlarged, however, in which case it may extend downward below the lower margin of the pleura.

In position, the longitudinal axis of the fetal spleen is always

D The Thoracic and Abdominal Viscera. 377

oblique, but it usually approaches the vertical more nearly than the horizontal. When the colon is distended, however, the lower extremity of the spleen is flattened and often pushed upward so as to throw the spleen into a position more nearly horizontal.

Wolffian Bodies (Mesonephros).

In the 11 mm. embryo, the volume of the Wolffian bodies is .000734 ec. (.6 per cent of the total body volume) ; at 17 mm., the volume is .00055 cc. (.124 per cent of total body) ; and at 31 mm., .00045 cc. (.0212 per cent of total body). It is therefore evident that the Wolffian bodies decrease in size, not only relatively but absolutely, from the beginning of the second month.

At the end of the first month, the Wolffian bodies extend from the lower cervical to the lumbar region (according to the observations of His and Mall). In the 11 mm. embryo (Figs. 1, 5, 9, 17, w), they extend along the posterior body wall from the 1st or 2d thoracic segment down to the lower lumbar region, lying just anteroexternal to the posterior cardinal veins. From the 3d to the 6th thoracic, they are also related internally to the suprarenal anlages. Anteriorly, they are related to the lungs, stomach, liver and sexual anlages.

In the 17 mm. embryo (Figs. 2, 6, 10, 18, w), the Wolffian bodies have apparently been drawn downward (being more firmly attached at the lower end), so that they extend from the 10th thoracic to the sacral region. They are separated relatively more widely by the expansion of the kidneys and suprarenals (cf. Figs. 17 and 18, w), and they no longer touch the lungs.

At 31 mm. the Wolffian bodies lie opposite the lower two lumbar and the upper sacral vertebrae. Internally they are related to the ovaries and lower part of the kidneys (Fig. 22a). At 65 mm,, the Wolffian bodies appear as rudimentary appendages of the testis, lying opposite the first sacral vertebra (Fig. 20).

Sex Glands. The early developmental topography of the sex glands is very similar to that of the Wolffian bodies, with which they are intimately

D 378 C. M. Jackson.

connected. They are likewise firmly connected with the body wall below, so that they are also dragged downward during the relative elongation of the lower portion of the trunk in development. In the 11 mm. embryo, the sexual anlages (sex not yet determinable) form narrow strips extending along the antero-intemal aspect of the Wolffian bodies, from the 9th thoracic to the 3d lumbar segment (Figs. 1, 5, sx). The relations are similar to those of the Wolffian l)odics in this region. At 17 mm, (Fig. 21a, ov), 31 mm. (Fig. 22a, ov), and C5 mm. (Fig. 20, t), the general position and relations are very similar to those already stated for the Wolffian bodies.

Suprarenal Glands.

In the 11 mm. embryo, the suprarenal glands (Figs. 1, 5, 17, sr, si) are ratlier ill-defined elongated bodies, extending on each side at tlie level of the 3d to the 6th thoracic segments, between the aorta and the posterior cardinal vein. Anteriorly, they are related to the lungs above, and to the liver (right side) and stomach (left side) hi low.

In the 17 mm. stage, the suprarenals have descended so as to extend from the 10th thoracic to the 1st lumbar vertebra (Figs. 2, r>, 10, 18, 21a, sr, si). They are shorter, thicker, and flattened from within outward. Internally they are closely related to each other, being separated posteriorly by the aorta. Posteriorly they are related to the 10th, 11th and 12th ribs below the pleural cavity. Superiorly they are related to the bases of the lungs (Fig. 18). Anteriorly the right suprarenal is related to the liver, the left to the stomach and spleen, Inftriorly they are in contact with the kidneys, and externally with the sex glands and Wolffian bodies.

At 31 mm., the suprarenals (Figs. 3, 7, 11, 19, 22a, sr, si) extt^nd from the 11th thoracic to the 1st lumbar vertebra. Owing to the change in position of the vertebral column (previously discussed), the suprarenals are separated more widely from each other by the lx)dies of the vertebne, and their former internal surfaces are rotated so as to face postero-internally. The aorta is displaced forward, so that it here lies between the anterior margins of the suprarenals. The relations to the body wall, etc., posteriorly are nearly as in the

D D D The Thoracic and Abdominal Viscera. 379

17 mm. embryo. They do not now extend up to the lungs or 10th ribs, however, although the pleural cavity has extended down behind the upper half of their posterior surfaces. Anteriorly the right suprarenal is in contact with the liver ; the left with the stomach and body of the pancreas, and externally with the spleen and left lobe of the liver. Inferiorly and posteriorly, the concave base of each suprarenal is in close contact wnth the upper portion of the corresponding kidney.

At 65 mm., the suprarcnals (Figs. 4, 8, 12, 20, 23a, sr, si) extend from the 10th thoracic to the 1st lumbar vertebra. They are still more widely separated by the vertebral column, and have rotated so as to lie almost in the frontal plane. They are decidedly flattened antero-posteriorly, and show a distinct groove on the anterior surface (Fig. 23a). The relations are much as at 31 mm., except that the kidneys have pushed up behind them to a greater extent, and the basrs of the lungs are again related to them above. The left lobe of the liver has retracted, so that it is no longer in contact with the left suprarenal. The relations of the suprarenal glands found here persist nearly unchanged throughout the remainder of the fetal period.


The kidneys form a notable exception to the general rule in developmental topography, since they migrate upward (cephalad) along the vertebral column, instead of downward like the other viscera. Their relations therefore deserve especial attention.

As is well known, the kidneys (kidney-ureter anlages) arise as a dorsal outgrowth of the Wolffian duct on each side in the sacral region. By active growth, the anlage elongates, extending forward on each side along the line of least resistance in the loose mesenchyme of the space bounded by the aorta dorsally, the rectum ventrally, and the umbilical arteries laterally.

In the 11 mm. embryo (Fig. 9, u) the T-shaped upper extremity of the anlage (representing the renal pelvis) lies in the upper sacral region, under cover of the wide origin of the umbilical artery. It is extending upward, and is approaching the lower extremity of the Wolffian body. In succeeding stages, as the anlage elongates, the

D 380 C. M. Jackson.

kidney is pushed upward on either side of the aorta in the loose mesenchyme dorsal and internal to the Wolffian body and sexual anlage.

In the 17 mm. embryo, the kidneys are well developed, with a distinct capsule, and extend from the 1st to the 5th lumbar vertebra (Figs. 2, 0, 10, 18, 21a, kr, kl). They have come in contact with the lower ends of the suprarenal glands, and further extension upward (except to a very slight extent) is possible only later when the suprarenals diminish in relative size. Antero-lateral to the kidney are the Wolffian body and ovary.

At 31 mm., the kidneys (Figs. 3, 7, 11, 19, 22a, kr, kl) still lie below the 12th rilw, extending from the 1st to the 5th lumbar vertebra. Their upper ends reach almost to tlie lower pleural margin (Fig. 3). The separation and rotation of the kidneys in connection with the changed position of the vertebral column is very similar to that descrilx?d for the suprarenal glands. Between the kidneys at the floor of the vertebral groove are the aorta and vena cava inferior (Fig. 19, 22a, a, vc). Anteriorly the right kidney is related to the corresponding suprarenal and the liver and ovary below; the left kidney is related to the suprarenal above, and to the liver, pancreas, stomach and ovary below.

At 65 mm. (Figs. 4, 8, 12, 20, 23a, kr, kl), the kidneys have reached approximately their permanent fetal position, extending from the 12th thoracic to the 3d lumbar vertebra. Both kidneys are nearly at the same level, the higher level of the left kidney later being correlated with the atrophy of the left lobe of the liver. Posteriorly the kidneys reach the 12th rib above (and, on the left side, the 11th rib). The pleural cavity covers the upper portion of the posterior surface of each kidney (Fig. 4). Anteriorly the right kidney is in contact with the suprarenal gland above, and with the liver, and small intestines (including the duodenum) below; the left kidney, with the suprarenal above, and the intestines (l)eginning of jejunum and descending colon) below. The relations of the kidneys found here are but little changed throughout the remainder of the fetal period.

D The Thoracic and Abdominal Viscera,



Fig. 1. — Graphic reconstruction of an 11 mm. human embryo (No. GO) from the left side, showing the body outline, extremities, central nervous system, vertebral centra, viscera, etc. The parts con-esponding to the viscera in the model (Fig. 5) are indicated by stippling. The various regions of the vertebral column are indicated (ceph.-cerv., cerv.-thor., thor.-lumb., lumb.sacr., sacr.-cocc.).

A, ascending aorta ; a, descending aorta ; a3, a4, Sd and 4th aortic arches ; ac, anterior cardinal (jugular) vein ; c, anlage of the cecum ; cd, caudal aorta ; cl, cloaca ; co, colon ; dC, ductus Cuvieri ; 1, lung ; L, liver ; la, left auricle ; Iv, left ventricle ; pc, posterior cardinal vein ; ph, pharynx ; R, rectum ; sx, sexual anlage ; si, left suprarenal anlage ; sa, origin of subclavian artery ; th, anlage of thymus; tl, tm, lateral and median anlages of the thyroid; U, umbilical cord; ua, umbilical artery; uv, umbilical vein; w, Wolffian body; wd, Wolffian duct ; x, window cut into great omentum ; ys, attachment of yolk-stalk to intestinal loop.



C. M. Jackson.


Fig. 2. — Graphic reconstruction of a 17 mm. human embryo (No. 58) from the left side, showing the body outline, extremities, central nervous system, vertebral centra, ribs, etc. The parts corresponding to the wax model (Fig. 6) are indicated by stippling. The various regions of the vertebral column are indicated (ceph.-cerv., cerv.-thor., thor.-lumb., lumb.-acar., sacr.-cocc. ) .

a, aorta ; ac, anterior cardinal (jugular) vein ; c, anlage of cecum and appendix ; co, beginning of colon ; da, ductus arteriosus ; ic, ileo-cecal junction ; kl, left kidney; Is, li, sui)erior and inferior lobes of left lung; L, liver; la, left auricle; Iv, left ventricle; o. oesophagus; pi, pleural boundary; pp, pleuro-peritoneal foramen ; R, rwtum, s, stomach ; si, left suprarenal ; si. small intestine; sp, si)leen ; tr, trachea; U, umbilical cord; w, Wolffian body; 1 to 12, 1st to 12th ribs.

D The Thoracic and Abdominal Viscera.



Fig. 3. — Graphic reconstruction of a 31 mm. human embrj-o (No. 57) from the left side, showing the body outline, extremities, central nervous system, vertebral centra, ribs, sternum, viscera, etc. The parts corresponding to the wax model (Fig. 7) are indicated by stippling. The various regions of the vertebral column are indicated (ceph.-cerv., cerv. thor., thor.-lumb., lumb.-sacr., sacr.-cocc. ) .

kl, left kidney; Is, 11, superior and inferior lobes of left lung; L, liver; la, left auricle ; Iv, left ventricle ; o, cesoiihagus ; pi, pleural boundary ; R, rectum ; si, left suprarenal ; si, small intestine ; st, sternum ; tr, trachea ; U, umbilical cord; ua, umbilical artery; uv, umbilical vein; 1 to 12, 1st to 12th ribs.

D 384

C. M. Jacksoiu



— f iumb.-«acr

Fig. 4. — Graphic reconstruction of a (35 mm. human embryo (No. 55) from the left side, showing the body outline, extremities (cut off short), central nervous system, vertebral centra, ribs, sternum, viscera, etc. The imrts corresponding to the wax model (Fig. 8) are Indicated by stippling. The various regions of the vertebral columnn are indicated (ceph.-cerv., cerv.-thor., thor.-lumb., lumb. sacr., sacr.-cocc.).

CO, colon ; kl, left kidney ; Is, li, superior and inferior lobes of left lung : L, liver ; Iv, left ventricle ; o, cesophagus ; p, penis ; pi. pleural boundary ; R, rectum; si, small intestine: si, left suprarenal ; sp, spleen; st. sternum; T, lower extremity, cut off just below the hip joint; t, testis; th^ thymus; tr, trachea; U, umbilical cord; 1-12, 1st to 12th ribs.

D D 886 C. M. Jackson.

Figs. 5 to 8. — From photographs of wax models reconstructed by Bom's method from human embryos, showing the thoracic and abdominal viscera, with projections of the ribs, viewed from the left side.

( For Figs. 5 to 8. ) A, ascending aorta ; a, descending aorta ; a3, a4, 3d and 4th aortic arches; ac, anterior cardinal (jugular) vein; bw, body wall; c, anlage of cecum and appendix; ca, left common carotid artery; co, bt»ginning of colon ; da, ductus arteriosus ; dC, ductus Cuvierl ; hi, hind limb ; ia, innominate artery: ic, ileo-cecal junction; kl, left kidney; L, liver; 1, left lung; Is, 11, superior and inferior lobes of left lung; la, left auricle; Iv, left ventricle ; o, oesophagus ; pc, posterior cardinal vein ; ph, pharynx ; R, rectum ; r2-rl2, projections of 2d to 12th ribs ; ra, right auricle ; rv, right ventricle; s, stomach; sa, left subclavian artery; si, small intestine: si, left suprarenal : sp, spleen ; t, testis ; th, thymus, tl, tm, lateral and median thyroid anlages ; tr, trachea : ua, uv, umbilical artery and vein ; w. Wolffian body; x, window in great omentum; ys, attachment of yolkstalk to intestinal loop.

Fio. 5. — From an embryo of 11 mm. (No. 60), showing also the lower extremity, and the lower portion of the body wall.

Fig. 6. — From an embryo of 17 mm. (No. 58).

Fig. 7. — From an embryo of 31 mm. (No. 57).

Fig. 8. — From an embryo of G5 mm. (No. 55).

D Fig. 5.


/ i la


sa r PC

dC re



r la U ca


, Is



r8 I.

- si -- rl2

kl -- w

th .,

L -.





rl2 w



Pig. 6.

o r3



— n

Iv-- r7

rO -sp

rlO rs — Bl




Fig. 7. Fig. 8.

D 388 C. M. Jackson.

Figs. 9 to 12. — From photographs of wax models reconstructed by Born's method from human embryos, showing the thoracic and abdominal viscera with projections of the ribs, viewed from the right side.

(For Figs. 9 to 12.) A, ascending aorta (or arch) ; a, descending aorta; ac, anterior cardinal (Jugular) vein; al, allantois; a2, a3, a4, 2d, 3d, and 4th aortic arches ; c, anlage of cecum and appendix ; cl, cloaca ; co, beginning of colon ; du, duodenum ; gb, gall bladder ; kr, right kidney ; 1, lung ; Is, Im, 11, superior, middle and inferior lobes of the right lung; L, liver; n, notochord ; o, oesophagus ; po, posterior cardinal vein ; ph. pharynx ; r2-rl2, projections of 2d to 12th ribs; ra, right auricle: rv, right ventricle; sa, sulK*lavlan artery ; sc, spinal cord ; si, small intestine ; sp, spleen ; sr, right suprarenal gland ; th, thymus ; tl, tm, lateral and medium thyroid anlages ; tr, trachea ; u, ureter ; ua, umbilicirl artery ; uv, umbilical vein ; vv, vitelline vein; w. Wolffian body; ys, attachment of yolk-stalk to intestinal loop.

Fig. 9. — From an embryo of 11 mm. (No. (X)). In the lower part of the model, the body wall is represented as having been dissected away nearly to the mid-sagittal plane, so as to show a portion of the spinal cord, notochord, aorta, pelvic viscera, etc.

Fig. 10. — From an embryo of 17 mm. (No. 58).

Fig. 11. — From an embryo of 31 mm. (No. 57).

Fig. 12. — From an embryo of G5 mm. (No. 55).

D a3 th ph


tl tm



tr A


FlO. 9.

tr ac

FlO. 10. r2 o tr ac th

r4 Im

r«  11






Iv L ua

Fig. 11.

Fio. 12.

D 390 C. M. Jackson.

Figs. 13 to 10. — From photographs of wax models reconstructed by Bom's method from human embryos, showing the thoracic and abdominal viscera, with projections of the sternum and ribs, anterior view.

(B'or Figs. 13 to 1(5.) A, ascending aorta; ac, anterior cardinal (jugular) vein; al, allantois; a2, 2d aortic arch; bl, bladder; c, anlage of cecum and appendix ; ca, left common carotid artery ; co, colon ; da, ductus arterioRUR ; gb, gall bladder ; hi. lower extremity ; ia, innominate artery ; ic, ileo-cecal junction ; L. liver ; is, Im. li, superior, middle and inferior lobes of the lung ; la, left auricle ; Iv, left ventricle ; Ix, larynx ; o, oesophagus : ph, pharynx ; r2-r8, projections of 2d to 8th ribs, with corresponding costal cartilages joining the sternum; ra, right auricle; rv, right ventricle; sa, left subclavian artery; sc, spinal cord; si, small intestine; th, thymus anlage: tl, tm, lateral and median thyroid anlages; tr, trachea; ua, umbilical artery; uv, umbilical vein; va, vitelline artery (with accompanying velnj.

Fio. 13. — From an embryo of 11 mm. (No. 00). Lower extremity and a I>ortion of the lower body wall showni on the left side of the model.

Fig. 14. — From an embryo of 17 mm. (No. 58).

Fig. 15. — From an embryo of 31 mm. (No. 57).

Fig. 10. — From an embryo of 05 mm. (No. 55).

D a2










A <ta ae








C uv

Fig. 13.

Fig. 14.

ri ri In: n



1 1 1





Is, r2

I- ^


— -"»v

Im. rv




rs L r7





Fig. 15.

Fig. 1G.

D 392 C. M. Jackson.

Figs. 17 to 20. — From photographs of wax models reconstructed by Born's method from human embryos, showing the thoracic and abdominal viscera, with projections of the ribs and intervertebral disks, posterior view.

(For Figs. 17 to 2().) A, arch of aorta; a» descending aorta; ac, anterior cardinal (jugular) vein; ba. bifurcation of aorta; bw, body wall; co, descending colon; dcO-7. level of disk between 0th and 7th cervical vertebne; dt3-4, disk between 3d and 4th thoracic vertebra*; dtl, disk between 12th thoracic and Ist lumbar; dl3-4. disk between 3d and 4th lumbar; du, duodenum ; hi, lower extremity ; hp, head of pancreas ; kr, kl. right and left kidneys; L, liver; 1, lung. Is, Im, II, superior, middle and Inferior lobes of lung; la, left auricle; o, (esophagus; pb, body of pancreas; pc, posterior cardinal vein ; pco, pelvic colon ; ph, pharynx ; r2-rl2, projections of 2d to 12th ribs ; ra, right auricle ; sa, subclavian artery ; sc, spinal cord ; si, small Intestine; si, sr, left and right suprarenals; sp, spleen; t, testis; th, thymus anlage ; tr, trachea ; u, ureter ; vc, vena cava inferior ; w. Wolffian body ; X, window in great omentum.

Fio. 17. — From an embryo of 11 mm. (No. 00). rx)wer extremity and a portion of the lower body wall shown on the left side of the model.

Fm. 18. — From an embryo of 17 mm. (No. 58).

Fio. 19. — From an embryo of 31 mm. (No. 57).

Fig. 20. — From au embryo of 05 mm. (No. 55).

D D D tr ac pq

,ra /'

la ^ ' r5

o la—


rlO ..w sr kr




— ac


— dc6-7

— ra





-- w


--sr — L

r7_ 1L>






Fig. 17. * Fig. 18.

ao sa o

, dt3-4



,.-.11 r7




a, — no --:: — kr













rl2 Pb







■ dii

kl . kr si-- ___dI3-4

od- vo t

ba- w


Fig. 19. Fio. 20.

D 394 C. M. Jackson.

Figs. 21 to 23. — Posterior view of the models as shown in Figs. 18 to 20, excepting that the kidneys, suprarenal glands, Wolffian bodies and sex glands have been removed.

(For Figs. 21 to 23, in addition to the explanations given for Figs. 18 to 20.) cr, cardia; dj, duodeno-jejunal flexure; dls, disk between last lumbar and first sacral ver^ebrje ; gc, groove for inferior vena cava ; gk, groove for kidney ; go, groove for ovary ; gs, groove for suprarenal gland ; Ic, caudate lobe of liver ; s, stomach ; tp, tail of pancreas.

Fio. 21. — From an embryo of 17 mm. (No. 58).

Fig. 22. — From an embryo of 31 mm. (No. 57).

Fig. 23. — From an embrj'o of 05 mm. (No. 55).


la Is II Iv





sp tp

dt2-3 -PC

Is -ra


Im -r8



gs Ro



Fig. 21.

tr ac r2





8 d,



- U — r6

-li -r8


- gs

- gc rlO

- Ic hp

- du

- "si


- t


Fig. 22.

Fig, 23.

D 396 C. M. Jackson.



Fio. 21a.

Fio. 21a. — Anterior view of detached model showing the kidneys, suprarenal glands and sex glands of the 17 mm. embryo. A posterior view of this model Is shown In Fig. 18, but It has been detached in Fig. 21. a, aorta, showing origins of the coellac, superior and inferior mesenteric branches; ba, bifurcation of the aorta, with beginning of the caudal (middle sacral) artery ; ov, ovary ; od. oviduct ; sr, si, right and left suprarenal glands, below which a small portion of the kidneys is visible.

Fig. 22ii. Fig. 23a.

Fig. 22a. — Anterior view of detached model showing the suprarenal glands, kidneys, etc., of the 31 mm. embryo. A posterior view of this model is shown in Fig. 10, but It has been detached in Fig. 22. a, aorta ; ba, bifurcation of aorta; kr, kl, kidneys; ov, ovary; od, oviduct; sr, si, suprarenals; u, ureter; vc, vena cava inferior; w. Wolffian bodies.

Fig. 23a. — Anterior view of detached model showing the suprarenal glands and kidneys of the (>5 mm. embryo. A posterior view of this model Is shown in Fig. 20, but it has been detached in Fig. 23. a, aorta; kr, kl, kidneys; sr, si, suprarenals ; u, ureter.




From the Histological Laboratory, Cornell University Medical College,

New York City.

The physiologic necessity of leaving the parathyroid glands uninjured in operations upon the thyroid region is now well known. It has also been demonstrated that the administration of beeves' parathyroids will to some extent relieve the postmortem tetany in man which follows the removal of the parathyroid glands (Ilalsted).

The number of parathyroid glands typically present has been quite generally stated as four, two posterior superior and two anterior inferior. Yet frequently at autopsy the typical four parathyroid glands are not to be found even after most careful dissection. The number which can actually Ik? recovered varies from none to as many as six (Erdheim). This extreme variation has been ascribed to such causes as irregular numl>er of glands in the individual, irregularities in their position, the presence of a varying number of accessory parathyroid glands, and even to the possibility of postmortem destruction of the small parathyroid glands by autolysis.

The reports of surgeons would lead one to believe that during ojK^rations upon the thyroid, the parathyroids are recognized w^ith relative ease and can be preserved together with their blood supply. Parathyroid serums and extracts have been prepared, presumably without microscopical confirmation of the tissue used in their manufacture. If a relatively small percentage of the tissue so used is actually parathyroid, the serum or extract w^ill certainly contain a considerable admixture of the products of such other tissues as thyroid gland, lymphatic nodes and thymus ; and experimental results based upon the specificity of such extracts must be interpreted in the light of this possible source of error.


D 398 E. T. Rulison, Jr.

In view of these considerations further anatomical studies of the parathyroid glands for the purpose of providing a more firm and accurate foundation upon which to build and interpret physiologic experiment and surgical procedure appear most desirable.

Recognition of Paratiiyboid Glands.

The following notes, based upon autopsies of bodies in the best state of preservation, one excepted, indicate that only 41 per cent of probable parathyroid glands, as identified by gross inspection, actually contain parathyroid tissue wjien examined microscopically. This percentage does not widely differ from the results obtained by Rogers and Ferguson, who found 61 parathyroids out of 189 pieces examined, or in 32.4 per cent (all suspicious looking pieces being included). In certain cases the gross appearance is undoubtedly misleading, tissue not resembling the classical type proving to be parathyroid aftt»r microscopic examination. Less frequently tissue presenting the typical macroscopic features of parathyroid is found to be lymphoid or other tissue.

The results of tlie gross dissection of tlie several autopsies are given somewhat in detail because of their relation to the question of the* percentage of cases in which one may hope that the glands may be positively recognized by the unaided eye, without microscopical study.

Autopsy I.^ W. M., aged 35 ; five hours post-mortem. Mental diagnosis, sc^nile delirium. Cause of death, acute enteritis. In this case throe pieces of tissue, none bearing the characteristic stamp of the parathyroid, were dissc^cted, two from the posterior-superior aspect of the lateral lobes of the thyroid and one from the anterior surface of the esophagus on the right side. All proved to be adenoid tissue.

The failure to obtain the parathyroids in this case may have been due (1) to rapid autolysis, (2) neglect to secure all tissues bearing the slightest resemblance to a parathyroid, (3) the possibility of the

The material used was obtained In 1007 through the kindness of the Hudson River State Hospital, Ponghkeei)sie, N. Y. My thanks are due this institution for the opportunity offered.

D Parathyroid Glands in Man. 399

glands being situated beyond the limited field of dissection, or (4) the remote possibility of an unobserved location within the thyroid gland.

Autopsy II. S. K., aged 55 ; twelve hours post-mortem. Mental diagnosis, epileptic insanity. Cause of death, appendicitis with perforation. In this case the tissue had undergone extensive postmortem change. Only one very small piece of tissue resembling parathyroid was found, being located at the superior pole of the left thyroid lobe, posteriorly. Microscopical examination proved this to be thyroid tissue.

Autopsy III. D. W., aged 22 ; seven hours post-mortem. Mental diagnosis, epilepsy with insanity. Cause of death, septicaemia following carbuncle.

The two left parathyroids were readily recognized by (1) the yellowish brown color, (2) smooth, finely textured surface, (3) lack of the firmness which characterizes the thyroid and other tissue so often confused. Both left parathyroids were found on the posterior surface, the lower one, however, being anterior to the main branch of the thyroid artery. Six other suspicious looking piecx'S of glandular tissue were removed from the loose areolar tissue bordering on the right lobe and along the eso])hagus. One of these proved to be the right anterior parathyroid. The fourth parathyroid, if present, must have been outside the field of dissection.

Autopsy IV. L. W., aged 40 ; seven hours post-mortem. Mental diagnosis, general paresis. Cause of death, general paresis.

The two anterior pararthyroids were recognized by the characteristics above mentioned. Both were located between the branches of the inferior thyroid artery and upon the lateral lobes of the thyroid gland. Two other pieces of tissue were examined, one from the superior pole of the left lobe of the thyroid and another from the anterior surface of the esophagus, on the right side. The former proved to be thyroid and the latter adenoid tissue.

Autopsy V. M. B., aged 36 ; two hours post-mortem. Mental diagnosis, imbecility with insanity. Cause of death, pulmonary tuberculosis.

The two anterior inferior parathyroids were immediately recog

D 400 E. T. Rulison, Jr.

nized. The posterior glands were not found in the usual location, and having in mind the possibility of an internal position, large wedge-shaped areas of thyroid were excised from the upper posterior surface of tlie lobes. These were sectioned throughout, but no parathyroids were found. A search was made along the wall of the pharynx and esophagus, but no tissue resembling parathyroid was found.

Autopsy VI. A. P., aged 70; eight hours post-mortem. Mental diagnosis, melancholia simplex. Cause of death, carcinoma with multiple metastases.

The two posterior superior parathyroids were found, the right one being in direct relation to the esophagus. Just anterior to the main artery of the right thyroid lobe, a gland was removed, bearing all the characteristics of the parathyroid. On microscopical examination this proved to be adenoid tissue. This was the only instance in the ten autopsies in which a piece of tissue having the typical color, consistency and location proved not to be parathyroid. To secure the left inferior parathyroid a wedge-shaped area was excised from the thyroid and a small gland removed from the areolar tissue just external to the left thyroid lol)e. Xo parathyroid tissue was found in the thyroid and the small gland proved to be a bit of accessory thyroid tissue.

Autopsy VII. A. J. B., aged 55, eight hours post-mortem. Mental diagnosis, imbecility witli insanity. Cause of death, lobar pneumonia.

The fi(>ld of dissection was limited and but three bhx'ks of tissue were obtained. At the extreme inferior pole of the left thyroid lobe, beneath the branches of the inferior thyroid artery, a parathyroid was found, the gland extending over upon the anterior surface of the thyroid. Two other pieces of tissue were removed, one from the esophagus at the level of the superior border of the cricoid cartilage and the other from the surface of the larjTix on the left at the same level. Neither had all the essential attributes of parathyroid and both proved to be thyroid tissue.

Autopsy VIII. P. F., aged 40 ; twelve hours post-mortem. Mental diagnosis, epileptic insanity. Cause of death, asphyxia following convulsion.

D Parathyroid Glands in Man. 401

The tissue was badly congested and no parathyroids were found in the customary locations. Four wedge-shaped pieces of tissue were excised from the thyroid at the most likely points, but no parathyroids were found.

Autopsy IX. E. K., aged 32 ; twelve hours post-mortem. Mental diagnosis, alcoholic psychosis. Cause of death, chronic interstitial nephritis.

All four parathyroids were readily recognized and secured. The inferior parathyroids were both at the extreme lower poles of the thyroid. The superior parathyroids were situated rather laterally than posteriorly. But four blocks of tissue were obtained in this ease, the glands being characteristic. No search for accessory parathyroids was made.

Autopsy X. W. P. II., aged 48 ; twenty hours post-mortem. Mental diagnosis, maniacal depression. Cause of death, cardiac valvular disease with nephritis.

Three parathyroids were obtained in this case, all located upon the posterior surfaces of the thyroid lobes, two on the left, the other on the right. Of the two left parathyroids, one was found immediately l»eneath the superior pole, lying in the groove between the thyroid and esophagus, the other, three centimetres below in the same vertical line. The third was found in the same relative position on the posterior surface of the right lol)e. Xo other tissue was searched for parathyroid, as it was badly congested.

With one or two exceptions all of the glands were fixed in Van Gehuchten's fluid. They were then hardened in graded alsohol, embedded in paraffin, sectioned and stained with hematein and eosin.

The following conclusions are based upon the dissection and microscopical study of the 17 parathyroids obtained :

1. Size. The size varied from 2x4x6 mm. to 4 x 6 x 10 mm. Th<?

average size was about 3x5x8 mm.

2. Color. Typical yellowish brown 13 gland.-}

Deep red brown (congestion) 3 glands

Light yellow 1 gland

D 402 E. T. Rulison, Jr.

3. Consistency. Typical, flaccid 12 glands

Tense, hard 3 glands

Medium 2 glands

4. Number and location.

Number of autopsies in which 4 glands were obtained 1

Number of autopsies in which 3 glands were obtained 2

Number of autopsies in which 2 glands were obtained 3

Number of auptopsies in which 1 gland was obtained 1

Number of autopsies in which no glands were obtained .... 3

Total number of glands obtained 17

Numl)er of superior glands obtained

Number of inferior glands obtained 11

Numlx^r of glands irregularly situated 7

Su|)erior : found on esophagus 1

Inferior : found on esophagus 1

Inferior : found at extreme inferior pole 3

Inferior : found on posterior surface 2

D. A. Welsh (Jour. Anat. and Physiol., 1898, XXXII, page 383) says "the posterior-superior parathyroid of each side is much more constant in its position, and much more easily found than the anterior inferior glandule." Contrary to the experience of Welsh, the inferior parathyroids in the above series seem the more easily located, in all excerpt two cases being found anterior to the inferior thyroid artery, and in all except four cases being located at the first branch of this artery on the thyroid lobe.

5. Recognition of th^ glands. In many cases the parathyroids can be definitely recognized by gross inspection if the tissue is fresh and not badly congested or pigmented or otherwise altered. In only one instance in ten did a gland with a smooth surface, of yellowishbrown color and flaccid consistency, fail microscopical confirmation, while seven atypical pieces of tissue proved to be parathyroids. In none of the ten autopsies were more than two parathyroids found in

D Parathyroid Glands in Man. 403

their typical location, but in autopsy IX, in which four parathyroids were found, the superior were typically located while the two inferior were atypical only in that they were displaced toward the inferior pole of the thyroid and were inferior to the thyroid branch of the inferior thyroid artery. Although glands of typical parathyroid appearance rarely failed of confirmation, the converse was more frequently true, that in certain instances glands of variable color and consistency proved to be parathyroid.

Tissue resembling parathyroids on dissection would seem to be divisible into three groups: (1) in which the location, form, size and color appear wholly characteristic to the unaided eye — such tissue on sectioning rarely proves to be other than parathyroid; (2) in which location, form, size and color approximate the typical — a considerable portion of these prove not to be parathyroid on microscopical section; (3) in which the location, form, size or color are wholly atypical — such occasionally prove to be parathyroids.

The obvious surgical importance of the above seems to be two-fold, viz., the surgeon during thyroidectomy can be reasonably certain of his identification of certain of the parathyroid bodies when they can be found, but to insure the preservation of these glands they must leave intact much, if not all, of the glandular tissue found in the immediate vicinity of the thyroid gland, bearing in mind, also, that the number of parathyroids constantly present in each individual has not been demonstrated to be as many as the typical number, four.

To those studying the parathyroids it soon becomes evident that if four glands are to be obtained in each case very extensive search is necessary. Only very rarely can four glands be immediately and definitely recognized. In cases in which, for example, but three glands are identified one is obliged to preserve all possible tissue from the hyoid bone to the aortic arch before excluding the presence of a fourth gland. If this search fails to locate the missing gland a complete sectioning of the entire thyroid may produce it, owing to its occasional location within the capsule of the thyroid gland.

Anatomy and Histology. In man the parathyroids may be as many as four in number,

D 404 E. T. Rulison, Jr.

occasionally five or 'six (Erdheim). They vary in size from that of a grain of rice, or even smaller, to that of a small white bean. Their color is quite characteristic, a peculiar yellowish brown, in contrast to the deep red of the thyroid. Their surface is smooth and conveys the idea of a tissue of finer elements than that of the thyroid or other tissue, for example, lymphatic nodes, thymus rests, with which parathyroids may lye confustnl. In many instances the parathyroid is situated in direct relation to the thyroid gland, less often in relation to the esophagus or trachea. The glandules have been found supc^riorly as far as the hyoid bone and inferiorly as far as the bifurcation of the trachea or upon the arch of the aorta in relation to the thymus (Welsh). The last named position is stated by Welsh as being the most common in the cow. The parathyroid glands most readily located an* the ones known as the anterior inferior parathyroids, usually situated at the inferior poles of the lateral thyroid lol)es, although there is much individual variation in their position (Kogers and Ferguson). Welsh however states that in his series of cascis the superior were the more easily recognized, whereas in the present series of auto])sies the inferior parathyroids were much the more easily recognized and more constantly present. The inferior thyroid artery forms a guide to the two inferior glands which often rest in the bifurcation of the artery upon the anterior inferior surface of the thyroid on either side. The two remaining parathyroid glands are most frequcaitly found posteriorly at the superior pole of the thyroid. Very often one or both of the superior parathyroids is found upon the esophagus. In rare cases a human parathyroid may be embedded within the thyroid gland, and this location is quite the usual one for the inferior parathyroid of the dog, horse and goat (Rogers and Ferguson).

The parathyroid is described as an epithelial structure resembling the glandular portion of the hypophysis cerebri. The epithelial cells are of two varieties: (1) so-called principal cells, which have a distinct limiting membrane or wall and are outlined by the supporting stroma of reticulum, thus assuming a polygonal, often pentagonal shape ; they have a clear cytoplasm, and a centrally situated and often vesicular nucleus; the principal cells predominate: (2) oxyphile or

D Parathyroid Glands in Man. 405

acidophile cells, somewhat larger than the principal cells, and have distinct outlines, a finely granular and acidophile cytoplasm, and a centrally situated nucleus which is small in proportion to the size of the cell and often highly chromatic. The acidophile cells are irregularly placed among the masses of principal cells, either singly or in small gi'oups. These groups are often situated immediately beneath the capsule in wedge-shaped areas, or in relation to bloodvessels.

The stroma includes a capsule, ahvays present, consisting of delicate white fibrous connective tissue. From the inner surface of the capsule fine trabeculae are given oif, w4iich in some cases appear to divide the gland into lobules. In many glands there is a distinct hilum by means of which the arteries of supply — branches of the inferior thyroid — enter the glandular substance, supported by the trabecular.

There is no duct through which a secretion may leave the gland. The tubular cysts found in the parathyroids of the goat, horse, and occasionally in other mammals (Kohn, Edmunds, Rogers and Ferguson) can not be considered as in any sense ducts, for they do not receive the secretion from the parathyroid parenchyma nor do they open upon any free surface. They are closed cysts and more nearly than anything else they simulate the branchial cysts which are frequently found in the connective tissue about the trachea, and which are of a congenital type, probably due to the embryonal dislocation of primordial epithelial cells from the branchial clefts. The embryology of the parathyroids would suggest a similar origin for the ciliated epithelium lining the occasional colloid-containing cysts within their substance.

No peculiar histology of the larger parathyroid blood-vessels presents itself, the coats being properly proportioned. The arterioles empty into an extensive sinusoidal system of capillary vessels which are then collected by the radicals of the venous system. The arrangement of the epithelial cells seems dependent upon the blood-vessels, the intimate contact of parenchymal cells and vascular epithelium suggesting the escape of secretion into the sinusoidal vessels. It is believed that the arrangement of cells varies somewhat with the age

D 406 E. T. Rulison, Jr.

of the individual. In young subjects the columnar arrangement of the epithelial oells seems to predominate (Ferguson), Later in life the cells show either a diffuse arrangement or assume something of an alveolar grouping. In rare instances within the parenchyma there are alveoli containing a colloid material^ but in these cases the cells lining the alveoli are low columnar or cuboidal cells riesemUing the true glandular cella of the thyroid and bear no obvious resenablance to the parenchymal cells of the parathyroid. These alveoli would seem to form no essential part of the parathyroid structure.

Most parathyroid glands contain fat, either in discrete cells or in groups. The presence of fat, however, very probably does not account for the somewhat characteristic yellow tint of the yellowish brown parathyroid glands, w^iich is apparently an intrinsic property of the parenchymal cells, for many of the glands of my series which possessed a most characteristic yellowish tint proved on sectioning to contain very few or no fat cells. The remaining brownish tint appears to be due to the extreme vascularity of the gland.

The following is a summary of the histological findings in the seventeen parathyroid glands removed from the above described autopsies.

Capsule. — The capsule was invariably present, varying greatly in thickness and structure. In most cases the capsule was found to consist of white fibrous tissue apparently containing some smooth muscle. From the inner surface, septa composed of a delicate reticulum of fibrous tissue were found to penetrate between the parenchymal cells. These septa support the blood-vessels.

Ililum, — A more or less distinct hilum- was found in each gland, the arteries entering and veins leaving at this point.

Reticular Tissue. — After tryptic digc^stion for twenty-four hours of sections of the parathyroid gland — removing the white fibrous connective tissue, elastic tissue, muscle and epithelial cells — a delicate reticulum remains. Hence, it would seem that this is true reticular tissue ; it appears to form the ultimate framework of the parathyroid. The distribution of the reticular tissue corresponds to the outlines of the capsule, septa and parenchymal cells, principal and oxyphile.

Blood-vessels. — ^^The parathyroids were in all cases foimd to be

D Parathyroid Glands in Man. 407

highly vascular. The histological study of the arteries and veins revealed nothing indicating any possible structural relation of the vascular system to the glandular secretion, such as is found -in the excessive development of the longitudinal muscle of the media of the adrenal veins (Ferguson). The capillary network pervading the gland is very delicate and seems to determine the grouping of the cells, the resulting arrangement of the cells presenting either a columnar or alveolar form. The columnar grouping was not observed, this arrangement being more frequent on younger subjects. In five instances an acinar grouping was indicated. In one of these the acini seemed very distinct, and a few of them contained colloid. In many glands there were broad vascular spaces lined by a single layer of endothelium — so-called sinusoids.

Principal Cells. — The limiting membrane of the cell was. usually distinct. Each cellmost commonly had a pentagonal outline; its cytoplasm was clear, often taking a faint bluish tinge in sections stained with hematein and eosin. The nucleus was centrally situated and often vesicular in character.

Acidophile Cells. — In size these were somewhat larger than the principal cells. The limiting membrane was fairly distinct and the cells had a polygonal outline. The cytoplasm was finely granular and strongly acidophile. The nucleus was centrally situated and often deeply chromatic; occasionally multiple. In twelve cases the acidophile cells were found dispersed uniformly among the principal cells without evident relation to capsule, septa or blood-vessels. In the remaining five cases some of the acidophile cells were found in distinct, often wedge-shaped areas beneath the capsule or in relation to septa or blood-vessels.

Fat. — In five glands no fat cells were found either in the capsule or among the epithelial cells. In four glands the fat cells were found in discrete arrangement only. In two glands fat ocurred in large masses only, and in the remaining six the fat cells were found in both discrete and grouped aran^ement.

Color. — The distinctive yellowish brown color of the parathyroid has been attributed to the presence of fat (Welsh). A review of the five glands in which I found no fat shows that the color in each

D 408 E. T. Enlison, Jr.

instance was a characteristic yellowish brown, and it must, therefore, be an inherent property of the parenchymal cells ; possibly the color may be due to reflected light from the masses of principal cells, the clear cytoplasm of which, together with the extreme vasularity of the gland, may combine to produce the characteristic yellowish brown.


Nothing of importance from a pathological viewpoint was noted in any of the glands. In one case, in which there was extreme congestion of all the organs, some pigmentary deposit was found in the small areas of hemorrhage. No evidences of degeneration were found unkss the presence of colloid containing follicles in two glands may be considered. No new growths were found, not even a metastasis in the case of general carcinoma.

In closing, I desire to express my thanks to Dr. J. S. Fergiison, Assistant Professor of Histology, at whose suggestion and under whose direction the present study was undertaken.

Received for publication May 11, 1909.

BIBLIOGRAPHY. Erdheim. Beitr. z. path. Anat. u. z. aUg. Path., 1904, xxxv, 3G6. Febouson. Text-book of Histology, 1905, 45G. Halsted. Am. J. Med. So., 1907, cxxxlv, 1. Howell. Text-book of Physiology, 1906. KoHN. Arch. f. mik. Anat, 1897. xlvUl. 398. Rogers and Ferguson. Am. J. Me<l. Sc, IJMXJ, cxxxi, 811. Welsh. D. A. J. Anat. and Physiol., 1898. xxxil, 292, and 380.



STEPHEN R. WILLIAMS. Professor of Zoology, Miami University.

The specimens of Necturus rrmculosus (Raf.) to be described in this paper are adult animals, which have been injected through the ventral aorta with red starch mass. The injections may be called good, since the main arterial vessels and their branches are well distended. None of the specimens show any evidence that the pulmonary circulation has been disturbed through injury or pathological condition.

In the first case the arteries are arranged as shown in Fig. 1. The right side of the animal is normal. Here the pulmonary artery arises from the combined portion of the second and third efferent aortic arches. On its way to the lung it sends small branches to the muscles of the pectoral region.^ These are comparable with the large cutaneous branch of the pulmonary artery in the frog.^ After sending out the pectoral branches, the pulmonary artery of Necturus, as described by Miller, "passes across the dorsal surface of the lung to gain its dorso-mesial side, along which it runs, gradually diminishing in size, to its tip."^ This normal condition is found on the right side.

The left lung of the Necturus shown in Fig. 1 receives its blood at its extreme distal end. The supply is from a large seventh intercostal trunk, which appears at the level of the twelfth vertebra, posterior of the celiac artery. This intercostal vessel divides at once on

These branches have been described by W. S. Miller in "Tlie Vascular System of Neeturus 7naculatus." Bull. Univ. Wisconsin, No. 33, pp. 213-214.

•Ecker, A. The Anatomy of the Frog. (Translated by G. Ilaslam.) 1880, p 229, Fig. 149.

•Miller, William S. The Blood and Lymph Vessels of the Lung of Necturus maculatus. Amer. Journ. Anat, 1905, Vol. IV, p. 440.


D 410

Stephen R. Williams.

Ext. carotid

.t r

Efferent brnnchinls

PertornI branches of the pulmonary

Int. carotid

Afferent branch la Is

Doranl aorta

•- SubclaTian



7th Intercostal .




Pulmonary branch of 7th Intercostal




Fig. 1. — Ventral view of the arteries of an adult Nooturus. Natural size. The right lung receives its blood from the 2d and 3d efferent branchial arteries, as is normal ; tlie left lung is supplied by a branch of the seventh intercostal artery.

D The Pulmonary Artery in Necturus. 411

leaving the aorta and is so turned that both divisions are on the right side. The left branch, w^hich is much larger, turns towards its own side dorsal to the aorta. It is entirely free from the musculature until after it has branched. The larger portion then enters the left lung near its tip and runs forward along its ventro-mesial side. The remainder of the vessel, reduced to less than one-third of its previous size, penetrates the musculature of the left side of the body about 5 mm. laterally from the tip of the lung.

This pulmonary vessel appears as the reverse of the normal artery, since it runs anteriorly from the tip toward the base of the lung. It gives off lateral branches alternately, and grows smaller and smaller until the starch mass can penetrate no farther. Beyond this point a fine line, due to the presence of clotted blood, indicates the continuance of the vessel for a short distance. The diminution and stopping of the injected vessel while on the lung, and the absence of any starch -mass starting from the fully injected aortic arch, indicate that there is no vessel at all comparable with the normal proximal end of the pulmonary artery. It is quite possible, however, that it is represented by capillary connections.

This extraordinary anomaly may be readily explained through embryological development. It is generally believed that after the vascular system has become established in very, young embryos, all subsequent vessels arise from it as offshoots. Where there is a connective tissue pathway, a branch from a neighboring vessel is likely to enter it and to anastomose with such other vessels as it encounters. If a favorable channel results, the small vessel becomes a main channel.* A striking illustration of this mode of development has recently been supplied by Spalteholz.^ It has been known for some time that in certain reptiles a bridge of connective tissue extends from the tip of the heart across the pericardial cavity to the thoracic wall. Spalteholz has shown that a branch of the adjacent

Thls conception of vascular development was advocated by F. T. Lewis at the meeting of American Anatomists in 1903, and by H. E. Evans at their last session, in 1908.

•Spalteholz, W. Zur vergleichenden Anatomie der Aa. coronariae cordis. Verb. der. anat. Gesellschaft, 1908, pp. 169-180.IC


Stephen R. Williams.

internal mammary artery may enter this bridge and extend to the heart, thus becoming a coronary artery. The heart may receive its blood supply from its apex as well as from its base provided that the connective tissue pathway is present. In the lung of Xecturus the mes(»ntery is the connective tissue path, although a slender and apparently unfavorable one. In the specimen under discussion either a branch of the pulmonary artery passed out from the lung to anastomose with the intercostal artery, or a branch of the latter

Pulmonary artery


Recurrent branch

of the Pulmonary


Pulmonary artery


Pulmonary branch of the aorta

Fio. 2. Fig. 3.

Figs. 2 and 3. — Variations in the piilnionnry arteries in the left lung of Neoturus. 2/3 natural size.

entered the lung and joined the pulmonary branches. That both of these processes ocx?ur is suggested by the following variations.

Fig. 2 represents a left lung, 100 mm. in length, which is free from the mesentery along its distal 20 mm. At a point 9 mm. before the lung becomes free, at the level of the middle of the eleventh vertebra, a fairly large branch passes from the pulmonary artery toward the median line. It is enclosed in the mesentery. This branch tapers gradually without forking, and terminates without making demonstrable connections with other vessels. It clearly conveyed a stream of blood away from the lung. In the case 8ho\vn

D The Pulmonary Artery in Jfecturus. 413

in Fig. 3, however, the dorsal aorta sends a branch to the left lung which supplies its distal two-thirds.® Here the current of blood is toward the distal end of the lung instead of away from it as in Fig. 2.

All three of these variations may be explained on the basis of embryonic capillary connections between the dorsal aorta and the pulmonary artery. Although such connections have not yet been observed in Necturus embrj^os, they may be confidently predicted. They have already been found by Dr. Evans in chicks incubated from sixty to seventy hours. It will be remembered that in the human adult, branches of the bronchial artery, which comes from the aorta, have been said by several investigators to anastomose with the pulmonary artery in the lung. Miller and others are unable to find such connections."' But Nicolas^ has stated that "it appears well established that the pulmonary arteries (of man) anastomose with the bronchial arteries not only through a capillary network, but also by branches which are quite large, and which may attain a diameter of 0.5 mm. or more." Detached portions or "accessory lobes" of the human lung are sometimes su|)plied entirely by a branch of the thoracic aorta, and are drained by veins emptying into the azygos system. Simpson® described such a case in a child at birth, and found records of two others, occurring at three months and eighteen years respectively. In one case the artery to the accessory lobe left the aorta at the level of the seventh thoracic vertebra ; in the other two it was at the level of the tenth vertebra. This is c<.mparable with the arrangement in the abnormal Xecturus.

Abnormal vessels entering the lung of the frog, near its apex, have l)een found by several observers. The veins have receivon more attention than the arteries. T. W. Shore has figured two cases, in

•The dissection and sketch of this specimen were made by Dr. F. T. Lewis when studying with me in the zoological laboratory in Cambridge. I am indebted to him for the memorandum concerning it.

'Miller, William S. The Arrangement of the Bronchial Blood vessels. (Preliminary Communication.) Anat Anz., 1906, Vol. 28, pp. 432-436.

•Nicolas, A. See Poirier's Traits d'Anatomie Humaine, Paris, 1805, Vol. IV. p. 528.

•Simpson, G. C. E. A case of accessory lobe of the right lung. Jouru. Anat. and Phys., 1908, Vol. 42, pi). 221-225.

D 414 Stephen R. Williams.

one of which a branch of the renal portal vein, and in the other a branch of the heptatic portal vein entered the lung.*® E. Warren** had previously described and figured three similar cases, in one of which a branch of the renal portal vein joined a branch of the hepatic portal, and the resulting trunk entered the left lung near its apex. In this case it is recorded that a branch of the posterior mesenteric artery accompanied the vein into the lung, and the figure indicates that it anastomosed with a normal pulmonary artery. W^arren states that "this arrangement of blood-vessels is strikingly similar to that seen in a teleostean fish, where an artery runs from the mesenteric artery into the rete mirabile of the air-bladder, and from there the blood is carried by a vein into the portal system." But he hesitated very properly before interpreting the anomaly in the frog as a reversion to a fish-like ancestor. G. P. Mudge*^ recorded still another case occurring in the frog. A branch of the celiac artery entered the right lung and two branches of the superior mesenteric artery entered the left lung. On both sides the arteries were accompanied by branches of the portal vein. Mudge states that in Ophidia Hyrtl found secondary pulmonary arteries arising from the aorta and the arteries of the liver, oesophagus, and stomach; the corresponding veins entered the portal system. Mudge concludes with the following statement: "Whether the close correspondence between the remarkable condition herein described as abnormal for the frog and that apparently normal for certain ophidians be indicative of anything more than mere coincidence, further investigation can alone determine."

The results of further investigation indicate that the occurrence of pulmonary branches of the aorta in the various classes of vertebrates has an embryological rather than an evolutionary significance. The connective tissue pathway being provided, capillary branches from the adjacent vessels will enter it, and occasionally, as in these Necturus specimens, certain of them will become very large, and notable as anomalies.

"Jourii. Anat. ami Pliys.. 1001. Vol. 35, pp. 323-329.

"Anat. Anz., 1000, Vol. 18, pp. 122-123.

"Jouru. Auat. and Pbys., 1898. Vol. 33, pp. 54-63.




APRIL, 1909.


In the Record, Vol. II, No. 9, Dec, 1908, p. 425, a brief statement was made of the purposes of the committee of one hundred appointed by the Council on Medical Education of the American Medical Association to consider the medical curriculum in schools demanding at least one year of college work in physics, chemistry, biology and language for matriculation. The sub-conmiittee on anatomy consisted of Professors G. A. Piersol, F. P. Mall, I. Hardesty, G. S. Huntington, J. P. McMurrich, A. C. Eycleshymer, T. G. Lee, C. M. Jackson, G. Carl Huber and C. R. Bardeen, chairman. The questions outlined in the Record were discussed by correspondence with the various members of the conunittee and were considered at an informal conference held at the time of the December meeting of the Association of American Anatomists. In response to the notice in the Record replies were received from several anatomists, including Professors A. W. Meyer, B. L. Myers and Robert Bean. Prof. Irving Haynes furnished an extensive account of the work in gi-oss anatomy at Cornell. From the data thus gathered a report on the teaching of anatomy was compiled, submitted to the various members of the sub-committee on anatomy, revised, and presented in abstract to a joint committee composed of the chairmen of the various divisions of the Committee of One Hundred and of the members of the Council on Medical Education. It was, in part, finally read at the annual Conference on Medical Education held by the Council on Medical Education at Chicago, April 5.

Two meetings of the joint committee were held, one at New York, in December, and one at Chicago preceding the conference.


D 416 C. R Bardeeii.

Practically the whole session at each meeting was spent in trying to adjutit, to a four-year curriculum of reasonable proportions, the claims of the various sub-committees for time for their various subjects. The Council started out with the assumption that in general the medical curricula of our medical schools are overcrowded and are not in all cases well proportioned. The various sub-committees were therefore requested to consider the minimum number of hours in a 3600-hour curriculum which should be devoted to the subjects they were appointed to consider. At the first meeting of the joint committee it was found that the total number of hours requested by the various subcommittees instead of amounting to 3600 hours, amounted to over 4500 hours. It was not found possible to get a general agreement concerning the reduction of the lunnber of hours devoted to each subject so as to formulate a 3600hour curriculum and it was decided to proceed on the basis of a 4000-hour curriculum. A provisional number of hours, on this basis, was allotted to each subject. The provisional schedule was referred back to the various sub-committees and at the second meeting of the joint committee a schedule amounting to 4345 hours was discussed. This was finally reduced to 4100 hours. The 4100 hours were regarded as the maximum number that should be definitely prescribed for a four-year course. Instead, therefore, of defining a minimum curriculum the joint committee finally determined a maximum curriculum. It was the unanimous opinion of all present that it would be a grave mistake to try to prescribe identical curricula for all medical schools; that freedom in schedule making in different schools is an imperative necessity if progress is to be made and that the schedule adoj)ted by the joint committee is meant merely to be suggestive. At best it is a compromise arrived at between the chairmen of the ten subcommittees and the members of the Council during two brief meetings.

It may be of interest to compare the schedule finally adopted by the joint committee with the provisional scheduler of this committee, with the present schedule of the Association of American Medical C^^olleges and with the revised schedule proposed at the last annual meeting of the Association.


Number of Hours Allotted to the Chief Subdivisions of the

Medical Curriculum, in


1 13 Hi


1 I I


III I |?i

. S2 fi I 2 c

lel 131

Anatomy, including Histology and Kmbrvoloev




1 760






1 300







1 270


• Bacteriology and Pathology —







Pharmacology and Materia Medica





1 160







I 80


Medicine, including Pediatrics and Nervous Diseases





Surgery (including G — U)















' t




Eye, Ear, Nose and Throat





Dermatology and Syphilis






Public Health and Medical Jurisprudence




110 4000






♦In the schedule of the Association of American Medical Colleges, chemistry includes inorganic as well as organic and physiological chemistry. In that of the American Medical Association Inorganic chemistry Is not included since this is assumed to be required for matriculation.

t Included here with obstetrics.

D 418 C. R Bardeen.

In the majority of the better medical schools of the country over 800 hours, about a fifth of the curriculum, are devoted to the anatomical sciences. There is undoubtedly at present a desire on the part of those teaching other branches to curtail anatomy so that more time may be given to those branches. The curtailment of anatomy to 630 hours in the 4000-hour curriculum of the Association of American Medical Colleges has, however, not proved satisfactory, and in the new curriculum proposed at the last annual meeting the number of hours for anatomy has been raised from 630 to 750. The 20 per cent, leeway allowed in each subject in the curriculum of the Association of American Medical Colleges makes this 4000-hour schedule elastic when it is taken in the right spirit and not rigidly applied. With 20 per cent, leeway the minimum allowed for anatomy in the new schedule would be 600 hours and the maximum of specified work might be 900 hours. These figures seem to me reasonable.

I believe, however, in a curriculum in which specific requirements are made in as few subjects, and in these subjects in as restricted a manner, as is consistent with a fair breadth of training. It is obviously impossible for the student in four years to become really proficient in each of the great branches included in the medical curriculum. He should gain some conception of the principles underlying each of these branches and should gain real proficiency in some one or two of them. Only thus can he acquire depth of understanding and the habit of thoroughness, only thus can he be made efficiently self-reliant. The twice repeated three months' lecture course of a generation ago had the advantage that it left time to the better students to develop themselves freely and independently during the other nine months. Good men were thus produced in spite of limited facilities. Give the student good facilities for work, freedom and encouragement, and the problem at once becomes not how to get him to work more, but how to prevent him from overworking.

Another advantage of having the required work reduced to a minimum is that it forces the teacher to discriminate carefully between the essential and the non-essential, and to emphasize prin

D Eeport of the Sub-Committee on Anatomy.

eiples rather than details in the required courses. On the < hand the teacher is forced to keep up a keen, active interest in his science if he is to attract students to his department to do advanced work. He must be an investigator in order to keep up that spirit of scientific enthusiasm which alone makes a laboratory or a clinic attractive.

Believing thus in freedom for the de^'elopment of the student and of the teacher I took at the first meeting of the joint committee a stand for a schedule of minimum requirements. I suggested a 3000-hour defined curriculum in which anatomy should have 600 hours. In such a curriculum the student who desired could elect other courses in anatomy than those required. The 600 hours in anatomy was acceptable to the chairmen of most of the other subcommittees, but a curriculum limited to 3000 hours was not. The result was a 4000-hour schedule in which anatomy was limited to 600 hours. In such a schedule the student who desired to elect more anatomy would have little opportunity to do so. At the Chicago meeting the amount of time finally allotted to anatomy was 700 hours in a 4100-hour curriculum. I believe 700 hours a fair requirement for anatomy in a 3600-hour schedule permitting of elective studies and work outside of scheduled hours. I believe that in a 4100-hour curriculum where obviously little freedom is allowed for extra work, 700 hours is proportionately too small an amoimt of time and that 800 hours would be a fairer allotment for anatomy.

The proportionate amount of time which should be allowed anatomy depends upon the relative value of anatomy in medical education. This in turn depends upon the importance of anatomy in developing the capacity to solve the problems presented by disease.

Of the various subjects in the medical curriculum, gross anatomy is the most concrete, the most definite in its relations to practical medicine and the easiest to supply with abundant material. The student in a good course in practical anatomy forms a direct familiar acquaintance with the intricate structure of the human body and becomes skilled in the use of instruments. Familiarity with human structure is essential for physical diagnosis, and hence is funda

D 420 C. R. Bardeen.

mental for scientific treatment. The skill gained in dissecting is an aid in every branch of medicine. It is certain that if a student does not dissect enough to acquire skill and does not study anatomy enough to become familiar with the body, infantile and youthful as well as adult, he will be seriously hampered in medical practice.

Microscopic anatomy is important for real understanding of gross structure. It is essential for physiology, pathology and clinical diagnosis. It therefore is, like gross anatomy, of fundamental value in medicine. Its educational value is increased by the fact that for it, as for gross anatomy, an abundant supply of excellent material may be readily furnished the student.

Embryology is scarcely less important. It offers the medical student his best opportunity to get some understanding of the phenomena of growth, the most basal thing in life. It gives abundant opportunity for experimental work.

Neurolog;^' is so important that it has acquired the dignity of a separate branch. Since man is so essentially a creature of nerves and brain it is obvious that some real understanding of the structure of the nervous system is essential for the scientific physician.

These fundamental anatomical sciences, gross and microscopic anatomy, embryology and neurology can be readily well taught. They cannot be quickly assimilated. The student must have plenty of time to dissect, to draw, to think over his work, to compare one part with another and a dissected part with cross sections, if he is to acquire familiarity with human structure and to^ learn to think anatomically. He must have plenty of time to prepare and study microscopic specimens and to compare microscopic with macroscopic structure if he is to get much real benefit from microscopic anatomy. He must have time to watch ova develop into embryos as well as to follow microscopically successive stages of development in prepared specimens if he is to get some understanding of growth. He must devote long hours to patient study if he is to get any real insight into the structure of the central nervous system. If the student has become thoroughly grounded in these branches he will have a definite and firm foundation upon which to rear an understanding of human medicine. If he has not this foundation he

D Report of the Snl)-Committce on Anatomy. 421

is likely to lack security and stability in his subsequent studies and work.

The statement is sometimes made that knowledge of function is more important to the physician than knowledge of structure, and that therefore anatomy should be reduced in amount in order that more time may be devoted to physiology. One can study human anatomy in the laboratory. Opportunities for the study of human physiology in the laboratory are limited. The tedside during the study of clinical medicine often offers more. Even mammalian physiology must at present be studied by students in a greatly restricted way in most schools, since public opinion has not yet been educated to the value of vivisection in medical training. In this respect jphysiology stands to-day almost where human anatomy did a century ago in the times of body snatching. Could laboratory courses in mammalian physiology be offered with perfect freedom as to use of animals, it is possible that many anatomical facts now learned in the dissecting room could Iw picked up incidentally as a part of physiological experiments. As it is, a considerable part of physiology consists of deductions from anatomical facts, and much of the rest consists of knowledge gained by experiments not practical for repetition by the student. The field of practical physiology is at present very limited so far as laboratory work for medical students is concerned. .

On the other hand in the courses in the anatomical laboratory the physiological aspects of the subject may with great advantage be emphasised. In embryology the physiology of growth may be taken up. In the study of the skeleton, muscles and joints considerable attention may be given to the simpler mechanics of motion. In the study of the gross and microscopic structure of the viscera and the nervous system many ])hysiological data may be learned. Thus a good groundwork* for the more technical aspects of physiology is gained. The committing of text-book detail to memory should be given up in favor of these broader aspects of the subject.

I believe, however, that what we as anatomists should stand for is not the retention of a large number of required hours for the anatomical sciences in the medical curriculum, but for a curriculum

D 422 C. R Bardeen.

in which in each of the fundamental branches there is small amount of required work and abundant opportunity for the student to do much more than the required work. The value to the physician of thorough work in the anatomical sciences will, certainly manifest itself where such a curriculum prevails and will insure abundant enthusiastic work in anatomy. We must not, on the other hand, permit anatomy to be unduely curtailed by too extensive time requirements in other departments.

In the following report of the sub-committee on anatomy I omit the introduction, which covers much of the ground just reviewed. I also omit some details not likely to be of interest to professional anatomists. The report in full, together with a summary of the opinions of the individual anatomists who contribute to it, is to be published by the Council on Medical Education.

Report of Sub-Committee on Anatomy to thk Council on

Medical Education of the American

Medical Association.

(A) Essential pre-requisites for the anatomical sciences of the medical course. Required preliminary work of college grade in physics, chemistry, biology and modern language is of especial advantage to anatomy. Anatomy comes at the beginning of a student's course in a medical school, and unless the student has had adequate preliminary training he must first learn how to work before he can begin to make headway in the science. The well-prepared student can do more intelligent work a few weeks after he enters the course than most ill-prepared students can toward the end of the course. The latter are apt to become Jost in a maze of detail from which they can derive no meaning. Two years of preliminary college work is far preferable to one year.

Of the preliminary courses the one most directly important for anatomy is a laboratory course of college grade in biology. A wellequipped laboratory is essential for an adequate course. There should be facilities for keeping living plants and animals of various types and the equipment should include an outfit for teaching the microscopic as well as the gross anatomy of plants and animals.

D Eeport of the Sub-Conunittee on Anatomy. 423

While the course should include some training in botany, the main stress should be laid upon zoology. Courses in comparative anatomy and in comparative embryology are advisable, but not necessary. The physiological as well as the morphological aspects of plants and animals should be emphasized. If the student studies the structure of the lower animals with especial reference to their activities he will subsequently find it the easier to take up the study of human structure from the standpoint of function, the standpoint of greatest importance to the physician.

(B) The place of the anatonUcal sciences in the medical curriculum. The first half of the first year of the medical curriculum should be devoted chiefly to gross anatomy and to histology. These subjects should, so far as possible, be co-ordinated. In the second half year histology may be followed by neurology and embryology. Gross anatomy may be continued throughout the second half of the first year or may be taken up again in first half of the second year. Topographical and applied anatomy may be taught in the second and subsequent years of the course. The majority of the committee believe that the required osteology, dissection, histology, neurology and embryology should be completed during the first year, leaving the second year for topographical and applied anatomy and elective work.^

(C) Required and elective subjects. Courses in gross human anatomy and in histology should be required.^ Courses in embryology, neurology and topographical anatomy may be elective in an elastic curriculum, but should constitute a specific part of a 3600to 4000-hour defined curriculum. If courses in embryology, topographical anatomy and neurology are not required, the elements of these subjects should be included in the courses in histology and gross anatomy. Various advanced and special courses may be

^Professors McMurrlch and Piersol believe some dissection stiould be required In the second year. Professor Hardesty would place neurology in the second year.

•In addition, Professor Piersol would require courses in embryology, neurology and applied anatomy. Professor Eycleshymer would require a course In embryology. Professor Hardesty would require one in neurology.

D 424 C. R Bardeen.

offered as electives. It is certain that no department of anatomy can meet the minimum requirements for medical students if it is not prepared to give much more than minimum opportunities for meeting these requirements.

When the medical school is fortunate enough to be a real and integral part of a university the anatomical department, or institute, may well provide not only courses for medical students, but also premedical courses in comparative anatomy and advanced courses for zoologists, and the work of the university in vertebrate anatomy may be concentrated in this department. Such a department should be centrally connected with the medical school, but should have close affiliations with other colleges and departments. Concentration of this kind adds both to economy and efficiency.

The work offered may comprise courses in

(a) Comparative vertebrate anatomy.

(b) Human and comparative osteology, including special provision for courses for dental students and for students of anthropology and paleontology.

(c) Dissection and systemic human anatomy.

(d) Topographical anatomy.

(e) Anatomy as applied in medicine, surgery and the specialties.

(f) ilicroscopic anatomy and histology, human and comparative.

(g) Embryolog;)', human, comparative and experimental, (h) Neurology, human, comparative and experimental.

(i) Anatomical technique: (1) Gross, including injections, corrosions, etc.; (2) General microtechnique; (3) Special technique (blood, etc.); (4) Illustrative work, including drawings, reconstruction methods, etc.

(j) Investigation of anatomical problems, including methods of looking up the literature on a subject.

(k) Seminars, discussion of advances in anatomy.

(1) History of the development of the anatomical sciences.

(m) Artistic anatomy.

While an "ideal" anatomical institute with a director and provision for all the branches mentioned is desirable, it is not essen

D D D Report of the Sub-Committee on Anatomy. 425

tial for the teaching of anatomy in a medical school. It is, however, essential that the courses in gross anatomy should be closely co-ordinated on the one hand with those in microscopic anatomy and embryology, and on the other hand with topographical and applied anatomy. It is also essential that in addition to the work required of every student there should be elective courses and opportunity for advanced work.

{D) Qualificaiiofis of instructors. The supervision of work in a department of anatomy in a medical school should be in charge of persons who have had a thorough professional training in the various branches of anatomy and who have demonstrated ability in teaching and research. They should have acquaintance with anatomy as applied to medicine and surgery. They should devote their entire time to teaching and investigation, and should be provided with ample time and facilities for doing both. The traditional bad name for dulness which anatomy has born among medical students in this country has been largely due to the fact that it has too often been taught by men who have known it only as a dead science. Unless the teacher is playing at least a small part in the growth of the science he is teaching he is not likely to have an intimate acquaintance with its more vital aspects. The teacher should therefore have opportunity to devote himself to investigation. On the other hand it is important that the teaching in the main fundamental courses and especially the laboratory work, including dissection, be directly in charge of the leading members of the staff and not intrusted to inexperienced assistants. In addition to the professor in charge there should, in general, be enough instructors and assistants to provide one for each twelve to fifteen students in a laboratory course. A strong leader may get along with a smaller number of assistants, but an adequate teaching force is essential for thorough efficient work.

While the fundamental work should be directly in charge of the leading members of the staff, many of the elective courses can be put into the hands of less experienced instructors. It is desirable that each instructor give an independent course of this nature, since it serves to show his capacity and aids in his development as a

D 420 C. R Bardeen.

teacher. If successful, a young teacher may inspire much wholesome enthusiasm in his students.

Courses in applied anatomy may be taught by competent clinicians, but should be taught by them in the anatomical department.

(E ) Methods of instruction. The chief essential is the able instructor, professionally trained in anatomy. Various methods of instruction will yield good results when employed by capable teachers. Courses should be arranged «o that the student may concentrate the greater part of his energies on the anatomical sciences during the period when he is mastering the essential principles of human anatomy. This so-called "concentration method" was urged half a century ago by Von Baer and has since been recommended by several of the greatest teachers of anatomy, including Waldeyer (see Mall, "On the Teaching of Anatomy," Anatomical Kecord, 1908).

Abundance of good material is necessary. The equipment may be simple, but should be adequate. Lectures and quizzes may be utilized according to the pedagogical ideals of the instructor, but should be ancillary to the laboratory work. The student, however, should not be allowed to content himself with mere mechanical laboratory work and with committing details to memory. He should get some understanding of general principles, some insight into methods of classification, some idea of the development of anatomy as a science and some knowledge of the relations of the science to medicine. He should gain concrete ideas of structure so as to become able to "think anatomically." He should be made to feel that he is studying an intricate and delicate mechanism which subsequently he will be called upon to set right when it gets out of gear. To do this he must gain a thorough understanding of biological mechanisms in general and of the human mechanism in particular.

Careful, thoughtful dissection of the human body is the chief essential of the work in gross anatomy. Atlases and text-books should be kept close at hand by each student, and he should frequently turn to these for information and guidance. In most of the better American anatomical laboratories the method of coordinate systemic dissection is adopted. The skin is first removed

D Report of the Siib-Conunittee on Anatomy. 427

and then the superficial fascia. In the latter the superficial nerves and blood vessels are carefully studied. In the dissection of the deeper parts the vascular and peripheral nervous systems are worked out in conjunction with the organs to which they are distributed. While fat and areolar tissue are removed freely the attempt is made to keep the more definitive structures in as near their natural relations as possible. The text-book, models, special preparations and charts, recitations and the final study of the dissected part are relied upon to give the student a clear conception of each of the great organ systems. A brief study of osteology may precede the dissection, but during dissection the student is constantly referred to the skeleton as a topographical basis and thus he becomes better and better acquainted with it. At the end of the dissection the soft parts are removed and the articulated skeleton is studied. Reference to cross sections during the course of the dissection is a great aid in emphasizing structural relations.

In the course in microscopic anatomy it is usual to study a fairly extensive series of microscopic sections from the chief organs of the human body. In addition to this, in the better laboratories, fine dissection of fresh and hardened tissues and organs, under low as well as high magnification, forms an important part of the work. The aim is made to co-ordinate carefully the structure revealed by the microscope with the gross anatomical structure of tissues, organs and systems. The dissection of a small mammal or an embryo may be utilized to co-ordinate the various organs and systems with the body as a whole. The elements of microscopic technique should be taught, but the student's time should not be wasted by too great a devotion to routine mechanical procedures.

In embryology the work may begin with a study of the general processes of vertebrate development as illustrated by the eggs of the frog and the chick. For medical students, however, a rela? tively large part of the course should be devoted to organ differentiation and histogenesis in mammals and man. Some experimental embryology should be included in the course. Regeneration in some of the lower forms may also be studied.

D 428 C. R. Bardeen.

In neurology careful dissection of the formalin hardened human brain and cord should be associated with the study of a good series of microscopic sections of the central nervous system. Study of the organs of special sense may form a part of the course in neurology. The study of microscopic sections of the eye, ear and nose should be correlated with dissection of the organs of some mammal.

In topographical anatomy cross sections and special preparations of formalin hardened bodies should occupy the chief attention of the students. Regional anatomy studied on the living model adds much to the interest of such a course.

Drawing is a valuable aid in stimulating the attention and sharpening the powers of observation of students of anatomy. It is commonly required in courses in microscopic anatomy. Simple semidiagrammatic drawings, if accurate, are equally valuable requirements for courses in gross anatomy. Elaborate drawings, on the other hand, are aj)t to involve too much thoughtless expenditure of time in mere mechanical procedures. Clay modeling is found by some teachers of much value, especially in the study of osteology. It is apt, however, to involve the risk mentioned in connection with elaborate drawings.

(F) Necessary laboratory equipment Under equipment for the department of anatomy are included: (a) quarters, (b) furniture, apparatus and supplies, (c) models and charts, (d) special preparations and (e) library.

(a) Quarters. Ample space should be provided for practical work in gross human and microscopic anatomy. There should be a lecture room, rooms for the members of the staff, for advanced work, for museum and library, and for storage. Good toilet and dressing rooms should be provided. We may specify in more detail certain of these requirements.

1. Gross human anatomy. Enough laboratory space should be provided to enable the student to do careful dissecting. Arrangements should be made to prevent disturbance of a dissection during periods when the dissector is not in the room. The rooms should be finished, furnished and cared for in a manner to inspire a sense of refinement in the student. A series of small dissecting rooms,

D Report of the Sub-Committee on Anatomy. 429

accommodating from one to three or four dissecting tables, is now preferred by many instructors to a single large dissecting room. Quiet, orderly work is thus much fostered. The dissecting room should be provided with hot and cold running water and be furnished with locker space in which students can put their apparatus, books and dissecting room clothing at the end of the day's work. It is desirable, if possible, that these lockers be placed adjacent to, but not in, the dissecting room itself. The dissecting rooms should be well lighted both by sunlight and by artificial light.

There should be rooms for the preparation and preservation of cadavers, and for maceration and cremation of dissected parts.

A valuable adjunct to the dissecting rooms is a study room in which special preparations, models, etc., are kept. Such a study room should be of ample size for accommodating a class in topographical anatomy.

2. For microscopic anatomy and embryology well lighted laboratory space should be provided. It is desirable, but not absolutely essential, that the major part of the light come from the north. The laboratory should be sufficiently large to provide well lighted space for all students in it at any given time. ^ The work in microscopic anatomy and embryology lends itself more x-eadily than gross human anatomy to teaching classes in sections which in turn use the same laboratory.

A study room in which a reference collection of microscopic specimens, special preparations and models are kept, is a valuable adjunct to the general laboratories for microscopic anatomy and embryology.

In connection with these laboratories there should be a room in which the material for the class is prepared. Such a room should be well lighted, be provided with gas and hot and cold water, and with a good equipment for microscopic technique.

3. If desired, lectures in anatomy may be given in the laboratories, so that a special lecture room is not absolutely essential. It is, however, a convenient adjunct. The seats should be so arranged that demonstrations may be readily observed from any part of iho room.

D 430 C. R Bardeen.

4. The members of the staff should be provided with private quarters suitable not only for office work, but also for carrying on scientific investigation. Other rooms in which advanced work in anatomy and research can be carried on are a valuable adjunct. The preparation rooms can, however, be utilized for advanced students.

A photo-micrograi)hic dark room is of great value. This should be provided with a photo-micrographic outfit and a camera and len» for copying and enlarging, making lantern slides and photographing gross specimens.

5. In addition to the working collections provided for in the study rooms mentioned above, a large museum containing preparations illustrating human and comparative anatomy is an important, if not absolutely essential, adjunct to an anatomical department. It should contain merely well selwted specimens and not be '^a reservoir for dumping a miscellaneous lot of stuff."

6. The department should l)e supplied with ample library facili ties. Unless the general library of the medical school is readily accessible from the quarters occupied by the anatomical department, a special room should be set aside for a departmental library.

(b) Furniture, apparatus and supplies. There should be an .am])le sui)ply of furniture, including book-cases, museum-cases and cabinets, to accommodate the students and the members of the staff.

The ap])aratus required is in part general and in part special It should be sufficient not only for class room work, but also for making good anatomical preparations, for advanced work and research.

For gross human anatomy there should be enough disarticulated skeletons to provide the students freely with bones. There should be an ami)le supply of well-embalmed bodies, so that each student may be furnished half a body for dissection. Six students are as many as should dissect upon a single body at one time. There should be a sufficient su])i)ly of cross sections to offer every student an oi)portunity to use them for the study of regional anatomy.

The dissecting rooms should be provided with dissecting tables of convenient height and width and with reading stands for holding

D Report of the Siib-Coramittee on Anatomy. 431

text-books. The students should be furnished with chairs or stools, so that they do not have to stand all the time they dissect. They should be given wrapping cloth and preserving fluids with which to keep the parts dissected in good condition. The study room should be supplied with metallic boxes or other rec<?ptacles for moist and wet specimens, including cross sections, special preparations and dissected parts.

In the preparation room there should he an injecting apparatus of such a nature that the fluid may be forced into the body under an even pressure. This is afforded either by a water blast or by a gravity tank. The room should contain one or more embalming tables and a tackle and clasp for handling cadavers. The room for preserving cadavers may be provided with a cold storage apparatus, but this is not necessary unless a large number of bodies are handled. If the bodies are well embalmed they may be preserved in vats or metal-lined boxes or in water-proof sacks.

The apparatus for maceration of parts may be simple, but should be well ventilated. A furnace in which forced draft can be obtained either by gas or by other means is necessary for the proper cremation of parts.

For microscopic anatomy and embryology there should be a good supply of prepared and mounted sections which can be loaned students or demonstrated to them. In addition there should be a supply of material which may be given the students. There should be a laboratory assistant to prepare vsections for the classes, including all steps in technique except mounting of specimens on the slides. The members of the class are thus provided with specimens of uniform excellence. It is probably a mistake to try to give students to keep specimens requiring difficulty in preparation. Specimens of this nature should be merely loaned for study. There should be an adequate supply of compound and dissecting microscopes. The student should provide himself with a set of simple laboratory instruments.

For embryology, in addition to the apparatus required in histology, special sets of serial sections of embryos in various stages of development and a collection of mammalian and human embryos

D 432 C. K. Bardeen.

and fetuses, in part dissected, are essential. Frogs' eggs and pig embryos preserved in formalin make excellent material for the study of the grosser features of early development. For the study of the development of the hen's egg an incubator of some sort is required.

For neurology a special section cutter for the brain is of great value.

(c) Models and charts. For gross human anatomy models of papier mache, of plaster and of wax are of considerable value. Such moflels are essential for the study of structures difficult to appreciate in gross dissection because of their minuteness, such as the finer structure of the larynx, brain, eye, ear, tongue, central nervous system and the nerves of the head. Models illustrating the comparative anatomy of various structures are also of value.

In histology models are useful in illustrating the minute structure of intricate regions, such as the organ of Corti and various parts of the central nervous system. They are also of great value in giving an idea of the third dimension of various other microscopic structures, but unfortunately accurate models of this kind are not yet readily obtained.

In embryology the scarcity of material makes it impossible for the student to get first hand knowledge of the structure of young human embryos. The Ziegler models therefore form an indispensable adjunct to a course in human embryology. Models illustrating the development of the frog, chick and other lower vertebrates and of various organs are likewise of great value.

Charts and lantern slides are an important adjunct to lectures in embryology, histology, neurology and gross anatomy.

(d) Special pre imrat ions. To illustrate gross human anatomy there should be numerous special preparations. These should include skeletal preparations, cross sections of the body, and special dissections of various parts of the body.

The skeletal preparations should include well articulated male and female skeletoUvS, adult, youthful, infantile and fetal; special preparations of the skull and bones of the head, and fixed and pliable preparations of the joints. Pathological specimens illustrating frac

D Report of the Sub-Coinmittee on Anatomy. 433

tures and abnormalities of the bones and joints are of considerable value.

The cross sections should pass through the body in various planes, and should illustrate infantile and youthful as well as adult conditions.

The special dissections should include regional preparations of the head, neck and extremities, and dissections of the eye, ear, nose, mouth, larynx, pharynx, cranial nerves, biliary system, the genitourinary system and the central nervous system. There should be a series of preparations illustrating the viscera in infancy and youth. Specimens illustrating the distribution of lymphatics are of importance owing to the difficulty of dissecting these in the average cadaver.

For organology and histology corrosion preparations of the lungs, liver, spleen, kidneys and other organs are of great value. For embryology specimens cleared in caustic potash and glycerine are useful in illustrating the development of the skeletal system and* in case of specimens which have been previously injected in illustrating the development of the vascular system and other organs.

(e) Library, The anatomical department should be provided with a library which should include the standard monographs and journals devoted to the subject. As already mentioned, if the main library of the medical college is not readily accessible to the quarters occupied by the department of anatomy the latter should have a special library in wliich files of current journals, text-books and works of reference are kept. Such a library should be readily accessible to students and members of the staff. Tn general, valuable sets of periodicals are best kept in the main library, where, as a rule, they will receive better care.

((r) Auxiliary facilities. See under F.

{H) The proportion of didactic to laboratory teaching. The committee is unwilling to set a fixed standard. Laboratory work should consume the major portion of the time devoted to each of the anatomical sciences. Lectures may perhaps justly take a greater portion of the time in courses in embryology, histology, neurology and applied anatomy than in gross anatomy.

D 434 C. R Bardeen.

(/) The proportionate number of hours to be devoted to the anatomical sciences in a 3600- to 4000-Aour curriculum. In most American medical schools of the better grade from 800 to 900 hours, about a fifth of the curriculum, are devoted to the anatomical sciences. In the present schedule of the Association of American Medical Colleges 630 hours out of 4000 are allotted to anatomy, but the committee on medical education of that association has recommended that this be increased to 750 hours. Your committee believes that 700 hours in a 3600-hour curriculum or 750 to 800 hours in a 4000-hour curriculum would represent a fair proportion of time for anatomy. The majority of the committee believe in a medical curriculum in M^hich the required work is kept at a minimum which will give students considerable time for elective work and independent study. They would prefer to see the defined required work for the whole curriculum kept within the original estimate of 3600 hours.^

The time allotted to anatomy will be sub-divided according to the arrangement of courses. Thus when a separate course in neurology is devoted to the gross and microscopic anatomy of the central nervous system and organs of s[)ecial sense an equivalent amount of time may l)e taken from the courses in gross and microscopic anatomy. In the present schedule of the Association of American Medical Colleges 00 hours are given to histology, 90 to embryology, 30 to osteology and 420 to gross anatomy. In the schedule recently proposed 540 hours are given to gross anatomy, 135 to microscopic anatomy and 75 to embryology. The following sub-division of time agi-ees approximately with the majority of the schedules proposed by the various members of your sub-committee on anatomy for a 3600hour curriculum.

In a 700-hour schedule the course in topographical anatomy should be taken from this list and ten hours added to the course in gross anatomy.

'Professors Ilardosty and .Tackson i)ref(T u ciuTicuhnii of at least 4000 hours. Professor Plersol believes it difficult to Iceep within a 4000-hour schedule, but that in sudi a schetiule anatomy should have 871 hours, of which 60 should come in the third year.

D Report of the Sub-Coininittee on Anatomy. 485

Gross Anatomy 370 hours

Histology , 140 hours

Neurology 90 hours

Embryology 90 hours

Topographical Anatomy 70 hours

Total 760 hours

(K) Cost of maintaining an anatomical department. A few years ago the chairman of the sub-committee on anatomy obtained from those in charge of the departments of anatomy in several of our leading universities an estimate of the cost per year of maintaining their departments. Below a summary is given of the average cost of maintenance of the anatomical departments of five endowed and of four state universities. When gross and microscopic anatomy are taught in separate departments the budgets of the two departments are added together in making up the estimates. The average number of students in each class at the five endowed universities was 85, in the four state universities 80.

Table Showinq the Average Yearly Expenditures for the Anatomical


At five Endowed Universities. At four State Universities. . .


14,000 8,000

Technical j Apf>aratuR,


etc. * I Total.

2,000 I 3,600 19,600

1,000 , 2,600 11,600

BIBLIOGRAPHY. A list Is appended of some of the more imiwrtant recent articles relating to anatomical teaching and to the equipment of anatomical laboratories.

American. Allen. H. Mori>hology as a Factor in the Study of Disease. Proceedings

Ass. AnL Anat., 1804. . Addresses on Anatomy. I. Comparative Anatomy as a Part of the

Medical Curriculum. Proc. Am. Ass. Adv. Sc, 1880. II. On the Teaching

of Anatomy to Advanced Medical Students. Medical News, Phlla., 1891. Baker, F. The Rational Method of Teaching Anatomy. Medical Record,

XXV, pp. 421-425, 1884. Barker, Ju F., and Bardeen, C. R. Outline of Course in Normal Histology

and Microscopic Anatomy. Johns Hopliins Hospital Bulletin, XII, No.

62-63, 1806.

D 436 C. R. Bardecn.

Babkeb, L. F., and Kyes, P. On the Teaching of the Normal Anatomy of the Central Nervous System of Human Beings to I^rge Glasses of Medical Students. Proc. Ass. Am. Anat, 1900.

Babkeb, L. F. The Study of Anatomy. Journal Am. Med. Assoc, March, 1901.

BevAfT, A. D. What Ground Should Be Covwed in the Anatomical Course in American Medical Colleges. Proceedings Assoc. Amer. Anat., VI, pp. 47-40, 1804.

Bbownino, W. W. Remarks on the Teaching of Practical Anatomy. Brooklyn Med. Jour., VIII, pp. 329-341, 1894.

Campbell, W. F. Teaching of Anatomy. Brooklyn Medical Journal. 1905.

CoBSON. The Value of the X-Itay in the Study and Demonstration of Normal Anatomy. I*roc. Ass. Am. Anat., 1900.

I) WIGHT, T. The Scope and the Teaching of Human Anatomy. Boston Medical and Surgical Journal, CXXIII, 337-340, 1890.

. Methods of Tea<*hing Anatomy at the Harvard Medical School.

Ihld, CXXIV, pp. 475-477, 1891.

. Problems of Clinical Anatomy. Massachusetts Med. Soc., June,


Flint, J. M. The Anatomical I^lwratory of the University of. California.

The Johns Hopkins Bulletin, Vol. XVI, 1905. CiFBBiSH, F. H. The Best Order of Topics in a Two Years* Course of Anatomy

in a Medical School. Proc. Ass. Am. Anat. 1894, VII, 8-17, Wash. 1895.

Huntington, G. S. T\\e Mon>hological Museum as an Educational Factor in

the University System. Proc. Ass. Aul Anat.. 1900.

. The Teaching of Anatomy. Columbia University Bulletin, 1898.

Jackson, C. M. A Method of Teaching Relational Anatomy. Proc. Am. Med.

Assoc, fifty-second annual meeting. Journal A. M. A., Sept., 1901.

. Report on (iross Anatomy. Proceedings American Medical Colleges. Cleveland, lOOa

Kkkn, W. W. On the Systematic T'se of the Uving Model as a Means of IlluHtration in Teaching Anatomy. Intern. M. Cong., Ijondon, I, 174, 1881.

Kkiller, W. The Teaching of Anatomy. New York Med. Journal, IX, 1894, r)p. 281), 513, 545.

. On the Preservation of Subjects for Dissection by Injection With

Formalin and Carbolic Acid Solution and Storage by Immersions In Similar Solutions. American Jour. Anat., II, 1902-3, Proceedings Ass. Am. Anat., p. 7.

Kkrr. AnsAM T. On the Preservation of Anatomical Material in America by Means of Cold Storage. Johns Hopkins Hospital Bulletin, Vol. XII, 1901.

I.i'SK, W. C. An Injecting Fluid for Preserving Cadavers. Anat. Record, III, p. 47, 1909.

D Report of the Sub-Coirunittee on Anatomy. 437

McMuBRicH, J. P. Present Status of Anatomy. American Naturalist, Vol.

XXXIII, Mar., 1809. . Conservatism in Anatomy. President's address. Proc. Ass. Am.

Anat, Anat. Record, 1909. Mall, F. P. The Anatomical Course and Laboratory at the Johns Hopkins

University. Johns Hopkins Hospital Bulletin, 1896.

. Liberty in Medical Education. Philadelphia Medical Journal, 1899.

. On the Teaching of Anatomy. Johns Hopkins Hospital Bulletin,

1905. . On Some Points of Importance to Anatomists. Anat. Record, Vol.

I, 1907, also Science, 1907. . On the Teaching of Anatomy. Anatomical Record, 1908.

Mabks, G. H. Tlie Study of Anatomy : Its Position in Medical Education in England and America. Boston Med. and Surg. Jouru., CXIII, 104-107, 1885.

MiNOT, C. S. The Unit System of Laboratory Construction. Phil. Med. Journ., 1900.

Moody, R. O. On the Use of Clay Modeling in the Study of Osteology. The Johns Hopkins Bulletin, Vol. XIV, 1903.

Shepherd. Method of Preserving Bodies. Proceedings Assoc, of American Anat., May, 1900, p. 23.

Shiels, G. F. a Plea for the Proper Teaching of Anatomy. Jour. Am. Med. Ass., XXIII, pp. 110-112, 1894.

SoucHON. Embalming of Bodies for Teaching Purposes. Anatomical Record, II, p. 244, 1908.

Taskeb, D. L. What Is a Practical Examination in Anatomy? California State Journal of MeiUclne, Oct., 1908. Abstr. Jour. Am. Med. Assoc, LI, p. 1818, 1908.

Terry, R. J. A Method of Sectioning the Whole Decalcified Body With a Knife. Proceedhigs Ass. Am. Anat, May, 1900.

. Class Work In Practical Anatomy. Quarterly Bulletin Med. Dept.

Washington Univ., Vol. Ill, No. 2, 1904.

Waite, F. C. Outlines of I^aboratory Course in Microscopical Technique, Histology and Microscopical Anatomy. Cleveland, 1903. Outline of Laboratory Work in Vertebrate Embryology. 1907.

. The Teaching of Histology and Embryology. Report Ass. Am.

Med. Colleges, 1908.

Zappe. Report Ass. Anh Med. Colleges, 1908.

Foreign. Addison, C, and Waterson, D. Teaching and Examination in .Anatomy.

British Metlical Journal, Aug., 11K)8. VoN Baer, K. E. Nachrichten liber Leben und Schriften. St. Petersburg, 1866.

D 438 C. R. Bardeen.

Cleland, J. Lecture on Anatomy as a Science and in Relation to Medical Study. Lancet, Tx)ndon, 1892, II, pp. 928, 982.

Cooke, T. The Teaching of Anatomy : Its Aims and Methods. Lancet, I>ondon, 1893, II s. 1135, 1350. Med. Times and Hospital Gaz., London, XXII, 621, 1894.

Griffith, F. New Method of Anatomy Study (Modelling). New York Medical Journal, May, 1908.

FlASSE, C. Die Lehrsammlungen der Breslauer Anatomie. Archiv f. Anat. u. Phys., 1899.

Hertwig, O. Der anatomische TJnterricht. Jena, 1881.

. Auflforderung zur Ueherlassung von mikroskopischen Praparaten

ftir das wissenschaftliche Museum der vergl. u. experim. Histologic u. Entwicklungslehre am anat.-biolog. Instltut zu Berlin. Anat. Anz., 321, 1900.

IIis, W. Ueber die Bedeutuug der Entwickelungsgeschlchte ftir die Auffassung der organischen Natur. I^eipzig, 1870.

. Ueber die Aufgaben und Zielpunkte der wlssenschaftllchen Anatomie. Leipzig, 1872.

Humphrey, G. M. An Address on the Study of Human Anatomy. British Med. Jour., London, I, 1030, 1887.

Keith, A. Anatomy in England During the Nineteenth Century. Lancet, I^ndon, Jan., 1908.

VON KoLLTKER, A. Die Aufgabcu der anatomisohen Institute. Wiirzburg, 1884.

KoLLMANN, J. Handsammlung f. die Studierenden in den anatomischen Instituten. Verb. d. anat. Gesellscb. auf der 9 Vers, in Basel, 1895.

Krause, W. Ilistologische Institute. Berl. klin. Wochenschr., XXXIII, pp. 571-573, 1895.

Macalister. On Methods of Anatomical Study. Brit. Med. Jour., London, II, 879-883, 1893.

. Introductory Lecture on the Province of Anatomy. British Med.

Journal, Ix)ndon, 1883, 808-811.

. Some Characteristics of Anatomical Teaching in Great Britain.

Internat. Monatscbr. f. Anat. und Histol. Berlin, I, 299-304, 1884.

Merkel, F. Ueber die Darstellungsweise der allg. Anatomie. Verb. anat. Ges. 16 Vers. Halle, 1903.

Morris, H. An Address on the Study of Anatomy. British Med. Journ., London, II, 1337, 1895.

Primrose, A. Study of Anatomy by Frozen Sections. Canadian Pract, Toronto. XIX, 319-a30, 1894.

Radber, a. Ueber die Einrichtung von Studiensalen in anatomischen Instituten. I^ipzig, 1895.

D D D Report of the Sub-Committee on Anatomy. 439

Stewabt. Ch. Catalogue, Descriptive and Illustrated, of the Physiological

Series of Comparative Anatomy in the Museum of the Royal College of

Surgeons. London, Vol. 1, 1900. Stohb. Ueber Demonstratlonsmlttel, Verh. d. anat. Gesellsch. auf d. 5 Ver sammluug, 1801. Testut. Qu'est-ce que Thomme pour un anatomiste. Revue Sclent., Paris,

XIII, 05-77, 1887. ToNKOFF, W. Ueber die Einrichtung der anatom. Lemsammlungen. Anat.

Adz., XXIX, pp. 479489, 1906. Waldeyer, W. WIe soil man Anatomie lehren und lernen? Berlin, 1884. . Die Mazerations-einrichtung an dor Anat. Anstalt zu Berlin. Anat.

Anz., XXXI, 240-251. 1907. WiNDLE, B. C. A. On the Study of Topographical Anatomy by Means of

Frozen Sections of the Cadaver and Foetus. BIrmiugham M. Rev., XIV,

145-155, 1883.


Early Ontogenetic Phenomena in Mammals and their Bearings ON Our Interpretation of the Phylogeny of Vertebrates. A. A. W. Hubrec'ht, Quart. Journ. Micr. ScL, Vol. 53, 1, 1908.

This important memoir consists of 170 pages embracing 37 plates with a total of 100 illustrations. As the title indicates, it represents an effort to determine more accurately the ancestral history of the vertebrates by aid of a close detailed study of the initial stages of mammalian development. It is, moreover, a first attempt toward ascertaining a more direct phylogeny of the Monodelphia — for the author does not regard the didelphian mammals as representing a transition phase l)etween Ornithodelphia and Monodelphia, but as descendants from monodelphian placental ancestors.

The value of the paper Vws perhaps more in its suggestiveness than in its contribution of new facts. It brings together the results of the author's own investigations and those of contemporary embryologists scattered throughout various periodicals and publications more particularly during the last quarter century. This wealth of facts the author attempts to collate in the form of several generalizations offered as good working hyi)otheses. The presentation of this mass of data in such easily accessible form is a very real service to embryologists; and the several novel views advanced — more or less tentatively — disclose the disputed ground and the obscure points, and thus indicate the lines for further research.

The author invites younger embryologists "to tackle wherever they can the early developmc^ntal stages of mammals or amphibia in preference to the cartilaginous fishes or Amphioxus, however much more easy the latter material can be obtained," having no doubt that mammalian embryology has yet very many surprises in store for us.

The fetal structures are regarded as invaluable guides in the


D Book Reviews. 441

study of the evolutiou of the Mammalia. The most delicate indications for the determination of natural affinities are furnished by the details of ontogeny and placentation. In the light of these assumptions llubrecht considers specially (a) the embryonic envelopes (trophoblast, chorion, and amnion) and appendages (umbilical vesicle and allantois) ; (b) placentation; (c) gastrulation (continued as notogenesis and cephalogenesis) ; and (d) the origin of the meso-. derm. A re-classification of the Vertebrates is suggested (in harmony with comparative anatomical and pal aeon tological evidence), and the hypothetical ancestor (vermactinian) sketched.

In all vertebrates except Amphioxus — and more clearly in mammals — the didermic stage of the blastocyst ("gastrula") arises by a process of delamination of entoderm from the embryonic knob. The delamination gastrula of mammals generally enters upon the later stages of ontogeny without the appearance of a distinct blastopore, though "atavistic" attempts have been described in Tarsius (Hubrecht), the rabbit (Keibel), the mole (Heape), the opossum (Selenka") and the shrew (Hubrecht).

All vertebrates pass through a process of notogenesis. This follows upon gastrulation and brings about the formation of the notochord and mesoblastic somites. Notogenesis is initiated by the appearance of a median ectodermal proliferation on the embryonic shield, the protochordal wedge (Ilenseii's knot). The latter occurs in the identical spot where the rudimentary blastophore was located, and recalls the blastoj)ore of invertebrates because cell proliferations commence here which give rise to mesodermic structures.

Hubrecht regards the trophoblast as a larval envelope of great antiquity present in all mammals, somewhat hidden in sausoj)sids and recognizable as reminiscences in Amphibia, Dipnoi and Teleosts as the so-called "Deckschicht." The trophoblast is regarded as the mother-organ of the chorion and amnion, and as such indicates a more natural line of separation among Vertebrates between Elasmobranchs and Teleostomes rather than between Amniota and Anamniota. This conclusion is supported by facts from comparative anatomy: (a) phenomena of ossification which reveal a close detailed homology between elements of the skull and the visceral arches of

D 442 The Anatomical Record.

bony fishes and higher mammals; (b) various anatomical differences between reptiles and amphibia break down in the case of very numerous important fossil forms; (c) presence of an air-bladder or lung from Teleostomes upward and their absence in Elasmobranehs, The suggestion is made that many of our Dipnoi, Ganoids and Teleosts may perhaps also have had terrestrial ancestors just as Cetacea are regarded as descendants of terrestrial nmmmals. The improved classification suggested to express more accurately degrees of consanguinity divides the vertebrates into four superclasses: (1) Cephalochordata (Amphioxus) ; (2) Cyclostomata ; (3) Chondrophora (Elasmobranchii) ; (4) Osteophera (all the other higher vertebrates). The idea that the mammals are descended from reptilian-insectivorean ancestors is abandoned. Both Sauropsida and Mammals are thought to trace their phylogenetic history back through very early protetrapods of the C.^arboniferous period (amphibian-like animals) and their still earlier aquatic progenitors to vermiform predecessors of coelenterate pexHgree.

A definitive chorion (Primates) or diplotrophoblast (Sauropsida) and amnion appear in all vertebrates above Amphibia. In sauropsids these two larval organs are often closely related in development. The chorion (^^amniogenetic chorion," Bonnet) seems to arise from the outer plait of the amnion-fold. Many mammals (some rodents, monkeys and man) exhibit no amnion-fold. The single fact that the chorion (trophoblast) appears long before the amnion argues against deriving the amniogcnesis of some mammals by a process of coenogenetic modification of phenomena as they obtain in sauropsids. It seems nearer the truth to think of the phylogenesis of these layers in the reverse order, or as originally unrelated. The question remains as to how the amnion has developed out of or along the side of the trophoblast.

Contrary to the opinions of Kolliker, Selenka, Ziegler, Keibel and others, Hubrecht would derive the mesoderm from both the ectoderm and entoderm. In the shrew, which furnishes the clearest case, the mesoderm arises from four independent sources. Homologous mesoderm sources are present under varied aspects also in sauropsids and ichthyopsids. These are (a) annular zone; (b) protochordal

D Book Reviews. 443

plate; (c) protochordal wedge; (d) ventral mesoblast. The first two are entodernial, the last two ectodermal in origin. The several sources soon become confluent. The annular zone produces the material for the area vasculosa (angioblast). The hinder part opposite the protochordal plate aids in the formation of blood vessels and blood from which the vascularisation of the connective-stalk (embryophore) of Primates is derived. The presence of such an annular zone of mesenchyme is denied by Rabe, Keibel and 0- Hertwig.

The protochordal plate functions as a median mesenchyme-producing area in the entoderm. Among other things it furnishes the endothelium of the heart (Tarsius). The protochordal wedge appears at the spot which coincides with the anterior lip of the evanescent blastopore of the didermic stage. This area of proliferation builds the notochord. No evidence appears in Tarsius that the protochordal wedge undergoes a forward extension to be identified with the "head process." The apparent forward growth is explained as an elongation by material being added poster ioly to form the notochord and primitive somites concomitantly with increase in length of the embryonic shield. The ventral mesoblast source is separated from the protochordal wedge by the potential blastopore. It gives rise to the early extra-embryonic coelom and the visceral and parietal layers of the mesoderm. It also contributes to the formation of the connective-stalk.

Facts gathered from the works of various authors bespeak a complete homology between Amphibia and Manmialia in respect to the presence of four areas of proliferation for mesoderm. Hubrecht claims, moreover, to find a great amount of correspondence between these groups and the Sauropsida concerning the general featun^s of very early mesoblastic formations. In Amphioxus, Legros ('07) has identified a protochordal plate and a protochordal wedge. In Elasmobranchs the four centers can be discerned. The ventral mesoblast is here identified with the so-called "Schwanzknopf." In Teleosts similar relations prevail except that the protochordal plate is here detected in a portion of the periblast.

The method of placentation as it appears in man, the anthropo

D 444 The Aiiatoniieal Record.

nior])hae and the hedgehop is lx4ieved to Ik* the primitive type. A cogent argument against the contrary j)ositi(m is the fact that a <Hffuse placenta is found in Lemurs, Cetacea, Edentata and Ungulates, groups anatomically widely separated. Placeutatiou here is interpreted as degenerate or secondarily modified. Accordingly, then, there seems to be little justification for the attempt to derive the intricate and highly efficient placental arrangement of Primates and Insectivores out of the so-called primitive (diffuse) placenta of Ungulates and Lemurs. Ilubrecht explains the absence of transition stages in the process of simplification between the more primitive (ape) and the secondarily modified (lemur) types, by the fact that these an» secrets taken into the grave by very old probably mesozoic Afammalia. The primarily primitive type of placenta thus remains unknown; all the types of placentation represent coenogenetic modifications and simplifications of varying degree (Apes, Tarsius, Lemurs). Other lines of simplification ended in the Ungulates (horse) and Edentata (Maiiis). In general the primates seem to contain the most primitive type of mammals; and man seems to have retained ontogenetic characters even more primitive than the ap(»s or Tarsius. Thus an umbilical vesicle exclusively hematopoietic in function, a small allantois confined within a connective-stalk, an amnion in the form of a closed cavity from the iK'ginning. a decidua capsularis and a restricted and highly specialized placenta, all characteristic of human ontogeny, are regarded as most ])rimitive.

The following quotation may be said to summarize the main conclusion of the paper: "Viviparity and placentation have gone hand in hand with the development of allantois and amnion, and only after the two latter had ap])eared in the early viviparous Tetra])ods of the ])alaeoz()ic period did certain side lines of development diverge from that which led up to modern Mono- and Didelphia. In these side lines oviparity again came to the front, and on them we meet the parent forms of the Ornithodelphia, Reptilia and the Aves."

//. E. Jordan. Received for publication, April 12, 1909.


There is every promise of a great development of the Anatomical Department in the University of Minnesota. Prof. Thomas G. Lee has been made head of the Department of Anatomy, which now includes histology, embryology and neurology as well as gross anatomy. Dr. Robert Retzer, at present Associate at the Johns Hopkins University, has been made Assistant Professor and is to have charge of the dissecting room. The staff in gross anatomy will inchide Prof. Erdman, Prof. Retzer, Dr. E. R. Hare, and Dr. Disler, while that in histology and neurology will include Prof. Lee, Prof. J. B. Johnston and Dr. W. S. Nickerson. Through the action of the legislature there will be a general development of the entire university. During the past tw^o years half a million dollars have been expended in enlarging the campus and a sum of throe hundred and fifty thousand more is available for the same purpose. For new buildings a sum of over a million has been appropriated ; the medical group of buildings will occupy a separate portion of several acres of the newly acquired territory on the bluffs along the bank of the Mississippi River. The new anatomical laboratory will be erected at a cost of two hundred thousand dollars. Prof. Lee is to go abroad in the early fall to visit the various laboratories and study the plans for the new department.




Vol. III. AUGUST, 1909 N^^ 8



R. H. WHITEHEAD. From the Anatomical Laboratory, University of Virginia.

The specimen which I am about to describe was very generously given to me by Dr. Glasgowl Armfltrong, of Staunton, Va. To him I am indebted also for the following history :

The mother of the monster was a negress 31 years of age, who had had two previous labors, both normal in all respects. Her personal history was negative. She menstruated last on March 20, 1908. At 8 a. nL, December 28, 1908, the amnion suddenly ruptured while the woman was going upstairs. She went to bed, but labor pains did not come on until about 10 o'clock. At 4 a. m., December 29, a head, a pair of shoulders, and one arm were delivered. The pains increased in severity, but as no further progress was made the attending physician called Dr. Armstrong in consultation at 10 a. m., with the statement that the child was thoroughly wedged, and could not be moved. Dr. Armstrong found one head, the corresponding shoulders, and three arms delivered; and he could feel a fourth arm. On making pressure upon the buttocks, the body of one of the components was delivered, and he could then determine definitely that he was dealing with a thoracopagus. The second head followed easily and quickly the hips of the first component, and labor was soon terminated. The patient made a normal convalescence,


D 448 R H. Whitehead.

and was out of bed in ten days. Thus Dr. Armstrong adds another to the list of eases of thoracopagi which have been delivered without a mutilating operation.

The specimen (Fig. 1) consists of two female fetuses at full term united by a bond of union which extends from the level of the lower borders of the manubria stemi to jthe umbilicus. The latter is single, measures 6 cm. in diameter, and is closed by a thin membrana reuniens containing no muscle tissue. This membrane had been torn, and a long piece of small intestine had escaped through the opening. The umbilical cord is also single, but bifurcates just before it reaches the umbilicus to send a prong to each component of the monster. A plane passed vertically through the center of the bond of union divides the monster into two symmetrical components; the one to the right of thia symmetry plane I shall call A, and the one to the left B. The bodies of the two components do not face each other exactly, but their median planes form an angle with the symmetry plane, so that the bond of union is wider on one surface of the monster than on the other. This wider surface, since the faces look obliquely towards it, I shall call, for purposes of description, the anterior surface, and the opposite one the posterior surface. Thus, the distance between the nipples on the anterior surface is 6 cm., while on the posterior surface it is but 4 cm.

Steuctuee of the Walls of the Bond of Union.

In the symmetry plane in front are the second and third pieces of the sternum connected with the ribs as usual. A similar, but much narrower, sternum is present behind. Developmentally, of course, these two sterna are compounded out of equal constituents furnished by each of the two components. Thus, in the case of the anterior sternum, the half to the right of the plane of symmetry is furnished by A, the left half by B. And so also as to the muscles in the bond: the rectus to the right of the synmietry plane in front is the right rectus of A, and the left one is the left rectus of B, etc. In this specimen, owing to the large size of the umbilicus, there is a wide triangular inter\^al between the corresponding recti.

D Description of a Human Thoracopagus. 449

The Abdominal Viscera. Each component possesses a stomach, normal in appearance and position, which is succeeded by a duodenum. The two duodena, however, fuse at their terminal portions to form a common jejunum, which, after a course of 20 cm., bifurcates

Fig. 1.

and sends a division to each component. Below this point both alimentary canals are normal.

The liver is single, and lies with its long diameter in the plane of symmetry. It is formed by contributions from both A and B, and

D 450 R. H. Whitehead.

consists approximately of the right lobe of each, the two left lobes being almost unrepresented in the completed organ- The attachment of the liver to the diaphragm, which muscle is also single, is along a plane at right angles to the symmetry plane and divides the organ into two synmietrical parts, which are, as said before, essentially the right lobes of the two components. Each lobe is crossed by a falciform ligament containing in its free margin an umbilical vein. There is no umbilical fissure, however, but the vein plunges directly into the substance of the lobe through its superior surface. On the inferior surface of each lobe is a short porta which contains the usual structures, portal vein, hepatic artery and but one hepatic duct. There is a gall bladder attached to the inferior surface of each lobe, the fundi of which look in opposite directions. The bladders taper in the usual manner to form cystic ducts, which unite with the corresponding hepatic ducts to form common bile ducts ; the latter, however, open into each other making a loop under the liver, and have no connection with the duodena. Just why the connections with the duodena have been lost is not obvious. All the other abdominal and the pelvic viscera seem quite normal.

The Thoracic Viscera. The lungs, two in each component, appear normal both in position and configuration ; they contain no air. The heart is the most interesting feature of the monster and is unlike any that I have found described in the literature. Inspection (Fig. 2) shows that it is compounded out of constituents from each component, and that its long diameter is transverse to the plane of symmetry. While four auricular appendages can be made out, it is seen that the auricles of the two components are united with one another. Furthermore, the auricular appendages do not occupy their normal position, but are too low, that is to say, the auricles have not been drawn as far upwards onto the base of the ventricles as happens in normal development. The portion of the heart contributed by A is dextrocardiac in position. Inspection also shows that the ventricular portions of the two hearts are united, but does not disclose the relationship of the various cardiac cavities to one another. This latter point could not be determined definitely except by studying a series of gross sections made transverse to the long diameter of the


TllK Anato.MUAI. Ukcoud. — Vol. III. No, S.

D D \

D Description of a Human Thoracopagus. 451

heart. This study discloses the following facts: the two auricles of A are connected with those of B by a large thick-walled channel, which we may call the interauricular sinus, with the result that there is virtually one huge auricle imperfectly divided on each side of the symmetry plane into two by an interauricular septum. The arrangement of the ventricular cavities and their connections with this auricle are shown in the accompanying diagrams. Fig. 3 represents the condition in the anterior portion of the heart The right ventricle of A and the left ventricle of B, both small, are united by a septum of muscular and connective tissue. Both open freely into the auricular cavity by means of a large auriculo-ventricular canal.

Fig. a

which is not guarded by valves. In the posterior portion of the heart (Fig. 4) the fusion between the two ventricles is much more complete, and there is practically but one ventricle, composed of the left ventricle of A and the right ventricle of B, opening into each other and into the common auricle. The left ventricle of A is closed at its base, having no opening into the aorta of that side, as will be described later.

The Blood-vessels at the Base of the Heart. — (Fig. 2.) In the right component (A) both the venae cavse are formed normally and open into the right auricle. But in B the superior vena cava is formed on the left side, and both the superior and the inferior venae cavae open into the left auricle. The vena azygos major of B also lies to the left of the vertebral column. The relation of the large

D 452 E. H. Whitehead.

arteries to the heart is different in the two components. In the case of B the pulmonary artery arises from the left ventricle, the aorta from the right ventricle ; in other respects these two vessels are normal. In A the pulmonary artery springs normally from the right ventricle and, after giving off a branch to each lung, continues as the ductus Botalli to join the aorta in forming an arch, from which the left subclavian artery takes origin; this arch then proceeds downward as the descending thoracic aorta. The ascending aorta is represented by an impervious cord extending from the base of the left ventricle, but not connected with its cavity; just before joining the arch mentioned above it develops a lumen, and gives off a short trunk which divides into the innominate and left common carotid arteries.

FiQ. 4.

The pulmonary artery in each component and the aorta of B are provided with normal valves.

It will be noticed that there is a curious bilateral symmetry in respect to the connections of the large blood-vessels with the heart. In both components the large veins empty their blood into that auricle which is the more anterior, namely, the right in A, the left in B; and in each the pulmonary artery arises from the ventricle which is the more anterior, the right in A, the left in B.

It is evident that the circulation in this heart was quite simple. The ventricles are little more than narrow clefts unprovided with auriculo-ventricular valves, through which the blood was forced by the large auricle directly into the aorta and the pulmonary arteries without stopping in the ventricles. It is also clear from what has

D Description of a Human Thoracopagus. 453

been said concerning the composition of the common auricle and the openings of the veins therein that the arterial blood brought to the heart in the inferior vense cavse was mixed in the auricle with the venous blood from the superior venae cavse, and that, accordingly, mixed blood was distributed to all portions of the body by the fetal arteries. This mixed blood was quite sufficient to bring the two fetuses to full intrauterine maturity — a fact which supports the view of Pohlman^ and others that mixing of the arterial and venous bloods in the heart is characteristic of the normal fetal circulation. And, finally, it also appears that the blood of each component was mixed with that of the other in the common auricular cavity.

Considerations as to Genesis. This specimen sheds no new light upon the causal genesis of thoracopagi or other monsters, and I shall not attempt to discuss that question. Indeed, it seems that the data at hand are far too insufficient to allow anything approximating positive conclusions. Wilder's^ attractive theory that monsters are merely germinal variations from the normal may or may not be correct ; it does not seem possible either to affirm or to deny it. And so also as to Mall's^ suggestion, I find it very difficult to work out a causal connection between the existence of an endometritis and the formation of a thoracopagus.

With respect, however, to the formal genesis of such monsters we are in much better position. Beginning with such experiments as those of E. B. Wilson and others which showed that the egg of Amphioxus and of amphibians could be induced by separating the blastomeres in the two-cell stage to develop two embryos, and ending with the observation of two blastoderms, or two primitive streaks, or very young double monsters on the same yolk in the fowl egg, we have a series of well-attested observations tending to show that under conditions not always understood the fertilized ovum is capable of

A. G. Pohlman. The Course of the Blood through the Heart of the Fetal Mammal, Anatom. Record, Vol. Ill, No. 2, 1909.

■H. H. Wilder. The Morphology of Cosmobia. Amer. Jour. Anat., Vol. VIII, No. 4, 1908.

■F. P. Mall. Study of the Causes Underlying the Production of Human Monsters. Jour. Morph., Vol. XIX, No. 1, 1908.


Fio. 5.

Fio. 6.

Fio. 7.

D Description of a Human Thoracopagus. 455

giving rise to two centers of embryo fonnation. Upon these facts as a foundation schemes for the reconstruction of the development of monsters have been advanced by various students of teratology, the latest of which is that proposed by E, Schwalbe.* It consists, essentially, in the assumption of two formative centers in the same ovum, each of which if uninterfered with by its neighbor, would produce a perfect embryo ; but under the conditions of development portions of the two embryos are brought together at a very early stage of development and more or less fusion takes place. In the remainder of this paper I wish to test this theory upon the thoracopagus under consideration.

Assuming that two centers of formation appeared in this ovum at a very early stage, we may pass at once to a consideration of the probable position of the primitive streaks. It seems clear, as has been emphasized by Kaestner,* that the relative position of these structures must be of much importance in determining the configuration of the future mx)nster, since the streak determines the position of the longitudinal axis of the embryo and of various organs. In our case, as the median planes of the two components formed an angle with the symmetry plane, it is probable that the streaks were not parallel, but were placed at an angle to the future symmetry plane. The diagram. Fig. 5, represents this condition in a later stage when the medullary grooves have formed. As is well known, the heart and pericardium are developed from a paired anlage, the two parts of which are soon brought together and fused in the median line, by which process there is also formed a floor for the foregut (Figs. 6 and 7). Starting with the condition shown in Fig. 5 it is seen that the heart regions of A and B — regions which bulge prominently on the ventral surface of young embryos — would soon be brought in the progress of their development into very intimate contact, and might easily fuse in the same way as the halves of the normal heart anlagen unite with each other. Fig. 8 represents this process of fusion as affecting only the pericardia — a condition which obtains

B. Schwalbe. Die Morphologie der MissbUdungen. Tell II, 1907. ■S. Kaecitner. Doppelbildungen an Vogelkeimscheiben. Arch. f. Anat. u. rhyslolog., Anat. Abt., 1901.

D 466 K. H. Whitehead.

Fio. a

in those thoracopagi which possess two separate hearts in one pericardium. It is dear, however, that in the case under consideration fusion had occurred between the hearts as well as the pericardia, a process represented at its beginning in Fig. 9. It is evident, further^ more, that the single heart tube must have been formed both in A and B previous to fusion. For, if in Fig. 6 one imagines a similar

Fig. 0.

D Description of a Human Thoracopagus. 457

diagram placed opposite to the one there drawn, it will be seen that fusion, of the paired anlagen of the two components would probably result in two hearts and two pericardia; and that these two hearts would lie the one behind the other in the plane of symmetry, a position which is just opposite of that found in our case. The intimate fusion of the auricular portions of the hearts in this case would indicate that these portions of the primitive hearts were brought into closest contact; and the more intimate fusion between the two posterior ventricles might be explained in the same way, the greater propinquity in this situation being due to the oblique position of the two bodies with reference to the plane of symmetry.

In connection with the heart we have yet to account for the anomalous relations of the pulmonary artery and the great veins to the left side of B's heart. These may be explained in some such way as follows: We have seen that in A the heart had a decided dextro position. It seems reasonable to suppose that, during the complicated twistings and rotations undergone by the developing hearts, the heart of one component might prevail over that of the other. It might well be that what is, so far as position is concerned, the left side of the heart in B was originally the right side, which was rotated into its abnormal position by the heart of A.

The conditions found in the other viscera are readily acounted for on this hypothesis of Schwalbe : the two f oreguts would develop normally, being separated by the fusing hearts and livers; but below the level of the liver there would be nothing to prevent the midguts from fusing over a greater or lesser portion of their extent.

I wish to thank Mr. F. P. Smart for the photograph of the specimen shown in Fig. 1 ; and Prof. C. E. Meloy for the excellent drawing of the heart.

Received for publication, May 12, 1909.



ALBERT KUNTZ. With 2 Figubes.

In studying embryos of the pig for the purpose of tracing the development of the sympathetic nervous system, the writer has been interested in the evidence for the migration of nervous elements from the neural tube along the fibers of the peripheral nerves. In a recent paper^ I have described the migration of medullary cells, among which are to be recognized cells of an indifferent character and neuroblasts, into the dorsal and ventral nerve-roots. In transverse sections of embryos of the pig 6 and 7 mm. in length breaches in the external limiting membrane of the neural tube occur quite frequently in the region of the dorsal nerve-roots. Through these breaches lines of cells practically touching each other end to end may be traced from the mantle layer into the proximal part of the dorsal nerve-roots (Fig. 1, dnr). Further evidence for the migration of medullary cells into the dorsal nerve-roots is afforded by the fact that in many sections of embryos 6 and 7 mm. long, where no breaches occur, cells are found in contact with the external limiting membrane inside the neural tube in the region of the dorsal nerve-roots. In embryos 9 mm. and over in length this region, as shown in Fig. 2, dnr., is always occupied by fibers of the dorsal nerve-root and rarely are cells found among them.

In transverse sections of embryos of the pig from 9 to 13 mm. in length breaches occur in the external limiting membrane in the region

From the Laboratories of Animal Biology of the State University of Iowa. •The Migration of Nervous Elements into the Dorsal and Ventral Nerveroots of Embryos of the Pig. Proceedings of the Iowa Academy of Science, Vol. 16.


D Histogenesis of the Nervous System. 459

of the ventral nerve-roots. Through these breaches medullary cells may be observed migrating into the ventral nerve-roots (Fig. 2, vnr.). Migration of medullary cells into the ventral nerve roots of embryos of the pig has recently been described by Carpenter and Main ('07). I have been able to substantiate their observation that cells may be found "just inside the external limiting membrane, in an intermediate position half in and half out of the neural tube, and in the base of the ventral nerve-root just outside the external limiting membrane." But, whereas they have described only elongated cells which they recognize as the indifferent cells of Schaper, I have observed cells of a distinctly pyriform type migrating with the indifferent cells.

According to the researches of Schaper ('97) the germ cells (Keimzellen) of His (cells of epiblastic origin undergoing mitotic division near the internal limiting membrane of the neural tube) give rise to cells which he characterizes as indifferent These indifferent cells migrate toward the mantle layer and are there transformed either into neuroblasts or into embryonic supporting cells. In the higher vertebrates some of these indifferent cells undergo further division by mitosis in the mantle layer.

The cells which migrate from the neural tube into the dorsal and ventral nerve-roots are of two general types; elongated cells which are to be regarded as the indifferent cells of Schaper, and pyriform cells which are to be regarded as the neuroblasts of Schaper. The neuroblasts are much fewer in number than the indifferent cells, but are distributed indiscriminately among them. When observed passing out of the neural tube the tapering end is usually directed peripherally. This also is in accordance with the usual position of the neuroblasts in the mantle layer.

The orientation of the cells in the neural tube is such that two general courses of migration into both dorsal and ventral nerve-roots may be recognized. In the ventral region some of the cells move directly outward from the ventral zone toward the base of the ventral nerve-root ; others tend ventro-laterally from the region in which the lateral horn of the gray matter arises. In the dorsal region the chief course passes quite directly from the dorsal zone toward the proximal

D 460 Albert Kuntz.

part of the dorsal nerve-root, with some cells moving from the most dorsal region along the inner surface of the external limiting membrane. The other course tends dorso-laterally from regions ventral to the dorsal nerve-root. The cells of the latter course probably originate in the same r^on as those which move ventro-laterally toward the ventral nerve-root.

Further observation has shown that the cells which migrate from the neural tube into the spinal nerve-roots wander peripherally along the spinal nerves and visceral rami into the anlagen of the sympathetic ganglia. As the fibers of the ventral nerve-root emerge from the neural tube they are accompanied by medullary cells which may be distinguished from the mesenchymal cells by their size and form. Very often they also take a slightly deeper stain. As the fibers grow peripherally these cells wander along their course, while others emerge from the neural tube to take their places in the base of the nerveroot. Similar cells detach themselves from the distal ends of the spinal gan^ia and wander down along the sensory fibers. Whether these represent cells which have migrated from the neural tube into the dorsal nerve roots could not be determined since it was not found possible to trace cells through the spinal ganglia. Beyond the point of union of the sensory and motor roots it is no longer possible to distinguish the cells which wander down from the spinal ganglion from those which migrate out from the neural tube along the fibers of the ventral nerve-root. The majority of the cells thus distributed among the growing fibers of the spinal nerves are cells of the elongated type. Frequently, however, pyriform cells are found among them.

In transverse sections from the dorsal region of embryos of the pig 7 mm. in length the spinal nerves have extended peripherally a little beyond the level of the dorsal aorta. The fibers are loosely aggregated. Numerous elongated cells and a few cells of the pyriform type are found among the fibers as well as at the surface of the bundle. Fibers are not yet present in the visceral ramus, but at a point a little above the level of the aorta, cells, either singly or in groups of two or three, are seen to bend from their course nearly at right angles and wander through the mesenchyme toward the dorso

D D D Histogenesis of the Nen^ous System. 461

lateral angle of the dorsal aorta along the path later occupied by the fibers of the visceral ramus (Fig. 1 pvr.). At this stage the anlagen of the sympathetic ganglia are already present as loose aggregates of cells along the dorso-lateral angles of the dorsal aorta. Cells of the pyriform type were observed along the paths of the visceral rami and among the loosely aggregated cells along the dorso-lateral angles of the dorsal aorta (Fig. 1 nb.). Thus pyriform cells which are to be

Fio. 1. — Transverse section of neural tube and sympathetic anlage of 7 mm. embryo of the pig x 1^- &Q-* sympathetic anlage ; da., dorsal aorta ; d. n. r., dorsal nerve root ; gc, germ cells of His ; ic, indifferent cells ; Ics., indifferent cells in syncytium ; nb., neuroblasts ; pvr., path of visceral ramus ; spg., spinal ganglion; spn., spinal nerve; vnr., ventral nerve-root.

regarded as the neuroblasts of Schaper have been traced from the neural tube along the spinal nerves and visceral rami into the anlagen of the sympathetic ganglia. It may further be observed that the elongated cells, both along the spinal nerves and visceral rami, may often be seen joined together in small groups by protoplasmic processes or in small syncytia. The pyriform cells, however, are always free.

D 462 Albert Kuntz.

In transverse sections of embryos 10 mnu in length fibers appear in the visceral rami but do not yet extend into the anlagen of the sympathetic ganglia, the cells of which have become more numerous and more closely aggregated. The distribution of the medullary cells along the paths of the nerve fibers remains about the same as in the preceding stage.

In embryos 12 and 13 mm. in length the embryonic nervous system has assumed more definite form. In transverse sections of 12 mm. embryos (Fig. 2) the fibers of the visceral rami are seen to extend into the anlagen of the sympathetic ganglia. These are still loosely aggregated cell columns but begin to show evidence of their future segmental character. As in the preceding stages numerous elongated cells are found among the growing nerve fibers and at the surface of the bimdles, and pyriform cells may be observed all along the path from the neural tube into the anlagen of the sympathetic ganglia.

The destiny of cells which migrate out from the neural tube and spinal ganglia has already occupied the attention of not a few investigators. Harrison ('01) suggested the possibility that certain medullary cells which he observed migrating into the ventral nerveroots of embryos of the salmon may wander farther peripherally, i. e., into the sympathetic ganglia, and there give rise to sympathetic motor neurones. Bardeen ('03) suggests that the cells which wander out from the spinal ganglia and cord along with the bundles of axiscylinder processes may take some part in the formation of the neurilemma. He, however, believes with Vignal and Qurwitsch that in mammals the neurilemma is derived largely from mesenchyme. KoUiker ('05), though formerly of the opinion that the neurilemma is of mesoblastic origin, came to the conclusion in his later researches that the elongated cells which wander out from the spinal ganglia give rise to the neurilemma of the sensory fibers and that the neurilemma is everywhere of epiblastic origin. Dohrn and Neal have expressed the opinion that the cells which compose the neurilemma of the motor fibers have their origin in the neural tube and brain. Carpenter ('06) has shown that in embryos of the chick cells of an indifferent character migrate out from the ventral wall of the mid

D Histogenesis of the Nervous System 463

brain along the oculomotor nerve and become transformed into nerve cells of the ciliary ganglion. Carpenter and Main ('07) "feel sure" that some of the medullary cells which escape from the neural tube become cells of the neurilemma and there subserve a supporting f unc

Fio 2.*— Transverse section of neural tube and sympathetic anlage of 12 mm. embryo of the pig x 165. an., sympathetic anlage; da., dorsal aorta; dnr., dorsal nerve root; gc, germ ceUs of His; nb., neuroblasts; spg., spinal ganglion; spn., spinal nerve; vnr., ventral nerve-root; vr., visceral ramus.

tion similar to that of the neuroglia cells in the central nervous system.

Thus far only cells of an indifferent character have been considered. There are, however, as ?hown above, two distinct types to be

D 464 Albert Kuntz.

recognized among the cells which migrate from the neural tube into the spinal nerve-roots. In the light of Schaper's researches we must conclude that the pyriform cells which migrate peripherally from the neural tube have already undergone differentiation and must develop into neurones. These cells having been traced along the spinal nerves and visceral rami into the anlagen of the sympathetic ganglia obviously develop into sympathetic neurones. Thus there is established a direct genetic relation between the sympathetic and the central nervous systems.

It is not the writer's purpose in this paper to discuss the fate of the elongated medullary cells found among the fibers of the peripheral nerves. Inasmuch, however, as frequent reference has been made to them they may not be passed by without brief consideration. Occasionally one of these cells is seen undergoing mitotic division. My observations do not preclude the possibility that transformations may take place along the paths of the peripheral nerves similar to those which, according to Schaper, take place inside the neural tube. However, proliferation of these cells outside the neural tube is probably not sufficient to be of any considerable importance. A large majority of the elongated cells which wander out from the neural tube and sensory ganglia probably enter into the formation of the neurilemma.


1. Medullary cells migrate from the neural tube into the dorsal and ventral nerve-roots of embryos of the pig.

2. These migrating cells are of two general types: (a) elongated cells which are to be regarded as the indifferent cells of Schaper; (b) pyrifom\ cells which are to be regarded as the neuroblasts of Schaper.

3. These migrating cells seem to have. their origin in more or less definite regions in the neural tube.

4. Both, the indifferent cells and the neuroblasts wander peripherally and may be traced along the spinal nerves and visceral rami into the anlagen of the sympathetic ganglia.

5. The neuroblasts which migrate from the neural tube into the anlagen of the sympathetic ganglia develop into sympathetic neurones.

D Histogenesis of the Nervous System. 465

Thus, there is established a direct genetic relation between the sympathetic and the central nervous systems.

6. A large majority of the elongated cells which wander out from the neural tube and sensory ganglia probably enter into the formation of the neurilemma.

Received for publication, May 15, 1909.


Basdeen, C. R., *03. The Growth and Histogenesis of the Cerebro-splnal Nerves in Mammals. Amer. Jour. Anat, Vol. II, No. 2, p. 231.

Cabpenteb, F. W., *06. The Development of the Oculomoter Nerve, the Ciliary Ganglion, and the Adducent Nerve in the Chick. Bull. Mus. Comp. Zool. Harvard College, Vol. 48, No. 2, p. 141.

Cabpenteb, F. W., and Main, R. C, W. The Migration of Medullary Cells into the Ventral Nerve-roots of Pig Embryos. Anatomischer Anzeiger, Vol. 31, No. 11/12, p. 303.

QuBWiTSCH, A., *00. Die Histogenese der Schwann'schen Scheide. Arch. f. Anat. u. Physiol., Anat. Abt, Jahrg. 1900, Heft 1-2, p. 85.

Habbison, R. G., '01. Ueber die Histogenese des peripheren Nervensystems bei Salmo salar. Archiv. f. mlkr. Anat., Bd. 57, p. 354.

KoLLiKEB, A., *05. Die Entwickelung der Elemente des Nervensj-stems. Zeitschrift flir wissenschaftliche Zoologie, Sonderabdruck, Vol. 82.

ScHAPEB, A., *97. Die frtihsten Differenzierung8vorg3,nge im Centralnervensystem. Arch. f. Entwick.-Mech., Bd. 5, Heft 1, p. 81. Abstract in Science, N. S., Vol. 5, No. 115, p. 430.


Text-Book of Embryology. By Frederick Randolph Bailey and Adam Marion Miller. New York : William Wood & Co., 1909 ; pp. 672; 516 illustrations.

Like other recent text-books of embryology, this one was written especially for the medical students with emphasis "upon those features which bear directly upon other branches of medicine." The authors of this book have, however, undertaken to "broaden its scope and make it of greater value to the student of embryology and allied sciences."

The volume is a good piece of book-making, as it is done in America, being printed on heavy clay paper well adapted to halftones, in large, clear type. The illustrations are unusually clear and the references are usually printed in full instead of being represented by abbreviations explained by a legend below.

The authors have conscientiously described the development of every organ of the body with no important exceptions and have added a final chapter on teratogenesis. Each chapter is followed by practical suggestions giving the technique for the study of the subjects described in the chapter, and by a list of the more important literature for further study. To avoid repetition, there is an appendix dealing with general technique. In the chapters on organogenesis there is included a brief account of the more important abnormalities of development of the systems described. A very complete index of 33 pages concludes the book.

These are the bare statistics of this ambitious undertaking; one would wish to be able to say of such a book, at the least, that it is reliable and may be safely trusted as a guide to the subjects of which it treats; but it is necessary to record, with regret be it said, that it is not a safe guide even in many matters of fact, and that many of the generalizations are hasty and uncritical, or even absolutely


D Book Reviews. 467

wrong. The general excellence of the illustrations, some 90 per cent of which are borrowed, weighs but lightly in the balance.

The chapter on the nervous system, which is the work of Dr. 0. S. Strong, is written with critical judgment aAd shows evidence of good scholarship on every page ; the only criticism to be passed on it in the opinion of the writer is that it is rather too abstract to be readily intelligible to the class of students for whom it is intended ; but they cannot fail to profit by its study, and to the more advanced student it is delightful reading. If the other chapters had been up to this standard the book would deserve nothing but praise.

But the other chapters are characterized as a rule more by industry than by insight. This is especially shown in the first part of the work, and, indeeed, in every place where knowledge of the general principles of biology and of comparative anatomy and embryology is required. When the authors reach the ground of anatomy, in the narrower sense, they move more surely.

Some examples of the above criticisms should be given. In Chapter II it is said of the frog's egg that "the dark side indicates an excesg of deutoplasm" ; this is evidently a slip, the exact reverse being intended. The authors then go on to say that "inasmuch as deutoplasm is heavier than cytoplasm an egg with polar differentiation, if left free to revolve as in water, will assume a definite position with the protoplasmic or animal pole above and the deutoplasmic or vegetative pole below." If the authors had studied the pelagic teleost egg, where the reverse is the case, they would not have made such a sweeping statement ; their rule is honored in the breach as well as in the observance. The "membrana undulatoria or wavy membrane of Birds," p. 15 (referring to the spermatozoa), is a discovery of the authors ; but perhaps Amphibia were meant.

The discussion of maturation in Chapter III is far from being illuminating or even clear. It is hard to see why if "the maturation of the male sex cells in the vast majority of forms is much more difficult of demonstration than the maturation of the female sex cells" the number of studies on the former subject should exceed by so many those on the latter. Nor can one understand why "it is necessary to consider all the generations of cells from the mature

D 468 The Anatomical Record.

spermatozoa back to the spermatogonia" in studying the maturation of the male cells^ if the same is not the case in the female sex cells. This statement is repeated twice as a reason for the assumed difficulty of studying the maturation of the male sexual cells. The classification of types of reduction into those with tetrad formation and those without, which differ only in the fact that "each chromatin mass does not show a differentiation into four pieces" in the latter case, but does in the former, might be characterized as naive.

In the chapter on fertilization, the occurrence of normal polyspermy "in some insects" is noticed, but its typical occurrence in some vertebrates (Selachia, Beptilia, Aves) with which the book deals, is not mentioned. The exploded theory that the so-called fertilization membrane is a protection against polyspermy is upheld.

The chapter on cleavage is inexcusably careless in many respects, but it is in the part dealing with some general features of cleavage that the best gems of thought are found: we are told that "in a spherical holoblastic egg the division plane may be in any direction, but must bisect the mitotic figure at right angles to the long axis of the spindle"; the fact of polarity, and the relation of the first cleavage plane to it, is ignored ; and the first part of the above statement is absolutely wrong. The distinction between radial and spiral cleavage appears always in the second cleavage, and may appear in the telophase of the first, a fact that the authors do not appear to know. Moreover, the third division in spiral cleavage is inclined usually to the right, not to the left as the authors state. The term "morula" is no longer used by students of cleavage, except perhaps as applied to the cleavage of the mamnialian ovum, and the statement that "after the morula has become fully formed there appears in it a cleft or cavity due to separation of its cells" is wrong or meaningless, as the reader chooses. "In eggs in which the cells resulting from segmentation show greater inequality in size (due to difference in yolk-content) as in the frog, the segmentation cavity is surrounded by several layers of cells" — ^this is true of the frog, but the generalization is absurd (cf. Clepsine).

The subject of the germ-layers (Ch. VI) is taught by comparison, homologizing the layers and the modes of their origin from Araphi

D Book Keviews. 4G9

oxus to man. This is one of the most difficult subjects in vertebrate embryology; it involves more divergent views than any other. It it perhaps unfortunate to have placed before the student an account which a trained embryologist cannot follow, but the student will at least be spared the pain of noting the frequent errors, e. g., the description of the origin of the primitive streak of birds from a crescent which is said to be marginal, but was not so described by, its discoverer, and is not so figured by the authors; recent exact accounts of the primitive streak are ignored, and Duval, Hertwig and Bonnet are quoted, the two former at least incorrectly; the reptiles are denied a primitive streak (p. 67). The archenteric invagination of reptiles is wrongly stated to be marginal (p. 67) though figured differently. The statements about the origin of the mesoderm in reptiles and birds is extremely confusing, leading to the conclusion that the mesoderm cells may be considered as derivatives of the protentoderm, whereas the figures show them correctly as derived from the ectoderm.

The statement that "The amniotic folds from the beginning involve the splanchnopleure" (p. 100) is another oversight; of course somatopleure is meant. In the chick the head-fold of the amnion does not begin on the first day as stated, but on the second.

P. 108. "The allantoic sac in most mammals is a very rudimentary structure"; even if the ungulates, in w^hich the allantoic sac is very large, had been excluded, this would be a very unsafe generalization.

P. 108. "In all cases where the embryo is retained in the uterus it (the chorion) forms a most highly specialized and complex structure which in connection with the allantoic vessels establishes the communication between the mother and the embryo"; on p. 112 we are told that monotremes and marsupials are dependent for their food upon the yolk stored up within the egg and that "in these two orders the foetal membranes present essentially the same condition as in Birds and Eeptiles." If we put these two statements together are w^e to infer that in the marsupials the embryo is not retained in the uterus ? Is the student also to infer that marsupials in general have large yolk-bearing ova? Is the chorion of marsupials most highly

D 470 The Anatomical Record.

specialized and complex, or does it present essentially the same condition as in birds and reptiles ? It is hard to say which is worse here, the confusion or the errors.

To give further citations would be tedious, and perhaps unnecessary to prove that the book requires a most thorough revision before it can be recommended as a text-book of vertebrate embryology. Apart from the above reasons for criticism, the writer cannot but feel that the method of introducing students to the study of embryology which consists in abstracting the salient points from the field of comparative embryology is pedagogically wrong, although it is the method of the great majority of text-books. If it is necessary to consider the problems of comparative embryology in an elementary text, and it may be admitted that it is at least desirable, the development of some single form should run through the book as the main stream of discussion, and it should be given in sufficient detail to stimulate the critical insight of the student; the comparative statements may be made as side branches contributing to, and expanding the main topic, and leading up to the generalizations.

F. R. Ullie.

Edingeb's Intboduction to the Study of the Nebvous System. (Edinger, L., Einfiihrung in die Lehre vom Bau und den Verrichtimgen des Nervensystems.) Leipzig, F. C. W. Vogel, 1909.

The seventh edition of Edinger's Lectures on the Central Nervous System has quite outgrown the original purpose for which the lectures were prepared, viz., an introduction to the internal organization of the central nervous system adapted for medical students and practitioners. The work has become, in fact, a treatise on the comparative anatomy and phylogeny of the vertebrate nervous system not well adapted for the use of beginning students of the subject. Accordingly, the author has just published a smaller and much more elementary student's manual.

The text of this little volume of 190 pages is largely rewritten; it is not a mere condensation of the larger work. The treatment of the subject is in plan similar to that of the earlier editions of the

D Book Keviews. 471

Vorlesungen, but with better arrangement of the matter and excellent summaries of the physiological and pathological significance of the organs described. There are a few new figures, but most of the 165 illustrations are taken without change from various editions of the Vorhsungen. The book has some of the elements of weakness of the seventh edition of the larger work, which has recently been reviewed at length in these columns (Anatomical Recobd^ Vol. II, pp. 273-283) ; but it is nevertheless a very successful elementary text-book. The graphic literary style, the wealth of pictorial illustration and the use of comparative, physiological and pathological data as aids to the exposition throughout, combine to make it a very useful manual.

The reviewer believes that a more thorough-going application of the functional analysis which in chapter 3 is applied to the spinal cord and nerves would greatly simplify and clarify the exposition of the medulla oblongata and its nerv^es for the elementary student. Dr. Edinger has elsewhere used this principle and it is unfortunate that he has not taken advantage of it here in his elementary textbook, where its usefulness as a pedagogical aid is greatest.

The book is sure to prove very valuable as a reference w^ork in medical courses in neurology and it should have a wide circulation in this country as well as in Germany.

C. Judson Herrich.


At the College of Physicians and Surgeons, New York, the subdepartment of Histology and Embryology, hitherto administered under the direction of the chair of Pathology, has been merged in the Department of Anatomy. This change was determined upon in the interests of a closer correlation of the work in these two branches of. morphology, and in order to avoid the disastrous dissociation in the student's mind of the facts of gross and minute anatomy. A beginner is always more alive to the manner than to the matter of a study and is prone to accept an administrative form for an actual province of science; nor does he inevitably recover from this false impression ingrained at the outset of his career. In view of the increasing f requentation of our medical schools by men preparing themselves to become investigators and laboratory workers along special lines, their early orientation upon a broad basis of morphological study becomes of pressing importance. The recognition, therefore, of the essential unity of the study of structure, it is felt, ought not to be jeopardized for the sake of administrative convenience in the handling of material and instruments. In morphology it is as •important as it is difficult to ^^see life steadily and see it whole." For the beginner the acquirement of an extensive basis is as important as is an intensive method for the investigator. The direction of the work is at least as significant as its momentum. Accordingly the course is planned to take advantage of the large collection of the Department of Anatomy to present the details of the finer structures hand in hand with demonstrations of gross anatomy; especially will use be made of the comparative series to illustrate by the adult conditions of lower forms the embryonic stages of the higher. It is. we believe, in the ontophyletic relations revealed in the comparison of evolution and recapitulation that the larger problems of morphology are conceived, and their educational value developed. For this purpose lectures will be given illustrated by means of an epidiascope in addition to the usual apparatus of dissections, models and charts.

The course begins with cytology. The general properties and


D Note. 473

structure of living matter are considered from the standpoint of the protozoa. The structure of the cell is described, its processes of nutrition, movement, multiplication and necrobiosis, and combination with other cells to form individuals of a higher order. The study of the ovum affords a natural transition to the early stages of development of the embryo and the specialization of its cells to form simple tissue — ^histogenesis, followed by organogeny, the development of organs and apparatus which in turn is succeeded by the structure of the adult organs. Here it is essential that brief demonstrations of the gross anatomy of each organ should precede the microscopic study of its details. The course will conclude with the study of the central nervous system with especial emphasis laid upon its histo-architectonics interpreted upon the basis of the component theory.

Ten hours per week for thirty-two weeks have been allotted for this work. This time has been distributed as follows : lectures, three hours ; laboratory work, six hours and conferences one hour weekly. Under the direction of Professor Huntington these courses will be given by H. von W. Schulte, M.D., Adjunct Professor of Anatomy, with A. H. Miller, M.A., and O. H. Strong, Ph.D., Instructors in Anatomy, and C. H. Smith, M.D., Assistant in Anatomy.

Received for publication, July 10, 1909.



Vol. III. SEPTEMBER, 1909. No. 9



LYDIA M. DeWITT, Instriiictor in Histology, University of Michigan,

With Three Figures.

The literature of the so-called atrio-ventricular bundle of His has been so recently and so exhaustively reviewed by Tawara, Monckeberg and others that it seems unnecessary at this time to do more than give a very brief survey of the history of the so-called bundle, reviewing somewhat more in detail the work of a few writers who have investigated the aspects of the subject with which my own work

has especially dealt.

In 1845, as Is well known, Purkinje first called attention to the presence in the subendocardial layer of the sheep's heart of a network of gray, flat, mucoid fibers, probably muscular in nature, partly on the papillary muscle, partly on other fibrous bundles and partly bridging folds and clefts in the heart wall. These fibers, which have always been known as Purkinje's fibers, were described by numerous writers in the half century following Purkinje's obseryations and they were noted, not only in the sheep heart, but also in that of pig, calf, goat, horse, dog, cat, rat, mouse, goose, hen, dove, and even in the human heart in the newborn infant by Henle and as late as 15 years old by Gegenbaur. They were usually found subendocardial, but were described also in the myocardium and by Hoffman in the pericardium.

In 1893, His, Jr., in mouse, dog and man discovered a bundle of muscle fibers arising in the posterior wall of the right auricle near the auricular septum in the auriculo-ventricular furrow, running along the upper edge of the ventricular septum musculature forward and dividing into a right and left branch which extend down into the ventricular septum and soon end by fusing with the ventricular muscle, thus connecting the auricular and ven (475)

D 476 Lydia M. DeWitt.

tricular musculature. In 1904, tbis observation was confirmed by Retzer, who saw the bundle in cat, rabbit and rat, as well as in dog and man. He found the course varying slightly, but always connected with auricular muscle and merging into the ventricular at a short but variable distance from the point of division of the bundle. He estimates that in the human adult the bundle Is about 18 mm. long, 2.5 mm. wide and 1.5 nmi. thick. In 1906, Tawara made the most Important contribution to our knowledge of this connecting muscular link between auricle and ventricle when he showed not only the presence of the bundle in every one of a large number of species of mammalia investigated, these including human embryo, child and adult, but also showed that the bundle ended, not as previously thought, by fusing immediately with the ventricular muscle, but by branching and spreading out Into a complicated system of terminal fibers throughout the entire lower portion of the ventricles, these being the well known Purklnje fibers ; that the bundle recognized by His and Retzer and the Purklnje fibers were but parts of a great complicated system of muscular fibers connecting auricle and ventricle. His results were confirmed in all Important points by Keith and Flack In 1906, by Retzer, Fahr and M25nckeberg in 1908. Retzer showed that the system developed In the pig from the sinus which originates In the right and left venous valves and grows down in crescent-shaped lamellse into the lumen of the right atrium. The left venous valve attaches itself to the atrial septum, the right divides Into the Eustachian and Thebesian valves. The sinus fibers then grow down through the septum intermedium to the right and left sides of the Interven. tricular septum where they become the highly differentiated structures — the Purklnje fibers. He suggests the term slno-ventrlcular conducting system, which is divided into the sino-ventrlcular bundle and the Purklnje fibers. In 1907, Fahr published a communication in which he confirmed Tawara's findings In the sheep heart but not In the human heart, in which he stated that the right and left limbs ended without branching, both in the embryo and in the adult, thus confirming the earlier findings of His and Retzer. However, in 1908, he modified his views, bringing them more in accord with the findings of Tawara. In this last work he reconstructed a portion of the interventricular septum with the bundle fibers of a three-year-old child. His model shows the left limb branching and sending twigs toward papillary muscles and growing down toward apex of heart and losing -Itself in the trabecular network. The right limb, however, he still represents as unbranched. In a twelve-year-old child however he finds slight but not extensive branching of the right limb also, but explains the difference between his findings and those of Tawara as due to a possible individual variation. Fahr was unable to recognize the bundle earlier than in a 16 cm. embryo and could see no branching of the bundle until later than this. Mon(^eberg however (1908) traced the bundle and its two main limbs clearly In a 7.5 cm. human embryo and in a 16.5 cm. human embryo notes histologic differences between the bundle fibers and those of heart muscle and also notes the division of the left limb into several branches which he could trace through the trabeculie to the papillary muscles. The right limb, however, showed no

D Sino- Ventricular System. 477

branching in this series. According to MOnckeberg's investigations, this conducting system is found constantly in the human embryo from 7.5 cm. on. It varies slightly in its course in the atrium and with respect to its relation to the left and right surfaces as it passes through the annulus fibrosus and later in its terminal branching. The development, therefore, proceeds from sinus region downward into ventricle and later into the complicated system of Purkinje fibers.

An especial interest has attached to the study of this sino-ventricular system (1) because of the importance assigned to it by physiological experimentation and (2) because of its relation to certain well known but little understood pathologic and clinical conditions. The heart physiologists were divided into two classes, the neurogenists and the myogenists, the former believing that the contraction and rhythm of the heart were due to nerve Influence, while the latter ascribed them entirely to muscular action. The neurogenists based their theory (1) on the well known and generally accepted fact that nerve cells are the natural originators of all impulses and nerve fibers the natural conductors of such impulses; (2) on the generally accepted absence of all muscular connection between the auricles and ventricles, at least in the higher mammalia, by means of which impulses could be conducted. The myogenists, on the other hand, rested their case on the early contractility of the heart, before the development of nerve elements and on the contractility of the heart in certain invertebrate forms in which a nerve mechanism in the heart seemed to be entirely lacking. The absence of a muscular connection between auricles and ventricles was an important missing link, however, in their chain of reasoning, by which they attempted to show that the impulses began In the sinus region and from there were carried to all parts of the heart in regular sequence and rhythm, the entire mechanism bein^ muscular. This missing link was supplied by the discovery of this small connecting bundle of muscle fibers and it was at once seized upon by the myogenists to complete the chain of evidence in support of their theory. Humblet and Bering, by cutting the bundle, showed a dissociation of the auricular and ventricular rhythms, while EIrlanger, by clamping the bundle, was able to cause either a partial or complete heart block, gradual or sudden, and also to bring the heart back to normal by loosening the clamp. His brilliant results seemed to establish beyond cavil the function of the bundle as conductive and as co-ordinating the auricular and ventricular rhythm and the muscular nature of its action seemed proved by Retzer's statement that no nerve fibers were included in the clamp and also by the declaration of His that no nerves run in the bundle. However, the presence of nerves in the bundle has been demonstrated conclusively by Tawara and others and later admitted by Retzer.

Clinicians and pathologists had long been interested In a disease described by Stokes and Adams and known as Stokes-Adams disease, in which the cardiac manifestations were similar or identical to those caused by experimental cutting or clamping of the bundle. It was therefore assumed that a pathologic affection of the bundle was responsible for this clinical affection

D 478 Lydia M. DeWitt.

and numerous pathologic hearts have been subjected to investigation with reference to the condition of this connecting bundle. Among the most noteworthy contributions to the literature along this line are those of Tawara, Aschoff, Fahr and Monckeberg.

Soon after the publication of Tawara's monograph, at the suggestion of Dr. Huber, I undertook an investigation of the sino-ventricular connecting system, especially with reference to verifying by reconstructions Tawara's findings regarding the extent and complexity of the system and its connection with the Purkinje system of fibers. Later publications have perhaps diminished the necessity for such verification,' but the growing importance of the subject and the growing interest in its embryology, morphology, physiology and pathology will justify the presentation of my results. My investigations have been carried on by careful gross dissections of the system in sheep, calf and man, by wax reconstructions in man, calf and lamb and by histologic study in man, dog, sheep, calf and cat after diflFerent methods of fixation anl staining, including the structure of the fibers themselves and of the connective and elastic tissue surrounding them, and the nerve distribution and supply.

My first model, shown in Fig. 1, represents the upper portion of the bundle in the adult human heart reconstructed by the Bom plate method as modified by Mrs. Gage, paraffined absorbent paper of the required thickness being used instead of wax plates. The model is 25 times the size of the bundle and is reduced in the figure to about one-tenth the size of the model. The posterior extremity of the main trunk of the model represents the anterior portion of the Knoten, which will be described later, the remainder of the Knoten and its posterior extensions, where the outlines were less clear because of fusions with the auricular muscle, not being represented in the model. The actual length of the undivided trunk of the bundle, so far as shown in the model was 11 mm., the diameter varying considerably, being 1.6 mm., 2.8 mm. and 4.4 mm. in different parts. The division into two limbs taken place, as is usual, near the point of union of the posterior and median aortic valve flaps and at the upper extremity of the ventricular muscular septum, so that the right and left limbs begin as subendocardial strands, separated by the

D Sino- Ventricular System. 479

narrow cone of ventricular muscle which is usually found at this place in the human heart The left limb is broad and thin, broadening more and more as it passes downward following the curve of the septal wall and always subendocardial. It divides at a distance of 21.2 mm. below the point of bifurcation of the main bundle into k —

Fio. 1. Wax plate model of human si no- ventricular bundle 25 times magnified, reduced in figure to one-tenth size of model, k is Knoten ; rt.l., the slender right limb; 1.1., the broad left limb with first branching, into ant., post., and median branches.

three branches, the anterior and posterior branches being broad, while the mesial branch is much slenderer. These all continue subendocardial to the end of my series of sections and hence to the end of the model. The right limb is much smaller than the left and almost cylindrical, varying in diameter from .8 mm. to 1.6 mm. It is at first subendocardial, but, as the septum thickens, this limb becomes surrounded by septum musculature and passes at first downward and forward and but little outward deep in the septum musculature for a distance of about 25 mm. It then turns sharply outward and its course is then downward and outward and but little forward for a distance of about 10 mm., when it reaches the endocardium. Here my series ends, near the point where the moderator band, or as Eetzer more appropriately terms it, the trabecula supraventricularis, leaves the septum and carries the right limb of the bundle with a large mass of septum musculature to the anterior right papilla. The model represented in Fig. 2 represents the upper portion of the bundle in the lamb's heart 60 times magnified, but it is reduced in the figure to about one-twelfth the size of the model. It was made by the Bom wax plate method and begins in the Knoten- Posterior to this point, the connection with the auricular muscle was so close and the outlines of the bundle so poorly marked on the auricular side that it was difficult to outline it with sufficient accuracy for reconstruction. The Knoten was from 1 to 1.5 mm. in diameter and 4 mm. long. At the neck where it passed through the annulus fibro-cartilagineus, it changed rather abruptly to a diameter of from .3 to .4 mm., enlarging again somewhat in the ventricular portion of the bundle. The narrowed neck portion was about 1 mm. in length and the bundle then extended downward and forward in the muscular interventricular septum about 4.6 mm., before it divided into the right and left limbs. The left limb then extended directly downward and outward to the endocardium on the left side of the septum, a distance of 3.4 mm., where it divided into two branches, an anterior and a posterior, the latter showing a small median branch near its beginning. These remain subendocardial to the end of the model. The unbranched subendocardial portion of the left limb was from 1.4 to 1.8 mm. in length and from .1 to .2 mm. in thickness. The right limb passes downward and forward and outward, a distance of 5 mm., until it reaches the endocardium where the series of sections ended. The right limb measures from .8 to 1.2 mm. in width and from .4 to .6 ram. in thickness, so that it approaches more nearly to the cylindrical form. The further course of the right and left limbs and their branches will be discussed further a little later in this paper. These two models show the course of the main trunk of the bundle, the Knoten and the upper portion of the right and left limbs to be very much as had been represented by His, Eetzer and Tawara. The block of human heart tissue serially sectioned for the model was something over 35 mm. in length and I realized at this time that a reconstruction of the sino-ventricular bundle with its terminal expansions would require serial sections of the entire ventricle with a portion of the auricle ; that these sections would need

Fio. 2. Wax plate model of lamb's sino-ventrlcular bundle, k is the Knoten with main bundle. rt.l. is the upper portion of right limb, and 1.1. the left limb with upper portion of branches.

to be relatively thin and to be magnified from 100 to 400 times in order to bring out the fine terminal ramifications of the bundle with any accuracy. Unwillingly recognizing the practical and economic difficulty and even impossibility of such a task in the larger hearts, I attempted to dissect out the bundle with its terminal ramifications and found that by the constant aid of a hand lens and the frequent aid of the microscope, the sino-ventricular system could be followed, not only in its subendocardial course, but also for some distance into the muscle, while it was comparatively easy to dissect out the main

D 482 Lydia M. DeWitt.

limbs and to trace them back to the trunk and Knoten and to the beginning in the sinus region. Still desiring to make graphic as accurate a representation as possible of this entire connecting system. I conceived the idea of making a reconstruction from sudi a dissection. With the patient and skillful aid of Miss Gertrude Welton, one of my students^ the model represented in Fig. 3 was constructed in the following manner. Choosing a young beef heart in which the differentiation between heart muscle and that of the sino-ventricular system was well marked, the entire heart was fixed in Kaiserling's solution, the ventricles being first filled with the solution and the heart then suspended in the Kaiserling's solution, in order to prevent the distortion which frequently follows fixation in the ordinary way. The apex and outer wall of the left ventricle were then removed, leaving the anterior and posterior left papillse in situ. The right ventricle was then opened carefully from above dovni posteriorly, without cutting the trabecula supraventricularis or destroying any of the apical connections between septum and outer wall. Starting with the easily seen left limb, removing the superficial layer of the endocardium, the dissection of the main bundle, Biioten and upper portion of the main limbs was relatively easy. The dissection of the terminal ramifications was then carried on with needles, constantly using the hand lens and, when in doubt, the microscope, until the entire system was exposed, so far as it was possible to differentiate it from the cardiac muscle. Then lengths, diameters and angles were carefully measured at every point, the measurements multiplied by ten, a wire skeleton was constructed and covered with wax, all angles and measurements being retained, as nearly as possible in the same proportion. Hence the model represents a ten-fold magnified sino-ventricular system of the beef heart, as accurately as a hand lens and gross dissection can show it. The connective tissue sheath was not always entirely removed, since that would often mean the complete disintegration of the bundle; hence the threads and especially the nodes represent these portions of the system with the most closely adherent connective tissue. As seen in this model represented in Fig. 3, which should be viewed through a stereoscope, the main "•ndle begins in the region of the coronary sinus by the union

D D D Sino-Ventricular Svstoni. 48.'5

Fig. 3. Stereoscope pbotograpb of model of sino- ventricular system in calf's heart. To be viewed through stereoscope.

of two portions, the origin of one of which seems to be lost in the mass of fat, connective tissue, nerve fibers and ganglion cells, which are found in the region of the sinus, while the other can be traced upward for a short distance and is then lost or merged in the auricular muscle. Some connections with the auricular muscle at the sides and upper part of the Knoten are not represented. The actual length of the bundle to the point of division was 33 mm. The point of division lay at the upper margin of the interventricular septum as in the human and not deep in the septum musculature as in the sheep heart. The left limb was therefore subendocardial from the beginning, while the right limb lay buried imder about 1 mm. of muscle. After passing about 38 mm. almost vertically downward from the point of division, the left limb divides into two main branches, an anterior and a posterior. The anterior and posterior branches pass each into a fibro-muscular strand or false tendon, these crossing the ventricular cavity to the anterior and posterior papilla?. About 2 mm. beyond the point of division, the posterior branch gives ofF two long slender branches to the septal wall, which anastomose

D 484 Lydia M. DeWitt.

with branches from a septal branch of the anterior division to form a septal plexus. The posterior division proceeds downward and backward about 20 mm. when it gives off an internal branch, which soon forms a characteristic node, from which five branches are given off, three to the septal wall and two to the inner side of the posterior papilla. About 12 mm. further down, the main posterior branch reaches the anterior side of the posterior papilla, where it divides into two branches, each of which again subdivides, the resulting branchings frequently fusing with each other or with strands from other divisions into characteristic node-like forms, which again break up, thus forming a complicated network, surrounding and penetrating the papilla, sometimes on the surface and often deep in the muscle. One long slender branch runs around the inner side of .the papilla and could be traced nearly to the apex of the papilla, most of it buried deep in the papillary muscle. Occasional branches are given off from the papillary network to join the septal network, which will be more fully described later. The paiTillary branches could be traced as independent strands nearly to the apex of the papilla, where they were lost or became continuous with the papillary muscle. The anterior papillary division, about 3 mm. below the point of its formation, gives off a large septal branch, while the remainder passes downward and forward without further branching to the anterior papilla, where it divides into three main branches, one of which is slender and passes around the septal side of the papilla near its apex; the other two are of nearly equal size, the one external and the other internal. Both subdivide a number of times, the resulting twigs anastomosing with each other at intervals to form the characteristic node-like forms which were before described, in which the component fibers lose their identity in a mass of connective tissue, containing large numbers of the so-called Purkinje fibers. This then breaks up again into separate independent fibers, some of which again enter into a node and again divide, the whole forming a plexus-like arrangement over the entire posterior surface of the anterior papilla. In the heart from which this model was constructed, the outer wall of the left ventricle was removed for convenience of dissection, so that the network which we know to have been present on this wall and to have been derived from branches of the papillary divisions, is not shown in the model. The septal wall of the left ventricle both between the papillse and where it cun^es around behind the papillse is covered by a network of Purkinje fibers, similar to the network described on the papillse and formed of the anastomosing subdivisions of septal branches of the anterior and posterior divisions of the main limb, which have already been described. These form the node-like forms described, which, with their connecting strands make up the network. Occasional small branches pass back from the anterior and posterior papillary plexuses to the septum and join with the septal branches to form the septal plexus. The plexus described covers well the septal surface to the level of the branching of the main left limb and even sends some fine filaments higher on the septal surface. The right limb is narrower and more nearly cylindrical than the left and approaches the endocardium slowly. It runs at first downward and outward and a little forward and then downward and forward and a little outward. After an imbranched course of about 42 mm. in the septal wall, it enters the moderator band or trabecula supraventricularis, which in this heart was about 23 mm. long, making the total length of the unbranched right limb 65 mm. I was unable to find any branches given off from this limb before it reached the anterior papilla, which it entered on its anterior surface about 28 mm. below its apex. Here it bifurcated, one branch passing forward to the outer wall of the right ventricle, while the other passed backward on the papilla. Both of these branches break up into a large number of smaller branches, which form nodes and redivide and form a plexus similar in all respects to that described on the left side. Some of these branches could be traced upward on the exterior wall nearly to the base of the ventricle, some running quite deeply in the ventricular muscle, but most of them in the endocardium just under its superficial layer. One branch passed around the papilla on its septal side nearly to its apex, while another passed on its peripheral side, the two nearly encircling the papilla near its apex. From the plexus described in the outer wall of the right ventricle and the anterior papilla on that wall, three main branches pass back to the right surface of the

D 486 Lydia M. DeWitt.

septum and the broad flat posterior papilla which is closely attached to the septum and presents only a short blunt head projecting forward and upward from its septal attachment. These branches reach the posterior part of the septal wall, redivide, form nodes and again break up, so that the entire right surface of the septum to about 20 mm. from the base is covered with an intricate network \>t these terminal branches of the sino-ventricular system. While this model represents the ramifications and distribution of the system in only one heart, careful gross dissections of other calf hearts and also of human and sheep hearts have shown beyond question that the distribution shown in the model may be regarded as typical. The main bundle may be longer or shorter, may be a little more to the left or right of the position shown in model and in the sheep, as before stated, passes down some distance into the interventricular septum before dividing. In all cases investigated, however, the bundle divided into two limbs, the right passing unbranched to the anterior papilla, where it divided into either two or three branches, from which a network similar to that shown in the model spread out over the papilla, the outer wall and the septum, while the left limb divided into either two or three branches, the anterior and posterior passing to the anterior and posterior papillae and forming networks, while the median, if it exists, or if not, septal branches of the anterior and posterior divisions form a septal network. The general course was the same, but slight modifications were found, not only in the hearts of different species, but also in those of different individuals of the same species.

In- the complicated course described and modeled, branches of the system often cross, not only narrow clefts between columnie carnefie. but also the entire ventricular cavity by means of longer or shorter trabeculse, which resemble connective tissue and are known as pseudotendinous threads. Tawara and Monckeberg have made these threads the subject of exhaustive research, especially with reference to their abnormal course in the left ventricle of the human heart in various pathologic conditions. Tawara defines these abnormal tendon threads as "bundles of the atrio-ventricular system which have become loosened from the walls and pass freely through the ventricular cavity." Monckeberg, however, after study of numerous pathologic

D Sine- Ventricular System. 487

conditions, concludes that not all trabeculse crossing the ventricular cavity between the septum and the papillse carry branches of the atrio-ventricular bundle. He divides them into three main classes: A. Threads having no muscle at all, but consisting of connective tissue — true tendon threads. B. Threads carrying both myocardial fibers and fibers of the atrio-ventricular system. C. Threads carrying only fibers of the atrio-ventricular system with their connective tissue sheaths. In order to determine whether this classification in abnormal human hearts might apply also to normal animal hearts, I examined microscopically all threads, long and short, crossing the ventricular cavity in two beef hearts. Since the results seemed uniform, with very slight variations in the number and course of such threads, I will describe in detail the threads found in only one of these hearts.

A. From the upper portion of the septal wall in the neighborhood of the valve insertion, three or four fine threads passed to the papillfiB, being inserted in or near the apices of the papillae. These threads consisted of connective tissue only and contained no muscle fibers, so that they may be regarded as belonging to Monckeberg^s Class A.

B. In the left ventricle, the following threads were found :

1. Between middle or lower part of posterior left papilla and septum below valve level.

a. Large strand from about 24 mm. below septum membranaceum to about 8 mm. below apex of papilla, with two fine subbranches back to septum.

b. 6 mm. lower, a smaller strand with two subbranches, one to septum and one to lower trabecula.

c. Short slender thread 7 mm. lower, with one subbranch to septum and one connecting with thread No. 2.

2. Between middle and lower part of anterior left papilla and septum below valve level.

a. Large strand from 30 mm. below septum membranaceum to 15 mm. below apex of papilla. No subbranches.

b. 4 mm. lower, smaller thread with two subbranches to septum.

c. Small thread 4 mm. lower with one subbranch.

D 488 Lydia M. DeWitt

All of these threads were examined microscopically and all were found to contain connective tissue, blood vessels and nerves and larger or smaller fibers belonging to the sino-ventricular system. All except two very fine ones contained also some myocardial fibers.

3. On the right side, the main moderator band or trabecula supraventricularis was the only connection between the anterior papilla and the middle part of the septal wall and this had no branches. The posterior right papilla is a broad flat papilla closely attached to the septum and between it and the outer wall of the ventricle two main strands could be seen, each of which had several subbranches. All of these showed both myocardial fibers and fibers of the sino-ventricular system, as well as numerous nerves and blood vessels and connective tissue.

C. The two threads found on the left side, which contained no myocardial fibers, but only sino-ventricular fibers, connective tissue and nerves and blood vessels.

From the limited number of investigations I have thus far been able to make on these threads, therefore, I would conclude that in the heart of the calf at least, as Monckeberg has shown in the human heart, true tendon threads are found connecting the upper part of the septum with the apices of the papilte ; that below the branching of the main limbs of the bundle, all the pseudo-tendinous threads crossing the ventricular cavity carry branches of the sino-ventricular system.

I find also that the great majority of them may be regarded as belonging to his Class B, carrying both myocardial and sinoventricular fibers, while only one or two very fine, short ones belong to Class C, carrying no myocardial fibers. I also find that all the threads of Classes B and C carry, in addition to the muscle fibers, small blood vessels and bundles of nerve fibers, the latter running in the connective tissue around the sino-ventricular branch and often sending finer nerve bundles between its individual fibers. Eetzer regards the so-called false tendons in the pig as always bridges for the conductive system, this including both Purkinje fibers and nerve fibers.

D Sino- Ventricular System. 489

Histology of Sino-ventbigulab System. The earlier writers on the histology of the Purkinje fibers were divided into two great classes, (a) those who, with Purkinje, described two kinds of elements in the fibers, a network of cross-striped fibers resembling heart muscle, In the meshes of which were imbedded peculiar, clear mucoid cells, containing nuclei ; b, those who believed that the Purkinje fibers consisted of short broad cells, whose center was clear and mucoid in character and contained from one to several nucleoid bodies, while the periphery showed crossstriped fibrils similar to those In heart muscle, thus having only one kind of element. Tawara, however, in his monograph, has described in detail the histology of the entire sino-ventricular system in sheep, dog, and man and his description with few modifications has been followed by all later writers. In all forms described, he divides the atrioventricular system structurally into two main portions, the auricular and the ventricular, each of which is subdivided into two parts: the auricular into the trunk and the Knoten, the ventricular into the upper undivided part and the terminal expansions or Purkinje fibers, the differences in structure being most marked in the sheep and least marked in the human heart, the dog being intermediate in the differentiation of this system. The Knoten in the sheep is described as a complicated network of small irregular fibers, which change as the bundle passes through the annulus fibro-cartilagineus, into large clear fibers, with peripheral fibrillation and central nuclei. This type becomes more pronounced when the limbs reach the endocardium and spread out in the terminal ramifications, which are known as Purkinje fibers and which later fuse with the myocardial fibers. Tawara describes the fibers of the main limbs and terminal expansions in the sheep heart as consisting of broad, short, irregular, polymorphous cells, two or three or more of which side by side are united to form cell str&iiLds, which at intervals anastomose with other cell strands forming often node-like structures, the whole appearing as a complicated network in the meshes of which are found connective tissue, fat, blood vessels and nerves. The fibril bundles are mostly at periphery of the cells and run uninterruptedly from one cell to another in the cell strand. In the terminal expansions, the clear centers are more pronounced and the meshes of the network longer and narrower. In spite of the continuity of the fibril bundles, Tawara regards the strands as consisting of separate and independent cells connected by intercellular fibrils comparable to those in the epithelial cells of the epidermis. In dog and man, he finds the auricular portion of the system similar to that in the sheep, while the ventricular portion is less sharply differentiated by reason of the more extensive development of the fibrillar structures. In the new-born dog, the fibers of the bundle consist of short broad cells with clear nucleated centers and fibrillated borders, very similar to those in the sheep, while in the adult dog he finds "sarcoplasmic territories" with distinct limiting zones, so that in the dog also he regards the Purkinje fibers as consisting of cells with connecting fibrils, the fibrillar structures being more developed and more uniformly scattered throughout the cells than in the sheep. In the human heart, the Purkinje fibers are even less clearly differentiated from the myocar

D 490 Lydia M. DeWitt.

dium than in the dog's heart, but he regards them as cellular rather than syncj'tial because of the presence of cross lines which are arched or wave-like, rather than step-like as in the myocardium and completely divide the fiber, between two nucleated areas ; he also finds some of the fibril bundles stopping at the cross lines while others pass through. The careful and detailed description given by Tawara has been accepted by other authors with very little modification. Retzer denies the existence of a Knoten in the pig's heart, stating that he finds the network-like structure ascribed to it by Tawara throughout both the auricular and ventricular portions of the system. Monckeberg, in his excellent monograph on the atrio-ventricular system in the human heart, questions Tawara's view that the cross lines seen in the Purkinje fibers of the human heart represent cell outlines, although he admits the presence of separate independent short polymorphous cells with clear nucleated centers and narrow peripheral fibrillar zone in the sinoventricular system of a 16.5 cm. human embryo. He states that the development of fibrils results in the formation of solid star-like or fusiform structures with anastomosing processes forming a network, with vesicular cells forming the meshes. Later the solid portions unite into tube-like fibers with occasional connecting branches, this process being carried further in the limbs than in the Knoten, where the network structure is preserved. An interesting communication by Aschoff reporting the work of his student Nagayo, calls attention to the glycogen content of the sino-ventricular system. In the beef, calf and sheep heart, he finds glycogen constantly present and characteristically distributed, either diffusely scattered in the interflbrillar portion or as numerous large granules in the perinuclear sarcoplasmic area, except in the Knoten, where there is but little. In the goat also it is constantly present, but mostly diffuse. In the pig, corresponding with Retzer's finding of the extension of the Knoten structure into the ventricle, he finds very little glycogen and almost limited to the terminal expansions. In the human bundle, the results were variable; thirty hearts were investigated, of which three showed distinct glycogen content, nine very small amount of glycogen, and eighteen almost none. The dog's heart with reference to the glycogen content of the sino-ventricular system, resembles the human heart, if we regard the large amount of glycogen as the normal condition. M5nckeberg also investigated the glycogen content of the sino-ventricular system in the human heart at different ages and under different pathologic conditions and, like Aschoff, finds it variable, a fact which he ascribes to pathologic conditions or to poor condition of tissues. He concludes that myocardial fibers are always glycogen-free in post-uterine life, while in the Purkinje fibers it is constantly and characteristically present, though diminished in atrophic and cachectic conditions and that glycogen stains may be used in properly fixed hearts as a means of differentiation between Purkinje fibers and myocardial fibers. The main points of variance then are (1) as to the presence and histologic differentiation of such an auricular structure as is designated by Tawara as the Knoten; (2) as to whether the ventricular jwrtion of the system and especially the Purkinje fibers are syncytial or cellular in structure.

D Sino- Ventricular System. 491

In my investigation of the histology of the sino-ventricular system I have studied the hearts of sheep, both young and adult, of calf, of dog, including a 6 cm. embryo, a 3-day old puppy and an old dog, of cat and of man. I have used the ordinary hematoxylin and eosin and hematoxylin and Van Gieson stains, also the Schultze chrom-hematoxylin, the Heidenhain iron hematoxylin, the silver, the Mallory connective tissue stain, the Weigert elastic tissue stain, and also the methods recommended by Heidenhain for the study of heart muscle. In all these types the most constant and typical structure found was the network in the auricular portion of the systeita designated by Tawara as the Knoten. This in all cases consisted of an intricate network of fibers, which varied greatly in size, but the average size of which was much less than that of the auricular muscle or of the rest of the sino-ventricular system ; the nuclei are smaller and more numerous than in the auricular muscle. The anastomosis of the fibers takes place, not by simple fusion of two uniting branches, as is usual in heart muscle, but by the formation of nodes or star-like forms, into which two or three or more fibers become •merged and from which a variable number of fibers emerge. In the iU)de, the fibers completely lose their identity, fibrils from the different fibers commingling confusedly. Surrounding the Knoten and in the meshes of its network are found connective tissue, blood vessels and many nerve fiber bundles, especially in the lamb and calf heart, with numerous ganglion cells in the caLPs heart and a few were seen in the sheep's heart. In the sino-ventricular system of the sheep and calf, this small-fibered network changes rather abruptly, shortly before the bundle passes through the annulus fibro-cartilagineus into a large-fibred network, the change beginning in the central portion of the bundle. The ventricular portion of the sinoventricular system is markedly different in the different species examined. In the sheep and calf, where the fibers are most typical and most clearly differentiated from the myocardial fibers, the fibers are much larger than the myocardial fibers, with fewer fibrils and much more sarcoplasm. The fibrils are grouped into larger and smaller bundles, which run partly in the direction of the fiber and partly cross it in devious directions, forming a network which

D 492 Lydia M. DeWitt.

encloses certain clearer areas in which are found one or several nuclei. Tawara regards these as independent cells joined together to form cell strands, but even in the sheep, in which the apparent cell outlines are most distinct, the fiber appears to me as a syncytium, a continuous mass of sarcoplasm, through which run the bundles of cross-striped fibrils in different directions, dividing up the sarcoplasmic mass into clearer non-fibrillar areas, containing one or several nuclei and three to six of which make up the transverse diameter of most of the fibers. The fiber bundles themselves anastomose at intervals, forming a network in the meshes of which are found connective tissue having many more elastic fibers than are found between the myocardial fibers, blood vessels and nerves. These rarely penetrate the fiber strand and then only a short distance. In the calf, ganglion cells were found in these mesljes throughout the entire distribution of the system, but I have been unable to duplicate this finding in the hearts of other species examined. My reasons for believing that the fibers of the sino-ventricular system are syncytial in character are (1) the fibrils and fibril bundles pass uninterruptedly through the fiber in different directions, forming a more complicated fibrillar network throughout the fiber than could be accounted for by any other hypothesis than that the fiber is the unit of the system; (2) I have been unable to find white or elastic connective tissue fibers penetrating the fiber, or dividing it, except a few peripheral strands which are easily explained as following the irregular contour of the fiber strand. Also no blood vessels or nerves seem to penetrate the fiber bundle; (3) I have occasionally seen clefts such as were mentioned by Tawara, but believe them to be due to shrinkage of the sarcoplasm during fixation. The cell-like forms or sarcoplasmic territories are very variable in form and size and separated simply by strands of cross-striped fibrils. In the terminal expansions of the system, the fibers, if cut longitudinally, appearoblong and occasional cross bands are seen, not unlike those in the myocardium, but either straight or slightly curved or arched, rarely step-like as in the myocardium. In dog, cat and man, the fibers of the ventricular portion of the sino-ventricular system more nearly resemble the myocardial fibers, being differentiated from them by being

D Sino- Ventricular System. 493

I>roader and paler and less fibrillar and more vacuolated^ large^ clear nucleusrcontaining areas resembling those in the sheep being not infrequent The nuclei are smaller and less frequent than in the myocardium^ although occasionally two or more are seen in a single sarcoplasmic territory. The cross lines are distinct and often cross the entire fiber, but may occur at frequent intervals, the short space between them being non-nucleated, or they may be widely separated, several nucleated spaces occurring between two such bands; this finding was noted also by Monckeberg. The fibers are very variable in size and nodal points showing a commingling of fibrils from different fibers are frequent. The appearance varies somewhat with the direction of the section, a section tangential to the endocardial surface showing much broader fibers with larger perinuclear areas than one vertical to the surface, showing that the fibers are broader than they are deep, the broader dimension lying parallel to the endocardial surface. Connective tissue and especially elastic fibers are much more abundant than in the myocardium. In the 3 cm. dog embryo and also in the 3-day old puppy the fibers appeared as independent cells, sharply outlined by blue lines in the Mallory stain, an outlining which was never brought out in the adult condition. Hence my findings in the embryo and newborn agree with those of Tawara and Monckeberg as to the fact that Purkinje fibers are composed of single short clear cells, but in later life, at least in man, dog, and cat, and probably also in sheep and calf, the fiber bundle seems to me syncytial.

Transitions from the fibers of the sino-ventricular system to the myocardial fibers are rarely seen in my sections; this is explained by the fact that the transition is very gradual, so that only rarely would a section follow a fiber through the entire transition from a typical Purkinje fiber to a typical myocardial fiber. Beginnings and endings of the transition are frequently seen, and in one or two instances I have been able to trace the entire process. It is impossible to say from my sections, however, whether all the fibers of the sino-ventricular system end by fusion with the myocardial fibers or not.

D 494 Lydia M. DeWitt.

Nerves, Tawara states that he finds the sino-ventricular system of the calf accompanied throughout by numerous large nerve bundles, which are intimately associated with the muscle bundles and even in the ventricular portion of the system contain ganglion cells. He also finds small nerve bundles in the sheep heart, but no noteworthy nerve bundles in man, dog, or cat, though he admits that quite fine nerve bundles may accompany the system in all species of animals. MQnckeberg states that in the human sino-ventricular system he finds no nerve elements or forms resembling such, though he used many special methods for their demonstration. Wilson, after examining the atrioventricular bundle in calf, sheep and pig, states that he finds in it, in addition to the special form of muscle fiber, an important and intricate nerve pathway, consisting of: —

(1) Numerous ganglion cells, mono-polar, bi-polar and multi-polar, whose processes pass either to other ganglion cells in the bundle, to the muscle fibers of the bundle, or through the bundle so far as examined.

(2) Abundant nerve fibers running through the bundle in strands and either ending in ganglion cells in the bundle, or in the muscle plexus, or passing through the bundle, so far as examined.

(3) An intricate plexus of varicose fibrils around and in close relation to muscle fibers of the bundle.

(4) An abundant vascular supply with well-marked vaso-motor nerves and sensory endings.

Retzer, on account of the presence and close relation of nerve fibers to bundle fibers in pig*s heart, regards the system as a neuromuscular end-organ, a conclusion with which Wilson does not agree, since he finds the essential anatomical structure of a neuro-muscular spindle, — shape, lymph spaces, lamellar capsule, as well as nerve endings, — lacking in the sino-ventricular system, while ganglion cells are present in the system and absent in the neuromuBCuar spindles.

In the calf's heart, the nen-e bundles are so large and prominent, not only in the Knoten, but also throughout the distribution of the sino-ventricular system, that they are readily seen in preparations stained by ordinary methods, and it would be impossible in this heart to question the prominent part which nerves and ganglion cells play in the structure and probably in the function of the sino-ventricular system. They are not only in the connective tissue sheath, but smaller bundles run between the fibers and appear to stand in more or less intimate relation to the fibers of the system. In the hearts of young lambs, and to a less marked extent also in those of older sheep, the nerve fiber bundles also follow the entire distribution of the system, although few ganglion cells could be seen in it, after the bundle passed through the annulus fibro-cartilagineus. In the

D D Sino- Ventricular System. 495

hearts of dogs, cats and men, larger nerve bundles were more rarely seen, and then only in the connective tissue sheaths of the system, and very small fiber bundles or single fibers only are seen between the single fibers and it is usually difficult to distinguish any nerve fibers between the muscle fibers of the terminal expansions. I have studied the nerve distribution in lambs hearts with the intravitam methyleneblue method and found larger nerve bundles in the sheath, breaking up into smaller bundles and finally into a network of single fibers, which appeared to end on the muscle fibers of the bundle, although nerve terminations were not satisfactorily demonstrated. It would appear that the nerve supply is independent of that in the immediately surrounding myocardium, since in some of my preparations very few of the nerve fibers surrounding the myocardial fibers were stained, while the nerve fibers of the sino-ventricular system were well stained throughout its entire distribution in the ventricular as well as in the auricular portion. Ganglion cells also were found, so that my findings corroborate those of Wilson in every particular. The question of the nerve and blood vessel distribution, however, I desire to leave for further investigation, the results of which will be published with those of investigations on other phases of the subject in a later communication.

Eegarding the function of the sino-ventricular system, my investigations have had little to do. That it has a function independent of that of the myocardium seems assured from the constancy of its presence in all species studied, the relative constancy of its structure and distribution, the probable independence of its blood and nerve supply, the relatively constant and typical glycogen content^ as shown by Aschoff and Monckeberg, and its independent pathology, as shown by Tawara, Monckeberg, Fahr and others. That its function is conductive, or at least coordinative, seems probable, both from physiologic experimentation and from pathologic study. Whether it may also originate impulses and thus account for the independent ventricular rhythm established after severance of the connection between auricle and ventricle, and whether its function is muscular, nervous or neuro-muscular, must be left for further study and experimentation. Because of the large number of nerves and their intimate

D 496 Lydia M. DeWitt.

relation to the musdes, at least in some of the species studied, the neuro-muscular hypothesis seems to me the most probable and it may be possible that it is called into action by the distension of the ventricles and the stretching of the endocardium.

In conclusion I desire to thank Dr. Huber for suggesting the subject of this investigation and for the kindly interest he has shown in its progress ; also I wish to express my appreciation of the skill and patience with which Miss Gertrude Welton aided me in the construction of the model ; 1 am grateful also to Drs. Novy and Streeter for their aid in photographing the models.

Received for publication July G, 1909.


AscHOFF. Ueber den Glykogengehalt des Relzleltungssystems des SSlugetierherzens. Verhandl. d. D. path. Gesellsch., XII, 1908.

Eblangeb. Physiology of Heart Block In Mammals with Especial Reference . to the Causation of Stokes-Adams Disease. J. of Exp. Med., VIII, 1906.

Fahb. Zur Frage der atrioventrlkuiaren Muskelverblndung im Herzen. Verhandl. d. D. path. Gesellsch., XII, 1908.

Fahb. Ueber die Verblndung zwischen Vorhof und Ventrlkel (das Hlssche Btlndel) im normalen Herzen und beim Adams-Stokes Symptomkomplex. Arch. f. path. Anatomle, Bd. 188, 1907.

Geoenbatjb. Notlz iiber das Vorkommen der Purkinjeschen Fftden. Morphologisches Jahrbuch, Bd. Ill, 1877.

HEinjc. Handbuch der GefS^sslehre des Menschen. 1876, p. 63.

W. His, Jb. Die Th&tlgkeit des Embryonalen Herzens und seine Bedeutung fUr die Lehre der Herzbewegung beim Erwachsenen. Arbeiten aus d. med. Klinik zu Leipzig, 1893. Seen only in abstract. Also Herzmuskel und Herzgangllen. Wiener med. Blatter, 1894, p. 653.

Hoffman. Beitrag zur Kenntnlss der Purkinjeschen F^den im Herzmuskel. Zeitschr. f. wlssenshaftliche Zoologle, Bd. LXXI, 1902.

Keith and Flack. The Atrioventricular Bundle of the Human Heart The Lancet, 1906.

Keith and Flack. J. of Anat. and Physiol., Vol. XLI, 1907.

MoNOKEBEBG. Uutersuchungen fiber Atrioventrlkuiarbttndel im menschlichen Herzen. Fischer, Jena, 1908.

D Sino- Ventricular System. 497

M5NCKEBERG. Ueber die Bog. abnormen Sehnenfaden im linken Ventrikel des menschlichen Herzens und ihre Beziehungen zum Atrioyentrikularbiindel, Verhandl. d. D. path. Gesellsch., XII, 1906.

PuBKiNJE. Mikroskopische-neurologische Beobachtung. Arch. f. Anat, Physiol, und wlssenschaftUche Medizin, 1845.

Retzeb. Ueber die muskulose Verbindung zwlschen Vorhof und Ventrikel des Saugetlerherzens. Arch. f. Anat u. Pkyslol., Anat Abth., 1904.

Retzeb. The Atrio-ventrlcular Bundle and Purklnje Fibers. Anatomical

Record, I, 41. Also II, 144. Retzeb. Johns Hopkins Hosp. Bull., 1808, p. 208.

Retzeb. The Moderator Band and Its Relation to the Papillary Muscles with Observations on the Development and Structure of the Right Ventricle. Johns Hopkins Hosp. Bull., June, 1909.

Taw ABA. Das Reizleltungssystem des SSugetierherzens. Fischer, Jena, 1906.

Ta^waba. Ueber die sog. abnormen SehnenfUden des Herzens. Zlegler*s Beitrage, XXXIX, 1906.

Wilson, J. Gobdon. The Nerves of the Atrioventricular Bundle. The Anatomical Record, April, 1909.






HERBERT M. EVANS. From the Anatomical Laboratory of the Johns Hopkins University.

With Twenty-one Figures.

In explanation of the somewhat comprehensive title of this report, I would say that I have omitted entirely, for the present at least, the development of the heart, first arch and cephalic aortse. But, on the other hand, the injections which I shall describe speak with great clearness on the primitive form and development of all that portion of the aorta below the omphalo-mesenteric arteries, of the entire upper portions of the anterior cardinal veins, and quite fully, of the posterior cardinal and umbilical veins. In addition to these primary trunks, I have explored other branches of the vascular tree in embryos of different ages and in various portions of the body and find everywhere the same story.

The patient and thoroughgoing analysis of serial sections through vertebrate embryos has, in the last twenty years, given us a fairly accurate idea of the position, size, and relations of all the chief vascular trunks present in successive stages. But such knowledge, even though complete, can tell us little or nothing about the method of development of the vascular system. When a large trunk is described as having extended to a certain level in the body wall or on a viscus we have still no idea of how it reached its position. Many processes could be concerned.

Read at the Twenty-fourth Session of the Association of American Anatomists, December, 1908. (498)IC

Blood Vessels of Vertebrate Embryos. 499

If the vessel in question was derived from neighboring vessels, and in some instancea even this has been disputed, have we then to conceive of it as having grown out as siLch to its future territory ? Certainly such a conception dominated the descriptive anatomy of a century ago and even to-day has not altogether lost its influence. Arteries are still described as dividing dichotomously and otherwise in accordance with the unconscious simile afforded by plant growth. To such an idea many of the vascular anomalies proved an utter enigma. Double vessels in place of the usual single one were disposed of by considering one vessel as the normal one and its brother as the interloper or "aberrant," It is easy to understand the curiosity with which anastomoses between strong and usually separate arterial channels were looked upon. Why indeed should vessels which had grown out to their destination, send communicating branches to their neighbors ? Thus the surprise with which Hunter discovered the phenomenon of a collateral circulation. Such phenomena in the adult body are themselves splendid evidences for quite another conception of the development of the vascular system, a conception which was first partially expressed by the anatomist Aeby in his "Der Bau des menschlichen Korpers," in 1868. Aeby contributed a mere hypothesis, but it has only recently been recalled and can now be convincingly supported by actual studies on the early developing vessels.

Both arteries and veins, said Aeby, arise from netlike anlagen; the veins retain this the more, whence the numerous venous anastomoses; the arteries hardly at all, whence their greater rarity here. He conceived of the whole question as one of functional adaptation. Not all the members of the vascular net are retained ; the victors in the struggle are the trunks as we know them in the adult.

Such an illuminating conception explained adequately all vascular anomalies. From the preceding network, the persisting vessel could course in practically any direction and with practically any connections. Though Krauae, who wrote the admirable section on the variation of the vascular system in Henle's Anatomy,^ readily saw the advantages of Aeby's explanation and adopted it, the conception was not recalled in any form for many years.

Henle. J. "Handbuch der Gefasslehre des Menschen," Braunschweig, 1876.

D 500 Herbert M. Evans.

In the anatomy of the adult vascular system nothing approaching a netlike condition exists until we reach the capillary bed, where it forms a characteristic feature. Is it not indeed a capillary net, this primitive net postulated by Aeby ?

The answer to this was given by the research of K. Thoma^ in 1893. In his attempt to solve the question of ancestry of arteries and veins Thoma selected the chick's yolk vessels where he soon observed stages so early that only a plexus of capillaries existed. Here no channels of the net were conspicuously larger or smaller than their neighbors, but from among them, in later stages, the chief vitelline arteries and veins are formed. Thoma's idea of how this transformation is brought about, his conception of the determining influence of the velocity of the circulation, is of the greatest interest, though it has not been adequately tested, nor does it concern us here. On anatomical grounds alone, however, it should be possible to establish clearly or disprove the more general statement that arteries and veins exist originally in the form of a capillary mesh. If this is a fact of general value, it will apply to the developing vascular system within the body of the embryo quite as well as to the extraembryonic circulation which Thoma had studied. The work of three recent investigators has indicated that this is indeed the case. I refer to the papers of Miiller, Rabl, and Sterzi. The latter observer has studied the vascularization of the spinal cord and some of his descriptions are splendid evidence of the capillary plexus ancestry of these vessels, though Sterzi himself has not realized their wider significance in this light. As I shall show later, the vessels of the cord and brain illustrate particularly well the method of origin and growth of the vascular system. The work of Miiller and of Kabl has been referred to in a previous communication"* in which I also call-ed attention to Curt Elze's emphatic criticism of their contention. Elze,*^ in a research done chiefly in Hochstetter's laboratory, has

Thoma, R. "UDtersuchungen ttber die Hlstogenese und Histomechanik dee Gef&sssystems/' 1893.

^Eyans. "On an Instance of Two Subclavian Arteries to the Early Arm Bud of Man," Anat Record, II, 9, Dec., 1908.

'Elze. "Beschreibung eines menschlichen Embryo," Anat.> Hefte, Bd. 35, 1907.

D Blood Vessels of Vertebrate Embryos. . 501

seen fit to deny emphatically such an origin for the blood vessels. One were as justified, he says, in the absurdity of considering the aortic arches as arising from capillaries, and yet many of the injections in this series show exactly this origin for the arches.

Moreover, it has now been possible to show that the main vessels of the limbs, the subclavian and the sciatic arteries, are the single persisting channels of a distinct plexus of capillaries which springs directly from the lateral aortic wall opposite the earliest indication of the limb buds.®

Surely, however, the whole question can be put to the most convincing test possible if the primary vessels themselves, the aortse and the cardinal veins, can be shown to be formed in this way. I have accordingly set to work to decide this question through injections of very young embryos. The injection method had already shown itself to be by far the most efficient way to demonstrate the fonu and entire extent of the vascular system in any area or organ. The methods employed in the injection of these minute vessels have already been indicated in a previous publication and a portion of the injected specimens shown at the last two annual meetings of this association, at New York, 1906, and at Chicago, 1907.

Chick embryos were selected for the study, not only for the convenience in securing and controlling material but also because the presence of early and easily accessible vitelline vessels furnishes a good portal of entry for the injection.

Development of the Lowee Aoetae.

In embryos of the chick and duck possessing from twelve to fifteen sonlites, the aortae begin to be in free communication with the extraembryonic vitelline capillary net near the most caudal of the series of somites. Here also the character of the aortse becomes entirely changed. They are no longer the distinct and fairly straight tubes which they constitute in the upper portion of the body, but instead

"Evans. "On the Earliest Blood-vessels in the Anterior Limb Buds of Birds and their Relation to the Primary Subclavian Artery." Am. Jour. Anat., IX, 2. May, 1909.

D 502 Herbert M. Evans.

begin to be most irregular, connecting with the vitelline capillaries at as many points as would obtain in any capillary mesh and soon becoming resolved completely into the general extra-embryonic net from which they are entirely indistinguishablfe. In but a short distance, corresponding to the length of some five or six somites (were they present here), the vitelline capillaries no longer gain the median region of the embryo but surround the latter in a wide detour which always characteristically encircles the caudal extremity — ^the region of the primitive streak.

By the time the embryo possesses some twenty-four somites, the two aortflB extend entirely through it, fusing with the vitelline capillaries only when the caudal tip has been reached. In what manner were the lower aortse developed ?

A study of the intermediate stages in injected specimens makes it possible to give a very clear answer to this question.

By the stage of twenty somites marked changes have occurred from the earlier condition described. The aortse no longer appear • to terminate in the region from the twelfth somite on, but are continued as strong vessels to the level of the twentieth segment, the caudal limit formerly reached by the most medial portion of the extra-embryonic plexus. Evidently the plexus formerly here has given place to the stronger single channel, but it has also continued to grow caudally in the tissues of the embryo, for it now reaches a considerable distance further caudad, to a point corresponding in position with the future twenty-fifth somite. In other wprds, there has been a continuous caudad invasion of the embryo by a plexus continuous always laterally with the extra-embryonic or vitelline net.

Fig. 1. — Ventral view of the posterior part ol an injected chick embryo of 17 somites, showing plexiform character of the aortae opposite the 17th somite.

Fig. 2. — Ventral view of the posterior part of an injected chick embryo of 20 somites, showing an extension of the plexus out of which the aorta develops. The large nonvascular area surrounding the region of the primitive steak is here much reduced in extent.

Fig. 3.— Ventral view of the posterior part of an injected chick embryo of 23 somites, showing the completion of the down growth of the capillary plexus out of which the aortae are formed.


Fig. 1. Fig. 2.


D 503

from till L.1 margii bu*: thi • Tinection the step presen portionj ^<^<ientvLSLtec i:i.^ious com .-ddally this lost strand

~~T find was The singh ^•^e the clu( ^^nizing th( ^^=>thers have ^^e given in ^»t«.

— the aortae

- for a lonjg ^*x vitelline ^ion of this -XX is estab "these tiny ^ with this

Fig. 3. ^^=^^ actively

"^^--^^^rimites, sigIHere the ^ ^^•"^-:>es of the Tst aortic '" opposite


Fig. 1. Fig. 2.

"lATOMiCAL Record. — Vol. Ill, No. 9.

D Fig. 3.


ti from the lal margin ^, but this lonnections E the steps >w present jr portions iccentuated luous com udally this iost strand

[ find was The single ve the clue jiiizing the >thers have VQ given in rtse.

the aortffi for a long n vitelline

ion of this in is estab, these tiny B with this

)re actively omites, sigHere the isues of the first aortic es opposite


Thb Anatomical Recoi

D Blood Vessels of Vertebrate Embryos. 503

That portion of the plexus formerly occurring in the region from the fifteenth to the twentieth somites has had its most medial margin enlarged into the continuation of the aorta on each side, but this has involved also the elimination of its former frequent connections in this region with the remainder of the mesh. Some of the steps in this process are to be seen beginning in the mesh now present below the level of the twentieth somite, for in the upper portions of this the medial margin of the plexus is distinctly accentuated above its fellows, with which, however, it is still in continuous communication.

From now on, in successive stages, there is continued caudally this invasion of capillaries and the conversion of the innermost strand of the plexus into the continuation of the aortse.

Such a conception of the development of the aorta, I find was clearly indicated by His and especially by Vitalleton. The single figure of His showing one of the stages in this process gave the clue to the story, but, doubtless due to the difficulty in recognizing the limit of capillaries by ordinary methods, Riickert and others have not accepted these views. Our preparations, however, have given in all its details this method of the formation of the lower aortse.

It is interesting to note that the many connections of the aortse with the plexus from which they have been formed persist for a long time in the area below the level of origin of the main vitelline arteries. They may be said to form the primary circulation of this portion of the intestine, for, when the caudal vitelline vein is established and comes to encircle the posterior intestinal portal, these tiny vessels, coursing in the splanchnopleure, connect the aortae with this vein.

Anteeiob Cardinal Veins.

At the same time that capillary growth begins to more actively extend the aortse caudally, i. e., at the stage of fifteen somites, significant changes also begin in the region of the head. Here the earliest capillaries to grow out independently into the tissues of the embryos have arisen from the cephalic convexity of the first aortic arch and, extending dorso-laterally, formed a few meshes opposite

D 504 Herbert M. Evans.

the constriction between fore and mid-brain. From here the capillaries spread forwards and backwards, growing somewhat more rapidly in the latter direction, so that a small plexus is soon formed at the side of the mid-brain. Anteriorly, the sprouts tend to encircle the stalk of the optic vesicles. At other points more caudad the dorsal aort® give rise to capillary sprouts which grow forwards and join those just mentioned and growing also in the opposite direction, coalesce vrith the vitelline veins near the junction of the latter trunks with the heart. Thus a slender but continuous chain of capillary vessels extends from the head region to the vitelline veins. Evidently enough of a circulation exists through this minute head plexus, fed as it is at several points from the aortse, to fashion a venule from the more caudal capillaries, i. e., those opposite the hind brain, so that at a very early date we have the picture of a long slender venule leading back from the plexus at about the region of the isthmus between mid and hind-brain to the vitelline veins near the heart.

It is out of this capillary plexus which has begun to grow up about the mid and fore brain vesicles that the head veins are all ultimately formed. These veins are the chief tributaries of the anterior cardinal trunk and consequently extend the latter vessel much forward into the region of the head.

In all vertebrate embryos which I have studied, a portion of this capillary plexus opposite the mid-brain soon lies more superficially than the remainder and it is from these capillaries, enlarging soon, that the main vein is destined to be continued. This interesting stage in the development of the head vessels is seen in Figs. . 4, 5, and 6.

It is thus possible to trace the history of all the head tributaries of the anterior cardinal. Out of the capillaries connecting the more superficial ones just mentioned with those surrounding the sides of the optic vesicle, are formed the ophthalmic veins. And at the same time the caudal margin of the plexus covering the mid-brain is enlarged to form a prominent drainage channel, a vein which is thus situated at the isthmus between hind and mid-brain or at the caudal edge of the latter vesicle.

D D 504


the const laries sp rapidly ij at the sidi the stalk dorsal ao] join thos6 coalesce M with the vessels e:i dently en< fed as it from the : so that at venule lei isthmus b heart.

It is ou the mid ai formed, dinal trun into the re In all V capillary ] than the r that the t stage in t 5, and 6.

It is thi of the ante superficial the optic V' time the clarged to i situated at edge of the

Fio 6.

lliB Anatomical Recokd. — Vol. Ill, No. 9.

D Fig. 3b.

Fig. 4.

Fig. 5. IC

D D D Blood Vessels of Vertebrate Embryos. 505

PosTEMOB Cardinal Vein. The origin and development of the posterior cardinal vein whose entire history can be followed, is the result of the activities of two systems of capillaries, a chain of capillaries arising from the duct of Cuvier and growing caudally in the splanchnopleure, and a row of capillaries, the intersegmental vessels, which in simple loops spring from the aorta and annex themselves successively to the former chain. What guides the course of these particular capillaries — ^the segmental vessels — so accurately into these loops is at present unknown, but it is no doubt a direct influence of the segmental structure of the neighboring mesenchyme which only favors endothelial proliferation at the inter-somitic spaces. Hence it is that at these intervals, the segmental capillaries (to become later the segmental arteries) grow out at right angles to the main axis of the embryo and after a dorsolateral loop or bend are free to extend longitudinally. Such a longitudinal extension involves their union with the capillary chain which

Fig. 3b. — ^Lateral view of head of injected chick embryo of 15 somites, showing primary head capillai*y plexus. The plexus takes origin from the convexity of the first aortic arch at several points and is continued posteriorly as a slender capillary chain which eventually joins the main vitelline vein near its junction with the heart This slender capillary chain has arisen at several points from the dorsal aorta on each side, and two of these points of origin are still preserved, opposite the region of the hind brain. The delicate capillary path from the head plexus to the vitelline vein is destined to form the anterior cardinal vein.

Fig. 4. — Lateral view of head of injected chick embryo of 17 somites, showing the primary head capillary plexus partially covering the lateral sides of the fore and mid-brain vesicles. It will be seen that a portion of the plexus lies more superficially than the remainder, and it is this superficial portion which is destined to become the main trunk of the vein in this region. The artery is shown darker than the vein.

Fig. 5. — Lateral view of head of injected chick embryo of 20 somites, showing the further development of the anterior cardinal vein out of the primary head capillary plexus. The capillaries bordering the groove between mid and hind brain have formed a prominent tributary of the main vein.

Fig. 6. — Lateral view of head of injected chick embryo of 25 somites. The lateral surfaces of the fore and mid brain vesicles are now completely covered by the capillary net, which is extending dorsally but is still far from the middorsal line. There is seen a corresponding great growth of the anterior cardinal vein and its system of tributaries.

D 506

Herbert M. Evans.

has grown down from Cuvier's duct and thus this channel is extended. Beyond the first four or five segments, this channel consists often of a single longitudinally coursing capillary and it is now further

- p. c V.

Pig. 7. — View of a total mount of an injected chick embryo of 17 somites, showing the duct of Cuvier and subadjacent region. iX^) a = anterior cardinal vein; b = duct of Cuvier; c = 10th segmental vessel; P. C. V. = capillaries from which the posterior cardinal vein is formed; x = endothelial sprout representing the 11th segmental vessel.

Fig. 8. — ^Vlew of a total mount of an injected chick embryo of 21 somites, showing the duct of Cuvier and subjacent region. (X^O.) The lettering is the same as in the preceding figure, with the exception of x, which repreaents the 14th segmental vessel. One sees the segmental capillaries bifurcate often into anterior and posterior sprouts, the union of which makes the continuation of the vein.

extended caudally solely by the longitudinal sprouts of the segmental capillaries, the cephalic sprout of the last of these joining the caudal sprout of the next preceding one as Figs. 7 and 8 show. Coin


Vnatomical Record. — Vol. Ill, No. 9.

D ■

Fig. 11

D D Blood Vessels of Vertebrate Embryos. 507

cident with the extension of this vessel its upper section becomes larger, for the increased number of segmental aflFerents gives a considerable drainage territory; thus it is that it soon becomes a vessel of more than capillary size and recognizable as the posterior cardinal vein.

Umbilical Veik.

Those irregular capillary meshes which border the duct of Cuvier in embryos of from fifteen to seventeen somites and which aid in the formation of the upper end of the posterior cardinal vein, after a considerable interval, again sprout caudally, this time in the somatopleure, and are developed later into the veins which we can recognize as the umbilicals. The history of these veins in the chick is fraught with the greatest interest, for by the injections we can follow them in the assumption of several roles in the embryonic circulation, long before the establishment of their ultimate function in connection with the allantois. They are successively the drainage channels of the arm, the body wall, and the leg before the allantois* has arisen. The latter sac indeed attains some little size before its vessels are in connection with the umbilical veins.

By the stage of twenty-three somites the first capillaries of the later umbilical vein form a simple mesh work in the uppermost portion of the somatopleure. Soon the cell mass constituting the future anterior limb becomes evident and its growth stimulates the

Fig. 9. — ^Injected chick embryo of 23 somites to show the orlsln of the umbilical vein from a capillary plexus situated in the angle between the posterior cardinal vein and the duct of Cuvier. A. C. V. = anterior cardinal vein; P. C. V. = posterior cardinal vein; U^ capillaries destined to form the umbilical vein.

Fig. 10. — Injected chick embryo of 24 somites to show the extension in the somatopleure, of capillary plexus forming the umbilical vein. The lettering is the same as Fig. 9.

Fig. 11.— Injected chick embryo of 30 somites. The capillaries destined to form the umbilical vein have reached the region of the future arm bud where they are joined by a direct capillary sprout from the aorta (subclavian artery).

FiG. 12. — Injected chick embryo of 35 somites, showing establishment of umbilical vein as the main drainage channel of the anterior limb.

D 508 Herbert M. Evans.

outgrowth from the aorta of a whole series of capillaries which unite to form a delicate plexus. These capillaries find and imite with those which have grown down from the duct of Cuvier and thus is

F.o.,3 X "%:^y

Fig. 13. — Injected chick embryo of the third day showing extension of capillaries from which the umbilical vein is formed, as far as the posterior limb bud. The reduction is much greater than In the preceding figures In order that the entire embryo can be shown.

established the earliest circulation in the limb bud, a circulation consisting of many afferent capillaries streaming from the lateral

D Blood Vessels of Vertebrate Embryos. 509

aortic wall^ forming in the limb tissue a few simple meshes and draining headwards into the capillary chain, already somewhat enlarged and venous in character, which is the later umbilical vein.

This capillary net still continues to grow caudally in the somatopleure, below the level of the upper limbs. At the same time another mesh of capillaries, that which has arisen in the posterior limb buds, has begun to grow upwards and the union of these two plexuses establishes a narrow continuous mesh in the somatopleure,

Fiff. 14

Fig. 14. — Caudal end of an injected chick embryo showing the sublntestinal vein draining the tail, allantois and posterior limbs. a, b. = allantoic branches; J. b. = limb branches; S. V. = siibintestinal vein; P. C. V.= posterior cardinal vein; Ac. = aorta; C. V. V. = caudal vitelline vein; U. A. = umbilical artery ; U. V. = umbilical vein.

into which the vessels of both limb buds and the body wall now drain. The capillaries of the hind limbs have also acquired connections with a more ventrally placed vein — ^the sub-intestinal vein, which has arisen in connection with the drainage of the tail and the allantois. This vein has hitherto been entirely overlooked and its presence in any of the embryos of the higher vertebrates is entirely unknown save for a few sentences announcing its occurrence in the ninety-six hour chick in the recent work by Lillie.*^ His remarks on the discovery of this vessel can be very appreciably extended now

'Ullie. "The Development of the Chick," 1909.

D 610 Herbert M. Evans.

from the injections. As has just been indicated, the sub-intestinal vein in Aves forms the primary drainage channel for the tail, hind limbs and allantois. Its position and chief tributaries can be seen from Fig. 14. Somewhat later, and at about the time the allantois approaches a millimeter in diameter, the umbilical system of capillaries has united with its vessels and begins to fimction as a means of drainage for the allantoic circulation. The uppermost portions of the umbilical have now enlarged appreciably, its connections with both limb buds are eventually lost and its last territory supplies it with such a volume that it becomes a relatively huge channel, the allantoic or umbilical vein.

In the Mammalia, although the umbilical veins precede the limb buds in time of appearance, nevertheless here also, when the limbs arise, they are at first drained into the umbilical veins.®

Dr. F. T. Lewis informs me that he had observed and demonstrated this drainage of the mammalian limbs into the umbilical veins at the meeting of this association in 1903. Unfortunately no record of this was made in the proceedings for that session, but Dr. Lewis has been kind enough to send me sketches and notes made at the time, showing this fact for rabbit embryos. I have recently been able to confirm these findings on the human embryo also, so that there is little reason to doubt its general applicability for the Mammalia.

Many other prominent vessels in the body have been traced to a similar origin from a capillary plexus, but time will now permit the mention of only a few of these. In reconstructions of the vessels of the head which have been made by various investigators, it appears as if the tip of the anterior cardinal veins grew forwards in a dorsomedial position, in place to form the future sagittal sinus. It has been possible to trace the formation of this vein quite completely in injected mammalian embryos. In pig embryos five and six millimeters in length the primitive capillary plexus which grows up over the sides of the mid and fore brain has not yet reached the dorsal surface, though sprouts can be seen along its upper margin. By the

•Evans. Am. Jour. Anat. IX, 2, 1909.

D Slood Vessels of Vertebrate Embryos.


»t Card, or ^Itr V««n

Fig. 15 a

Fig. 15 a. — Lateral view of upper portion of pig embryo 8 mm. Icmg, showing location of mid dorsal non-vascular area, the extent of which has been purposely exaggerated laterally. It will be noted that the capillaries have fused dorsally over the mid brain and upper portion of the hind brain.

Fig. 15 b. — Dorsal view of fore and mid brain region of the pig of 8 mm. shown in Fig. 15 a, showing the limit of extension of the capillary plexus here. The mesh work which has grown dorsally from either side halts sharply in two parallel lines between which is the narrow non-vascular strip. Anteriorly is seen the earliest indication of the superior sagittal sinus, which is formed from either margin of the capillary mesh, and consequently at this stage paired.

D 512 Herbert M. Evans.

time the embryo has attained a length of nine millimeters, the capillary mesh has covered the top of fore and 'tween-brain vesicles save in the median line. Here the two meshes are as yet unfnsed, and confront each other along two parallel lines, which thus bound a median dorsal non-vascular strip, across which no connecting capillaries have ventured to grow. It is these two medial margins of the plexus which, in the region of the cerebral hemispheres, are enlarged to form the superior sagittal sinus, thus originally paired. (Figs. 15a and 15b.)

One of the most beautiful and evident instances of the conversion of a capillary mesh into an arterial channel is afforded in the history of the anterior spinal artery. Here too we have the best possible axis of reference, for the mid-ventral line of the spinal cord is constant. On the ventral surface of the cord we can observe all the steps in the first invasion of a plexus of capillaries there, their later coalescence and enlargement in the mid-line as an irregular, illdefined channel, and eventually, the further conversion of this into the very definite artery of regular contour and calibre — the anterior spinal. Fig. 18 illustrates the development of this vessel in the pig.

I may be permitted to instance one more vessel, in this case one of the very largest in the body, though not the earliest to develop, which can easily be seen in the young embryo in the form of a capillary mesh. I refer to the pulmonary artery. The endothelial sprouts which later form this trunk spring from the sixth aortic arch as true capillaries. In fact they reach the lung bud as a chain of capillary meshes and retain this character for some time, as Fig. 21 shows.

Besides the history of many individual blood vessels of the body, these specimens have given weighty evidence towards a number of general laws or phenomena of blood vessel development and these will be briefly mentioned. They concern

(1) The presence always in the embryo, of a united vascular system, so that the blood vessels form a single though irregularly branched endothelial tree whose branches are in no case added after an independent formation but arise by sprouting from the parent trunks.

D Blood Vessels of Vertebrate Embryos, 513

(2) The place and maimer of spread of the first capillaries through the body.

In discussing these briefly, we may say

(1) Serial sections of perfectly injected embryos show no evidences of vessels which have not received the injection mass and are hence unconnected with the general system. Investigators, working with uninjected material, have repeatedly reported such vessels. Their findings are in all probability to be explained by a collapse of the connecting vessels, since

a, injected specimjens show these connecting vessels and

b, injected specimens fill other vessels previously unrevealed by ordinary methods, thus furnishing a far more complete picture than is otherwise obtainable.

The recent accoimts by Riickert and MoUier in Hertwig's Handbuch, on the subject of the first blood vessels are perhaps the most conspicuous of the claims of vessel origin in situ. Their evidence has come from Riickert's studies of serial sections through selachian embryos. His statements can doubtless be successfully attacked by injecting selachian embryos and studying carefully the areas in question.

With the light which such specimens have shed, the statement that any vessels in the embryo arise at first unconnected with the vessels in that region can be now challenged. If an instance be given it can doubtless be speedily disproven, providing complete injections of that area can be secured.

(2) The spread of the first vessels through the body. Whatever may be the first source of the endothelium in the body of the embryo, after the earliest stages, the injections have furnished a complete history of the further capillary proliferation and outgrowth into the tissues of the embryo.

Inasmuch as the first vessels lie somewhat centrally in the embryonic body, the general direction of growth is from center to periphery. The center consists of the upper dorsal aortse together with the first arch and Cuvier's duct ; the periphery comprises the various viscera and central nervous system as well as the body wall, but the ultimate periphery, the skin, is supplied late.

D 514 Herbert M. Evans.

In spreading outward, the capillaries do not grow uniformly in all directions, thus successively invading various zones, but are apparently governed by the character and needs of the various tissues, reaching some of them early and some remarkably late. Hence there are present during all the early stages in the embryo's growth, vascular and non-vascular areas.

The method of injection reveals such a wealth of small vessels whose existence we had not hitherto known, that at first thought one is inclined to suspect the universal presence of the vascular net, throughout the tissues of the embryo. This, however, is as much an error as was the former notion of the scant extent of the embryonic vessels. Injections made under the best possible conditions and afterwards explored in serial section have all shown the existence of definite non-vascular areas bordered by a margin of true capillary sprouts. The position of such non-vascular areas is as constant as is thai of any vascular channel in the body and the more funda/mental of them are probably represented at homologous stages in all vertebrate embryos.

Among the tissues, the central nervous system receives the first investing capillary net, but even here the capillaries do not at once surround the neural tube but occupy only the lateral aspect, gradually growing ventrally and dorsally. At the top of the brain, the capillary mesh is some time in fusing from either side, so that there exists here relatively late the narrow non-vascular strip in the middorsal line already mentioned. (Fig. 15.) In the case of the hind brain there is an especially conspicuous lack of much capillary proliferation in a dorsal direction, so that in comparatively late stages of all vertebrate embryos the roof of the hind brain presents a characteristic large non-vascular zone. Indeed, while in pig embryos ten millimeters in length the lateral capillary beds have completely fused dorsally, in the fore and mid-brain region, the nonvascular area on the top of the hind-brain persists until the embryo has attained a length of over twenty millimeters. (Figs, 16, 17a, and 17b.)

In the cord also the ventral and dorsal surfaces are invaded only secondarily and are at first entirely non-vascular. The dorsal sur


Fig. 16.


■ m. b.


ncAL Record. — Vol. Ill, No. 9.IC j

Blood Vessels of Vertebrate Embryos. 515

face is bridged last of all and so the spinal axis presents for a time the remarkable sight of a close capillary investment everywhere save on its upper aspect, at the margins of which the two long parallel borders of invading capillaries and their sprouts have halted sharply in their spread. (Fig. 19.) This narrow non-vascular zone is maintained for a long time, but when the time comes for its obliteration, quite suddenly, capillary sprouts push out and bridge the gap. This bridging occurs successively from above downwards and embryos which have the dorsal surface thickly covered with capillaries in the upper half will show the first bridging capillaries in the caudal region, as Fig. 20 shows.

Other examples of vascular and non-vascular areas may be mentioned. The center of each sclerotome is, on its upper surface, supplied by a sheet of closely anastomosed capillaries ; but the outer divisions of the sclerotome are not so supplied. There capillaries are absent for a considerable time, so that the vertebral column presents a succession of vascular and non-vascular zones, the former areas in each case overlying the segmental vessels.

Furthermore, in the gro^vth of the embryo, tissue at one time permeated with a quite uniform capillary mesh may in its further growth show a later differentiation into vascular and non-vascular areas. This arrangement of its vascular mesh is of course coincident with corresponding changes in the nature of the tissue at various

Fig. 16. — Dorsal surface of hind and mid brain of a pig embryo 8.5 mm. long, showing fusion of the primary head plexus across the mid line, except the three non-vascular areas shown.

Fig. 17 a. — Dorsal surface of fore and mld-braln vesicles of Injected chick embryo of 32 somites.

Fig. 17 b. — Dorsal surface of fore and mld-braln vesicles of Injected chick eiuoryo at the end of the 3d day. c. h. = cerebral hemisphere; th. = thalameneephalon ; m. b. = mld brain. In the earlier stage (Fig. 17 a) the primary head capillary plexus has fused across the mid dorsal line only at one point, between the two divisions of the primitive fore brain. In the later stage the mesh quite completely invests the mid dorsal surface of the head, but the cleft between the cerebral hemispheres Is non-vascular, as is also the zone surrounding the pineal organ. At the mesial margins of the two prominent lobes of the mid brain are seen the two mesencephalic veins which have been formed from the plexus.

D 516

Herbert M. Evans.


Jto. Jl



Fig. 19.

Fig. 20.

The Anatomical Record. — Vol. Ill, No. 9.

D Blood Vessels of Vertebrate Embryos. 517

areas and it is often the most positive evidence of these changes. Thus pre-cartilage and pre-muscle tissue are characterisitically non-vascular and wherever these condensations of the mesenchyme occur, they


Fig. 21 Fig. 21. — Camera lucida tracing of the fourth and sixth aortic arches of an injected pig embryo 12 mm. long to show the early character of the pulmonary arteries. P. A. = pulmonary arteries; L. = lungs; P. V. = pulmonary vein; VI = sixth arch ; IV = fourth arch ; S. A. = subclavian artery.

form islands of tissue sharply circumscribed by capillaries but uninvaded by the latter for a considerable interval. Of this the limb buds furnish splendid examples, for the blastema of the arm, which is at first furnished with a uniformly distributed capillary net, later

Fig. 19. — A series of views of the dorsal surface of the spinal cord at the level of the lower cervical segments in injected pig embryos, a, from an embryo 8 mm. in length. All the capillaries seen are limited to the lateral surfaces of the cord and the spinal ganglia, b, from an embryo 10.5 mm. long, showing the first capillary bridges across the cord in this region.

c, from an embryo 15 mm. long, showing the type of plexus establiiUied here.

d, from an embryo 20 mm. long, illustrating the farther growth of the plexus. Fig. 20. — Tail of an injected pig embryo, 13.5 mm. long. In the cervical

and thoracic regions, the plexus investing the dorsal surface of the cord has become a close mesh ; in this region, however, the first capillary bridges, and, at the end, the first sprouts, are shown.

D 518 Herbert M. Evans.

begins to exhibit areas which the capillaries appear to avoid — areas corresponding to the later masses of cartilage or muscle groups.

In most cases the capillaries tend to anastomose at greater or lesser intervals forming a loose or close mesh and this plexus formation is doubtless one of their most characteristic and fundamental properties. It has perhaps been better termed their tendency to grow in every direction, yet influences often check this tendency successfully and in some areas permit their growth from the very first only in a certain definite direction. The best example of this is furnished by the dorsal segmental vessels which, as is well known, are rigidly governed in position by the presence of the primitive segments, between which they course.

All of these examples clearly indicate that the behavior and character of the capillaries is even from the very first intimately influenced by the tissues into which they grow. A new set of problems confronts us, problems which can aim more than ever before at the causes at work in the developing organism, for now that we may recognize with certainty vascular and non-vascular areas and the relation of each of these to the tissues, there come up at once questions concerning the differences in chemical nature of the tissues and a closer determination of the real stimulant for vascular growth.

The story of the development of the vascular system is primarily the story of the spread of the capillaries, the form and relation of their plexuses and the role of these in the elaboration of the trunks of the adult.

We have to do always with the extension of a fimctioning system not the blind outgrowth of vessels to their ultimate territory. Such a system extends itself by capiUary sprouts and as the capillary bed increases, its supplying and draining channels, the arteries and the veins, grow and rearrange themselves concordantly.

Received for publication July 6, 1909.




WILLIAM H. HOWELL. Deem of the Medical Faculty, Johns Hopkins University,

Whenever, since Mr. Gilman's death, I have attempted to formulate for myself an estimate of his services, I have finally summed it all up in an expression of his own, which I heard him use on the occasion of the memorial exercises to the late Professor Rowland. I remember the occasion well. As he advanced to the edge of this platform to open the exercises, looking silently at his audience for a few seconds, he began his remarks by the simple sentence, made impressive by his manner, "A great man has fallen in our ranks." I am confident that this estimate, applied to him, is shared by every one in this audience and by all of our fellow alumni of the Johns Hopkins University. He was a great man, and above all a great college president. He was a great president by virtue of the fact that he was a man of ideas and high ideals which reacted like a stimulus upon all who were brought into contact with him; he was a great president because of his masterly genius for organization; but he was a great president chiefly, in my judgment, because he possessed in such large degree the rare power of getting the best out of those who worked with him and under him. He led and guided them by the all-constraining force of his enthusiasm, his sympathy, and his tact. The kind of executive who drives things before him by the mere force of his personality, is liable, in accordance with the law of action and reaction, to create round himself an atmosphere of opposition and discontent. Such an executive may be needed in some of the affairs of life, but he is not the type most suited to develop the greatest efficiency of a university faculty. This University was most fortunate in possessing in Mr. Gilman a leader and executive who, by reason of a happy


D 520 William H. Howell.

combination of genial qualities of mind and heart, was able to inspire a general and enthusiastic spirit of co-operation among his official subordinates. We must never forget, nor allow others to forget, that the great success which this University attained, almost from the beginning, was in a large part, in chief part, due to him. The creation of a university of a new type was not a game that played itself. On the contrary, there was opportunity in abundance for mistakes and disaster, and if, instead, there came, on the academic side, a train of successes and renown, we owe it largely to his ability and experience as a leader and administrator.

I have been asked to speak of Mr. Gilman, especially in regard to his connection with the medical school. In truth the medical department owes as much to his wise and stimulating leadership as its older comrade, the philosophical faculty. It is well known that the subject of medical education interested Mr. Gilman deeply. What circumstances gave this direction to his thoughts I am not able to say from personal knowledge. I know only that it antedated his connection with this institution. That a special interest existed is evident from his published addresses, as well as from the record of his services while President. In his inaugural address the subject of the formation of a worthy school of medicine comes up first, and the hope is expressed that at no very distant day a medical faculty may be organized. So also, in describing the purpose and aims of the biological department, which constituted a novel feature in the newly-established university, he laid great emphasis upon its importance in relation to the study of medicine. Indeed, from the beginning of the University there was organized a pre-medical course along the lines which had been laid down by Huxley, a course which in its general features, has since been endorsed and imitated by many of the leading schools of the country. As a matter of fact, medical education among us at the time of the founding of the University was in a deplorable condition. Deprived of adequate financial support and without the uplifting aid of an academic connection, most of our medical schools had sunk to a very low level. They demanded practically no educational preparation on the part of their matriculates, and they made little or no effort

D Relation of President Gilman to Medical Education. 521

to give their students an adequate training in the theory and science of medicine. The training, in fact, resembled that of an apprentice rather than that of a candidate for admission to a learned profession. Mr. Gilman, with his wide interest in education in general, must have been impressed, as many other thoughtful men were, with this very undesirable state of affairs. With the prevision characteristic of a great leader, he seems to have selected medical education as one of the great opportunities which the new university might utilize to do a needed service to the country at large. For reasons over which he certainly had no control the realization of his plans was deferred for some seventeen years. It was not until 1893 that the medical school, as we now know it, was founded. It was and is a graduate school in the sense that it accepts as students only those who are college graduates. At the time of its foundation its requirements for entrance seemed almost absurdly high. It was supposed that only a few students each year would be veiling to meet these requirements, considering that in the other leading schools the conditions for entrance were so much less difficult ; and the idea ,that our standards would ever be adopted generally by other schools was scarcely reckoned among the probabilities. Yet, to-day, this school has three hundred students upon its rolls, and for many years past there has been a steady approximation on the part of other good medical schools toward the standards established here. Many agencies have undoubtedly contributed to the great improvement in medical education which has taken place in this country during the last generation — ^volunteer organizations among high-minded physicians, the effective action of our State Boards, etc.,- — ^but I believe it will be admitted that the actual example held before the eyes of the medical public, in the successful experiment carried out here under Mr. Gilman's direction, has been the most potent influence of all in strengthening the weak faith of those who doubted the feasibility of such a reform.

Many speakers and writers have commented upon the timeliness of the foundation of the Johns Hopkins University. The University was started at a time when the country was ripe for the opportunity to obtain genuine graduate instruction. Certainly

D 622 William H. Howell.

the same observation may be made with even more justice in regard to the appropriateness of the movement inaugurated by the foundation of the medical school. The country was prepared, indeed had been prepared for some years, for a development of this kind. Mr. Gilman and his colleagues had the wisdom to understand this, and the courage to make the experiment on a scale befitting the reputation of the University and worthy of the unique opportunity afforded by the existence and close affiliation of that splendid sister institution, the Johns Hopkins Hospital.

Mr. Oilman's devotion to the affairs of the medical school in its early history was unfailing. He gave to it on the administrative side an ideal organization which has been the envy of other schools, and which will eventually, I believe, be generally adopted. The central feature of this organization is that it places all power in the hands of a small but representative body, composed of the heads of departments, the president, and the superintendent of the hospital. Over the deliberations of this body he presided constantly during his incumbency, and it is needless, for those who knew him, to add that he was a most admirable presiding officer. Courteous, considerate, and informal he invited a free expression of opinion from all, but he knew well the art of controlling gently but firmly all tendencies to useless and diffuse discussion. The routine business was dispatched with promptness, while matters of importance from the standpoint of policy or precedent were treated with care and circumspection. A more harmonious and effective board it would be hard to imagine, and, indeed, how could it have been otherwise with a man like Gilman as presiding officer and a man like Welch as dean and secretary. Our foundations were well laid, and I am sure that the great s^iccess of the school, acknowledged everywhere, was a source of the deepest gratification to Mr. Gilman. It may be fairly claimed that it constituted his second great contribution to the educational development of this country. I hope that the future historian of medical education in the United States will not make the mistake of supposing, because Mr. Gilman was not a member of the medical profession, that therefore his connection with this medical school was in any sense perfunctory. On the

D D D The Relation of President Gilman to Medical Education. 523

contrary, it was real, it was vital, and it was continuously maintained.

He had a clear comprehension of the actual conditions and the needs of medical education, and, I believe, a definite idea of the special traditions which he wished to see established here. He took a direct part in the discussions regarding appointments upon the staff, appropriations for the various departments, the standards for admission and graduation and other matters, great and small, which arose during the formation period of organization. I do not believe that this fact of his constant active participation in the details of administration was a matter of common knowledge outside the small circle of the governing board. I am quite sure, in fact, that the students and graduates of the medical school and many of the members of the faculty have assumed that the labor and credit of the successful foundation of the school belong chiefly to the leading members of its faculty, who by their position naturally represented the department in the eyes of the medical public. But I am also quite confident that these same members of the faculty are ready, without exception, to acknowledge and to insist upon the importance of Mr. Gilman's influence throughout the early years of the school's history. This influence was exerted in many ways and its result may be summed up, I believe, in the statement that there was established in the Medical Faculty a distinctly academic spirit. In many of our strong medical schools it may be said, without injustice I think, that the administration of affairs had absorbed something of the methods of compromise, expediency and personal gain which are so evident in the commercial and political worlds. Considerations of this kind press close upon the administrator, of course, and it is difficult for him to ignore them, but the individual or the institution which keeps its eyes focussed too constantly on such methods suffers in the end from a sort of spiritual myopia. The academic spirit takes the larger view beyond the immediate advantage of the present toward that which is fundamentally true and right, and for such a measure of this nobler spirit as we are fortunate enough to possess we are indebted very largely to the personal influence of Mr. Gilman.

Received for publication Mnrrh 8, 1009.



Vol. III. OCTOBER, 1909. No. 10



ROBERT J. TERRY. From the Department of Comparative Anatomy, Harvard Medical School,

With Two Fiqubes.

The name "vomer," given to the unpaired plowshare-shaped bone of the human cranium, has been applied in the works on comparative anatomy to a pair of bones in the skulls of the Sauropsida and Ichthyopsida. This homology, maintained by the earlier writers and by most osteologists up to the present time, is founded largely on the relations of parts in the adult skull. The unpaired vomer of the mammals was explained by assuming it to be the equivalent of a pair fused together, a theory supported by observations on the origin and development of the single vomer in certain teleosts and birds and in man. According to Gaupp^ true paired Anlagen of the vomer have not, however, been seen in the lower mammals.

In 1884 Sutton^ proposed a new homology for the mammalian vomer by claiming its presence in the lower- animals in the parasphenoid, an unpaired bone in the base of the cranium which exists in all classes from the fishes on, except the mammals. The parts in the mammalian cranium which, according to the theory, should correspond with the ichthyopsidan paired vomers, were found in the palatine

^Gaupp, E. Die Entwickelung des Kopfskelettes. Hertwig's Handbuch der Entwickelungslehre der Wirbeltiere. 1906, Bd. Ill, Zweiter Teil, p. 850.

•Sutton, J. B. Observations on the Parasphenoid, the Vomer and the Palato-pterygoid Arcade. Proc. Zool. Soc, 1884, p. 566.


D 526 Robert J. Terry.

processes of the premaxillaries. These processes have been observed by Albrecht, Sutton and others to arise independently of, and subsequently to fuse with, the tooth-bearing portions of the prejnaxillaries.

In later years, Broom^ has contributed much to our knowledge of the comparative anatomy of the vomer, and with evidence adduced from the investigations of Turner, W. K. Parker, Wilson and Symington, he strongly supports the homology of the mammalian vomer and parasphenoid. As to the comparison of the paired vomers of the lower forms w4th the palatine processes of the premaxillaries, he does not agree entirely w^ith Sutton. Broom has suggested the term '^prevomer'^ for the category of bones represented by the paired vomers (of other authors) in the lizard, and finds its homologues in the paired vomers of the Ichthyopsida. But in the great majority of the higher mammals the prevomer does not exist, its place being taken by invasion of the palatine processes of the premaxillaries. These are regarded as true portions of the premaxillaries and not independent elements w^hich Sutton considered them to be. In the dumbbell-shaped bone of Ornithorhynchus and in a median ossification in the nasal region of Miniopterus, Broom identifies the prevomer. These bones, although azygos in the adult, are both derived from a fusion of a pair of splints underlying the cartilages of the vomeronasal organs.

An objection to the comparison of the mammalian vomer and the non-mammalian parasphenoid lies in the fact that the latter presents in the series of animals a history of retrogression ; in the lowest forms the parasphenoid reaches forward to the ethmoidal region, w^hereas in most reptile? and birds its anterior end is far back and away from this region. A more serious obstacle to the new homology is the circumstance, already mentioned, of the single vomer developing from a pair of centers. The one instance in mammals might wvll be taken to be an exception to the rule of single origin, if

"Broom, R. On the Iloinology of the Palatine Process of the Mammalian PremaxUlary. Proc. Linn. Soe. N. S. W., 1895. Vol. X, p. 477-485.

On the Occurrence of an Apparently Distlnet I'revomer In Gomphognathus. Jour. Anat. and Physiol., 1896, Vol. XXXI.

On the Mammalian and Reptilian Vomerine Bones. Proc. Linn. Soe. N. S. W., 1902, Vol. XXVII, part 4, p. 545-,')G0.

D Development of the Mammalian Vomer. 527

single origin were known to be the rule. But how many studies have been made by modem methods to determine this matter ?

During the reconstruction of the cranium of a cat embryo there was observed a tendency to bilateral formation of the vomer, and my attention was thereby directed to the question of the origin of this bone in the mammals. A review of the series in the Harvard Embryological Collection resulted in finding one instance of paired origin of the vomer, and tliat in a marsupial. The discovery by Fuchs* of the remains of the parasphenoid in a Didclphys embryo and

' B

Figs. 1 and 2. — Transverse sections tbroiigh the nasal region of a 17 mm. specimen of Caluromys philander; 1, through the anterior ends of the vomers ; 2, through the middle of the vomers. Harvard Emb. Coll., Series 707, Sections 245 and 228. X 39 diam.

A, cartilaginous nasal septum; B, palate; C, vomer; Z>, palate process of premaxilla ; £7, vomeronasal organ of Jacobson.

its bearing on the homology of the mammalian vomer, induced me at this time to communicate the observation. Through the courtesy of Professor ilinot I have been enabled recently to review the sections of the heads of three pouch-specimens of Caluromys (Didelphys) philander in which the paired origin of the vomer had been noted.

Fuchs, Hugo. Ueber einen Rest des Parasphenoids bei einem rezenten Saugetiere. Anat. Anz., 1908, Bd. 32, p. 584-590.

D 628 Robert J. Terry.

In a specimen 18 mm. in length there is present a pair of vomers. These elongate ossifications lie approximately parallel with the ventral edge of the cartilaginous nasal septum and extend from its caudal end forward as far as the middle of the vomero-nasal cartilages of Jacobson. Here the septum is continuous ventrally with the palate, and in this region the vomers are connected with one another across the median line. The connection is a feeble one, consisting of a few delicate bony trabeculse which are present in only three of the section.^. Beyond this place in the caudal direction, the septum and palate are separated by a space, so that the nasal cavities are in communication with each other from side to side. In this region the two vomers are seen to diverge as they are followed backward. Each bone for the most part is compressed, with sharp edges and surfaces directed more or less obliquely — ventrally toward its anterior end, ventrolaterally in the middle of its extent. Anterior to the vomers lie the paired palatine processes of the premaxillaries, adapted to the convex surfaces of the vomero-nasal cartilages. A parasphenoid ossification center was not observed.

The conditions here described were found to be essentially the same in the two other specimens examined which were from the same pouch.

The study of younger specimens may decide whether or not the bony bridges are secondary connections between a pair of independent vomerine ossifications. The large size and advanced state of ossification of the lateral parts is indicative of an earlier origin for them than for the insignificant median ossification. The osteogenetic tissue in which the vomers are developing is disposed in two lateral masses of niesenchyma, connected here and there by strands of the same tissue stretching across the middle line ventrad of the nasal septum. In sections passing through its anterior end, the vomerine ossification tract is found to be unpaired and to be situated beneath the nasal septum, from the perichondrium of which it is well separated. This median mass of osteogenetic tissue is, however, of small extent in comparison with the lateral masses of the tissue, and, except anteriorly, presents itself in strands and not as a continuous bed of mesenchyma.

D Development of the Mammalian Vomer. 529

Kelics of the pair of plates which fuse to form the vomer in man are to be found in the ake, projecting conspicuously at the caudal end of the bone. This suggests the probability of the alse of the cat's vomer and of the vomers of other mammals having an origin from paired parts of the developing bone. The dumbbell-shaped bone of Ornitliorhynchus is bifid caudally, a condition which seems to follow from the original paired state of this element. It is not, however, my intention at this time to enter into the question of the phylogeny of the vomer. The object of this communication is to record the paired origin of the vomer in a low mammal, in which class generally it must be insisted that further study of the development is necessary before the bone can be regarded as azygos in its beginning.

Received for publication August 9, 1909.



THOMAS DWIGHT. Parkman Professor of Anatomy, Harvard Medical School.

With Two Figubes.

The following observation is reported not only because it is perhaps unique, but because it is a very striking instance of what might be called vital readjustment

The foot is that of a white man aged 78, and apparently had not attracted any particular notice either before or during dissection. By an unfortunate mischance, the other foot has been lost sight of. The hands showed nothing remarkable beyond disease of one joint of a thumb. The foot consists — apart from the phalanges, which are free and apparently healthy — of the following pieces : the astragalus ; the OS calcis ; the scaphoid, to which the bases of the first and second metatarsals are fused; the cuboid, with which are fused the third and fourth metatarsals and the distal dorsal portion of the external cuneiform; the fifth metatarsal. There is no trace of an internal cuneiform. The external cuneiform is represented by a small part of the dorsal aspect fused with the third metatarsal. It is very doubtful whether the middle cuneiform is represented at all, but it is possible that a small prominence in the sole may represent its distal end. This part of the tarsus is very pathological. The line of the joint between the astragalus and the scaphoid on the dorsal aspect is obscured by irregular bony growths, more or le^s interlocking, which probably interfered with motion, though the joint has persisted. The distal borders of the scaphoid cannot be made out accurately. This bone may be said to be amalgamated with the bases of the first and second metatarsals. Not only is there no sign of an internal or middle cuneiform bone on the dorsum, but the distance from the astragalus to the bases of the metatarsals seems if


D Suppression of the Cuneiform Bones. 531

anything smaller (certainly no greater) than that usually occupied by the scaphoid. On the inner aspect there is a great prolongation of the scaphoid into an uncommonly sharp tuberosity, which appears on the plantar aspect. The line of the joint with the first metatarsal may be followed in a general way, and there is no hint of any remnant of a cuneiform. A rounded ridge in the sole of the foot, from the base of the second metatarsal to the scaphoid, suggests vaguely a small part of a cuneiform, but nothing can be identified certainly.

The culx)id is distinctly shorter than a normal one, and is inextricably mixed with the bases of the third and fourth metatarsals. The reason for accepting a remnant of the external cuneiform bone is furnished by the position of the joint between these bony mas^ses and the scaphoid. This joint, seen from the dorsum, is proximal to the apparent level of the second metatarsal, which would not be the case were the third metatarsal the only element. The joint of the fifth metatarsal shows some pathological changes, especially on

D 532 Thomas Dwight.

the dorsal aspect. It is much more oblique than is normal, and the tuberosity projects laUrally to an uncommon degree. Accurate measurements of any of the metatarsals, with the exception of the fifth, are out of the question ; but it is quite certain that this one is the longest of all. I believe that Pfitzner never obser^^ed this when making his studies of the relative lengths of the metatarsals. The illustrations show that the normal outlines of the foot have been remarkably well preserved, although the outer side of the foot is longer than it should be, and the tarsus forms too small a part of the whole. The most striking defect in the proportions is the great breadth of the foot. This is not caused by any error of articulation, for the nature of the pieces is such as to admit of practically no choice. The length of the articulated foot, measured from the back of the os calcis to the tip of the second toe, is a little over 21 cm., which is certainly small for a male foot. The greatest trans verse breadth, at the proximal end of the phalanges, is a little more than 9 cm., which is abnormally large. It is a case of very advanced "flat foot." The illustrations make further description unnecessary.

An interesting question is — what caused this condition? That the foot is pathological is very clear ; but nothing is more certain than that extraordinary anomalies are associated with pathology in a way that does not allow us to determine what is cause and what is effect. It does not srem possible that this condition is the consequence of a resection in early life. All we can say is that probably the internal and middle cuneiform bones, as well as the greater part of tiie external one, failed to develop; and that the organism, adapting itself to uncommon circumstances, attempted to preserve the general outline of the foot. The underdevelopment of the cuboid, and the obliquity of the joint between it and the fifth metatarsal are elements in this process.

All one has to do to appreciate how successful the effort at reparation has Ix^en is to put together the bones of a foot, leaving out the inner and middle cuneiforms and the greater part of the external one, and to compare the curves made by lines connecting the heads of the metatarsals or of the terminal phalanges with similar lines drawn on these illustrations. The rnore one thinks of it, the more

D Suppression of the Cuneiform Bones. 533

remarkable does it appear that so good a foot should have been formed under the circumstances. Through what agency has this been brought about ? It seems to me that it is a clear instance of the act of the vital principle regulating growth, and repair; the same by which the amputated leg of a newt is reproduced. I incline to agree with Driesch that it should not be called a vital energy, for it is rather something regulating the energy. He would call it entelechy; a something "which bears the end in itself."^ I find no fault with this, but prefer the other term.

Received for publication July 9, 1909.

The Science and Philosophy of the Organism. By Hans Drlescli, Ph.D. Gifford lectures, 1907-08, Vol. T, p. 144.



I. Primates, Cabnivora, Rodentia, Uxgulata and Marsupialia.


CHARLES F. W. McCLURE AND, CHARLES F. SILVESTER. From the Laboratory of Comparative Anatomy, Princeton University.

With Three Text-Figures and Ten Plates.

Huntington and McClure^ have shown in the adult eat {Felis domestica) that the communication between the lymphatic system and the systemic veins may normally occur on each side of the body, within either one of two or within two typical districts. These two districts include, approximately, the angle of confluence formed by the union of the external and internal jugular veins (common jugular angle) and the angle of confluence formed by the union of the external jugular^ and subclavian veins (jugulo-subclavian angle). An examination of a large number of adult cats proved conclusively that neither one of these two districts predominates as the place of communication between the lymphatics and the veins, but that either

IIuntiDgton and McClure, The Anatomy and Development of tlie Jugular Lymph Sacs In the Domestic Cat (Fells domestica). A paper read before The Association of American Anatomists in Chicago, in 1907, published in The Anatomical Record, Volume II, 1908, and soon to be published in a more complete form in The American Journal of Anatomy.

"This vein, strictly speaking, is a common jugular vein in the cat, but on account of its large size, as compared with the internal jugular, is usually spoken of as the external jugular vein of which the internal jugular Is a tributary.


D Lymphatico- Venous Communications. 535

one of the two or both may serve equally in this capacity and for this reason both districts must be regarded as constituting normal points of communication between the lymphatics and the veins.

In following the development of the jugular lymph sacs in the embryonic cat Huntington and McClure were able to establish the basis upon which the duplex character of the lymphatico-venous communication in the adult rests. They found that the right as

Internal J

Lymph Sao

JagaUur .

Juguur V

SubolATian Thyrootnri

Lympluuio SuboUTian ,

Innomuate V

Text-fig. I. — A reconstruction of the left jugular lymph sac of a 11 mm. cat embryo (Felig domestica) showing the relations of the thyrocervical artery to the jugular and subclavian approaches through which the two typical adult communications are established between the lymphatics and the veins. Ventral view. Drawn from a reconstruction made by Huntington and McClure after the method of Born.

well as the left jugular lymph sac in the embryonic cat invariably presents two caudally directed processes or prolongations which they termed, respectively, the Jugular and Subclavian approaches (Textfig. I). These two processes, on each side of the body, are directed

D 536 Charles F. W. McClure and Charles F. Silvester.

toward and approach the district of the common jugular angle and the district of the jugulo-subclavian angle, respectively, and they observed that it is through either one of the two or through both of these processes that the adult communication is established, a circumstance which accounts not only for the presence of a double communication in the adult cat but also establishes it as a character of morphological significance.

In view of the uniform conditions which prevail in the adult domestic cat concerning the presence of two typical districts of lymphatico-venous communication on each side of the body, the present writers have undertaken to determine to what extent this same uniformity may prevail in adult mammals in general.

We have thus far examined twenty-five (25) species distributed among fifty (50) adult mammals (24 primates, 4 carnivora, 12 rodents, 5 ungulates and 5 marsupials). These mammals Avere chosen at random from the Princeton Collection so that the conditions observed in them represent fairly well the average conditions which one might expect to find in any other similar group chosen in the same manner. The lymphatic system of each mammal was injected with gelatine and then carefully dissected out in the appropriate regions on each side of the body. A drawing to scale was made of each dissection to facilitate comparison. All of the figures in this paper therefore represent accurately the arrangement of the lymphatics as met with in the regions of communication and it is worthy of notice that the lymphatics present a marked variability, more so than the veins, not only in the different species examined but among different members of the same species. These variations will not be dealt with to any extent in the present paper except in so far as it becomes necessary to speak of them in connection with the communications which exist between the lymphatics and the veins.

We may state at the beginning that we are warranted in drawing the conclusion from the adult mammals thus far examined that the lymphatic system normally communicates with the veins in these forms as in the adult cat, either at one of two or at two typical districts (common jugular and jugulo-subclavian districts) and that a commimication at the two typical districts is the commonest of theIC

D D Lymphatico- Venous Communications.


three possibilities which may normally occur on either side of the body. This is clearly .shown in the following table (Table I) in which it is seen that a communication between the lymphatics and the veins occurred at the two typical districts on the right side of the body in sixty-two (62) per cent and on the left side in seventyfour (74) per cent of the mammals examined; also, when a communication was present, on either side, at only one of the two typical districts, it occurred more frequently at the common jugular than at the jugulo-subclavian district.


Showing the Relative Frequency with which the Lymphatico- venous ck>mmunicati0n8 occub on each side of the body at ettheb one of the two ob at both of the typical distbicts of communication in the Fifty Mammals undeb Consideration.

Communication at Common Jugular District only. . .

Communication at JuguloSubclavian District only. .

Right Side.

Left Side.

2 ' .2

t ! g

5 « 

1 3


Communication at both

Districts ;14' 3 8' 2 4 31 ! 62 17 ' 4

— —


a> &»

t ? I •= . ft

S , "O ) ^ I «  t ' O C I t


1 116 32 I 7 01 3 2

,12 ' 24

3 6' O' 1 Oi 0' 1 , 2

I I 1

81 3 5

37 74

Text-fig. II is a diagram of the precaval system of veins showing the two typical districts of lymphatico-venous communication, encircled by rings, which are met with on each side of the body. Since we have found that the general plan of the communication is fimdamentally the same on each side of the body, in that it normally occurs at either one of the two or at both of the typical districts of communication (Table I), our interest has been largely centered upon the determination of the combinations in which the communications may occur in the fifty mammals examined when both sides of the

D 538

Charles F. W. McClure and Charles F. Silvester.

body are taken into consideration. Since a communication may be normally established in one of three ways on each side of the body, it is evident, when both sides of the body are considered, that the lymphatico-venous communications may occur in nine possible combinations and that each combination may be regarded as a type of

Zai«rB«l Jncater V.

«— @i

Text-fio. II. A diagram of tlie precaval system of veins showing the two typical districts of lymphatico-veiious communication on each side of the body and the nine possible combinations in which communications may occur when both aides of the body are taken into consideration. Ventral view.

communication which characterizes the lymphatico-venous communication of a particular individual.

Each of the nine possible combinations is indicated in Text-fig. II by a series of arrows which radiate from a small circle enclosing the Roman numeral (I-IX) applied to the combination and is also

D Lymphatico- Venous Communications.



Showino the Nine Possible Combinations (Types of Lymphatico- venous Communication) in which Lymphatico-venous Communications may be Normally Established in an Individual when Both Sides of the Body abe taken into Consideration.

Right Side

Left Side

TvnA Common Jugular Jugulo-Sub- I Common Jugular Jugulo-Sub^' District clavlan District, i District. I clavlan District.





III. ,


1 IV. j

V. ■




VII. 1













Tlie word TAP, under the district designated, indicates the presence of a communication at this district between the lymphatics and the veins.


Showing the Distribution among the Fifty Mammals Examined of the Nine Possible Combinations or Types of Lymphatico-venous Communication which may be Normally Established in an Individual.




Carnivora 3
















1 Number and Percentage




of Ind

ivid per



29 or 58



1 9 or 18



6 or 12

2 or 4

2 or 4

1 1 or 2

1 1 or 2




' 50

D 540 Charles F. W. McClure and Charles F. Silvester.

shown in Table II, in which the word tap indicates the presence of a communication under the district of communication designated.

The following figures illustrate examples of the diflFerent types of lymphatico- venous communication as met with among the fifty mammals examined :

Type I, Figs. 6, PI. Ill (Papio porcarius), 16, PL VI (Putorius vison), 22, PL VII (Cavia porcellus), 26, PL VIII (Sus scrofa domestica) an'd 31, PL X (Didelphys virginiana) .

Type II, Figs. 12, PL V (Anthropopithecus troglodytes) y 21, PL VII {Lepus cuniculus) and 27, PL IX {Sus scrofa domestica).

Type III, Figs. 5, PL II {Papio anubis), 15, PL VI (Canis familians) and 30, PL IX (Didelphys virginiana).

Type IV, Figs. 4, PL II (Papio anubis) and 1, PL I {Nycticebus tardigradus) .

Type V, Figs. 13, PL V {Anthropopithecus troglodytes)^ and 8, PL III (Macacus rhesus).

Type VI, Fig. 19, PL VII (Lepus cuniculus).

Type VII, Fig. 20, PL VII (Lepus cuniculus).

As shown in Table III the lymphatico-venous communication of every mammal examined fell within one of these nine combinations (Types I-IX). This circumstance, together with the fact that in twenty-nine (29) individuals or fifty-eight (58) per cent of those examined (Table III) the communication occurred on both sides of the body at the two typical districts of communication, and that this type of communication was commonly met with in each of the five orders of mammals examined (Type I, see Text-fig. II and Table III), indicates that the embryonic basis for the establishment of two typical communications on each side of the body must be fundamentally and potentially the same in the fifty mammals examined by us as that described by Huntington and McClure for the domestic cat. Although our present deductions do not lead us beyond a consideration of the conditions observed in the fifty mammals (25 species) it is evident, if this conception of two primary and

■It is significant to note in the two chimpanzees examined, in which the precaval system of veins resembles that in man, that Types II and V are represented.

D Lymphatico- Venous Communications. 541

typical districts of communication on each side of the body can be generalized for mammals as a whole, that a consistent description of the adult lymphatico-venous communications should rest upon a morphological interpretation of their development, and not upon the ill-defined and variable conditions which are at present described in works of anatomy as constituting the points at which the lymphatics communicate with the veins.

It is interesting to note that Type VIII, in which a communication occurs on both sides of the body only at the jugulo-subclavian district (angle of confluence formed by the union of the external jugular and subclavian veins), a region commonly assigned by anatomists as the point at which the thoracic and right lymphatic ducts tap the veins, vxis not met with in a single one of the mammals examined.

As shown in Table I and Table III, when a communication is present on each side of the body at only one of the two typical districts it is usually found at the common jugular (Type II) and not at the jugulo-subclavian district.

The assignment of the jugulo-subclavian angle as the point at which in general the lymphatics communicate with the veins in the adult, appears to us to be a correct interpretation for the cat and for the mammals we have thus far examined, only in the sense that the two districts of confluence formed by the union of the external and internal jugular and by the union of the external jugular and subclavian veins, respectively, be regarded as constituting a single district of communication.

Tables IV to VIII, inclusive, show in tabulated form the different species of mammals examined by us as well as the type of lymphatico-venous communication presented by each individual.'*

The word tap in each column, under the district designated, indicates the presence of a communication at this district between the lymphatics and the veins, while the absence of the word tap indicates that a communication is wanting.

The number after the sex sign of each species indicates the cata

The nomenclature of the species mentioned in this paper follows that given in Trouessart's Catalogus Mammalium, Quinquennale Supplement. 1904.

D 642

Charles F. W. McClure and Charles F. Silvester.


Nyciicebus tardigradus

9 2374 (Fig. l.PLI.).... Callithrix jacchus

9 2457

Cebus capucinus

9 2481 (Fig.2, PL I.)... Cehu8 hypoleucus

9 2472

Ateles hybridua


Ateles hybridua


Ateles vellerosua

9 2476 (Fig. 3, PI. I.)... Papio anubis

9 2368 (Fig. 4, PI. II.).. Papio anubis

c^ 2432 (Fig. 5, PI. II.) . . Papio porcarius

d' 2453 (Fig. 6, PI. III.) . Papio porcarius


Macacus speciosus

d 2376 (Fig. 7, PI. III.) . Ma>cacu8 irhestis

9 2487 (Fig. 8, PL III.) . Afacacus wmestrinus

d'2458(Fig. 9, PI. IV.).. Alacaciis nemestrinus

9 2459

Macacus nemestrinus

d* 2461

Macacus nemestrinus

d 2483

Cercocfbus fuliginosus

9 2394

Cercopithecu^ callUrichus

d 2434

Cercopithocus cMlitrich u s

d 2464 (Fig. 10, PI. IV.). Cercopithecus cnllitrichus

9 2477(Fig. 11,P1. V.).. Cercopithecus cnllitrichus

d 2478

Anthropopithrnis troglodytes

d2467(Fig. 12, Pl.V.).. A nthropopithpcus troglodytes

9 2468(Fie. 13, Pl.V.)..

Right Side.


Common Jugular District.















































Jugulo Subclavian











Left Side.

Common Jugular IHstrict.


Jugalo Subclavian






D Lymphatico- Venous Communications.




Right Side.

Left Side.

Common Jugular DiBtriot.


Jugulo Subclavian


Common 1 Jugular 1 District.





Jugulo Subclavian


Felis domestica

d'2452(Fig. 14,P1.V.).. Canis familiaris

$2451 (Fig. 15, PI. VI.). Putorius vison

6^2384 (Fig. 16, PL VI.). Mephitis putida

6^2362 (Fig. 17, PL VI.).







Right Side.

Lepus cunictUus

(^2363 (Fig. 18, PL VII.)., Lepus cuniculus

(5^2364 (Fig. 19, PL VII.). Lepus cuniculus

d^2366 (Fig. 20, PL VII.). Lepus cuniculus


Lepus cuniculus

$2431 (Fig, 21, PL VIL). Cavia porcellus


Cavia porcellus

6^2360 (Fig 22, PL VII.). Fiber ziheihicus


Fiber ziheihicus

(^2365 (Fig. 23, PL VIII.). Fiber zibethicus


Marmota monax

$2469 (Fig. 24, PL VIII.). Seiurus hudsonius

$2466 (Fig. 25, PL VIII.).

Type. ,

Common Jugular District.











Jugulo Subclavian






Left Side.

Common Jugular District.

Jugulo Subclavian























D 544 Charles F. W. McClure and Charles F. Silvester.


! Type.

Su8 scrofa domestica

9 2436 (Fig. 26, PI. VIII.). Sus scrofa domestica

9 2437 (Fig. 27, PI. IX.) . Sti8 scrofa domestica


Sus scrofa domestica


Stis scrofa domestica

9 2441 (Fig. 28, PI. IX.)..


I "



1 I

Richt Side.

Common Jugidar District.


Juinilo Subolavi&n


Left Side.



Common Jugular District.


Juffulo Subclavian







Riffht Side.

j Left Side.

Common Jugular District.

Jugulo Subclavian


Common 1 Jugular 1 District.

Jugulo Subeiavian


Didelphys virginiana

d^2388(Fig. 29,P1.IX.)... Didelphys virginiana












Didelphys virginiana

9 96 (Fig. 30, PI. IX.).... Didelphys virginiana





Did4phys virginiana

d^5r(Fig 31,P1.X.)


logue number of the individual in the Princeton Collection. The individuals figured in this paper are also designated in these tables.

Before proct^eding to a detailed description of the lymphaticovcnous communications observed in the fifty mammals, we will first consider the general character of the two typical districts of lymphatico-venous communication.

Text-fig. Ill is a composite diagram of the prccaval system of veins constructed on the basis of the conditions actually observed in the fiftv mammals under consideration and should be constantly referred to in connection wath the following description of the external and internal jugular and cephalic veins.

D Lymphatico- Venous Communications.


1. The External and Internal Jugular Veins.

The external jugular may be larger (Fig. 16, PI. VI) or smaller (Fig. 12, PI. V) than the internal jugular vein, or, of practically the same size (Fig. 5, PI. II).

The common jugular district of lymphatico-venous communication may lie on the same level and in close proximity to the jugulo-sub


Exteraa.1 Jugular V

Internal Jugular V

LEFT External Jugular V

Cephalio V.


Ccpbalio V.


Text-fig. III. — A composite diagram of the precaval system of veins constructed on the basis of the conditions observed in the fifty mammals examined and with especial reference to the relations of the two typical districts of lymphatico-venous communication to each other. Ventral view.

clavian district as in Papio porcarius (Fig. 6, PI. Ill) or, as is usually the case in the domestic cat, it may lie somewhat cranial to the jugulo-subclavian district as shown on the right side of Fig. 14 (PI. V) and on the right side of Text-fig. III. In either case, however, the two districts of communication are separated from each other by the external jugular vein.

In the domestic cat the internal jugular vein at times gives up

D 540 Charles F. W. McClure and Charles F. Silvester.

its original connection with the external jugular and then drains into the innominate through the inferior thyroid vein. In a case obser\'ed by us in which this occurred (Fig. 14, PI. V, left side), the common jugular district of lyraphatico-venous communication was not transferred to the new angle of confluence formed by the union of the inferior thyroid and innominate veins but remained on the external jugular at the point where, as on the right side of the same individual (Fig. 14, P1..V, right side), the internal jugular normally joins the external jugular vein.

2. The External Jugular and Cephalic Veins.

As is well known, the cephalic vein presents considerable variability in its relations to the external jugular and subclavian veins, not only in mammals in general but even upon opposite sides of the same individual. The complex of vessels connected with the external jugular and subclavian veins on both sides of Text-fig. III. represents a composite picture of the conditions observed in the fifty mammals examined by us and is an attempt to explain, from the standpoint of comparative anatomy, the variable conditions of the cephalic vein, as well as those sometimes presented by the external jugular, the transverse scapular and deep transverse scapular veins. One might infer from a study of comparative anatomy that this complex of veins may possibly represent a ground-plan arrangement and that the elements of which it is composed are capable of serving in variable capacities, not only in different individuals, but even upon opposite sides of the same individual.

In explanation of the basis upon which the above observations are made the following description of the conditions actually met with, to be compared with the diagram (Text-fig. Ill), is given.

One of the commonest terminations of the cephalic vein met with is its connection with the external jugular, as in Felis dome^tica (Fig. 14, PI. V), Cams familiaris (Fig. 15, PI. VI), Putorius vison (Fig. 16, PI. VI), Fiber zihethicus (Fig. 23, PI. VIII), Cat>ia parceUus (Fig. 22, PI. VII), and Sits scrofa dom^stica (Fig. 26, PI. VIII), where it is commonly established through the vessel C or D in Text-fig. III. It may be formed, as in Maca^us rhesus (Fig. 8,

D Lymphatico- Venous Communications. 547

PL III), Cercopithecus callitrichus (Fig. 11, PI. V) and Cebus capucinus (Fig. 2, PI. I), through the retention of the vessels B and D in Text-fig. Ill, as is also the case in Didelphys virginiana (Fig. 31, PI. X, left side). It may also be formed in Didelphys virginiana (Fig. 31, PL X, right side), as is more commonly the case,^ through the retention of the vessels A and D in Text-fig. III. In Sciurus hudsoniiLs (Fig. 25, PL VIII) it may be formed through the retention of the vessel C in Text-fig. Ill, but vessels B and D may also be present where they, together with the external jugular vein, form a loop or ring through which the clavicle passes. In one specimen of Ateles vellerosus examined (Fig. 3, PL I), in which there is an asynmietrical arrangement of the veins on opposite sides, the cephalic vein is formed through the retention of the vessel A in Text-fig. Ill on the right side and by the vessel B on the left. Although the conditions observed in Ateles are probably abnormal, they sen-e as a good general illustration of the variability manifested by termination of the cephalic vein.

The elements which compose the complex of veins ordinarily entering into the formation of the cephalic also appear to be capable of functioning as the terminals of veins other than the cephalic. For example, in Lepus cunicuhis (Fig. 20, PL VII) the transverse scapular vein appears to have been established through the vessel B in Text-fig. Ill which in Ateles vellerosus (Fig. 3, PL I, left side) and in part in Macacus rhesus (Fig. 8, PL III) constitutes the terminal of the cephalic vein. Also in Marmota monax (Fig. 24, PL VIII), as shown by its relation to the clavicle, the external jugular vein has largely transferred its drainage from its conventional pathway to the vessel B in Text-fig. Ill which now functions as the chief terminal of the external jugular vein.

At first sight one might regard the boundaries of the two typical districts of lymphatico-venous communication as being fundamentally modified in accordance with the variations presented by the veins in the region of the lymphatico-venous communications. Such does

•^C. F. W. McClure, A Contribution to the Anatomy and Development of the Venous System in Didelphys marsuplalis. Part I, Anatomy. Amer. Jour, of Anatomy, Vol. II, 1903 (see Fig. II, p. 377).

D 548 (^harles F. W. McClure and Charles F.. Silvester.

not appear to be the case, however, since the primary relations established by the union of the external and internal jugulars and by the external jugular and subclavian veins, respectively, definitely mark out the two typical districts of communication as evidenced by the constancy with which communications occur in connection with these districts, regardless of the variation presented by the veins.

We pass now to a consideration of the general character of the lymphatico-venous communications as met with in the fifty mammals examined.

When a communication occurred on either side, at either one of the two typical districts, it was usually single in character as shown in Table IX. In only a few instances were the lymphatics found to communicate with the veins within either of the two typical districts by more than one opening. A multiple communication between the lymphatics and the veins was found at the common jugular district of communication in Ateles vellerosus (Fig. 3, PI. I, right side), in Cercopithecus callitrichus (Fig. 11, PI. V, both sides), in Felis domestica {Y\g. 14, PI. V, right side), in Macacus nemestriuus (Fig. 9, PL IV, both sides) and the the jugulo-subclavian district of communication in Cards familiaris (Fig. 15, PI. VI, left side) and in C ercopiihecus callitrichus (Fig. 10, PI. IV, right side).

As shown in the following table, only twelve (12) instances of a multiple communication at the two typical districts were met with and it occurred more frequently on the right than on the left side of the body.

One hundred and seventy-eight (178) points of communication between the lymphatics and the veins were observed by us on both sides of the body in the fifty mammals under consideration (89 on each side). Of these, one hundred and forty-two (142) were found either at the angles of confluence formed by the union of the external and internal jugular and by the union of the external jugular and subclavian veins, respectively, or w-ere included within these two angles, as illustrated by Figs. 12, PI. V, 13, PL V {Anthropopithecus troglodytes), 8, PL I (Ateles vellerosus), 7, PL III

D Lymphatico- Venous Communications.


{Macacus speciosus), 8, . PI. Ill (Macacus rhesus), 1, PI. I {Nyticehus tardigradus) y 4, PL II, 5, PL II (Papio anubis), 22, PL VII (Cavia porcellus), 19, PL VII, 20, PL VII, 21, PL VII (Lepus cuniculus), 24, PL VIII (Marmota monax), 25, PL VIII {Sciums hudsonius) and 27, PL IX (Sus scrofa domestica).

TABLE IX. Showing the Pbedominance of a Single over a Multh^le Communication


Right Side.

Left Side.

Single Communication at Common Jugular District 19

Multiple Communication at Common Jugular District

Communication Wanting at Common Jugular District


11! 3 5

[ I 2

4 12 1 5; 5


! I

o I 6

41 22 4 11

6i' 2!

3|. ij^ 60 ' 24 4 I 12

5i 6 47 2

5 6

1 50

Single Communication at Jugulo- 1

Subclavian District 13 ' 3 9 2,4

I , I

Multiple Communication at Jugulo-' ! i



Subclavian District 3

i 1

Communication Wanting at Jugulo- I

Subclavian District ISil 3 31


I 24 4 I 12 . 6 5


1 3 9




37 1






4 12

i 5 5


Three (3) points of communication met with did not fall either within the angle of confluence formed by the union of the external and internal jugular nor within the angle formed by the union of the external jugular and subclavian veins, but, as in Figs. 2, PI. I {Cehus capucinus, right side), 28, PI". IX (Sus scrofa domestica.

D 550 Charles F. W. :McClure and Charles F. Silvester.

left side) and 14, PL V {Felts domestica, right side), they occurred on the veins slightly caudal to the angle of venous confluence (caudal to the common jugular angle in Cebus and Felis, and to the jugulosubclavian angle in Sus).

Also, in addition to these three points of conununication just mentioned, thirty-three (33) others were met with which did not fall within the common jugular nor jugulo-subclavian angles but which were dorsally or ventrally situated on the veins in dose proximity to either one of the two of these angles.

Of these dorsal and ventral communications between the lymphatics and the veins, the former proved to be the more common of the two.

A dorsal conmiunication between the lymphatics and the veins on both sides of the body at the conmion jugular district of communication is shown in Figs. 15, PI. VI {Cards familiaris) and 17, PL VI (Mephitis putida). A dorsal communication between the lymphatics and the veins on the right side of the body at the common jugular district of communication is shown in Figs. 10, PL IV {Cercopithecus callitrichus) and 9, PL IV {Macacus nemestrinus) and at the jugulo-subclavian district of conmaunication in Fig. 23. PL VIII (Fiber zibethicus). A dorsal communication between the lymphatics and the veins on the left side of the body at the jugulosubclavian district of communication is shown in Figs. 29, PL IX (Didelphys virginiana) and 26, PL VIII (Sus scrofa domeslica).

A ventral communication between the lymphatics and the veins on both sides of the body at the common jugular district of communication is shown in Figs. 11, PL V (Cercopithecus callitrichus) and at the jugulo-subclavian district in Fig. 17, PL VI (Mephitis putida). A ventral communication between the lymphatics and the veins on the left side of the body at the jugulo-subclavian district of communication is shown in Fig. 6, PL III (Papio porcarius) and on the right side of the body in Fig. 11, PL V (Cercopithecus cailitrichus) .

In consideration of the large number of commimications observed within the common jugular and jugulo-subclavian angles (142), the presence of these thirty-six apparently variant forms appears to

D Lymphatico- Venous Communications. 551

find its explanation in the circumstances that the communications established between the embryonic jugular lymph sac and the veins are not confined exclusively within these two angles of venous confluence but may vary about the same in a sphere which we have designated a district of communication (Text-figs. II and III). The circumstance that angle, dorsal, and ventral communications may be found in the same individual, in which they hold definite relations to either one of the two typical angles of venous confluence and that, in some cases, dorsal and ventral communications are alone present (Fig. 17, PI. VI), seems conclusive evidence that all of the points of communication observed by us between the lymphatics and the veins must have been established in the embryo in fundamentally the same manner as in the domestic cat and in definite relation to two typical districts of commimication.

Received for pubUcation July 9, 1909.


Figs. 1 to 31, inclusive, were drawn to scale from dissections of adult mammals and represent ventral views of the veins and lymphatic vessels in the regions where the lymphatics communicate with the veins.

The veins are draAvn in outline, the lymphatics are colored. The word TAP indicates a point at which the lymphatics communicate v^th the systemic veins.

The name of the species represented is given under each figure while a complete list of the mammals dissected and studied in connection with this paper may be found in Tables TV, V, VI, VTI and VIII.



Int«maL Jugular V



SuboUTian v


AiygM V.

Fio. 1 (Type IV)

Slow Loris ?

yifcticchus tanlipradus, Linn.



BIQBT External Jugular V

Internal Jugular V

Ceptaalio V

LEFT External Jugular V

Fig. 2 (Type I)

Capuchin Monkey $

Cchus capucinus, Linn.

AIQflT External Jugular V


Internal Jugular V


LEFT External Jugular V



Thoraoio Duot

SubolaTian V

'Bronchomediastinal Trunk

Fig. 3 (Type I) Spider Monkey $ Ateles vellerosus, Gray


Fic. 4 (Type IV)

Aiiubis Baboon 5

Papio anuhis F. Cuv.

The Anatomical Record. — Vol. III. No. 10.

D BIOflT External Jogalar V

Internal Jus^ilar V


qplUfcUO V.


Aaygos V

Thoraoio Dnot.

Thoracic Duct.

Fig. 5 (Type III)

Auubis Baboon (^

Papio antihis, F. Cuv.


Fig. 6 (Type I)

Chacama Baboon J

Papio porcarius, Bodd.


TfaoTMsie Doc

Fig. I


Thb Axatomical Record. — Vol. Ill, No. 10.


Internal Jugnlar V

aiOHT External Jagular V.

LEFT External Jugular V.

Thoracic Duct.



«, Cuv.

Fig. 8 (Type V)

Rhesus Monkey J

Macacus rhesus, Audebert

Cephalic V


Fig. 4 (Type IV)

Anubis Baboon 5

Papio anubis F. Ciiv.

Tub Anatomical Uecord. — Vol. III. No. 10.


RIOflT Bztenud JogaUu* V

Internal Jugular V

\ _

LBFT External Jugular V

BobolaTiaa 1

Thoraoio Dnot.

Tboraoio Duot

Fig. 5 (Type III)

Auubis Babo(3ii ^

Papio anuhis, F. Cuv.


BIQflT Bztanial Jagalar V

Inumal JusuUr V

Cephalic V.

Fig. 6 (Type I)

Chacairia Baboon J

Papio porcariHs, Bodd,


Thoracic Dtt(

Tub Anatomical Record. — Vol. Ill, No. 10.

Fig. 7


Miicacus « 


Internal Jugular V

EIGHT External Jugular V.

LEFT External Jugular V.

Thoracic Duct.

II) K«, CuV.

Fig. 8 (Type V)

Rhesus Monkey J

Macacus rhesus, Audebert

Cepbalio T

)igitized by IC


BIOfiT Bxtemal Jugulftr V

IsEFT MxtantalJngalmr V

Fig. 9 (Type II)

PIg-talled Macaque c?

Macacus nemestrinuSy Linn.

The Anatomical Record. — Vol. Ill, No. 10.


Internal Jugular V

BIGHT External Jugular V

LEFT ternal JuguJar T

Subclavian V.

- Thoracic Duel.

Aaygot V.

Fig. 10 (Type I)

Green Monkey J

Cercopithecua callitrichus, Geoflf.


Exeernol JnguUr V.


P y Extcnial Jogalmr r

SnbcUnaa T.


Internal Jugular V.

RIGHT External Jugular V,

LEFT External Jugnlar V.

Thoracic Duoi.

Fig. 13 (Type V)

Chiiupanzee $

Anthropopithecus troglcHfytes, Linn.





D Internal Jugular V.


LEFT Sztemal JugnUr V.

Subclavian V.

ymph Nodes



(Type II) panzee d iS troglodytes, Linn.

)igitized by IC


aiOBT Sxtonul Jugular V

lAtMiial JagnlAr V.

olftvian V.

o IHiet.


-— Aorte

Fig. 16 (Type I)

Mink c?

Putorius vison, Brlsson





LEFT Extornal Jugular V.

Subclavian V.

Thoracic Duct.

'. 15 (Type III)

Dog $ f familiarise Linn.





BIOBT External Jugular V

LEFT External Jngalar V

Internal Jugular V.

Trans. Sea


Internal Jugular V.

BIGHT Ti-ana. Soap. V.

I ^4^ PrMsava

Thoraoio Duct.

Fig. 18 (Type I) Rabbit c? Lcpus cuniculus, Linn. TOMicAL Record,— Vol. Ill, No. 10.


alar V


V uiceniBi

Fig. 19 (Type VI)

Rabbit d

Lepus cwiiculus, Linn.

Subola Thor*


D Aortaj

Fig. 20 i

Lcpus rutm




BFT rogular \

Fig. 21 (Type II) cephaUo v Rabbit $

^'•^^ .Trana. Scap. V.^^PW* CUniCUlUS, Linil.

Subclavian V.


LEFT External Jugular V

Cephalic V


Lymph Hodes

VII) > Linn.

Subclavian V.

Subclavian V

Thoracic Duot


Fig. 22 (Type I)

Guiiiea-Pig ^

Cavia porccUus, Linn.


BIGHT Eztornal Jugular V.

Cepbalio V.

SIGHT External Jngalar V.

Intornal Jugular V

Captaalio V.

Cepbalio V

Fig. 24 (Typk I)

Ground Ilog $

Marmota momix, Llnii

LEFT External Jugular V

Fig. 23 (Type I)

Musk' Rat c?

Fiber zibethUvH, Linn.

Thoraoic Duot

Asygoa V

'TOMicAL Record. — Vol. Ill, No. 10.



LETT sternal Jugular V.

Uo V.



BIGHT Sztomal Jugular V


LEFl Szternal Jn

Internal Jugular V


Cepbalio V.

phaiiov. Fig. 26 (Type I)

Pig 2 S!u8 scrofd domcstica. Gray


Fig. 25 (Type i) Red Squirrel $ iurus hudsonius, Erxleb.


uwrcMJAU J^UO(.




Internal JagnUr V External JaculAr V

Bnbolai .^



rig ?

Su8 8(rofa domes tica. Gray

LBPT BIGHT External Jugular V


nidi The Anatomical Kkcord — Vol. 111. No. 10.

D RIGHT Internal Jugular V

J)i<Jeli)hifS rirgiiiiaiia. Km*

Plate IX.



Plate X.

BIGiiT BxMnud JofiiUr V

LEFT Bxt«rn«l JoguUr V




Fig. 31 (Type I)

Opossum (J

Didelphys virginiana, Kerr

The anatomical Record. — Vol. Ill, No. 10.


Doctor Irving Hardesty has been appointed Professor of Anatomy in Tulane University. The department, which formerly included only gross anatomy, will now have charge of histology as well. Professor Hardesty will be assisted by Assistant Professor Henry W. Stiles, M.D., formerly of the University of Michigan ; Assistant Professor Henry Bayou, A.M., M.D., and Mr. H. H. Bullard, A.B., M.S., formerly of the University of Missouri, who will be Instructor in Anatomy. The following physicians will be assistant demonstrators : Dr. Sidney P. Delaup, B.Sc. ; Dr. Marion S. Souchon ; Dr. John F. Oechsner; while Dr. Charles A. Wallbillich, Dr. John F. Points and Dr. M. H. McGuire will be junior assistant demonstrators. Professor Hardesty, Professor Stiles and Mr. Bullard, as well as the technical assistant, Mr. Linstaedt, will give their entire time to the department. Dr. Edmond Souchon has been made Professor Emeritus of Anatomy and Curator of the Museum.

Eichard E. Scammon, A.M. (University of Kansas), has recently been awarded the degree of Doctor of Philosophy by the Faculty of Arts and Sciences of Harvard University, for studies in medical sciences,- particularly in embryology. He is thus the first candidate to avail himself of the new arrangement whereby this degree may be obtained by study and investigation conducted in the Medical School. Dr. Scammon's thesis will be published as the Normentafel zur Entwichlungsgeschichte des Squalus acanthias in Professor KeibePs series.




Vol. III. NOVEMBER, 1909 No. 11





H. VON W. SCHULTE and FREDERICK TILNEY. From the Anatomical Laboratory of Columbia University,

With EleVek Figures.

This paper is an attempt to formulate a few general propositions having reference to the organization of the venous system as a whole, and further to indicate some of the underlying hydrodynamic factors incident to the formation of the major lines of drainage. Broadly speaking, the veins of the mammal between the peripheral capillaries and the heart fall into two fairly definable regions, a central district of large venous trunks and a distal region of smaller plexiform vessels. The circulation in the adult differs from that of its embryo largely in the reduction of these plexuses to form larger single veins, the zone of plexiform veins retreating farther toward the periphery.

The result of the substitution of large trunks for plexuses is the reduction of the impediment offered to the venous return by surface friction, consequently either a reduction of cardiac work or, the work performed by the heart remaining the same, a more rapid circulation and potentially a higher rate of metabolism.

We conceive that it is the general competency of the circulation as a whole, rather than the topographical situation of the lines of


D 556 H. von W. Schiilte and Frederick Tilney.

venous drainage, which is of evolutionary significance. Natural \ selection would easily be imagined to operate to destroy an animal whose venous system offered too great a resistance to the flow of the blood, while it is by no means obvious that, given a circulatory competency, the exact topography of a vein can often be of moment to its possessor. The high variability of veins is common knowledge and lends support to this view.

A cursory examination of the variations of the venous system in the three forms most extensively studied, man, the opossum^ and the cat,^ suffices to show that while variations in the situations of the individual veins and even in the topography of the major trunks are wide in range and frequent in occurrence, yet points at which plexus formations replace single veins are subject to relatively little change — ^the anatomic plan varies widely within limits rigidly fixed by physiologic efficiency. It might then fairly be expected that the evolution of the venous system, in its broad outlines, should be in the direction of organization and higher physiological efficiency, rather than the formation of a series of morphologic types. From this standpoint the venous system of the monotremes appears to us a generalized type of low organization, comparable to the embryonic veins of marsupials and placentals.

In Omithorhynchus the plexiform arrangement involves even the postcavflB to the renal level (Figs. 1 and 2). The two vessels are connected dors'ally by massive anastomoses. Traced caudad to about the lumbo-sacral junction each postcava resolves itself into two extensive plexuses, one dorsal and the other ventro-mesial to the psoas minor. At the lateral border of the muscle a wide channel connects the two plexuses ; the dorsal plexus is composed of tributaries, enumerated cephalo-caudad as follows:

1. An ilio-lumbar plexus.

McClure, C. F. W., *03. "A Contribution to the Anatomy and Development of the Venous System of Didelphys marsupialis." Part I. Amer. Jour. Anat, Vol. II, No. 3.

•Darrach, W., '07. "Variations in the Postcava and its tributaries as observed in 605 Examples of the Domestic Cat." Amer. Jour. Anat., Vol. VI, No. 3, page 30.


Organization of the Venous Eetum. 557





Fig. 1. Ornithorbynchus paradoxus. From a dissection in the study collection of the Department of Anatomy, Columbia University. Showing the postcavfe and their tributaries Ventral view, semi-diagrammatic. The aorta and branches have been omitted to allow the dorsal anastomoses between the postcavfie to come into view. The psoas minor of the left side Is also omitted.

D Fig. 2. Ornithorhynchus paradoxus. Dorsal view. From a dissection in tlie study collection of tlie Department of Anatomy, Columbia University. After injection the vessels were removed in mass.

D Organization of the Venous Return. 559

2. A large trunk from the dorsum of the thigh (saphena parva).

3. Veins coming from the fat pad of the groin.

4. A femoral plexus.

It is noteworthy that the drainage from the panniculus and the subpannicular fat of the trunk is accomplished by large veins (vide supra 2 and 3), while that from the deeper parts is given by the plexiform vessels accompanying the arteries. The ventro-mesial plexus is composed from without inward of the following plexiform vessels :

1. A deep epigastric which receives a considerable plexus from

the thigh.

2. An internal iliac, receiving the obturator and vesico-pudendal


3. A caudal plexus.

Across the median line the plexuses of the two sides anastomose freely dorsad of the aorta, and ventrally more or less completely across its large branches.

It is, of course, arguable that the circulatory system of Omithorhynchus is not primitive, but highly specialized in adaptation to the animal's semi-aquatic habits. Besembling, as it does, the venous plexuses of the cetacea, though it differs in degree and constitution, no one would deny that it stands in relation to the creature's habits. But it by no means follows that it is highly specialized, since a retention of the embryonic characters in the adult may have as high adaptive value as the development of new characters. In the extensive plexus formation, the thin walls of the vessels and the excess in lumen of the veins over the arteries, we have a picture closely resembling the vessels of the embryos of higher forms, — a system of vessels in which a relatively small column of blood reaches the capillaries through the small and often plexiform arteries, accumulates in the veins and sluggishly returns to the heart. This condition is undoubtedly one of low physiological organization, fitted by the multiplicity of its venous paths to serve as a composite schema of the variants of the venous system in higher mammals. Further in Echidna, which is not an aquatic animal, but little reduction of the plexus has

D 560 H. von W. Schulte and Frederick Tilney.

occurred. It is significant that this reduction affects the postaortic anastomoses.

In turning from the monotreme to the marsupial, we find in the latter an immense advance in the organization of the venous return. Not only the postcava but also the iliac and often the caudal vein are free of plexus formation, though there is not infrequently a remnant of the plexus in the form of "venous islands" in the region of the promontory, yet even these are fewer than in such placental forms as the edentate or even the cat, and they are far less extensive than in cetacea.

If the plexuses of monotremes are primitive, as we take them to be, it becomes a problem to deduce from them the trunk drainage of the marsupials and placentals. A few details of the morphol(^y of the marsupial veins are a necessary preliminary.

In a specimen of Trichosurus (Fig. 3) the cava is formed by the confluence of a number of radially disposed vessels, the ilio-lumbar, external iliac and internal- iliac veins. The left internal iliac receives the caudal. The specimen shows, perhaps, a slight tendency to the formation of a common internal iliac trunk. Compared to this tbe monotreme presents a great excess of vessels in its fan-shaped plexuses, while here we have, as it were, only the ribs of the fan, the intervening web reduced to small tributaries of the major trunks. Kadial hydrodynamic lines have developed and there has followed, as must follow, a reduction of the network. It is obvious that if two converging veins of equal size are connected by a homogeneous reticulum, the blood flows from all parts of the reticulum under the vis-a-tergo of the arteries and the suction through the veins of the cardiac diastole (Fig. 4). It follows that the blood will flow from the center of the reticulum toward the large veins, and that the periphery will have not only to transmit the blood directly reaching its meshes but also to drain the central areas. Its function, therefore, is at a maximum near the large veins, and diminished toward the center where, as it were, a watershed is formed, dividing the reticulum into two drainage areas. The peripheral parts of the plexus persist as small tributaries of the veins favored by the hydrodynamic lines (Fig. 5), and resolve themselves into smaller vessels and capillaries


Organization of the Venous Return. 5G1





FiQ. 3. Trichosurus vulpecula. From a dissection in the study collection of the Department of Anatomy, Columbia University. Postcava and pelvic veins showing radial arrangement

D Fig. 4.

Fig. 5.

D Organization of the Venous Return. 563

as the divide is approached. Our knowledge of the ontogeny of the pelvic vessels in marsupials is unfortunately small, yet it seems fair to argue from the redundancy of fcetal vessels everywhere, that some such reducing factor as we have mentioned is active mechanically in occasioning the substitution of trunk vessels for plexus formation.

The radial arrangement of the pelvic vessels described above appears to be unstable even in Trichosurus. (Fig. 6.) Besides the specimen of Trichosurus figured, it occurs, so far as we are aware, only in Pseudochirus where it is complicated by the presence of venous rings about the large arteries. Elsewhere the arrangement is disturbed by a tendency of adjacent, ultimately confluent trunks to form a conmion vessel of greater or less extent, transforming the V shape of their union into a Y, and giving rise to an apparent distad recession of the angle of union. The trunks thus affected are the external and internal iliacs with the resultant common iliac — an almost constant formation in the Australian marsupial — or the two internal iliacs to produce a common internal iliac as often in Didelphys virginiana. Trichosurus appears as the starting point of these two types, inclining, however, markedly to form a common iliac In either case the result is the same, the reduction of friction and a displacement of the angle of confluence distad. This phenomenon seems capable of mechanical explanation. Given two equal veins inosculating proximally and connected distally with drainage areas which are increasing in size, an increasing venous return demands an increase in the size of the veins. The trunk proximal of the union tends to lie in the prolongation of the axis of the angle of union (Roux). At this point, therefore, the blood stream changes its direction. Its momentum may be decomposed by the parallelogram of forces into a moment acting in the axis of the resulting vessel and a moment at right angles to this, tending to push the walls of the tributaries into closer approximation and to form a spur at the angle of confluence (Fig. 8). Thus more and more the uniting vessels would tend to have their proximal segments parallel, with their walls in apposition. This spur will sustain the pressure of the b'lOod stream upon both sides, which constitutes an abnormal

D 564 H. von W. Schulte and Frederick Tiln^.






Fig. 6. Trichosurus vulpecula. From a dissection in tlie study collection of tlie Department of Anatomy, Columbia University. Showing common iliac type.

D Oi^anization of the Venous Return.


environment for its cells, tending to its ultimate reductionr McMurrich* has reported cases in man of partial persistence of such formations in the iliac veins. Against this mode of accounting for the Y type, an alternative explanation may be argued. It might be held that the recession of the angle was apparent only, that actually a new confluence had been formed by the development of a cross

RADIAL TYPE Individuals of Triehesurws \ •Ml PMudochiruK

COMMON ILIAC Australian marsupial in B«n«rai



Fig. 7.

Fig. 8.

channel through a more distal portion of the reticulum. As will appear subsequently, we are far from denying this possibility, but interpret it as the disturbing result of factors extrinsic to the circulatory system itself, in fact as an example of the establishment of a

■McMurrich, J. P., *06. "The Occurrence of Congenital Adhesions In the Common Iliac Veins and their Relation to Thrombosis of the Femoral and Iliac Veins." Brit. Med. Jour., II, page 1699.

D 666 H. von W. Schulte and Frederick Tilney.

collateral circulation following interference in a hydrodynamic line (Fig. 9).

In our figure, should such a factor operate upon the segment B tending to its destruction, flow would be reversed in the reticulum previously draining into it, and a new channel such as C would result from the enlargement of portions of the reticulum responding to the


Fig. 9.

increased function required below the obstacle. But apart from such external interference, the line B would tend to be retained, for the inlet from B' is freer into B which is in line with it, than into the diverging channel C. A good illustration of the displacement of the angle distad is afforded by the vessels in the blastoderm of the chick. The caudal vein in marsupials is subject to a wider range of varia

D OiTganization of the Venous Return. 567

tion than the iliac vessels. In Phascolomys (Fig. 10) in addition to a mouth in the left common iliac it is connected by three pairs of transverse branches with the internal iliacs^ forming a sort of grill pattern. In one individual of Phascolarctos a closely similar arrangement was found. In both of these forms the tail is rudimentary. In other marsupials it usually opens by a single mouth into one or other common or internal iliac, only occasionally retaining the remnant of a plexus in multiple points* of debouchment. Followed distad it soon breaks up (usually at the root of the tail) into a plexus surrounding the caudal artery. A distinction can thus be drawn between the large-tailed forms and those having short tails, in reference to the caudal vein, which in the latter forms retains more of its plexiform character. Evidently the size of the drainage area, the volume of blood to be transmitted, is the determining' factor in the evolution of trunk veins as against plexuses. The existence of traces of plexus formation in some individuals of the Macropodidse does not militate against this view, as here a large part of the caudal return is provided by subcutaneous channels.

In general the larger the area drained — the greater the length of the trunk in the proximal portion of its drainage line — ^the farther distal is the point at which the plexuses occur. A familiar example is the comparison of the V. femoralis and the V. poplitea in man with his Vv. brachiales which are plexiform. The caudal vein is, however, a more convincing example, as here such disturbing factors as might result from the upright position of man may be excluded.

The arrangements of the common and external iliac veins among the marsupials are of considerable theoretic importance. Two types may be distinguished in the relation of the vein and artery. In Didelphys* the external iliac vein lies lateral to the artery. In the Australian marsupial the leg is drained by channels lying mesial to the artery (Figs. 3 and 10), external iliac and common iliac veins. Intermediate forms, however, occur. In Trichosurus and Phascolomys, for example, there is sometimes in addition to the large

McClure, C. F. W., '03. "A Contribution to the Anatomy and Development of the Venous System of Didelphys marsupialis." Part I, Amer. Jour. Anat., Vol. II, No. 3. (See plates.)




Fig. 10. Phascolomys Mitchell!. From a. dissection in the study collection of the Department of Anatomy, Ck>lumbia University Showing vena iliaca externa lateralis, vena iliaca externa medialis and grill pattern of caudal veins.

D Organization of the Venous Return. 569

eritally placed vessel a smaller lateral one, varying in development and inosculating distally with the external iliac near the groin, but not at a constant level. It receives tributaries from the psoasiliacus and may show a plexiform character. In the elephant (Darrach) double external iliacs accompany the artery, one lateral, one mesial. The same arrangement occurs in the 20 mm. cat embryo (Huntington). In monotremes a plexus accompanies the arteries. The evidence, while admitted fragmentary, appears to us to warrant the conclusion that the external iliac vein results from the solution of a plexus.

While in the case of the internal iliac vein, reduplication of the vessel has not been observed among the marsupials, so far as we are aware, yet the variety in its relation to the artery — it may be dorsal or ventral, lateral or mesial — su^ests a similar origin. And again the monotreme has the plexus.

The corollary follows that these homonymous great veins are not morphological equivalents; the V. iliaca externa lateralis is not morphologically the same as the V. iliaca externa medialis, but results from a specialization of a different area of plexus. The veins are homodynamous, agreeing in function, that is in the drainage of similar areas ; and it thud appears that the anatomical names of veins designate not morphological but physiological units. The hydrodynamic line of drainage is far more constant than the morphological structures which compose it, and unconsciously this drainage line has been the subject of our nomenclature. A more striking illustration is afforded by the term postcava, as already pointed out by Dwight' and Lewis.® This vessel is composed of morphological, distinctly defined elements, sinus venosus, vena hepatica communis, hepatic sinusoids, subcardinal, supracardinal and postcardinal veins. In the marsupial the cardinal collateral replaces the supracardinal

•Dwight, Thomas, '01. "What constitutes the Inferior Cava." Anat. Anzelg., Vol. XIX, pages 29-30.

•Lewis, F. T., *02. "The Development of the Vena Cava Inferior.'* Amer. Jour. Anat., Vol. 1, No. 3.

^McClure, C. P. W., *06. "A Contribution to the Anatomy and Development of the Venous System of Didelphys marsupialls." Part II, Vol. V, No. 2. (See page 194.)

D 570 II. von W. Schulte and Frederick Tilney.

in this list. The post-renal element may be cardinal, supracardinal or cardinal collateral ; it may be double or single to the right or left in front of the aorta. Apart from its parallelism to the aorta, its only constant character is that it drains the tail, the hinder extremity and one or both of the gonads according to its site, at one side or in front of the aorta. It is obvious that we are dealing with a definite line and area of drainage, which may be aflFected indifferently well by any one of a series of vessels. The term postcava only indicates this hydrodynamic line. The case here is essentially the same as we have attempted to show for the external iliac. A venous plexus surrounding the aorta is antecedent to the formation of trunk vessels. The variously named cardinals are merely dilated portions of chis reticulum along the major hydrodynamic lines, which, responding to the large volume of blood they transmit, dominate the picture. The smaller transverse channels were long treated as of little morphological importance, except where from external factors in the course of development, the flow became retarded in one of the longitudinal vessels, when they entered the field of consciousness under the term anastomosis as a means of accounting for the emergence or enlargement of another longitudinal line. The early investigators of the development of the venous system rarely figured these plexuses, and their schemata showing only the longitudinal hydrodynamic lines, still illuminate our text-books and adumbrate the subject. Recent workers give more complete figures (Lewis,® ; Miller,® ; Huntington and McClure,®).'

The plexiform arrangement of the postcaval line disappears earlier, both in development and phylogeny, than is the case in the more peripheral regions. Ornithorhynchus alone shows the plexus in any marked degree in the postrenal cava ; in Echidna it has almost disappeared. The determining factor in the formation of these large

•Miller, A. M., '03. "The Development of the Post Caval Vein in Birds." Amer. Jour. Anat, Vol. II, No. 3. (See Fig. 5, page 289.)

•Huntington, G. S., and McClure, C. F. W., W. "The Interpretation of the Variations of the Postcava and Tributaries of the Adult Cat, based on their Development." Amer. Jour. Anat, Vol. VI, No. 3, page 33.

This paper was illustrated by reconstructions which have not yet been figured, showing clearly the plexiform nature of the periaortic vessels.


OiTganization of the Venous Return. 571

trunks, by the enlargement of a part of the plexus, we believe lies in the large volume of blood which must pass through the centrally placed vessels. The well-known facts of collateral circulation, following ligation, abundantly prove that a vein responds by growth to an increased flow of blood, that is, its size is determined by its drainage area. Evidently the vessels proximal to the heart have more blood to transmit than any of their tributaries and receive, therefore, a greater stimulus to growth. The hydrodynamic factor operates most intensely at the center, and the development of the venous trunks proceeds from the center toward the periphery by the enlargement of capillaries and plexuses along favorable lines and the resolution of the remaining reticulum into small tributaries. The considerable range in the variation of the post-cava appears due to the approximate parallelism of the channels about the aorta. When vessels inosculate at a very acute angle the freedom of the out-flow into the common trunk must be very nearly equal for both. When both have a common drainage area, as these vessels, it must be a very nice balance that determines which of them is to survive. In many cases it is some external factor, such as the pressure of a muscle, or the development of some organ, e. g., the mesonephros, or the migration of the kidney in the case of the postcardinal, which gives decision when the internal factors seem so nearly in equilibrium. The hydrodynamic lines once established, the plexus resolves itself into tributaries of small size, as in the case of the pelvic vessels.

It remains to determine, if possible, the origin and direction of the major hydrodynamic lines. The earliest vessels in the vertebrate appear as a capillary network which increases at the periphery by the formation of new capillaries in the growing region, while the central areas are constantly being resolved into larger vessels. Beautiful demonstrations of this reticulum have recently been given by Clark^^ and Evans. ^^ The development of the vessels in the chick's

Clark, E. R., '09. "Observations on the Living Growing Lymphatics in the Tail of the Frog Larva." Anatomical Record, Vol. Ill, No. 4.

"Evans, H. M., '09. "On the Earliest Blood-vessels in the Anterior Limb Buds of Birds and their Relation to the Primary Subclavian Artery." Amer. Jour. Anat., Vol. IX, No. 2.

D 672 H. von W. Schulte and Frederick Tilney.

blastoderm is too well known to require description, especially in the light of Thoma^s^^ extensive work, but it illustrates admirably a number of points which we desire to emphasize; first, the capillary reticulum ; second, the formation of veins outward from the center^ and third, the presence of a continuous channel at the periphery — the vena terminalis. The drainage lines at first radiate from the center like the spokes of a wheel — ^that is, they correspond to the direction of the growth of the area vasculosa. Later a reduction in their number occurs and larger branching vessels are formed. This seems to be merely a case of the recession of the angle of confluence and the expression of the already cited tendency of V-shaped confluences to develop a Y formation. A very similar phenomenon occurs in the development of the middle cerebral in man. Mall's figures*' show the same changes from reticulum to reticulum with hypertrophy of its radial lines, corresponding to the radial growth or expansion of the pallium, and finally the emei^ence of the brandied veins. The pelvic vessels show the same series of types. It would appear that the primitive hydrodynamic lines conform to the direction of growth. A further illustration is afforded by the early veins of the longitudinally growing body. Lewis*'* has shown that the first veins in the rabbit are the umbilical and the piecardinal, both of them longitudinal.

We regret that the Important article of Thoma came Into our hands too . late to receive the attention It deserves In the body of our paper. In many places our observations overlap and our conclusions are closely similar. We would point out, however, that the material used Is largely different Thoma working on the chick's area vasculosa was able to demonstrate the emergence of vessels along hydrodynamic lines, proceeding centrlfugally by the enlargement of some of the capillary channels and the reduction of others. That he appreciated the general application of this important observation is shown by the following excerpt: "Auch die doppelten Begleltvenen der Arterlen des Menschen und die Entwlckelung des Venenplexus weisen auf solche Besonder "Thoma, R., '©3. "Untersuchungen ttber die Hlstogenese und Hlstomechanik des Gefasssystems." 1893, Stuttgart.

"Mall, F. P., *04. "On the Development of the Blood-vessels of the Brain In the Human Embryo" Amer. Jour. Anat., Vol. IV, No. 1.

"Lewis, F. T., '02. "The Development of the Vena Cava Inferior." Amer. Jour. Anat, Vol. I, No. 3.

D Organization of the Venous Eetum. 573

heiten bin, doch wftre m offenbar yerfrtibt, diese Formeigentbiimlicbkelten, bei denen slcher nocb andere Umst&nde mitwlrken, bier ansftibrlicber zu erortem."

Tbls principle we baye sougbt to apply to otber areas. In tbe continuation of bis paper be is interested mainly in tbe arterial system, wbile we baye sougbt to apply certain simple mecbanical views to tbe major Tenons lines. As regards tbe bydrodynamic factors tbemselves, wbile admitting freely tbe Importance of Tboma's findings and tbe ingeniousness of bis deductions, we bave ventured to depart somewbat from bis conclusions notably tbe first of bis bydrodynamic laws^ Witb tbe otber two our paper, from its limited scope, is not concerned. Tbis law is formulated by Tboma as follows : '*Das Wacbstum der Gef&sslicbtung, d. b. das Fl&cbenwacbstbum der Grefftsswand is abbftngig von der Stromgescbwindigkeit des Blutes." It must be borne in mind tbat tbe conditions under wbicb tbe arteries and veins develop are different as are tbe functions wbicb tbey perform, and tbat, tberefore, conclusions arrived at by tbe study of one system cannot be directly transferred to tbe otber, as, for example, tbat in general tbe velocity of fiow determines tbe size of tbe lumen; wbile tbis may bold true for tbe arterial system in itself, it is invalid in a comparison of artery and vein, e. g., compare tbe lumen of tbe aorta witb tbat of tbe postcava. Tbis rule would lead us to expect a larger lumen in tbe aorta tban in tbe cava ; the exact opposite is tbe case. And yet tbese vessels must transmit in one cardiac revokition tbe same volume of blood, unless congestion or anemia of tbeir commoa «rs%. is to result. We are inclined to consider the volume tbe determining factor. Now tbe volume is tbe product of tbe pressure and cross section, tbe smaller tube will deliver in a unit of time the same volume as tbe larger under sufficiently increased pressure. Accordingly we find vessels adapted in two directions to supply tbe volume determined by metabolism of tbe tissues ; first, under conditions of higher pressure, with thicker walls and smaller lumina ; second, under conditions of lower pressure with thinner walls and greater lumina. Tbe velocity of flow, we bold, to be conditioned by these moments, as in the adult by tbe difference between tbe a tergo and the a fronte factors.

In the chick of 36 hours (Fig. 11) is found an arrangement of^ veins closely parallel to the condition existing in the area vasculosa. A centrifugal formation of veins along the hydrodynamic lines which correspond to the direction of growth of the drainage area, and terminates peripherally in a reticulum, bounded by marginal vessels, the umbilical and postcardinal veins, which in this respect resemble the vena terminalis. Proceeding from the periphery to the center, that is, caudo-cephalad, we find first an abundant capillary reticulum, then an area where the retictilum has a somewhat ladderlike figure, the emerging cardinal and umbilical forming the




Fio. 11. Chick of the thirty-sixth hour. From a reconstruction in the study collection of the Department of Anatomy, Columbia University. Showing the character and disposition of the capillary reticulum between the umbilical and post-cardinal veins.

D Organization of the Venous Return. 575

uprights, while the intervening reticulum assumes a transverse disposition. The longitudinal vessels respond by increasing growth to increasing function; their diameter is determined by the length of the area drained, that of the transverse vessels by its widtli, both in rough proportion to the volume of blood they carry and this in turn to their respective drainage areas. Finally we reach distinct areas with lateral tributaries, trunks having emerged along hydrodynamic lines by the solution of a plexus. The formation of a divide between the lines of the umbilical and postcardinal veins is really a simple example of the principle we tried to show as operative in the case of veins convergent at an angle. The marginal arrangement of this drainage system has been alluded to. At first the umbilicals predominate, later they largely lose their function as veins of the somatoplcure and the postcardinals usurp their territory, and becoming united by a reticulum across the aorta form important elements in the periaortic plexus. Here is an obscure instance of the substitution of axial for marginal drainage. Owing to the relatively large volume of blood seeking return along these lines — from the trunk and posterior extremities through the cardinal, and from the allantois through the umbilical — the longitudinal are so accelerated in growth and so dominate the picture that their relation to the reticulum is masked. The phenomenon of deep axial drainage replacing a superficial marginal type which is earlier in time, appears also in the limbs, in the substitution of the axillary and femoral lines for the primitive Rand-venen, while in the tail the lateral caudal veins may possibly be a persistence of the marginal type. Their apparent connection with remnants of the umbilical line in marsupials lends color to this view. The general problem appears worthy of further investigation, especially with reference to the underlying mechanical factors.

We have made but little reference to external factors modifying the development of the venous system, both because they have in general received more attention than conditions of flow which it was our purpose to estimate, and because we believe them to be modifying factors only, acting upon otherwise determined hydrodynamic lines. We believe that the retention of multiple points of debouchment by

D 576 H. von W. Schulte and Frederick Tilney.

a trunk vein^ of which the venous island or fenestra is a special case, must be explained in this way. It is a survival of a retrogressing plexus, but its retention increases the surface friction of the system. The mechanical factors we have instanced in the reduction of the plexus operate against its persistence, for in the case of a fenestra in a venous trunk, either the recession of the angle of confluence would tend to the fusion and absorption of the walls separating the arms of the loop, or else the arm that fell most nearly into the lines of the eflFerent and aflFerent vessels would have the freer in-let and out-let and so f eceive a greater stimulus to growth. An exact equilibrium, requiring that both arms should converge and diverge at the same angle to the parent trunk, or that one side of the loop should be favored by the entrant, the other by the emergent vessel in equal degree, could not be expected often to occur. One arm would increase, the other decrease, until the favored arm was lost in the continuity of the trunk, the other forming a minute tributary or two, or altogether retrogressing. Evidently an external factor must be sought in the motion of adjacent structures, favoring or impeding flow, now in one arm of the loop, now in the other. Obviously tiie passage of an artery or a nerve through a fenestra does not occasion the persistence of both of its arms. In the case of a muscle, the matter is different ; for example, in Omithorhynchus, the anastomosis lateral to the psoas minor affords escape for the blood when the psoas presses upon the dorsal plexus.

Our argument has, hitherto, been that veins develop out of a capillary reticulum under the influence of hydrodynamic factors. The genesis of the reticulum does not affect its reaction to these factors, yet if the argument which we have presented has validity, its eventual extension to the origin of the capillaries themselves out of inter-cellular spaces may not prove entirely mistaken. Such spaces serving as circulatory channels are described in a number of invertebrates.^* The flow through these spaces might be conceived to occasion a flattening of the surfaces impinging upon the blood stream — the inception of an endothelium, which would thus cease to be an

"IDahlgren and Kepner. "A Text-book of the Principles of Animal Hlstolq^y." (See Figs. 134, page 151.)

D Organization of the Venous Return. 577

entity and become merely a position modification of the mesenchyme. Its instability as a tissue-form after the ligation of a vessel is well known. The fact that the endothelium of the embryo spreads from the center to the periphery does not preclude the possibility that its characters are determined by its relation to the blood stream, for the flow is most rapid and voluminous at the center and consequently there should be sought its earliest and greatest effects. There also^ the mesenchyme giving rise to the muscularis and adventitia receives its greatest stimulus. We have ventured upon the debatable and very controversial ground of the vascular endothelium, in order to point out that the vital problem of the veins concerns not the great vessela but the capillaries.

Received for publication August 1, 1909.



CALVIN B. COULTER. From the Laboratory of Comparative Anatomy, Princeton JJniversity.

With Twei-ve Figxjbes.

Through the investigations of Rathke, Hochstetter, and others the development of the aortic arches of the vertebrates in general is very well understood. The existence in the mammals, however, of a fifth aortic arch, lying between the systemic and pulmonic arches, has been a matter of recent discussion, and the work on this paper was begun with the view of investigating the conditions as they are in the cat. Consideration will be given to the papers dealing with a fifth arch in the mammals when this arch is dealt with in the following pages. An extensive general bibliography may be found in Hertwig's "Handbuch" following Hochstetter's article on "Die Entwickelung des Blutgefasssystems."

The history of the arches in the cat was studied by means of wax reconstructions made after the method of Bom at enlargements of sixty-six and forty diameters. Twenty-six embryos were examined, from 3.6 to 16 mm. in length, and sixteen reconstructed, of which ten are reproduced here. The reconstructions represent casts as it were of the lumina of the blood vessels and the cavity of the pharynx, and do not indicate the thickness of the walls or the character of the glandular structures developed from the branchial pouches. No attempt will be made to describe fully the development of the pouches, which are sho^\Ti in the earlier stages to illustrate the consistent relations of the blood-vessels to the branchial arches.^

0. Hertwig, Handbuch der verglelchenden und experimentellen Entwickelungsgeschichte der Wlrbeltiere, Baud III, Teil 2. (578)

D D D Aortic Arches of the Cat.


List of Material Studied.

3 mm. 3 mm.

4.5 mm. 5 mm.

5 mm.

5.6 mm.

6 mm. 6 mm.

6 mm. 6.5 mm. 5 mm.

5 mm . 6.8 mm. 6.8 mm.

7 mm. 7 mm.

7 mm. 7.25 mm.

8 mm.

8 mm.

9 mm. 10 mm. 11.5 mm.

Series 188. Series 45. Series 93. Series 47. Series 11. Series 110. Series 84. Scries 126. Series 127. Series 131. Series 30. Series 31. Series 103. Series 105. Series 137. Series 138. Series 2. Series 13. Series 3.


19. Series 101. Series 29.

Columbia Princeton Columbia Princeton Princeton Columbia Columbia Columbia Columbia Columbia Princeton Princeton Columbia Columbia Columbia Columbia Princeton Princeton Princeton Princeton Princeton Columbia Princeton

University University University University University University University University University University University University University University University University University University University University University University University

Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection. Collection.

In the youngest embryo examined (3 mm., Fig. 1), in which the pharyngeal membrane has not yet broken through, there is a single aortic arch, through which on each side the primitive heart communicates directly with the dorsal aorta. There are two well-defined branchial pouches on each side, which appear to be homologous with the first and second pouches of later stages. Between these two there are slight protrusions from the dorsal aorta, and an irregular outgrowth from the ventral aorta, as shown in Fig 1. This ventral outgrowth is of a verj^ indefinite character, and has the appearance of a tissue-space which has become continuous with the ventral aorta in two places. An identical condition was found Ia another embryo of this litter. Series 188. Whether or not these cavities are to be regarded as tissue-spaces that are utilized in forming the second aortic arch cannot be discussed here; at any rate, they and the dorsal buds seem to be the ventral and dorsal anlages of the second arch. Behind the second pouch also there is on each side an evagination from the dorsal aorta, the anlage of the third arch.

An embryo of 4.5 mm. (Fig. 2) shows the second and third arches completed, and the ventral anlage of the fourth arch extending caudad from the middle of the third arch. On the left side, a

D 680 Calvin B. Coulter.

dorsal as well as a ventral anlage of the fourth arch has appeared. From the distal end of the first arch a vessel has grown forward, and when this arch degenerates carries the blood directly forward from the dorsal aorta into the head region. This vessel constitutes that portion of the internal carotid artery which is developed in

Arch P<mch


Pouch Arch

Dorsal A ah

Fig. 1. — Reconstruction of the aortic arch system of a 3 mm. cat embryo. Series 45, Princeton University Collection. Right side. X 66.


P<>^^^ AarHcBulb


Pouch t ^«* *

ArdiA Pouch S

Pouch A-6 Donal Aorta

Fig. 2. — ^Reconstruction of the aortic arches of a 4.5 mm. cat embrya Series 93, Columbia University Collection. Right side. X 60.

front of the first aortic arch. A small aortic bulb has been formed by the coalescence of the ventral ends of the first and second aortic arches. Behind the third arch a large branchial evagination has made its appearance, and has already begun to divide into the third pouch and the swelling from which the fourth and fifth pouches are subsequently formed by a similar division.

D Aortic Arches of the Cat. 581

In the next stage (5 mm., Fig, 3) the first two aortic arches are reduced in size, and from the ventral portion of the first arch capillaries extend out into the mandibular region. The ventral ends of the third arches have begun to fuse together, so that the aortic bulb is enlarged and shifted caudad. This is a stage in the progressive coalescence which takes place between the ventral ends of all the aortic arches, with the .subsequent formation of a large aortic bulb.

Archl Pouch 1 ArehM

Tr. Arterio9U9 Pouch g Aortic Bulb

Arch 3 Pouch S Arch 4 Pouch 4-6

Dcraal Aorta

Fig. 3. — Reconstruction of the aortic arches of a 5 mm. cat embryo. Series 47, Princeton University Collection. Right side. X ^•

The dorsal anlage of the fourth aortic arch is now present on both sides, and the separation of pouches 3 and 4-6 more distinct.

Fig. 4, of an embryo 5 mm. in length, shows the fourth arch completed. The first arch has lost its connection with the dorsal aorta, leaving a dorsal remnant which soon disappears. Just anterior to this remnant, and to the hypophysis which lies mesial to it, there is an anastomosis, not visible in the figure, between the two dorsal aortflB by means of a large cross-trunk, a peculiarity which was observed only in this embryo and in Series 31, another of the same

D 582 Calvin B. Coulter.

litter. The fusion of the ventral ends of the third arches has continued and now involves the bases of the fourth aortic arches. In this embryo (Fig. 4) the sixth arch makes its first appearance, as a spur extending caudad from the ventral portion of the fourth arch. The third pharyngeal pouch has become still farther separated from


Pouch 1 Archg

Pouch B Arch 3 Pouch 3

Arch 6 Arch 4 Pouch 4-6

Dortal Aorta

Fio. 4. — Reconstruction of the aortic arches of a 5 mm. cat embryo. Scries 30, Princeton University Collection. Right side. X 60.

the evagination caudal to it, which is a simple rounded structure that shows no evidence of division.

The 5 mm. embryo. Series 11 (Fig. 5), is a very important one, for it brings us to the question of a fifth aortic arch.^ Before presenting the results obtained in the cat the observations in regard to this arch in other mammals will be briefiy reviewed.

In view of the differences observed in the relative development of their arches, it is probable that the measurements of the 5 mm. embryos (Ser. 11, 30, and 31) are incorrect.

D Aortic Arches of the Cat. 683

Zimmermann^ (1889) described in the rabbit an artery arising from the truncus arteriosus and emptying into the dorsal aorta near the base of the pulmonic arch, separated from the systemic and pulmonic arches by distinct entodermal pouches. In an incomplete sheep series he found a vessel extending ventrad from the distal end of the pulmonic arch, but was unable to trace its ventral connection. The fifth arch as he described it in man represented a very different condition, as it arose from and terminated in the fourth arch, enclosing its middle third.

Tandlei^ (1902) found in two human embryos a vessel extending from the ventral aorta to the distal end of the pulmonic arch. In the rat he interpreted an anastomosis between the fourth and pulmonic ardies as a fifth arch, but could not discover a fifth pouch.

Lehman' (1905) found in the rabbit irregular vessels arising from the fourth and pulmonic arches, and in the pig a somewhat similar condition, but found in one case a complete vessel from the ventral end of the fourth arch to the dorsal aorta. This vessel was connected by a short stem with the pulmonic arch, and was separated from the fourth and sixth arches by distinct branchial pouches.

Lewis® (1906) in the rabbit and pig described only irregular vessels, and expressed the belief that none of these spurs or additional roots at the bases of the arches could be interpreted as a fifth aortic arch, and that the evagination described as postbranchial body is not serially homologous with the preceding pouches.

Locy*^ (1906), commenting upon the condition of the fifth arch in the mammals, states his belief in the existence of a fifth arch. He

•W. Zlmmermann. Ueber einen zwischen Aorten und Pulmonalbogen gelegenen Kiemenarterienbogen beim Kaninchen. Anat Anz., Bd. IV, 1889.

Rekonstruction eines menschlichen Embryos. Verb. Anat Ges., 1889.

Tandler, J. Zur Entwickelungsgescbicbte der Kopfarterien bei den Mammalia. Morph. Jahrb., Bd. 30, 1902. 'Lehmann, Harriet. On the Embryonic History of the Aortic Arches in Mammals. Anat Anz., Bd. XXVI, 1905.

•Lewis, F. T. The Fifth and Sixth Aortic Arches and the Related Pharyngeal Pouches in the Rabbit and Pig. Anat. Anz., Bd. XXVIII, 1906.

Locy, William A. The Fifth and Sixth Aortic Arches of Chick Embryos with comments on the condition of the same vessels in other Vertebrates. Anat Anz., Bd. XXIX, 1906.

D 584

Calvin B. Coulter.

thinks that its extreme variability and transitory character undonbt' edly explain the lack of definite information regarding it in some of the forms, and notes the individual differences in those forms in which a complete arch has been described.

Soulie and Bonne® (1908) in their paper on the arches of the mole describe a typical fifth aortic arch, arising separately from the

Pouch 1

Pouch 2

Pouch 3

Doraal Aorta


Arch ft

Tr. Arterioms

Aortic Bulb Arch 3

Arch 6 Arch 4 Arch 6 Right Pulm, Artery

Pouch 4-S

Fig. 5. — ^Reconstruction of the aortic arches of a 5 mm. cat embryo. II, Princeton University Collection. Right side. X 60.


aortic bulb or in a common trunk with the pulmonic, and emptying in every case into the dorsal aorta in common with the pulmonic arch. This fifth vessel in the mole occupies a distinct branchial arch, which lies somewhat lateral to the fourth and sixth arches.

The typical mammalian fifth aortic arch appears thus to be a vessel which arises from the aortic bulb and empties into the pulmonic arch near its junction with the dorsal aorta. The development is

•Souli^, A., and Bonne, C. L'Appareil Branchial et les Arcs Aortiques de FEmbryon de Taupe. Journ. de TAnat et de la Phys., No. 1, 1908.


Aortic Arches of the Cat. 685

most complete in man and the mole, in which an unbroken arch is the rule; in the cat, as will be described in the following pages, and in the pig, the same type of development is followed, but a perfect arch would seem not to be produced ordinarily. In the rabbit the condition is still more rudimentary, and one must agree with Lewis that evidence of a fifth aortic arch in this form is wanting.

n Thyroid


Arch 4 \

Arch £ .

Arch 6 ,

Arch 6 S

Dorsal Aorta

Left Pulm. Artery

FigT 6. — Reconstruction of the aortic arches of a 5.6 mm. cat embryo. Series 110, Columbia University Collection. Ventral view. X 50.

while the observations on the sheep and the rat are still incomplete, as giving evidence for a vessel of the type described above.

A condition very similar to that occurring in man and the mole, but, in general, more rudimentary, was found by the writer in the cat. In embryo Series 11 (Fig. 5) on the right side a spur extends dorsad from the aortic bulb, between the fourth and pulmonic arches (arch 6). This spur occupies the position from which a fifth arch


586 Calvin B. Coulter.

would develop and resembles in all respects the anlages from which the other arches arise. The sixth arch is complete, and gives off a short pulmonary artery on the right side. In addition to the spur of the fifth aortic arch, there is a short vessel connecting the dorsal ends of the fourth and sixth arches, very similar to the anastomosis between the two arches found in the rat by Tandler and to the vessel between the fourth arch and the root of the pulmonic in the pig described by Lehmann. On the left side there was to be found no


Arch 4

Pouch . , ^ ^ ^

iial Arch 6


Arch 6

Fig. 7. — Photomicrograph of a transverse section through the fourth, fifth and sixth branchial arches of a 5.6 mm. cat embryo. Right side. Series 110. Columbia University Collection.

trace of a fifth aortic arch. The fourth and fifth pharyngeal pouches have not separated in this embryo and consequently the fifth branchial arch is not clearly marked out.

In an embryo of 5.6 mm. (Series 110, Fig. 6) the second aortic arches have lost their connection with the dorsal aorta, and their ventral remnants are disintegrating. There is on the right side a spur of the fifth arch from the aortic bulb similar to that shown in the preceding embryo, and in addition, a spur from the dorsal root of the pulmonic arch, with a blind vessel between them, almost con

D Aortic Arches of the Cat 587

tinuous with the ventral spur, and running parallel to the arches on either side. Each pulmonic arch joins the dorsal aorta by three distinct roots, not clearly shown in the figure. On the left side two spurs project from the dorsal end of the pulmonic arch, the larger of which is directed ventrad between the fourth and fifth branchial pouches. The fourth or most caudal pharyngeal evagination has grown out, in its dorsal portion, into two divisions, the fourth and fifth branchial pouches, which are shown in section in Fig. 7, through the right side, and Fig. 8, through the left side. The photomicrographs show also the distinct character of the fifth branchial arch, and the two ectodermal grooves in the floor of the sinus precervicalis. The ventral portion of the fourth pharyngeal evagination remains undivided and as a result the fifth branchial aich is very short This stage marks the highest development of the fifth aortic and branchial arches in the cat ; in later stages the development is retrogressive.®

In embryo Series 138, 7 mm. in length (Fig. 9), the first aortic arches have entirely disappeared, and the second arches are mere stubs which break up into capillaries. There is no fifth aortic arch, but the fifth branchial arch is very clearly marked out by the ectodermal grooves on the outside, and as in Series 110 (Figs. 7 and 8) lies to the outer side of the fourth and sixth branchial arches. The dorsal end of the sixth aortic arch is very large, and on the right side shows a peculiar grooving which is suggestive of a division into two much longer roots than found elsewhere. In this and the pre

Since the completion of this paper, Tandler has published in the Anat Hefte, 115 Heft (38 Bd., Heft 2), a careful description of the aortic arches and related pharyngeal pouches to be found in human embryos. His account agrees remarkably with mine. In man, however, the fourth and fifth pouches are derived from the ventral portion of the last pharyngeal evagination, and become more widely separated and distinct structures than in the cat. Correspondingly, the fifth aortic arch attains a more complete development. The pouches of the cat have been made the subject of a careful study by Henry Fox, whose article on "The Pharyngeal Pouches of the Mammalia*' has appeared since the completion of the present work in the Am. Jour, of Anat, Vol. VIII, No. 3. His results are entirely in accord with mine, although be makes no mention of a fifth pouch, which I interpret as a division of bis "dorsal process of the fourth pouch." An indication of this separation into two pouches Is to be seen on the left side In his Fig. 60.

D 588 Calvin B. Coulter.

Branchial Arch ^•. Branchial Arch 6^^. n , .

Branchial ArchjS-^ - '^"^^ ^

Fig. 8.— Same as Fig. 7. liCft side.

Pouch 1

^^<^ * .Aortic Bulb

Pouch $


Pouch S

Arch 4


Right Pultn. Artmy

Pouch 4 Pouch 6

Dorgal Aorta

Fig. 9. — ^Reconstruction of the aortic arches of a 7 mm. cat embryo. Series 188, Columbia University Collection. Right side. X ^•

D Aortic Arches of the Cat. 589

ceding embryo the fourth and fifth branchial pouches are distinguishable, but their lumina are becoming obliterated, and their common connection with the pharynx cavity is being elongated and constricted off. Traces of the fourth pouch are to be found in embryos of 8 and 9 mm., but in later stages it apparently disappears completely.






^^^ Pouch A

Right Pulm. Artery Pouch S m* '««'^

Arch Left Pvlm, Artery


Fig. 10. — Reconstruction of the aortic arches of a 9 nun. cat embryo. Series 19, Princeton University Collection. Right side, x 30.

In an embryo of 6 mm. (Series 129, not figured) the dorsal root of the sixth aortic arch is very large, as in the Series 110 (Fig. 6), and a similar but longer spur arises from the base of the left pulmonic arch and ends blindly in the substance of the fifth branchial arch.

Whatever the significance of the arterial spurs in the cat may be, it is certain that we have here, outlined by the five entodermal pouches on the inside and the corresponding ectodermal grooves on

D 590 Calvin B. Coulter.

the outside, six branchial arches. The fifth is a diminutive structure and occupies a position relatively dorsal and lateral to the other branchial arches. The facts observed point to the conclusion that Ordinarily no fifth aortic arch is completely developed in the cat; and it seems more than probable that the incomplete development and

Arch Arch


Left Pulm. Artery. Right Pulm. ArteryArch

Doreal Aori

Vertebrc Svbclavia

Fig. 11. — Reconstruction of the aortic arches of a 10 mm. cat embryo. Series 101, CJolumbla University Collection. Right side. X 28.

uncertain character of the fifth aortic arch is merely an expression of the incomplete development of the fifth branchial arch. It may well be that the anastomoses and irregular roots about the base of the pulmonic arch which have been so generally described in the mammalia are evidence of an assimilation of the fifth aortic arch into the pulmonic, beginning at their dorsal extremities.

D D D Aortic Arches of the Cat 691

The ventral anlage of the sixth aortic arch appears firsts as a bud from the ventral end of the fourth arch (Fig. 4). Somewhat later a dorsal bud grows out from the mesial side of the dorsal aorta, and the completed arch pursues a curved or bent course around the fourth anl fifth pouches. The dorsal root of the pulmonic arch in every case, from its first appearance until after the buds of the pulmonary arteries arise, was found to be pierced by two or more "islands." The significance of this has been referred to above. At




Common Carotid' Arch 4


Doraal Aorta

Ductus Arterionu



Arch 6 Right Pulm. Artery

Fig. 12. — ^Reconstruction of the aortic arches of an 11.5 mm. cat embryo. Series 29, Princeton University Collection. Right side. X 27.

about the time that the rudiments of the fifth aortic arch appear, (Figs, 5 and 6) the pulmonary arteries begin to develop from the middle of the sixth or pulmonic arches. Their development is very similar to that described by Bremer^^ (1901) in the rabbit. They grow caudad, following the curve of the dorsal aorta, on each side of the trachea. The aortic bulb now begins to lengthen out between the fourth and sixth arches, and to divide into the short systemic

"Bremer, J. L. On the Origin of the Pulmonary Arteries in Mammals. Am. Jour. Anat, Vol. I.


592 Calvin B. Coulter.

and pulmonio trunks (Figs. 9^ 10 and 11). In this process, the pulmonic trunk is twisted from right to left, and so comes to lie on the left side of the systemic trunk. At the same time the ventral ends of the two pulmonic arches are brought into contact, and they fuse together up to the point where the pulmonary arteries are given off (Figs. 11 and 12).

The later history of the aortic arch system is too well known to require any comment, and I leave the description at this point.

Recelyed for publication July 9, 1909.



Vol. III. DECEMBER, 1909. No. 12


RALPH EDWARD SHELDON, A88i8tant Professor of Anatomy, University of Pittsburgh.

With Six Figures.

The structure of the human nervous system is so complex and is modified in so many ways from the normal vertebrate type that the interpretation of its morphology and function often becomes a matter of serious disagreement. The situation is further complicated by the fact that experimentation, open to the student of lower animals, is necessarily barred in an investigation of the highest. It is largely owing to these considerations that the question of the innervation of the tongue for the sense of taste is even now, after seventy years of study, one of the most disputed points in human anatomy. This discussion is for the purpose of calling attention to the ease with which a careful analysis of the data furnished by phylogenetic history will elucidate the most vexed questions of human morphology.

The tongue is innervated by two sensory nerves, the glossopharyngeal and the lingual. While the former takes its course direct from the posterior part of the tongiie to the petrosal ganglion, the latter is joined shortly by the chorda tympani from the facial nerve. (Fig. 6.) It would seem that the question as to which of these nerves fur ^Address given before a joint meeting of the Chicago Neurological Society and the University of Chicago Biological Club, March .30, 1909.


D 594 Ealph Edward Sheldon.

nishes the fibers for taste could be easily settled by their dissection in the adult, by sections of embryological material or by the observations made in clinical or other pathological cases. Such studies have, however, led to the most diverse results. Since the time of the researches of Claude Bernard in 184:3 most authorities have been agreed in assigning the fibers for taste for the anterior part of the tongue (one-third to four-fifths) to the chorda tympani and denying their presence in the lingual above its junction with the chorda. Lussana, Wolf and Halban found that section of the chorda led to the loss of taste in the anterior part of the tongue while Blau obtained taste sensations by stimulation of the same nerve. Disease of the middle ear, aflFecting the chorda, after ihe observations of TJrbantschitsch, Schlichting, Kiesow and Nadoleczny and Koster, likewise causes loss of taste. Destruction of the lingual V above its junction with the chorda almost invariably is without effect on the sense of taste, although Schiff in 1887 and recently Koster, consider that a part of the taste fibers go by way of the lingual either directly into the Gasserian ganglion or into the otic and thence to the brain.

There is likewise considerable unanimity in regard to the source of the taste fibers for the posterior part of the tongue. The work of Dana, Pope, Cassirer, Zander, etc., demonstrates conclusively that these belong to the glossopharyngeal, the lingual IX. Most authors are agreed that these fibers originate from cells in the petrosal ganglion and connect directly through the sensory IXth root with the fasciculus solitarius. Some few, as noted in Fig. 5, trace the glossopharyngeal fibers, however, into the brain by way of Jacobson's nerve and the small superficial petrosal to the trigeminus. Such a view is given little weight at present. The controversy, then, centers on the course which the taste fibers from the anterior part of the tongue take after they enter the facial nerve from the chorda tympani. The earlier workers, such as Claude Bernard, Lussana, Duchenne and Vulpian, believed that these fibers took the most obvious course and entered the brain through the pars intermedia of Wrisberg. Clinicians in removal of the Gasserian ganglion or resection of the second or third ramus of the trigeminus for facial neuralgia often noted a complete loss of taste on the anterior part of the tongue of the same side.

D The Facial Nerve and Chorda Tympani. 595

Similar conditions M^ere observed in cases of lesions affecting these nerves. Many such cases have been described, notably those by Erb, Ferguson, Gowers, Salomonsohn, Turner, Senator, Ziehl, Kron, etcL These cases have been so numerous and have been supported by such a weight of authority that even now the most widely accepted course for the chorda taste fibers is by way of the fifth nerve. Men holding this view differ, however, as to whether the taste fibers enter the brain by way of the maxillaris or the mandibularis branch of the trigeminus and likewise as to the pathway by which they reach these rami from the facial trunk. That most generally accepted traces the chorda fibers into the geniculate ganglion, thence through the great superficial petrosal and Vidian nerve to the sphenopalatine ganglion and thence into the maxillaris. Another course advocated is by way of the geniculate ganglion, the anastomotic branch to the small superficial petrosal, thence to the otic ganglion to the mandibularis nerve. It should be noted that practically all these cases rest upon clinical observations of pathological cases. As Scheier ('95) observes, many of these presented chronic lesions which are rarely local and might easily affect other nerve roots or ganglia. The same is true with the operations for removal of the Gasserian ganglion particularly as performed by the Hartley-Krause method. Few of the cases, moreover, were carefully studied for a long period of time by trained observers. The importance of this will appear later.

There are many cases recorded in opposition to this interpretation and it must be argued that a large number of such cases of removal of the Gasserian ganglion without interference with taste negative any number of cases accompanied by loss of taste. If the ganglion can be removed in its entirety without loss of taste the taste fibers certainly cannot go through it. Brunns, Tooth, Thomas, Tiffany^ Frankl-Hochwart, and Fasola all note such cases. The most convincing, however, is the report of Gushing ('03) who removed the Gasserian ganglion completely in eleven cases and partially in two. Observations on the sense of taste were made in practically all the cases before operations. There was a temporary diminution or loss of taste in most of the cases, but in all but one taste returned eventually. The exception was under observation for only six days after

D 596 Ralph Edward Sheldon.

the operation. These results offer the strongest kind of evidence that the taste fibers for the anterior part of the tongue do not enter the brain by the way of the trigeminal nerve. Probably in the first cases noted other nerves were affected or else the cases were not under observation for a sufficiently long time.

Some recent experimental work on the lower vertebrates (Sheldon, '09) indicates that these discrepancies may be due to another cause: viz., that the nerves of general sensation, as found in the trigeminal nerve, for instance, react to certain kinds of chemical stimuli. Under fluch conditions, therefore, destruction of the Gasserian ganglion or of the lingual above its junction with the chorda might affect what we call the sense of taste.

Of course it is possible that the taste fibers enter the brain by way of the glossopharyngeal nerve. Such a course has been advocated by Carl, Herman, and Cassirer, for instance. The usual course advocated is by way of the chorda tympani, geniculate ganglion, ramus anastomoticus to the small superficial petrosal and thence into the plexus tympanicus and Jacobson's nerve to the petrosal ganglion. Another course su^ested is by way of the lingual nerve into the otic ganglion and thence iJirough the small superficial petrosal into Jacobson's nerve. While there is some anatomical evidence that there are fibers in the small superficial petrosal running in the direction indicated, there is little support to the view that these are taste fibers from the chorda tympani. The fibers certainly could not be derived from the geniculate ganglion as will be pointed out later, and there is no conclusive evidence that glossopharyngeal fibers enter the chorda or the lingual from the otic as Carl advocates. If the chorda fibei^ do not enter the brain through the fifth or the ninth nerve they must pass by way of the seventh. The situation here is complicated by the fact that clinicians generally find no interference with taste in lesions of the facial nerve in facial palsy, provided the lesions are centrad of the geniculate ganglion. Lesions or fractures peripherally of the ganglion in the temporal bone usually destroy taste in the anterior part of the tongue on the same side. Such evidence is presented by Wachsmuth, or more recently and fully by Koster, Rosenfeld, Kopczynski and Scheiber. It may be pointed out here, however, that the evidenceIC

The Facial Nerve and Chorda Tympani. 597

is not conclusive in these cases that the portio intermedia is involved. It might easily happen that the motor root is destroyed without injury to the sensory. This is supported by the large number of cases of destruction of the facial centrad of the ganglion involving the sense of taste on the anterior part of the tongue. Such cases are cited by Claude Bernard, Brunns, Lehman, Scheier, Panski, Donath, etc. Summarizing the clinical and other pathological cases we may say that the evidence is quite conclusive to the effect that taste fibers do not enter the brain by way of the fifth nerve, equally conclusive that the fibers from the anterior part of the tongue do not enter the ninth nerve and inconclusive so far as the pars intermedia is concerned.

So far as the innervation of the soft palate for taste is concerned there is much difference of opinion. The weight of evidence however, as Dixon points out, is to the effect that it is innervated through fibers from the geniculate ganglion through the great superficial petrosal, the sphenopalatine ganglion and the palatine nerves. Part of these fibers may, however, come from the glossopharyngeal nerve through Jacobson's nerve and the anastomotic branch from the tympanic plexus into the great superficial petrosal.

Writers who deny that the taste fibers for the chorda originate from the geniculate ganglion usually argue that the ganglion is sympathetic and may be concerned with fibers of that type found in the lingual. There is little doubt that the chorda contains sympathetic fibers for the submaxillary gland. Sympathetic fibers are also undoubtedly present in the glossopharyngeal, the superficial petrosal nerves and Jacobson's nerve. Part of these fibers are probably postganglionic with their cells of origin in some of the many sympathetic ganglia of this region ; on the other hand a large proportion of such fibers found in the chorda tympani and associated visceral nerves are without question of the preganglionic type with their origin in the brain. Such include visceral efferent fibers of the secretory or excito-glandular type and vaso-motor fibers. The presence of such fibers, common to all the visceral nerves, in the pars intermedia of Wrisberg, cannot, therefore, be sufficient ground for assuming that this root or the geniculate ganglion are exclusively sympathetic.

Embryological and histological studies give more positive results.

D 598 Ralph Edward Sheldon.

Alexander ('02) found loss of taste after degeneration of the geniculate ganglion, which would show that the chorda fibers take their origin from it From the time of His it has been known that the course of conduction in nerves can be demonstrated through knowledge of their method of growth ; that is, an efferent nerve grows out from the brain while an afferent nerve grows, both toward the brain, and peripherally from its ganglion. His, Dixon and Streeter show conclusively that the facial is a mixed nerve, motor and sensory, and that its sensory part, the pars intermedia of Wrisberg, and the geniculate ganglion are similar in structure with the sensory parts of other nerves. Retzius, Martin, von Lenhossek and Ramon y Cajal, all working on mammals, have shown that the geniculate ganglion conforms to the cerebro-spinal rather than to the sympathetic type. It possesses the same kinds of cells as do other ganglia of the central nervous system and these cells send their processes both into the brain and peripherally. Roller, Turner, Huguenin, van Gehuchten, Ramon y Cajal and many others trace these central fibers into the fasciculus solitarius, known to be the center for taste in the human brain. Alexander obtained a peripheral degeneration of both the chorda and the great superficial petrosal. Sapolini traced fibers from the geniculate ganglion into the chorda, von Lenhossek believed that such was the course of the fibers, Weigner traced part of these fibers into the great superficial petrosal, Ramon y Cajal did likewise, while Dixon and Streeter found that both this and the chorda arise as outgrowths of the geniculate ganglion. This evidence shows that the geniculate ganglion is not sympathetic but cerebro-spinal in type, that its cells join centrally the taste centers of the medulla oblongata, and that its peripheral outgrowths form the chorda tympani and the great superficial petrosal nerves. Dixon shows that at the fifth week in the human embryo this latter nerve is free from any anastomoses with the fifth and that if any fibers grow into it from the fifth they must do so at a very late date. The anatomical evidence is, therefore, conclusive that the fibers for taste for the anterior part of the tongue are derived from the geniculate ganglion and enter the brain through the pars intermedia of Wrisberg. The question now arises as to the extent to which comparative neurology will clarify the question. It has been noted that the fas

D The Facial Nerve and Chorda Tympani. 599

ciQulus solitarius, the central connection of the fibers from the geniculate ganglion, is likewise the terminus for the taste fibers of the ninth nerve and is, therefore, the center for taste in man. Comparative studies show that the facial muscles of man are derived from the musculature of the hyoid arch of the lower vertebrates ; the facial is, therefore, the nerve of the hyoid segment, just as the ninth is the nerve for the first posthyoidean segment. Such studies have also shown that although the tongue musculature has grown forward from the postbranchial segmented mesoderm that the sensory surfaces of the tongue in man belong to more rostral segments. The case is, simply, that the sensory surfaces involved retain their primitive innervation with only a slight shifting while more caudal muscles grow forward when the tongue evolves in phylogenetic history. We, therefore, find that the mucous surface of the posterior part of the tongue in man belongs to the first posthyoidean segment or that of the ninth nerve, while the anterior mucous surface of the tongue is a part of the hyoidean and mandibular segments, or the segments of the facial and trigeminal nerves. It is known also that the Eustachian tube of mammals corresponds to the spiracular cleft of fishes in a general way, although other structures also enter into the formation of the Eustachian tube. In man the chorda passes over the tympanic cavity, but underneath the auditory ossicles. It has, therefore, long been considered a pretrematic or prebranchial nerve, although Froriep held the opinion that it is postbranchial. In 1904 Emmel, however, found that at an early stage the mammalian chorda passes beneath the spiracular cleft or primitive Eustachian tube and that its pretrematic position is taken up later. The nerve is, therefore, posttrematic.

The facial nerve of the lower vertebrates, leaving out of consideration the lateral line component of the neuromasts which is only apparently a part of the facial, is in every case a mixed motor and sensory nerve. As has been noted before, it is the nerve of the hyoid segment. As the musculature of this segment is, in all lower forms, derived from the lateral mesoderm, its motor component, as in man, is visceral motor. Primitively, as Johnston has shown in the lamprey, the facial nerve contained two sensory components, that is, it possessed fil)ers for the innervation of the skin for general sensation and

D 600 Ralph Edward Sheldon.

the region of the spiracle, roof and floor of the mouth for taste and other visceral sensation. In most forms above the cyclostomes the evidence is strong that the only sensory component remaining is the visceral, innervating mucous surfaces, while cutaneous sensation for the hyoid segment is served by fibers from the trigeminal or vagus nerves. J. Ramsey Hunt has, however, brought forward evidence to show that in man a part of the general sensory system of fibers still remains in the facial ; shown, for instance, by the persistence of tactile sensation on the anterior part of the tongue after section of the lingual above its junction with the chorda. In every carefully studied example among the lower vertebrates all taste buds in front of the glossopharyngeal segment are innervated through visceral sensory fibers from the facial nerve. They are likewise all derived from the geniculate ganglion and always end in the brain in the fasciculus solitarius which is, as has been noted, the center for taste in man.

Taking up the different groups of lower vertebrates more in detail the situation is made clearer. No attempt will be made, however, to consider the many mooted questions as to the comparative morphology of the different rami of the facial already ably discussed by Herrick, Cole, Coghill, etc In the selachians, as will be noted from Fig. 1 (Chlamydoselachus anguineus), the facial contains the usual motor component, the ramus hyoideus for the hyoid musculature. There is a small component for the skin, derived, however, from the trigeminus nerve in all probability, through the anastomotic rami from the Gasserian ganglion. There are several rami for the sense of taste. Close to, or else arising directly from the geniculate ganglion is the palatine nerve for the mucosa of the roof of the mouth. Either from this or from the main trunk, slightly more distad are given off prespiracular or pretrematic rami. These innervate the spiracle and occasionally a part of the floor of the mouth. Descending with the main branch of the facial, the hyomandibular, is a large sensory component for the mucosa of the floor of the mouth rostrad, the mandibularis internus. This is evidently the homologue of the chorda tympani. Practically all selachians, as is shown by Cole, Strong, Ewart, Green, Jackson and Clarke and other workers, exhibit these conditions. We may, therefore, say that in this group the

D The Facial Nerve and Chorda Tympani. 601

sensory fibers of the facial serve the sense- of taste of the roof of the mouth through the palatine rami, and the floor of the mouth rostrad through the pretrematic, and particularly the mandibularis intemus.

In teleosts (Fig. 2, Menidia; Fig. 3, the cod) conditions are quite similar. The motor nerve, the ramus hyoideus, innervates the muscles of the hyoid arch, accompanied by a cutaneous branch from the Gasserian ganglion. A visceral sensory palatine arises from cells in the geniculate ganglion for the taste buds of the roof of the mouth ; a large posttrematic ramus, the mandibularis intemus innervates the floor, including the mucosa over the bones of the hyoid arch. A few visceral sensory fibers in Menidia leave the geniculate ganglion with the maxillaris for the upper lip and mucous lining of the upper jaw. The glossopharyngeal nerve here, as in higher forms, sends a lingual branch to the taste buds of the floor of the mouth. In the cod, as shown in Fig. 3, a branch from the petrosal ganglion of the ninth nerve, Jacobson's anastomosis, joins the posterior palatine nerve which largely distributes to the mucosa of the roof of the mouth. In Pleuronectes, as shown by Cole, there is a similar anastomosis. We know, from Dixon, Streeter, Ramon y Cajal, etc., that the origin of the great superficial petrosal of man is similar to that of the palatine VII in fishes. It is interesting to note that in fishes there exists an anastomosis from the ninth quite similar to that between Jacobson's nerve and the great superficial petrosal in man. It should be emphasized, however, that such an anastomosis is absent in many cases, as in the selachians and Menidia, so that the palatine VII in fishes is actually a derivative of the facial, flacobson's anastomosis of fishes and Jacobson's nerve in man are apparently both palatine or pharyngeal rami of the ninth nerv^e carrying taste fibers for the roof of the pharynx.

An important special case is to be observed in the catfishes. Herrick has shown that the outer skin of Ameirus is covered with taste buds similar to those in the mouth. He has likewise shown, through extensive and painstaking experiments, that the fishes react to sapid substances on stimulation of these external taste buds in exactly the same way that they do when such substances come in contact with the mouth. In these fishes the external as well as the usual internal

D 602 Ealph Edward Sheldon.

taste buds are innervated through the facial nerve. Many special rami are developed for this purpose, leading to an enormous hypertrophy of the nerve and its center in the brain, the fasciculus solitarius.

In the amphibia conditions vary somewhat, but are essentially similar to those in fishes. Cutaneous fibers may be present in the Vllth but if so are derived from the Xth by way of a communicating branch. The ramus hyoideus in all amphibia is homologous with the motor facial of the fishes and man. A palatine ramus from the geniculate ganglion is always present and in the Urodela is joined by Jacobson's anastomosis, the fibers innervating taste buds in the roof of the mouth. The taste buds of the anterior part of the mouth are innervated by a nerve, called the ramus alveolaris in the Urodela and the mandibularis intemus in the frc^. It is, in both cases, derived from the geniculate ganglion as in the fishes. It is said to be, however, postspiracular or posttrematic in the Anura and pretrematic in the Urodela. (See Coghill, '02.) The origin and the region innervated are the same in both cases and it is without doubt the functional equivalent of the chorda tympani. The posterior part of the tongue is innervated by the lingual IX as in man.

In all these cases it will be noted that there are present the same three rami of the trigeminus as in man. There are, likewise, in the amphibia as in man, anastomoses between the terminal rami of the mandibularis and the nerves for taste for the rostral part of the tongue.

It is evident from the foregoing that the facial of lower vertebrates is both sensory and motor ; that its motor portion is homologous with the facial proper in man; that its sensory portion through nerves homologous with the chorda tympani of man innervates the taste buds of a region comparable to the anterior part of the human tongue and that another branch of this same sensory element of the nerve, the palatine, is homologous with the great superficial petrosal of man. The evidence of comparative neurology, therefore, offers the strongest possible confirmation of the view that the chorda tympani of man is the nerve for taste for the anterior part of the tongue and that its fibers are derived from the geniculate ganglion, entering the brain

D The Facial Nerve and Chorda TympanL 603

through the pars