Talk:Paper - On the vascularization of the spinal cord of the pig (1914)

From Embryology
Revision as of 00:38, 25 June 2020 by Z8600021 (talk | contribs)

oN THE VASCULARIZATIOX OF THE SPINAL CORD OF THE PIG

Hoskins EK. On the vascularization of the spinal cord of the pig. (1914) Anat. Rec. 8(7): 371 - 392.

E. R. HOSKINS From the Inslilule of Anatomy, University of Minnesota

FIVE FIGURES

HISTORICAL

Until 1904 little work had been done upon the development of the blood-vessels of the spinal cord, except that of His ( '86) who undertook to follow the growth of these vessels in human embryos. The observations of this author have been largeh- disputed by Sterzi ( '04) and Evans ( '09), the two writers who have done most of the work in this field.

By far the most comprehensive publication upon the development of the vessels of the spinal cord is that of Sterzi ('04j. In this he discusses the development of the vessels in the five higher classes of vertebrates. As a tj'pe of the ^Mammalia he uses the sheep. He brings out the following points:

The blood-vessels first approach the cord at the ventro-lateral border and spread over the ventral surface, then over the lateral, and finally over the dorsal surface. Each vertebro-medullary artery as it approaches the cord divides into a ventral and a dorsal ramus, the ventral and dorsal radical arteries. The ventral radicals from either side halt at the lateral edges of the floor-plate, and each divides into a cranial and a caudal branch. These anastomose with those of adjacent segments and form two longitudinal arteries on tlu^ ventral surface of the cord, the "tractus arteriosus primitivus. " Later they send out medial rami and through these become connected. Still later, alternate parts of the two tracts degenerate while other parts continue to develop. These enlarged segments are joined together through their medial rami and form a single ventral artery which Sterzi terms the tractus arteriosus ventralis, and which is tiie anterior spinal artery of most authors. From the primitive tract, dorsal rami enter the substance of the cord. Each dorsal ramus forms a loop and gives rise to a vein, which courses ventrally and enters the primitive sulcus. Later other vessels extend into the cord from the lateral, and still later from the dorsal surface.

The dorsal radical arteries, where they di\ide, form many small longitudinal capillaries just ventral to the i:)()ints of emergence of the tlorsal ner\'e roots. From these capillaries there is formed later a longitudinal artery on either side of the cord, in this plane (tractus arteriosus lateralis). The vessels entering the cord are f'rst solid and later become hollow.

Evans ( '09) shows by a series of injected pigs the early de\'elopment of anterior spinal artery In his figures the mid-ventral surface of the cord is shown to be free from vessels until the embryos are 8.5 mm. in length, and the mid-posterior surface until after the pigs are between 8 and 10.5 mm. in length. He does not take uj) the later stages,

MATERIAL AND MKTIIODS

For a study of this nature, injected embryos are indispensable, and they are best injected while living, with warm india ink diluted one-half with weak ammonia water. It is pi-ef(M'al)le to inject through the umbilical arter}^ rather than through the umbilical vein, because the arteries are less readily ruptured and because the route between them and the vessels of the spinal cord is much more direct.

Eml)ryos used for thin serial sections are better if they are congested instead of being injected. This congestion is accomplished as follows: The uml)ilical cord is tied while the embryo is yet living, thus causing an increase in the blood pressure in the aorta. One of the most direct outlets for this increased pressure is the system of segmental arteries, and through these the bloodvessels in and around the spinal cord soon l)ecome engorged. When this condition is i-eached, as evidenced by the increased redness of the dorsal region of the embryo, the li\'e embryo is droi;)]Ded into a fixing fluid which penetrates rapidl}^ so that the capillaries are fixed before they collapse. Bouin's picro-formo-acetic mixture serves this purpose very well. P^mbryos treated in this manner show the smaller vessels much more plainly than those fixed in the usual way.

Small injected embryos which have been cleared in oil may be dissected under the binocular microscope and all the external vessels of the cord demonstrated, or they may be sectioned in celloidin to show the internal vessels.

From pigs larger than 25 or 30 mm. the cord with its membranes may be dissected out and embedded in celloidin and cleared, or for temporary preparations may be cleared directly.

Serial sections of the cleared embryo or spinal cord can be kept permanently between two pieces of paper soaked in oil, or can be transferred to slides and mounted in damar gum. To make slide preparations, the sections are cut in oil and placed on a piece of thin paper in the same order they are to have on the slide. Another piece of oiled paper is laid down upon them and the whole inverted. The first paper is now peeled off and the other paper holding the sections is inverted upon the slide. This paper is then peeled off, leaving the sections on the slide in their proper order. They may then be washed carefully with xylol, and covered.

There are several advantages in a study of this kind, in section ing celloidin-emljcddcd embryos free-hand in oil. The cord can be turned so that sections may be cut through any plane. Sectioning is done more rapidly than with a microtome, and much time is saved. It is easy to make transverse, sagittal and frontal sections from any region of tlie same cord. The block can be examined witli a Umis. (hiring tlie sectioning, and any particular vessel or vessels included, in a section. Also, sections may be made of different shapes.

HLOOD-VKSSKLS OF TIIK XKAH Ffl.T.-TKlOI FKTIS

In order to determine as near as j>ossible the arrangement of the blood-vessels of the s|)inal cord in the adult comlition, a iHnnl)er of f(>tal pigs, neai' full-term, were injected and their spinal cords dissected out for study. Alt liough these showed some variation in the blood-vessels, as is to be expected, a general


374 E. K. HOSKINS

plan was to be olisorvod. The following descrii)ti()n i.s of the tj'pical condition found in the blood-vessels in and around the spinal cord of pigs of about 240 nun. in length:

The terminology used by Kadyi ('89) for the blood-vessels of the adult human cord will he rt^ferred to frequently.

On the surface of the cord at this stage are four main longitudinal arterial systems which are located, one, median on the ventral surface; one, on each dorso-lateral surface; and one median on the dorsal surface.

The vertebro-meduUary branches of the dorsal segmental arteries approach the cord laterally and each divides into a dorsal and a ventral branch termed the posterior and ventral radical arteries respectively (figs. 1 and 2). The latter reach the cord along the cranial surface of the spinal nerves and ganglia, and, on the cord, run cranially. The}^ may be equally distributed to the two sides of the cord and to the different regions of it, or some regions may ha\'e more than others. They average eighteen in number. The \Tntral and dorsal radical arteries have lost their connection wdth the vertel)r()-medullary artery in places and extend between the artery on the ventral surface of the cord and one of those on the dorso-lateral surfaces.

Soon after reaching the cord each ventral radical artery branches, giving off one ramus which courses cranially, and another which extends caudally. These divisions anastomose with those of neighboring ventral radicals, on the same or opposite side

Fig. 1' Ventral svirfarc of the iimldlr thoracic- regiou of the spinal eonl of a fetal pig, 240 mm. in length. .4..S..4., anterior spinal artery; A.R.W, ventral radical artery; T.A.P., accessory anterior spinal artery (remains of primitive arterial tract); V.R.V., ventral radical vein; V^..S..4., anterior s])inal vein; C.P., capillary plexus. X 0.

Fig. 2 Dorsal surface of I he same part of t he eoid shown in figure I. A .M .D., median dorsal artery; A.P.L.. dorso-iatcral artery; A.R.I)., dorsal radical artery; V.^f.I)., median dorsal vein; V .I'.L.. dorso-lateral vein and plexus; V.R.D., dorsal radical vein; \'.fj.. dorso-laleral venous plexus. X 10.

' The figures in this pajx')-, exc('i)t nunilKMs '.] and I, were drawn witji a camera lucida. The size of the eml)ryf)s is the greatest length, as iiieasured in 7.") per cent alcohol.


.J^^ft"


-*1^,


TA P-1


.-■)■■


I


tC.P


>, '



375


37(3 E. R. HOSKINS

of the cord, and form in this way the anterior spinal artery, which Ues in or near the median ventral line. Occasionally a radical artery, instead of dividing, goes across the cord and joins the cranial or caudal ramus of the one on the opposite side. The anterior spinal artery has a winding course, bending laterally to meet the vessels which form it, and making many smaller irregular bends to one side or the other. In some places it may lie to one side of the mid-line for several segments and where this occurs there are found numerous longitudinal rami of the anterior spinal artery, or the cranial antl caudal tlivisions of the ventral radicals. These may be called accessory anterior spinal arteries, and are sometimes present even where the anterior spinal artery lies in the mid-line. Here they are located between this vessel and cord, at the lip of the ventral median fissure. They are the remnants of the "tractus arteriosus primitivus" of Sterzi.

The dorsal radicular rami of the vertebro-medullary arteries are much more numerous than the \'entral ones. They course dorsally along the cord and in a slightly cranial direction, to a plane just ventral to the emergence of the dorsal roots of the spinal nerves, where they divide into two rami, one extending cranially and one caudally. Each of these rami anastomoses with the one of the adjacent segment, and thus there is formed on either side an irregular longitudinal vessel, the dorso-lateral artery (fig. 2; tractus arteriosus postero-lateralis of Kadyi). From this artery recurrent rami supply the dorsal nerve roots and si:)inal ganglia, and the lateral surface of the cord. Other rami, two or three in each segment, and much larger than the above, run dorsally and by longitudinal anastomoses with each other, and with similar rami from the opposite side, form an artery in the mid-dorsal line which may be termed the median dorsal artery.

Wry small rami from the dorso-lateral arteries run ventrally along the cord and unite with oth(M"s from the anterior spinal artery, forming a plexus on the ventral and lateral sides. These ventral rami of the dorso-laterals anastomose freel_y in a longitudinal direction and form one or more small longitudinal arteries between the ventral and dorsal nerve roots in some parts of the cord. These are the tracti arteriosi laterales, and ventro-later


VASCULARIZATIOX OF THE SPINAL CORD 377

ales, of Kadyi. Still other small rami from the dorso-lateral and median dorsal arteries form a capillary plexus on the dorsal and dorso-lateral surfaces of the cord. In a few places along the cord the dorso-lateral arteries are double, one divisif)n lying dorsal to the dorsal nerve roots, and corresponding perhaps to the '"tractua arteriosus posterior" of Kadyi.

The median dorsal artery is a very irregular longitudinal vessel formed by the dorsal rami of the dorso-lateral arteries, as described above. In places it is double or ma\' show a longitudinal capillary arrangement. Many of its lateral rami anastomose longitudinally forming small arteries parallel with the median dorsal artery (fig. 2). This is also true of the dorsal rami of the dorso-laterals. By the anastomoses of the rami of the various arteries just described, the entire cord is surrounded by an arterial vascular system, and from all parts of this network smaller arteries penetrate its substance.

The veins of the spinal cord are in three princijial longitudinal systems, and other smaller ones. Of the three, two are dorsal and one ventral. All three show evidence of their capillary origin. The anterior spinal vein is the smallest of the three (tig. 1). It lies between the cord and the arteries, in the median ventral line. It is larger than the accessory anterior spinal arteries, but never attains the size of the anterior spinal artery i)r()i)er. It is very irregular and in some regions is entirely replaced by a narrow network of capillaries.

On either side of the median ventral sulcus, the cord is covered with large venules, some of which lie between the cortl and the arteries, and some of which are external to the latter. They are often two or three times as large as the arterioles to which they correspond. They anastomose freely with the anteri«)r spinal vein and empty laterally into the ventral radical veins which are in close relation with. the ventral nerve roots and ventral radical arteries, but which are nuicli nu^v numerous tiian the latter, one being present on nearly every nerve root. They ilrain blood also from the lateral sm-face of the cord. Their ventral and dorsal rami often form short, small, longitudinal veins by anastomoses, some of which in otluM- animals have been named, antero-lateral.


378 E. R. HOSKINS

etc. The 1)1()()(1 from tho anterior spinal veins and venous capillaries of the gc'neral ventral surface form, in places, transverse channels which are perhaps large enough to be called veins.

On either side of the dorsal surface of the cord there extends longitudinall}' a large irregular vein, the dorso-lateral, about half way between the artery of the same name, and the median dorsal artery. Some parts of these vessels and their rami, like the ventral venous capillaries, lie external to the arteries and some internal to them. They are the largest vessels on the cord with the exception of the anterior spinal artery (compare figs. 1 and 2). Dorsally these veins are united through large capillaries, and blood leaving the cord in the median line may flow either to the right or left. Half way between two consecutive nerve roots the dorsolateral veins usually break up into many divisions so that each may be seen to drain blood from adjacent halves of two segments. Laterally' they empty through one or more divisions into the large dorsal radical veins, one of which lies upon each dorsal nerve root (fig. 2).

A fourth longitudinal venous system, smaller than the three described, lies in the median dorsal sulcus. It resembles the dorso-lateral veins except that it is more irregular, and in places it may be entireh^ lacking. Its lateral rami empty in the dorsal venous capillary plexus or directly through larger vessels into the dorso-lateral veins. It may be termed median dorsal venous system.

Some of the venous capillaries of the lateral surface drain into the dorso radical veins, some into the dorso-lateral veins, and some into ventre radical veins, and all these vessels together with the anastomoses of the veins on the ventral and dorsal surfaces already mentioned above, completely surround the cord with a venous system, corresponding to the system described for the arteries.

Of the arteries entering the cord, the largest are those in the ventral fissure, the ventral central arteries, which form two nearly parallel rows, but which are not paired. They arise from the anterior spinal arteries, or the accessory anterior spinal arteries. They show evidence of the capillary origin in longitudinal anas


VASCULARIZATION OF THE SPINAL CORD 379

tomoses found between vessels of the same side. These anastomoses are numerous in the fissure, particularly near the vessels from which the ventral central arteries arise.

The ventral central arteries vary considerably in size, some being as large as the vessels they arise from and others much smaller (fig. 3) . The course of the smaller vessels is usually more irregular than that of the larger. They pierce the substance of the cord at different levels, some entering near their origin and others extending some distance into the fissure. Their general course is dorso-lateral, but those entering near the mouth of the fissure may bend very sharply to the side and enter the ventral horn of the gray substance. The others course more dorsally nearly to the level of the central canal where they make a decided lateral bend, and divide into two or more rami, although sometimes they give off rami more ventrally than this (figs. 3 and 4i. The principal divisions of these arteries extend in a longitudinal plane, and anastomose with similar rami of adjacent vessels. They also give off smaller arteries and capillaries which ramify through the gray matter in all directions, helping to form a dense plexus. The longitudinal arteries tend to form loops after they have coursed in one direction for a short distance, as thej' do in young embryos (fig. 5). One artery may form several such loops, producing as many longitudinal vessels, each succeeding vessel lying dorsal or lateral to the last, and of a lesser cahber. These smaller longitudinal vessels anastomose with each other ventro-doi*sally and laterally by rami which usually leave them at right angles, and also anastomose with rami from vessels other than the ventral central arteries, as will be described later.

Besides the rami of the central arteries just described, other rami extend farther laterally into the gray substance before l)ranching. Some of these, instead of ft)rming longitudinal vessels, form small irregular ones which ramify through the gray matter in all directions, anastomosing with similar vessels from other arteries in this region and forming a dense capillary plexus in the ventral and dorsal horns.

Other arteries, smaller than the ventral central arteries, enter the c(ird from the dorsal median sulcus and course ventrallv and


380 E. R. HOSKINS

laterally to tlio dorsal horns of tho gray substance. Here they form still smaller vessels resembling to some extent those formed by tlie \'entral central arteries, but most of their rami are short and do not extend longitudinally. These may be called the dorsal central arteries, but they are more similar to the peripheral arteries from other surfaces of the cord, than to the ventral centrals, and perhaps should be called dorsal peripheral arteries. They give ofT many small lateral rami in the white substance and in the outer part of the gray substance.

In addition to these vessels, other small arteries enter the cord from all sides, from the arteries and arterioles which surround it. These are the peripheral arteries referred to above. The}^ are very numerous and in a single thick cross section as man}^ as fifty or sixt}' of them ma}^ be counted. They give off short rami in the white layer of the cord and extend into the gray layer. These rami branch and anastomose and form a loose capillary network. The vessels entering the gray substance enter into the longitudinal plexus already described and give oflf lateral rami which branch freely and anastomose.

The longitudinal vessels which arise from the ventral central arteries are quite large, but the other vessels formed from these trunks, and those formed from the dorsal central and peripheral arteries, are much smaller. A very thick section presents a l)icture of an inner core of longitudinal vessels with other vessels extending into it at right angles from all points on the periphery of the cord (figs. 3 and 4).


Via. 3 Saf!;ittal section from tlic lower thoracic region of the .si)inal coni of a 240 mm. fetal pig. A.S.A., anterior spinal artery; .l.f. /I ., ventral central artery; ,1./'., peripheral artery; A.C.P., dorsal central artery; r..S..4., anterior spinal vein; V.C.A., ventral central vein; \'.('.l'., dorsal central vein; l'./'.. jx-ripheral vein. X 3.5.

Fig. 4 Transverse section through the lower thoracic region of the spinal cord of a 240 mm. fetal pig. A.S.A., anterior s|)inal artery; A. P., peripheral artery; A., artery; A.C.A., ventral central artery; S.A.S.A., accessory anterior spinal artery; A.I'.L., dorso-Iateral artery; A.M.D., median dorsal artery; .t.N.r., anterior spinal vein; V'., vein; V.P., peripheral vein; V.P.L., dorso-latcral vein; 1'.. !/./>., median dorsal vein. X 35.



A VMD AMD AP



382 E. K. HOSKINS

DKNKLor.MKNT OF TH1-: HL(>()I)-\KSSELS

The early development of the anterior spinal artery has been described by Evans ('09) and Sterzi ('04).

Some pig embryos of 12 mm. show a fairly well developed anterior spinal artery, while in others of 14 or 15 mm. it is just beginning to form. Although, after this vessel is once formed, it does not undergo marked changes, there is some modification. For example, the ventral radical arteries meet it at right angles or nearly so, until the embryo is 40 or 45 mm. in length. Thereafter, the growth of the cord and the fixed position of the radicals seem to cause the artery to be pulled laterally by the radicals, and a gradually decreasing angle is formed at the points where the radicals meet it (fig. 1).

As the embryo grows, the number of radical arteries continues to decrease even after the anterior spinal artery is well formed. This seems to be true until the embryo reaches the length of about 100 mm.

Some of the arterial capillaries on the ventral surface of the spinal cord are continuous with the anterior spinal artery directly, or indirectly through remains of the tractus arteriosus primitivus, and others are continuous with the capillaries of the lateral surfaces of the cord.

A dorso-lateral artery is formed in the capillary plexus on each of the lateral surfaces of the cord, just ventral to the point of emergence of the dorsal nerve roots. The dorsal radical arteries branch in this region and give oiT dorsal and lateral rami, which are continuous with the lateral capillaries just mentioned. A very irregular longitudinal vessel develops where certain of these

Fig. .5 Transverse section through the mid thoracic region of the spinal cord of an 11 mm. |)ig embryo. T.A.P., primitive arterial tract; R.D.T.A.P., R.L. T.A.I'., R.M.. dorsal, lateral, and medial rami of the primitive arterial tract; A.V..\I., vertehro-medullary artery; A.R.V., A.R.D., ventral and dorsal radical arteries; C.R., Cr.R., D.R., V.fi., caudal, cranial, dorsal and ventral rami of the dorsal radical artery; I). P., D.L.P., dorsal and dorso-lateral capillary plexuses; D.L.C.G., V.L.C.G., dorso-lateral and ventro-lateral groups of peripheral capillaries; V.L.V.P., ventro-lateral venous plexus; V.R.V., V.R.D., ventral and dorsal radical veins; S.P.G., spinal ganglion; V.N.R., D.N.R., ventral and dorsal nerve roots; S.N ., spinal nerve. X 200


VASCULARIZATION OF THE SPINAL CORD


383



3S4 E. R. HOSKINS

capillarios iiuToaso in sizc\ perhaps on account of the increased pressure from the dorsal radical arteries. This longitudinal vessel is indicated in embryos of 12 mm. and is quite stronglj' developed in embryos of 15 to IS nnn. In these stages it seems to dip ventrally to meet the approacliing radicals, as pointed out by Sterzi for the sheep ('04). -Vs the embryo grows, this dorso-lateral artery becomes more and more regular. It is still somewhat irregular in embryos of (30 mm. but quite regular in those of 75 mm. The dorso-lateral artery never attains the size of the anterior spinal nor is it ever so regular in its course. In places it may develop as two or more vessels, but these are always smaller than the single artery. The dorso-lateral arteries are each continuous with the capillaries of half the cord in the early stages, but as the cord increases in size they supply directh^ only the dorso-lateral surface.

The capillary network on the lateral surface of the cord is at first continuous with tliat extending through the mesenchyma of this region as far, laterally and dorsally, as the myotomes and body wall respectively. In later stages when the membranes of the cord begin to develop, the connections between the vessels of the cord and those in the mesenchyma around it are lost.

The median dorsal artery is the last of all the vessels on the cord to develop. In pigs of 30 nun. it is still very irregular and indefinite, and is entirely lacking in places, although the vessels which go to form it, the dorsal rami of the dorso-lateral arteries, may be seen in embryos of 20 mm. In pigs of 45 mm. it is quite definite, lying in or near the mid line of the dorsal surface, as described above for the 240 mm. embrj'o It never becomes very regular, and in pigs of 100 nnn. it resembles the condition seen in the pig oi 240 mm. It is continuous with the arterial caj)illaries of the dorsal surface of the cord and with the dorso-lateral arteries.

In addition to these main arterial trunks there develop on various parts of the cord, especially on the lateral surfaces, short longitudinal arteries. These are never large or regular. The}' have been desci-ibcd in connection with adult human coi-d under the terms "tractus arteriosus; ventro-lateralis, posterioris, and


VASCULARIZATION OF THE SPINAL CORD 385

lateralis" (Kadyij. Of these, the "tractus arteriosus posterior" is the most prominent and corresponds to the description in this paper of parts of the dorso-lateral artery, where it sometimes has two divisions, one of which runs dorsal to the dorsal nerve roots and the other ventral to them. The dorsal divisions are evidently the same as this ' tractus. '

The veins on the cord develop in much the same way as do the arteries. The ventro-lateral surface of the cord in very young embryos is covered with capillaries, and these are continuous laterally with the capillaries in the mesenchyma round the neural tube. Medially they become continuous with the lateral rami of the primitive arterial tract. When this tract becomes separated from the cord by the ingrowth of mesench>niia, these capillaries send medial outgrowths between the tract and the cord, as seen in embryos of 12 to 15 mm. Dorsall}' the}' grow along the cord and spread over the dorso-lateral surface (pigs of 6.2 mm.) and later over the dorsal surface (pigs of 7.5 mm.). Laterally they spread over the ganglia.

From the ventral surface, the blood draining away through the capillaries soon establishes segmental vessels, the ventral radical veins, which course laterally along the nerve roots. Each radical vein on one side drains adjacent halves of two segments. These receive blood from the capillaries of the ventral, lateral, and ventro-lateral surfaces. Lying in the ventral median fissure in young embryos, small longitudinal veins may be seen in different regions of the cord, and in embryos of 25 to 30 mm. a fairly detinite longitudinal vessel may be found here. This vessel in still older embryos becomes a more definite trunk and may be called the anterior spinal vein. It never attains the size of the anterior spinal artery. Laterally it drains into the ventral radical veins.

Some of the ventral and lateral capillaries of the younger embryos, early become dilYcrentiated into veins. This is especially true of the dorsal vessels. From these, some of the bUxxl drains laterally out through vessels in the mesenchyma to the myotomes. .\ pig of (').2 mm. shows three planes in whicii this occurs, one on a le\el with tlu^ dorsal surface, one just al)ove the level of the ven


3S6 E. R. HOSKINS

tral surface, and one about half way between the other two. At the myotomes the blood drains Aent rally into the intersegmental veins. Some of the capillaries of the lowest of these three planes, which tlrain the blood from the lateral surface of the cord and from the ganglia, soon become large and are called the vertebromeduUary veins, one pair of which is formed for each segment. In older embryos they course along the spinal nerves with the vertebro-meduUary arteries. They recei\e the blood from the ventral and dorsal radical veins. The former have been described. The latter develop along the sides of the gangUa in the capillaries already mentioned. At firsi they carry only a part of the blood from the dorsal surface of the cord, but later (pigs of 25 mm.) they carry practically all of it. They are more numerous than the corresponding ventral radicals, and are found in every segment.

The venous capillaries of the dorsal-lateral surface on either side draining toward the nerve roots early establish longitudinal veins. These are onlj' about half as long as a segment of the cord. Figure 5 of an 11 mm. pig, shows an indifferent plexus on this surface, but in 15 to 17 mm. embryos, fairly definite vessels may be seen. These become more and more regular as the animal develops, and as embryos of 50 to 60 mm. show, they form a venous system on either side of the cord just dorsal to the dorsal nerve roots, much like that described for the 240 nun. stage. These systems constitute the dorso-lateral veins (fig. 2).

The first blood vessels entering the cord grow in as capillaries from the ventral surface. Sterzi ('04) reports vessels in the cord of a sheep of 5.5 nmi., but they were not apparent in the cord of pig embryos of less than 7.5 nmi. These vessels are the dorsal rami of the primitive arterial tracts, of the lateral rami of these tracts, and of the other capillaries near the median line. They are the first indications of the central arteries and veins. They form two nearly parallel rows, one on either side of the epend>nnal layer, f)r some of them may lie in this layer. They grow dorsally about half way to the dorsal surface of the cord. They exhil)it numerous longitudinal anastomoses and form a plexus along the ateral side of the ependymal layer in each half of the cord. These are true capillaries at first, but soon differentiate into arteries and veins.


VASCULARIZATION OF THE SPINAL CORD 387

Those coming directly from the primitive arterial tracts all become arteries, while those coming from the vessels lateral to the tract may become either veins or arteries.

In embryos of 9.5 mm. another group of capillaries may be seen to have entered the cord. These come from tjie lateral surface, extending medially nearly to the central canal. Later they anastomose ventro-dorsally and longitudinally, among themselves and with the vascular sprouts from the ventral surface.

The vessels in the cord of a pig of 11 mm. present the following characteristics, as shown in figure 5. Rami from the primitive arterial tract may anastomose with those from the ventral capillaries. Neighboring vessels of the same kind anastomose freely and give off lateral rami into the anlagen of the ventral horns of gray substance. These rami branch and anastomose with each other and form loops which anastomose with the central vessels from which they arise, or with neighboring vessels. In a plane just above the anlagen of the ventral horns each of the central vessels ends blindly, or divides into a caudal and a cranial ramus, which anastomose with adjacent similar rami and form irregular longitudinal vessels. By other anastomoses among the central vessels, a longitudinal plexus is formed, which covers very coinpletely the lower half of the lateral side of the ependymal layer.

A comparison of figures 3 and 5 shows how closely the form and arrangement of these capillaries corresponds to that of the future central arteries and veins. Besides these main capillaries two smaller lateral groups are present at this stage. These may be called the ventro- and dorso-lateral groups, and later fonn peripheral arteries and veins. Both groups enter the cord from the capillaries on the lateral surface between the dorsal and ventral nerve roots. The ventro-lateral group enters at the level of the dorsal extremities of the central vessels, and courses medially and anastomoses with them. Occasionally the ventro-lateral group gives off rami which extend into the anlagen of the ventral horns. The capillaries of the dorso-lateral group are confined to the dorsal two-fifths of the cord, and although they anastomose with each other at this stage, they do not anastomose with the central or ventro-lateral capillaries. They course medially and

THE ANATOMICAL RECOBO, VOL. S, NO. 7


388 E. R. HOSKINS

dorsally along the ependjana, ending blindly or forming loops, but do not reach the dorsal surface.

As development proceeds, the lateral groups of capillaries shown in figure 5 spread dorsally and ventrally and capillaries enter the cord from the periphery. With the exception of the above-mentioned dorso-lateral group of capillaries, all the vessels entering the sides of the cord grow toward a common center, namely, an area on the lateral border of the ependyma about half way between the dorsal and ventral surfaces. The dorso-lateral group of capillaries which are shown in the same figure send rami toward this center after the embryo attains the length of 14 mm.

The vessels from the dorsal surface grow ventrally along the ependyma and unite with the dorsal rami of the primitive arterial tract. This union continues the plexus on the lower part of the epend>Tna dorsally so that the ependyma except below the floorplate and above the roof-plate, is entirely surrounded by a capillary plexus. A thick transverse section of the cord of an embryo of 25 mm. shows this plexus with numerous vessels extending from it laterally at right angles. These lateral vessels are joined together by dorso-ventral rami. This picture is characteristic of the cord until the embryo reaches the length of 30 or 35 mm. when it is changed by other peripheral vessels meeting the ependymal plexus obliquely and by the branching of the vessels in the anlage of the gray substance.

By this time the central arteries from both the ventral surface (ventral central arteries) and from the dorsal surface (dorsal central or dorsal peripheral arteries) have become quite large, although the latter do not nearly equal the size of the former. The ventral central arteries have formed more longitudinal loops similar to those shown in figure 5. They are separated more and more from each other, owing to the growth of the cord, and as this separation continues the longitudinal vessels grow in length.

In embryos of 35 to 40 mm. in length the peripheral arteries from all sides together with the lateral rami of the central arteries have formed a dense plexus in the gray substance, although the white substance contains only the peripheral arteries running through it, and the short branching rami given off at right angles


VASCULARIZATION OF THE SPINAL CORD 389

from them. By the time the embryo reaches a length of 50 mm. the capillaries in the white layer have much the same appearance as those of the full term fetus, except that in the latter they branch and anastomose more freely and the growth of the cord tends to separate both the peripheral vessels and the central vessels. Embryos of 75 to 100 mm. in length show the arteries in the cord quite as completely developed as in the 240 mm. embryo.

The posterior rami of the primitive arterial tract in the ventral part of the cord of embryos of 12 to 15 mm. are more numerous than the central arteries in the 240 mm. embryo which are formed from them.

The veins within the cord develop in the same planes as the arteries, and from the same plexus of capillaries that form the latter. They may be called the central and peripheral veins corresponding to the similarly named arteries. They are shown in figures 3 and 4 in a fully developed condition.

SUIVIMARY

The dorsal rami of the primitive arterial tract, and other rami from the capillaries in its immediate vicinity enter the cord, forming an undifferentiated capillary plexus (fig. 5) and this plexus later becomes differentiated into arteries and veins. It was not found, as stated by Sterzi for the sheep, that each dorsal ramus of the primitive arterial tract grows into the cord, and forms a loop, giving rise to a vein which grows back along the artery to the ventral surface.

The dorsal rami of the primitive arterial tract are more numerous than the ventral central arteries which develop from them.

Sterzi reports solid blood-vessels in the cord of sheep of 5.5 mm. and hollow ones in those of 6.6 mm. In pig embryos the bloodvessels within the cord seemed to appear first as hollow vessels. These are seen first in embryos of 7.5 nmi. in length.

The "tracti arteriosi laterales" of Sterzi, are the dorso-lateral arteries of this and postero-lateral of other papers, and are the posterior spinal arteries of human descriptive anatomy. Evans shows these two tracts first united by medial anastomoses in a


390 E. R. HOSKINS

pig of 8.5 nun. in length, but many such anastomoses are to be found in embryos as small as 7.5 mm. in the cervical and thoracic regions, and one specimen of 6.2 mm. showed them in the cervical region.

The embryos described in this paper show the mid- ventral and mid-dorsal surfaces of the cord to be covered with blood-vessels at a somewhat earlier stage than has been described.

As reported by Sterzi ('04) and Evans ('09), blood-vessels first appear on the ventro-lateral surface of the cord, then on the ventral, then on the dorso-lateral, and finally on the dorsal surface.

The blood-vessels on the cord are continuous with those in the mesenchyma surrounding it until the membranes of the cord are formed.

It is generally stated in textbooks of human anatomy that the spinal artery arises from the vertebral arteries, and is reinforced by segmental spinal arteries. It il rather to be considered that this artery arises from the segmental spinal arteries, and anastomoses with, or is reinforced by, the vertebrals.

The term median dorsal is suggested for the artery present in places in the median dorsal line of the spinal cord.

My thanks are due to Dr. Richard E. Scammon for his constant interest in the progress of this work, and for his many helpful criticisms.


VASCULARIZATION OF THE SPIXAL CORD 391

BIBLIOGRAPHY

Adamkiewicz, a. 1881 Die Blutgefiisse des Menschlichen Riickenmarkes. I. Teil: Die Gefasse der Ruckenmarksubstanz. Sitzungsber. k. Akad. Wiss., Wien, Math.-Xaturwiss. Kl., Bd. 84, iii. Abt.

1882. II. Teil: Die Gefasse der Riickernmarksoberflache. Sitzungsber. d. k. Akad. d. Wiss., Wien, Math.-Xaturwiss. Kl., Bd. 84, lu. Abt.

DoRELLE, P. 1911 Rapporti tra encefalomeria e vascalarizzazione. Del cervello embrionale. Ricerche Lab. Anat. Xorm. R. L'niv. Roma, voL 15.

EvAXS, H. M. 1909 On the development of the aortae, cardinal and vunbilical veins, and other blood vessels of vertebrate embryos, from capillaries. Anat. Rec, vol. 3, p. 498.

1909 On the earliest blood vessels in the anterior limb buds of birds, and their relation to the primary subclavian arterj'. Am. Jour. Anat., vol. 9.

1912 Development of the vascular system. In 'Human embryology,' Keibel and Mall, vol. 2.

His, W. 1887 Zur Geschichte des Menschlichen Riickenmarkes und der Xervenwurzlen. Abhandl. der Konigl. Sachs. Gesellschaft, der Wissenschaften 22. Math.-Phys., Classe 13, Leipzig.

HocHE, A. 1899 Vergleichenden-anatomiches iiber die Blutversorgungen der Ruckenmarksubstanz. Zeitschr. Morph. u. Anthrop., Bd. 1.

HoF.MAX, M. 1900 Zur vergleichenden Anatomie der Gehirn und Riickenmarksarterien der Vertebraten. Zeitschr. Morph. u. Anthrop., Bd. 2.

Kadyi, H. 1889 Ueber die Blutgefasse des menschl. Ruckenmarks. Wien. (Also Denkschr. Math. Xatunv. Kl. Akad. Wissensch. Krakau).

Ross, J. 1880 Distribution of the arteries of the spinal cord. Brain, vol. 9.

Smith, H. W. 1909 On the development of the superficial veins of the body wall in the pig. Amer. Jour. Anat., vol. 9.

Sterzi, G. 1904 Die Blutgefasse des Ruckenmarks. Anat. Hefte, Bd. 24.


A COURSE OF CORRELATIONAL ANATOMY

EDWARD F. MALONE

From the Department of Anatomy, University of Cincinnati

During the first year and a half of study the medical student acquires a knowledge of certain aspects of the structure and function of the human body. Since these different aspects are studied in separate courses the tendency is for the student to acquire certain disjointed groups of facts which neither his ability nor the time at his disposal permit him unassisted to bring together. Such a mass of fragmentary knowledge falls far short of what the student is supposed to have acquired, namely, a reasonably good understanding of the structure and activities of the entire organism. It is true that in every good course the instructor brings the subject matter of his speciahty into relation with that of other courses; this correlation in the separate courses is indispensable and serves as a foundation for further efforts in this direction, efforts made by the student alone, or preferably under suitable supervision. But in attempting in each course to correlate the subject matter with that of other courses the results obtained are inadequate, not only on account of the lack of sufficient time but especially because the knowledge of the student is as yet too limited. It therefore appears advisable, after the various courses involved have been completed, to renew in a special course the effort to help the student bring together certain important facts already learned piecemeal. This article deals with such a course introduced this year by the Department of Anatomy of the I'niversity of Cincinnati.

The course in correlational anatomy is the logical result of the method of instruction in the Department of Anatomj'. The head of the department, Professor Knower, has in all the work of the department constantly insisted upon the correlation of

393


394 EDWARD F. MALONE

structure and function. And accordingly those portions of the body which are of the greatest functional importance have in the courses of the department received the gi'eatest attention. A further advance consists in the increasing correlation between the different courses of the department (gross anatomy, histolog>', neurology and regional anatomy). The work in gross anatomy and in histology is so related that the student studies the gross and microscopic anatomy of each organ simultaneously, as far as this is practicable; this is only one instance of the correlation of work within the department. In the second year the association of gross anatomj', histology, physiology, neurology and regional anatomy is much closer. This intimate relation between the separate courses within the department is maintained by the close association of the various members of the staff so that each member is familiar with the nature and progress of each course; this knowledge is attained not only by means of frequent conferences but also through actual experience in assisting from time to time in each course. On the other hand, such a correlation of courses within each department makes additional demands upon the staff, demands which are difficult to meet unless the department be supported in a more liberal manner than is customary. It is manifestly unreasonable to blame the student for regarding different courses as uru-elated when the instructors themselves behave as if this were true, and it is also unreasonable to expect instructors to establish such a correlation if they be already overworked. It is upon this intimate association of the different courses within the department that the course in correlational anatomy is founded. AMiile the author has designed this course and has conducted it alone, he does not claim the entire credit of originating it, since it is the result of the whole attitude which Professor Knower has impressed upon the department.

In addition to correlating its own courses and introducing a considerable amount of general biolog>', embrj^ology and physiology, the Department of Anatomy has repeatedly requested other departments to send students back for supplementary study whenever these departments demand special anatomical knowl


A COURSE OF CORRELATIONAL ANATOMY 395

edge which the student cannot and should not obtain during his regular courses in anatomy. Such supplementary work has proved of great advantage to the students, since it is undertaken by trained men who realize the need of the knowledge which they return to- acquire, and they gladly accept the opportunity. The advantages which an anatomical laboratory- possesses as against didactic teaching are evident. This arrangement is impossible if the student's time be completely occupied by required work; in addition it makes demands upon the time of the anatomical staff, and the question arises as to whether the institution is willing to pay for such advantages to the students or whether it will take the course of least resistance and of least expense.

The course in correlational anatomy is given at the beginning of the second semester of the second year. Before it begins the student has finished dissecting the body and has completed, beside other courses, those in histology', physiolog\' and neurolog\'; in addition, he has studied the various aspects of anatomy from a physiological standpoint. During this year and a half the student has amassed a large number of facts and has made some progress in bringing isolated facts together; moreover, he has attained to a considerable degree the abihty to recognize essentials and to work problems out for himseh. The course in correlational anatomy lasts eight weeks, and during the remainder of the second semester the student is at liberty to elect such work in the department as he may desire; he may take the course in regional anatomy, or he may spend the time reviewing the essentials of the body, working out for himself (with the assistance of the staff) certain important mechanisms not given in the course in correlational anatomy. This course is accordingly preceded by the study of the structure and function of the entire body, and is followed by a period during which the student has the opportunity of bringing together on his own initiative further disconnected groups of facts; in this manner the student is encouraged to fonn the habit of study outUned in the course just completed so that he may thus acquire a real knowledge of the human body. This result is aided by the fact that at the end of the second vear he


396 EDWARD F. MALONE

must pass an examination which includes all the subjects studied in the department and in which a fair but real knowledge of the essentials of the body is demanded.

At this point the author would like to enter a protest against the unfortunate tendency in some institutions to complete the work in anatomy during the first year. The student can undoubtedly finish the courses in histology and neurology and dissect the entire body in one year, but even if the time should be sufficient to permit the student to finish his task without undue haste the result would still be unsatisfactory. For a real knowledge of anatomy cannot be acquired all at once, but only when the study is prolonged to such an extent that the student has the repeated opportimity of thinking of the problems which the dissection of the body merely makes possible to study. ^Moreover, the student's knowledge of physiology and his capacity for independent constructive thinking is too limited to permit him to obtain an adequate knowledge of the body during the first year. Finally, this excessive concentration and hurry encourages the student to regard each of the anatomical courses (and each part of each course) as a separate task to be gotten out of the way in a definite length of time, and not as aspects of one great problem to be correlated with one another and with those aspects learned in other departments.

The course in correlational anatomy consists in the study of certain mechanisms of the body. New matter is introduced only when necessary, while on the other hand, unessential details are eliminated. The course therefore attempts to help the student rescue from the mass of details the really vital facts concerned in each mechanism, and by correlation of these facts to fix them firmly in his memory. Especial emphasis is placed upon the relation of the nervous system to the rest of the organism, and the various reflexes involved in the activities of each mechanism are studied thoroughly, the path of the impulse being followed throughout its entire extent. The course differs from one in physiology in that the anatomical structures upon which depend the activities of any mechanism become to the student realities; as far as possible the actual anatomical structures are


A COURSE OF CORRELATIONAL ANATOMY 397

studied in the gross and in sections, and in the nervous system the actual location of nervous centers and the course of fiber tracts are reviewed not only in diagrams but also in the specimens themselves.

For the most part the student is expected to work out, with the aid of specimens and books, the various problems for himself. Since all anatomical and physiological facts involved have been previously studied it is possible to assign at each exercise a large field to be covered, and to expect the student, with a certain amount of guidance, to select the most vital points and to ignore nonessentials; the value of this training is evident. 'S^Tiile lectures are necessary they are mostly informal, assuming the character of conferences, while at the beginning of each exercise the main results of the work of the preceding day are briefly summarized. Finally, the student is expected to hand in at the end of the course a complete account of the mechanisms studied, and to describe certain activities which have not been studied but with whose anatomical basis he is supposed to befamihar.

In selecting topics for study in a course in correlational anatomy the following points should be kept in mind:

1. The topics should be of importance.

2. They should necessitate the study of structures which form part of many other functional groups, and which thus involve a knowledge of large portions of the body.

3. They should involve functions which have a demonstrable anatomical basis.

The respiratory system meets these requirements in a most satisfactory manner. It is of great importance; it demands the study of the entire thorax, a part of the head and neck, the spinal cord and spinal nerves, the vagus and trigeminal nerves, the s>inpathetic system, and the main sensory and motor tracts and centers of the brain; and finally, the correlation between stucture and function can be shown in a most satisfactory manner. The almientary system also is satisfactory in most respects. Although the correlation of structure and function is not easily shown in the portion below the diaphragm, in the portion above the diaphragm this can bo shown vorv successfullv.


398 EDWARD F. MALONE

The study of the anatomical structures of the ahmentary tract involves many hnportant relations in the head, neck, thorax and abdomen. The relation of the nervous system to swallowing and to mastication (taken in a very broad sense and including prehension and other acts preparatory to mastication proper) involves many reflexes in which practically the whole nervous system is utilized; of course the regulation of secretion and of blood supply should be included. .Ajnong other mechanisms which suggest themselves as suitable for study may be mentioned the maintenance of the erect position (including the part played by the nerv^ous system in receiving, correlating and sending out impulses), the heart beat, the development and mechanism of speech together with the different forms of aphasia.

The course in correlational anatomy this year was limited to sixteen periods of three hours each; next year it will be extended. An outline of the course follows:

I. Respiration

1 February 19 The thoracic wall and its movements

2 February 20 Relations of the thoracic contents. Diaphragm

3 February 26 Gross anatomy of nose, pharynx, larynx and

trachea

4 February 27 Histology of respiratory system

5 March 5 Mechanics of respiration

6 March 6 Nerves and nervous centers

7 March 12 Nervous reflexes

8 March 13 Summary

II. Mechanisms of the alimentary system

A. Mastication

9 March 19 Gross and microscopic anatomy of the mouth

10 March 20 Nerves and nervous centers

11 March 26 Mechanism of mastication

B. Deglutition

12 March 27 Gross and microscopic anatomy of tongue,

pharynx and esophagus

13 April 2 Mechanism of deglutition

C. Movements of stomach and intestines

14 April 3 Gross and microscopic anatomy of stomach and

intestines

15 April 16 Movements and nervous mechanism of stomach

and intestines

16 April 17 Lecture

(a) Sympathetic system

(b) Innervation of viscera and of overlying muscles and skin

(c) Theory of emotions


A COURSE OF CORRELATIONAL ANATOMY 399

In order tc make such a course a success the instructor should carefully avoid each of two extremes. In the first place, the course may degenerate into a feeble attempt at a review in physiology' in which the student, neglecting to study in actual preparations the anatomical structures upon which the various functions depend, spends his time over a book or in theorizing. In the second place, the student may become lost in a maze of anatomical details, losing sight of really vital anatomical facts and failing to bring these into relation with the activities which depend upon them. The instructor should possess above all a thorough knowledge of the anatomy and physiology of the nervous system; in addition he should be familiar with the gross and microscopic anatomy of the entire body and with general physiology', while a knowledge of pathology', clinical neurology, psychology and psychiatry will be of much value. With such a background of knowledge he may be expected to guide the student in forming a real conception of the various mechanisms of the human body.


To Cornell University ]VIedical College, The Wistar Institute returns a grateful acknowledgment for a contribution of $500.00 towards current expenses of The American Journal of Anatomy and The Anatomical Record.


A USEFUL MODIFICATION OF 2^1 ANN'S :METHYL BLUE-EOSIN STAIN 1

FRAXKLIX P. REAGAX From the Laboratory of Comparative Anatomy, Princeton University

Mann's methyl-blue-eosin stain has recenth' been found to be very useful in the differentiation of embryonic tissues, especially in the study of the vascular system. Skillful manipulation of this stain gives a brilliant red to developing blood-cells, while other tissues are stained deeply blue. The stain is especially favorable for the study of haemapoesis.

There are, however, certain defects in the original method. Constancy in the amount of differentiation in caustic alcohol is difficult to obtain; even different parts of the same embr^'o should be left in this reagent different lengths of time if a proper balance of red and blue be maintained throughout a given series. In the caudal region particularly it may be found that sections must remain for too short a time in caustic alcohol if enough blue be left in the tissue to be of greatest differential value. The nonvascular tissue is likely to possess a reddish color which cannot be removed by any reasonable amount of washing in chstilled or acidulated water. With eosin present in all the tissues it is difficult to judge the amount of methyl blue which is being removed by caustic alcohol. These objectionable features may be largely eliminated by a few simple alterations of method. TMiile these modifications have not received exhaustive trial, and while the stains so obtained have not yet been thoroughly tested for permanency, it is nevertheless true that they have afforded an astonislung degree and brilUancy of differentiation.

Sections are cleared in xylol, transferred through the alcohols to water, and then stained from fortv-eight to ninetv-six hours in Mann's mixture of:

One per cent aqueous solution of methyl blue 35 parts

One ]->er cent aqueous solution of eosin ' 45 parts

Distilled water 100 parts

Sections are rinsed in distilled water, then thoroughly dehydrated by direct transfer to ab.soluto alcohol, after which they are differentiated in caustic alcohol made as follows: To each 30 cc. of absolute alcohol

'Mann, G. Physiological histology, 1902, p. 216.

Grubler's aqueous eosin labeled 'W. GELBL' was found to be quite satisfactory. 401


402 FRANKLIN P. REAGAN

are added five drops of a 1 per cent solution of caustic potash in absolute alcohol. Sections are removed from this solution when they have acquired a reddish-purple color. FolloA\'ing this they are rinsed in absolute alcohol, and then washed in distilled water, which, according to the original method should be allowed to remove the eosin from all the tissues except the blood-cells. In my ovm experience I have found that the eosin is sufficiently removed ^\'ithin five minutes or less; then they are transferred to the folloA\ing mixture until over-stained:

One per cent aqueous solution of methyl blue 40 drops

Glacial acetic acid 30 drops

Distilled water 200 cc.

Sections are then washed in distilled water to remove acid, transferred to absolute alcohol, dehydrated thoroughly, de-stained in causticalcohol until the desired amount of blue is left in the tissue, rinsed well in fresh absolute alcohol, cleared in xylol, and mounted.

This gives the mesenchyme a clear blue, and the blood-cells varjang degrees of bright red, depending on their developmental stages. Ganglionic and glandular structures tend to retain the purple obtained by the original method. Xerve-fibers and cartilage may assume a pea-green color. Endothelium of vascular plexuses may be made to take on a denser blue than the surrounding mesenchyme, rendering a plainness rivaling that obtained by injection. Also it might be mentioned that in the original method, an adequate amount of differentiation in basic alcohol tends to remove all blue from blood cells and their nuclei, so that a given field might exhibit cells all of which would apparently be eosinophilous. In the suggested modification such blue is restored to the basophile cells and nuclei.

Further modification of the amount of methyl blue added to the acidulated water or of the acidulation itself may effect further improvement, yet the present modification leaves little to be desired.


^


o'>


THE MORPHOLOGY AND HISTOLOGY OF A CERTAIN

STRUCTURE CONNECTED WITH THE PARS

INTERMEDIA OF THE PITUITARY BODY

OF THE OXi

ROSALIND WULZE.V From the Hearst Anatomical Laboratory of the University of California

SEVENTEEN FIGURES

Certain physiological experiments are now being conducted in the Rudolph Spreckels Physiological Laboratory of this University which necessitate the separation of a great number of ox pituitaries into their two main divisions. As an interesting anatomical feature was in this way brought to our attention the material was used in addition for this anatomical study. This feature has not been mentioned by the following who have written more or less fully upon the pituitary bod}' of the ox, Peremeschko ('67), Dostojewski ('86), Herring ('08), and Trautmann ('11).

The pituitary body of the ox, like that of other vertebrates, is composed of two distinct portions. One, the pars nervosa, is derived from the brain as an outgrowth of the hypothalamus. The other originates as a hollow buccal evagination which in time is completely separated from the digestive tract. That portion of this evagination which comes into contact with the pars nervosa is called the pars intermedia. It is a comparatively thin sheet of epithelium which spreads as a coating over nmch

' Material amounting to thousands of pituitary bodies was most kindly supplied by the Oakland Meat and Paeking Company through the courtesy of the Superintendent. It was derived from cows, bulls and steers. .\s the cone structure was present indifferently in these throe varieties, its appearance can have little to do with se.\ or castration.


\{V,


THE ANATOMICAL RECORD, VOL. S, NO. S ACOUST. 19U


404 ROSALIND WULZEN

AHHKi:\IATl()NS

P.X., Pars nervosa CI., Cleft (residual lumen)

P.O., Pars glandularis I.S., Infundibular stalk

P. I., Pars intermedia B.S., Blood sinuses C, Cone

Cephalic



P.G.


Hg. 1 Tracing of a mid sagittal section of ox pituitary, natural size. The pars glandularis is caudad and ventrad, the pars nervosa cephalad and dorsad. The pars intermedia lies between the two but is separated by the cleft from the pars glandularis. Note the cone upon the pars intermedia and the cavity of the pars glandularis into which it fits, also the blood sinuses coursing through the pars glandularis toward the cone.

of the pars nervosa. The cells in the remamder of the evagination proliferate greatly to form the pars glandularis, the bulkiest part of the pituitary body. Between the pars intermedia and the pars glandularis appears the remnant of the lumen in the original evagination. This is the residual lumen or cleft of the pituitary body. Figure 1 represents these parts as they appear in mid sagittal section. I have taken the side of the pituitary toward the brain to be 'cephalic,' that opposite 'caudal,' the side toward the nose to be 'ventral,' that opposite, 'dorsal.'

The pituitary body lies in the sella turcica roofed with thick dura mater which is perforated ventrally by the opening which transmits the infundibulum. Just underneath this covering is found the pars nervosa, a flask-shaped structure with a long narrow neck, the infundibulum, jiassing to the brain through the aperture in the dura. The pars glandularis surrounds the pars nervosa throughout its length and enfolds it on either side to such an extent that its only free portions are its dorsal extremity and a strip of its cephalic surface in contact with the dura. The cleft of the pituitary body is easily opened for examination. It


PITUITARY BODY OF THE OX 405

is found to have the same shape as the pars nervosa and thus the broadest portion is close to the dorsal extremity. In the greater number of cases the two walls of the cleft are closely approximated, but sometimes they are spread widely apart by the liquid or solid material which gathers in the cavity. Occasionally the main divisions of the pituitary are so extensively attached to one another that the cleft is obliterated except in its ventral extremity around the neck of the pars nervosa.

In the cleft there is ahnost invariably found a mass of tissue attached to the pars intermedia but very different from it. This structure is usually s}Tnmetrically placed in the mid sagittal plane one-third or less of the way from the dorsal to the ventral end of the cleft. Its general shape is that of a cone one side of which may be longer than the other. The cones differ in proportion; some are broad and low, even flat, others are tall and narrow at the base. Rarely there are tv»'o cones. The following are the dimensions of a few:

Length Breadth Height

mm. 7nm. mm.

2 2 5

7 5 4

4 3 2

2 3 4

3 3 2 3 2 2

They vary from such sizes as the above to specks so small as to be just visible. Rarely they are not to be seen. Thus out of 760 tabulated cases 38 showed no such structure. Probably almost all of the 3S would have revealed it liad a microscope been used. The writer has exaniineil at least five thousand of these pituitaries and has come to the conclusion that the cone structure in some fonn is practically always present. Sometimes the cone has the shape of a bulb which is connected by a slender stalk with the pars intorniedia, the whole structure being 8 or i) miii. in IcMigth. In these cases the bull) may be so deeply iinl)ed(le(l in the i)ars glandularis as to reach almost to its opposite extremity.


406 ROSALIND WULZEN

Figures 2 to 7 show typical fonns of the cone. The sections are approximately mid sagittal. By comparison with figure 1 their various parts may be easily identified. The bulkier and lower portion of each specimen is pars glandularis. The pars nervosa constitutes the larger part of the remainder. Between the two is the pars intennedia which is joined irregularly with the pars nervosa but is separated from the pars glandularis by the well developed cleft. Within the cleft is more or less colloid. Each spechnen possesses a well marked cone which occupies a corresponding depression in the pars glandularis but is in no way attached to it. In figures 2 to 5 the cone is finnly attached to the pars intermedia. In figures 6 and 7 a line of division runs all around the cone separating it from pars intennedia as well as from pars glandularis. Probably in these the cone is attached to the pars intermedia by a slight strand of tissue which would be shown in a different section. Many cones partly separated in this way from pars intermedia have been found.

Occasionally the cone is shifted in position to the dorsal end of the cleft. It thus comes into contact with the pars glandularis and may be finnly attached to it. Figures 8 to 13 are examples of this arrangement. The sections are similar to the preceding ones, the difference being that here the cone is firmly attached to pars glandularis. Figures 8 and 9 show a transition between the forms preceding and those following. The larger part of each cone is separated from pars glandularis by the cleft but there is nevertheless an area of firm attachment to the tissue of the pars glandularis.

In order that the remaining specimens may be understood, it is necessary to notice the arrangement of the blood sinuses of the pars glandularis. When a sagittal section of the pars glandularis is made it is seen to have a dark core arching toward the cleft. This is composed of large blood sinuses. Though present elsewhere in the glandular substance they are most prominent here. They often stretch through the gland as an almost flat sheet in the mid sagittal plane and go with surprising directness to the cleft where they come to the surface in the mesial line and distribute themselves with the greatest liberality <n^er the


PITUITARY BODY OF THE OX


407





Figs. 2 to 7 Approximately mid sagittal sections of ox pituitaries showing typical forms of the cone struct vire. Specimens wore hardened in Zenker, stained lightly with alum-oochineal, anil sectioned l\v hand. See description in text. Photograjjh X -. These and the following photographs were made by L. R. Newhart to whom the writer wishes to extend thanks.

surface of the depression into whicli the cone fits. In every specinuMi examined the rone projects into this most vascular portion of the pars glanihilaris. The hlood vessels are often so close to the surface in tliis region that they Meed with a touch.


40S


ROSALIND WILZEN





Mrs. fS to \:i SpcciiMcn.s prci):u('tl .similarly to those in liKiues 2 to 7. Tlicsc forms arc somewhat e\ee|)tional. See ilescri|)t ion in text. Photograph X 2.

Kig. 14 Mid sagittal section of ox pituitary in the region of the cone. Note the abrujjt transition between ])ars intermedia and cone, cellular debris in the cleft, and numerous blood sinuses in pars glan<lularis. Tlie band crossing cone and pars glandularis is an artifact. Section stained with Mallory's connective tissue stain. Photograph X 25.

Fig. 15 Description the same as for figure 14. Here the cone tissue does not project into the cleft but it is seen to be as distinct from the i)ars intermedia as in the former rase. Photograpli X 25.



14



15

400


410 ROSALIND WULZEN

In several of the specimens this core of l)lood vessels is distinctly seen as it travels to the exact location of the cone. In figures 10 and 11 the cleft appears only around the neck of the pars nervosa. The impression would be that there is no cone present were it not for the arrangement of the blood vessels. This is quite clear in both but is photographed most plainly in figure 10. Instead of arching to meet the pars intermedia at the apparent junction between pars intermedia and pars glandularis the l:)l{)()d sinuses come to an abrupt stop. In this way they mark out a large cone which is as well shaped as if it were separated completely from the pars glandularis by the cleft, no trace of which is here visible. In figure 11 the cone is marked out just as definitely. This also occurs in figures 12 and 13 but the outline of the cone is not so perfect, probably owing to incorrect section.

In color and consistency the cone is different from the pars intennedia. This difference which has been found to hold throughout the macroscopical examination is shown somewhat in the figures. The cone is composed of compact, creamy white tissue suggestive of the pars glandularis, whereas the pars intermedia is soft and more or less brown. The cone is often whiter and firmer than the pars glandularis itself.

HISTOLOGY

Pituitary bodies were fixed in Zenker's, Bensley's or Orth's fluids, were sectioned sagittally and stained with hematoxylin and Congo red, eosin and methylene blue, or Mallory's connective tissue stain. Ten specimens were examined. Of these eight showed the cone structure undoubtedly present, and in one it was probably lost through poor technique. Figure 14 shows the cone i)rojecting from the pars intennedia into the hollow of the pars glandularis. The cleft separates the two. That the tissue of the cone differs markedly from pars intermedia can be seen at a glance, the transition between the two being well marked. Figure 15 is a similar section of another specimen. The cleft runs through the center separating purs intermedia above from pars glandularis below. Here the cone tissue is raised only


PITUITARY BODY OF THE OX 411

slightly from the surface. It is, however, a fairly definitely circumscribed area which is striking!}' different from pars intermedia. Even when the cone was so small as to be composed of a few cells only, its apearance was unmistakable. Greater magnification reveals the reason for the difference. Cone tissue contains as its most striking characteristic, deeply staining acidophile cells with coarsely granular protoplasm which are apparently identical with those of the pars glandularis. As far as I have been able to detemiine, acidophile cells have never before been noticed so closely associated with the pars intermedia. Indeed, Herring ('08) says The intermediate portion, although derived from the same source as the main anterior lobe (pars glandularis), differs from it in adult mammals in that it contains no eosinophile cells." Tilney ('11) joins him in the remark, The juxta-neural epithelial portion (pars intermedia) is invariably basophilic. These statements have great weight in that both investigators have made careful examinations of the pituitary bodies in many different animals. The remaining cell elements of the cone are similar to those of the pars glandularis. But certain differences exist which, though perhaps of slight importance, make it possible to distinguish between pars glandularis and cone tissue. The connective tissue septa are finer and the interstices are ordinarily smaller and contain fewer cells in the cone than in the pars glandularis. On this account the grouping of the cone cells into acini may appear to be lacking, while the acini of the pars glandularis are apparent at a glance (figs. 10 and 17).

One of the specimens examined was found to resemble the condition shown in figures 10 and 11. The cone was finnly joined to pars glandularis as well as to pars intennodia. Microscopically it showed itself to be as typical a cone as any of the others. A definite connective tissue septum out fined the surface of the cone, adjacent to the pars glandularis. Its tissue differed in the same manner as the other cones from the surrounding tissue of the pars glandularis.

It is noteworthy that in the only specimen in which the cone was actually lacking the cleft was very small, appearing only


412 ROSALIND WULZEN

around tlio neck of the pars nervosa. The pars intennedia over-lapi)od the cleft at its dorsal end and spread irregularly some distance into the pars glandularis. Also large portions of the jiars glandularis had the characteristic a})pearance of cone tissue, that is, the acini had few cells and were separated by very fine strands of connective tissue. In no other specimen did the pars glandularis have this appearance.

jMicroscopically the vascularity of the cone appears to be slight but in spite of this when the cleft of a fresh specimen is opened the cone is often flushed with blood. The bloodvessels appear to be superficial and are spread mainly about the base of the cone.

This preliminary work leaves many problems unsolved. \Miat are the phylogenetic relationships of this very constant structure? Can any light be thrown upon it by the study of its development? Has it any physiological significance? If this study suggests anything it suggests a closer relationship between pars glandularis and pars intennedia than has hitherto been suspected. The distribution of the blood sinuses of the pars glandularis about the cone is often so striking that it is impossible to avoid the impression that such an appearance cannot be an anatomical chance. Thus when the cone penetrates deeply into the pars glandularis, the walls of the cavity into which it is so closely fitted are solidly packed wdth blood sinuses. This would a])poar to be an excellent arrangement for the absorption into the })lood stream of any secretions from the cone or pars intennedia. Certain color changes should also ])e mentioned in this connection. The pars intermedia is ordinarily distinctly tinged. It assumes all colors between dull white and bright orange or yellow or brown. The pars nervosa never shares its color. It is a uniform dull white. On the other hand the i)ars glandularis may have its characteristic creamy or

Figs. 10 and 17 iJctail from t lie six'ciincii sliown in tinurc 14. Figure 10 sho\v.s typical ti.ssuc of the cone, fip;uro 17 typical tissue of pars glandularis. Note the prominence of the deeply staining acidophile cells in both, but observe that in the pars glandularis the acini are more prominent and the connective tissue septa are broader. Photograph X 275.



16




' «*•


t^fj


17

413


414 ROSALIND WULZEN

grayish white appearance when the pars intermedia is brightly colored or it may share the coloration of the pars intennedia. Thus it too is sometimes a deep orange. It, however, has never been observed to be highly colored unless the pars intermedia as well is brightly colored. This also suggests a common activity of pars intermedia and pars glandularis or some interrelation between them. Work is being carried out along the lines suggested.

Acknowledgment is due to Dr. P. E. Smith for his kindly help.

CONCLUSIONS

1. A structure of more or less definite cone shape appears constantly upon the pars intermedia of the ox pituitary.

2. Its cellular elements resemble those of the pars glandularis. Numerous acidophile cells are its most striking feature.

3. It differs from pars glandularis through having in a general way finer connective tissue septa and smaller acini.

BIBLIOGRAPHY

DosTOiEwsKi, A. 1880 t^ber den Ban dcs ^'or(lerlapl)ens des Hirnanhangs. Arch. f. niikros. Anat., Bd. 26.

Herring, P. J. 1908 The development of the mammalian pituitary and its morphological significance. Quart. Jour. Exper. Phys., vol. 1. 1908 Histological appearance of the mammalian pituitary. Quart. .Jour. Exper. Phys., vol. 1.

Pkrkmkschko 1867 Ubcr den Bau des Hirnanhangs. Virchow's Arch. f. path. Anat. und Phys. und klin. Med., Bd. 38.

Tii.NKY, F. 1911 Contribution to the study of the hypophysis cerebri with especial reference to its comparative histology. Memoirs Wistar Inst. .\nat. and Biol., No. 2.

1913 An analysis of the juxta-neural epithelial portion of the hypophysis cerebri, with an embryological and histological account of a hitherto undescribod part of the organ. Intern. Monatsschr. f. Anat. und Physiol., Bd. 30.

Tratttma.w, a. 1909 Anatomie und Histologic der Hypophysis cerebri. Arch, f. mikroH. .\nat., Bd. 74.


OSSICULUM LUS

J. C. MILLER

Frotn (he Anatomical Department, Western Reserve University, Cleveland, Ohio

Some reference to this mysterious bone is found in all anatomies of ancient times and in those of the Middle Ages. The name itself is derived from the Hebrew 'luz' (almond), Arabian, 'lauz.' Its histor>' may be of some interest.

In the Old Testament (1) we find the following passage: "Castodit Dominus omnia ossa justorum — -unum ex illis non confrigetur." In the English version we read: Not one of them is broken," but there is no emphasis on the 'one,' nor need it mean, as the Latin renders it, "there is one which .shall not be broken," for had the Hebrew writer meant simply to sa}'. "none of his bones shall be broken," he would have stated it in the same manner. Reference to both Lsaiah, xxxn'. 16, and Psalm cvi, 11, .shows the use of the .same idiom. In neither of these passages is it to be understood that a particular one (bird or enem}') is missing.

Further, we may say that not merely may Psalm xxxiv. 20. be translated as in the English version, but that it must l)e so translated. The two clauses of the verse, as so often in Hebrew poetry-, contain parallel ideas, simply repeating the same thought in different words, thus:

He keepcth all his bones, Xot one of them is broken.

In this psalm tiiere is no thought of any life after death. It is in this life that God preser\'es the righteous. Xeverthele.ss, the possible alternative translation of the Hebrew affords a point for argument, and it is probal)le that tlio rabbinic mind, aj^t to controversy, has raised the que.>^tion whether or no the tran.-^lation sliould l)e 'one of them' with the emi)liasis on the 'one.' thereby stimulating the search for the 'unum os' now well-known as the 'os luz.

The learned Rabin Uschaia (2), who lived in the third century- .\.d., tlefined this bone as in fine octodecim vertebrarum." Xot only Uschaia. but other writers (especially the anatomists of the School of Salenio) have giv(M) the numlnM" of vertebrae as eigliteen (3): "Sunt autem octodecim s|K>n(lilia, in colic sex, in dorso duodecim." According to Haller (4) the lumbar vertebrae were not considered vertebrae proper, because tiiey do not enclose any part of the spinal cord, but only the cauda ecjuina of tlu> talmudists. Even at that, as Hyrtl (5) remarks, the number of vertebrae would be 19 and not IS.

' For help in the transh\tion and interpretation of the literal Hebrew. I woul Hebrews, in size but that of a. pea, touched by no decay, nor concjuered l)y the fire, will remain imblemished forever; from it, as a plant from the seed, f)ur ])odv shall arise hi the resurrection of the Dead," which


OSSICULUM LUS 417

sentence he ends with these remarkable words: Et hae virtutes non declarantur ratione, sed experientia;" ("Os minimum, fiuod Hebraei Luz appellant, magnitudine ciceris mundati, nulli corruptioni obnoxium, nee igne quidem vincitur, sed semper conversatur iUaesum, ex quo, velut planta ex semine, in resurrectione mortuorum, corpus nostrum repullascet").

An editorial in the Lancet (9) refers to this bone as the twelfth dorsal vertebra, "the turning point and centre of the spine." Although no reference is made to the source of this information, I think the quotation is taken from Galen, who in all proljability dissecterl onh' animals, and it would thus refer to the 'anticlinal' vertebra.

Since this remarkable bone could not be found in fine octodecim vertebrarum," search was begun elsewhere. The skull as an important part of the body attracted attention first, and the supernumerary- bone, frequently found at the junction of the sagittal and lambdoidal sutures (perhaps the os interparietale or os Incae?j was thought by some to be the 'os lus.' Many virtues were attributed to this bone, and its remedial powers were supposed to be great, especially when it came from the skull of an executed criminal. According to HyrtI (14), the pulverized bone was first used as a remedy for epilepsy by the Swiss physician Hochner, who latinized his name to Paracelsus, and hence the term, ossiculum antiepilepticum Paracelsi.

A similar term was used in reference to the os epiptericura ( Virchow ) OS epilepticum, which, according to Lombroso. is always to be found in delinquent and demented people (23).

But not all skulls have these epactal bones, and thus other writers looked for it at the l)ase of the skull, as Hieronymus Magius ( 10) tells us, without stating, however, which bone it is. According to Bauhinus (I.e.), the seventh cer\'ical vertebra, the vertebra prominens: according to Dassovius (11), the os coccygis was taken for the 'ossiculum lus.' The explanation of Dassovius is ver>' probably ba.sed upon the Arabian name for cocc^-x, 'al ajab' (al ajas), '^' which Mohammed stated to be incorruptible and to ser\'e as the basis for the future edifice" (9).

The same author (9) informs us that the os sacrum was thought to be the 'os lus' '"on account of its old name iepoi- boTtov but here the author makes the same mistake which many others made before him. Isidorius (12) gives the following explanation of the os sacrum: '"Ima spinae pars, quam CJraeci hpbv oarovv vocant, quia primum in infanti nascitur, ideoque et ho.stia a gentilil)us diis suis dabatur."

The word 'sacer' is explained l)v Festus ( 13) in the following manner: (lallius Aelius declares sacred (sacer) that which in any manner is dedicated to the state, whether house, altar, e:igle. riches or anything dedicated and consecrated to tlie gods." ("(lallius Aelius ait. sacrum esse quoquun(iue modo at(iu(> instituto civitatis consecratum sit. sive aedis, sive ara, sive sigiuun. sive p(>cunia. sive (|uid aliud diis dodic;itura atque consecratum sit'i.

^Iarcus Aurelius: .Vn^-thing set apart for the gods is tennetl sjicre«i (sacer);" and Sacred (^ sacer) are tho.>*e things which have been conse


418 J. C. MILLER

crated to the gods above;" ("Quidquid destinatum est diis 'sacrum* vocatur;" "Sacrao (res) sunt quae diis superis consecratae sunt:" Institutiones juris civilis).

All these deftnitions of 'sacer' do not explam the 'sacrum' in 'os sacrum;' there are, however, other possi])le explanations: The first is contained in the statement made by Hyrtl (14), that the 'os sacrum' is simply an erroneous translation of the (ireek lepop barkov.

1. The Greek term for 'os sacrum' was (nrovbvKo^ jikyas or tepos ioarkov ukya or lepbv), where up6% has tlie meaning of 'magnus;' thus Homer uses "IXtos Ipi) and iepos ttovtos (for "IXtos /jLeya'Kr] and iJikyas woutos). Spigehus (15) says: Graecis omnia magna 'sacra' vocabantur;" and Caelius Aurelianus (16): "Majora omnia vulgus 'sacra' vocat."

2. There is, however, in Latin itself an explanation of 'sacer' which seems to me preferable and more plausible, namely, its meaning 'detestable,' as we find it in the Leges xii talnilarum: "That man is anathema (sacer) whom the people have found guilty of a crime; and "an advocate shall be detestable (sacer) who defrauds his cUents"; (Homo 'sacer' est (luem populus judicavit ob maleficium;" and "patronus qui clienti fraudcm fecerit 'sacer' esto" (ibid).

According to this definition of 'sacer,' 'sacrum' would be the equivalent of 'detestandum,' and the bone received its name 'sacrum' (i.e., detestandum) from its lieing near the rectum (obscoena).

3. In addition to this Garrison (17) quotes Ramsbotham (18), who suggests that lepov in connection with the sacrum is not the Greek Upov, Ijut a corrupted form of the Hebrew 'iieron,' signifying conception, parturition, whence also Hera, the goddess of childbirth.

Garrison (1. c.) also produces some evidence that the external sesamoid bone of the great toe was thought by certain authors to be the 'ossiculum lus.'

Lastly, the inner, larger sesamoid Ijone of the Articulatio metatarsophalangea hallucis was selected as the os resurrection is on account of its real hardness and its form (seed of the sesamum) ; it is mentioned by Vesalius (17), Riolanus (18) and Bartolinus (19) under the name 'albadaram,' and as such it played a great role "apud magiae et occultae philosophiae cultores."

The OS sacrum, however, as the mysterious ossiculum lus, has found its place in history in the 'rump parliament,' as may be seen in the following quotation from Butler (22) :

The learned Rabbis of the .Jews Kroin whcnco the learned sons of art

Write there is a bone they call Luz Os sncrum justly style that part,

r the rump of man, of such a virtue, Then what can better represent

Xo force of nature can do hurt to: Than this Rump Bone, the Parliament

And therefore at the last great day That after several rude ejections,

.Ml th' other members shall, they say, And as prodigious resurrections,

Spring out of this, as from a seed With new reversions of nine lives

All sort of vegetals proceed: Starts up and like a cat survives?


OSSICULUM LUS 419

I^ut the naine lias disappeared from anatomical text-books, and the word remains in our dictionaries only as a reminder of theTinatomy of times past.


(1

(2

(3

(4

(5

(6

(7

(8

(9

(10

(11

(12

(13

(14

(15

(16

(17

(18

(19

(20

(21

(22

(23


REFERENCES Psalm xxxiv, verse 20.

Uschaia: Bereschit rabba (glossa magna in pentateuchum;. Magistuar Ricardus: Codex anatoinicus, p. 23. Haller: BibHotheca anatomica, T. 1, 2, p. 126. Hyrtl: Das Arabisc-he und Hebraische in der Anatomic, p. 166. Bauhinus: Theatrum anatomic-urn, Lib. 1, cap. 48.

Rolfink: Dissertationes anatomicae, Lib. 2, cap. 54: De ossibus sesamoideis. CoRXELius Agrippa: De occulta philosophia. Lib. 1, cap. 20. The 'Os Luz'. Lancet, vol. 2, 1910, p. 1029.

HiERONYMUS Magnus: De mundi exustione et die judicii, Lib. 5, cap. 1. Dassovius: Tractatus de resurrectione mortuorum, cap. 3, p. 23. LsiDORius: Etymologicorum. Lib. 2, cap. 1. P'estus: De verborum significatione (letters M-Vj ed. Miiilcr. Hyrtl: Lehrbuch der Anatomie, pp. 285 and 324. Spigeliu.s: De corporis huraani fabrica, Lib. 10. Caelius Aureliaxus: De morbis acutis, Lib. 1, ca[). 4. Garrison: The bone called 'luz,' New York Med. Jour., vol. 92, p. 147. Ramsbotha.m: Obstetric medicine and surgery, p. 698. Vesalius: De corporis humani fabrica. Lib. 1, cap. 28. RioLANus: Commentarius in Galeni librum de ossibus, cap. penult imum. Bartholixcs: Institutiones anatomicae. Lib. 4, cap. 22. Butler: Huriibras, Part 3, canto 2, 1. 1915. Le Double: Traite des variations des os tlu crane, p. 306.


THK ANATOMICAL RECORP, VOL. 8, NO. 5


BOOKS RECEIVED

The receipt of publications that may be sent to any of the five biological journals published by The Wistar Institute will be acknowledged under this beading. Short reviews of books that are of special interest to a large number of biologists will be published in this journal from time to time.

Psychology in D.\ily Lifk (Conduct of the Mind Series, edited by Joseph Jastrow), by Carl Emil Seashore, Professor of Psychology and Dean of the Graduate College in the State University of Iowa. 226 pages, New York. 1914, D. Appleton and Company', ?1.50 net.

Publisher's Announcement. This volume well represents the general purpose of the Conduct of Mind series which is, to present for the intelligent reader the several aspects of mental affairs which are involved in the regulation of practical interests. The volume comprises a selection of illustrative material with their interpretation, and may well serve as an introduction t* the stud}' of psychology. It proceeds bj' selecting a few general topics rich in application and about which a considerable range of mental principles may be grouped. The several chaptcr.s deal with toi)ics such as Play, The Law in Illusion, Mental Measurement, Mental Health and Mental Efficiency. The illustrations are in each case given a sufficient setting so that they become typical of the problems of psychology and at once suggest how competently the 'ssues of our daily life are conditioned b}' the psychological basis. The work is free from technical terms and presents a fresh and original arrangement of the material characteristic of modern interest in the laws of the mind.


420


PERSISTENT ARTERIAE BRACHII SUPERFICIALIS, ANTIBRArHII SUPERFICIALIS ET MEDIAXA

E. R. HOSKINS FTom (he Inslilute of Anatomy, University of Minnei^ota

OXE FIGURE

An unusual artery found in the left arm of a man of thirtj'-seven years seems? worthy of record.

The vessel emerges from the axillaris midway between the aa. subscapularis and thoracalis lateralis, on the median side. It runs in the deep fascia anterior and medial to the a. axillaris, the a. brachialis and the n. medianus, almost to the middle of the humerus, where it crosses the a. brachialis and the n. medianus, to enter the m. biceps brachii from beneath, through two large divisions.

It gives off two small cutaneous rami in the lower axillary' and upper brachial regions. In size the artery is about twothirds that of the nonnal subscapularis until it reaches the biceps muscle. At this point it gives rise to a small ramus almost at right angles to it. This courses lateral ami anterior to the biachial artery in the deep fascia, becomes superficial at the elbow, and continues anterior to the ulna, to the palm. Here it enters into fonnation of the arcus volaris superticialis. together with the a. ulnaris. The arch has no connection with the a. radialis.

There is no ranuis of the :i. ulnaris or a. interossea which may l)e called an a. mediana. The a. mediana descril>ed in this paper has no relation to the n. nK^dianus. wliich is placed deep in the forearm.

The embryological signilicance of the artery in question may be derived from Midler's' figiue of the arteries in the ann of an

'Miilirr '(W. Anal, llt'ftc. H.l. iJ. T.if. 2.")-2('.. fig. 0.

4J1


422


E. R. HOSKINS



Fig. 1 Persistent artcria; biachii superfiicialis, antibrachii superficialis et incdiaiia. The rami of the aa. brachialis, radialis and ulnaris are not shown in the figure, as they are normal excei)t for the two discrepancies noted.


11.7 111111. human embryo ('03). From this fijiuro it would soom that we have a persistent a. brachiahs superliciahs, giving rise to an a. antibrachii superficialis, which becomes an a. mediana, but all anastomoses with the a. brachialis have been lost. As stated by some texts of anatomy, this condition is one that is (luite rarely found.


THE MICROSCOPIC STRUCTURE OF MAMMALIAN

CARDIAC MUSCLE WITH SPECIAL REFERENCE

TO SO-CALLED MUSCLE CELLS

H. E. JORDAN

From the Anatomical Laboratory of the University of Virginia

EIGHT FIGURES

In recent histologic descriptions of the mammaHan heart the muscle is more conmionly regarded as a syncytium. The unifjue characteristic of heart muscle is the presence of intercalated discs. The earlier interpretation that these discs mark cell boundaries has been most recent h" championed by Zimmermann fl). That cardiac muscle, however, can not be regarded as composed of cells separated from one another by 'intercellular' intercalated discs I have attempted to prove in a recent series of papers (2, 3, 4, 5). The salient observation among the countervailing facts recorded concerns the occasional supernuclear position of these 'discs.'

In 1888 Apathy (cited from Lewis, 6) advanced the interpretation that striped muscle, inchuling heart nmscle. was structurally comparable witli the connective tissues, consisting of cells and extracellular bundles of myofibrillae. Recently Baldwin (7, 8) has attem])ted to establish this hypothesis u])on a basis of cytologic observations, and concludes that voluntary striped muscle generally, and cardiac muscle of the adult white mouse, consists of distinct nuiscle cells, and extracellular columns of muscle fibrils and sarcoplasm enveloped by sarcolenuna. The 'cells' are described as lying outside the sarcolemma.

If Baldwin's conception of cardiac nuiscle is correct, then his observations contribute one (^f tlic^ strongest objections to any interpretation that ccMisiders tlie intercalated discs ns intercellular structures marking cell l)oun(laries.

THE ANATOMICAI, UECDni). VOL. 8, NO. '.•

sv:rTv:Miu:u. I '.114


424 H. E. JORDAN

I shall concern myself Ikmo chief! 3' with Baldwin's conception of striped muscle structure as it pertains to cardiac muscle, my special interest l)eing enlisted by reason of its bearing on the nature of the intercalated discs.

Baldwin used only sectioned material. The essential point in his proof that there are 'muscle cells' pertains to the presence of a delicate 'cell membrane,' separating a nucleus with an envelope of cytoplasm from the myofibrils hnbedded in a distinct sarcoplasm. Such conclusion presupposes very delicate observations. One must guard against fixation artefacts and misleading appearances due to obliquity of section. Obviously from this standpoint macerated tissue is preferable to sectioned tissue. In macerated preparations one can examine considerable lengths of single muscle trabeculae, and by careful focussing thus view exact median longitudinal sections (optical) of 'fibers/ obviating all errors due to obliquity of section. Also, in properly preserved specimens shrinkage is prevented; and even a fair degree of dififerential staining can be obtained.

For the purpose of my study I employed principally dissociated tissue of fresh cat's heart; also macerated tissue of previously fixed (in Carnoy's fluid) heart of white mouse. I had on hand also abundant sectioned and stained material (treated according to Zimmenuann's technic) of various mammals, and of heart of aduh white mouse (Carnoy's fixation ; iron-hematoxylineosin stain) for additional study.

The cat tissue was macerated in a saturated solution of potassium chlorate in nitric acid and preserved in a mixture of equal parts of water, 95 per cent alcohol and glycerin. Many stains, both cytoplasmic and nuclear, and various combinations of stains were employed. The best results were obtained l)y use of borax cannine or eosin in xarious degrees of concentration.

That the method of treatment does not seriously injure the tissue for detailed observation is indicated by figure 1 which purports to be an accurate representation of actual appearances in lli(> median longitudinal plane. The .sarcolemma is well preserved as shown in the lower portion of tlie illustration where it is festooned between .succes.sivc tcl()})iiragmata (Krause's


MAMMALIAN CARDIAC MUSCLE


425



Fiji. 1 -Muscle fiber of cat's heart from macerated preparation, showing two nuclei connected by a continuous axial strand of coarsely granular sarcoplasm. The drawing represents appearances in the optical median longitudinal plane. There is no evidence of a cell membrane separating the central granular from the peripheral non- or finely-granular sarcoplasm. X 1500: reduced one-third in reproduction.

Fig. 2 Median longitudinal section of muscle fiber from ventricle of adult white mouse showing three nuclei imbedded in a continuous axial strand of coarsely granular sarcoplasm. The specimen was fixed in Carnoy's solution, and stained with iron-hematoxylin and eosin. X 1500; reduced one-third in reproduction. The destaining process is here carried too far to show the intercalated discs.'


' Fixation in Carnoy's strong solution (acetic alcohol with chloroform^ followed by iron-hematoxylin staining, is a good technic for demonstrating intercalated discs. This metiiotl brings out siiarply also the Q-substance and the Z-linos. The phase of contraction is thus clearly shown. For a study of the relation of the discs to 'contraction bands" it seems preferable to any of the special methods fi>r intcnubiftMl discs.


4'2() H. E. JORDAN

iiioinl)i-aii('s, of Z discs). In tlie ui)i)('r i)()rtion of the figure the festooning does not appear. From the standpoint of the niyofihrils the fiber differs in (hfferent regions. In the mid-portion, which is constricted (contracted), the fibrils are coarser and show a distinct alternation of Ught and dark bands (discs). In these same regions also the telophragmata are much coarser. The coarsei' telophragmata and the darker discs of the stouter fibers are identical, and represent 'contraction bands of Rollct.' The nuclei are clear with sharp contour, each surrounded by coarsely granular cytoplasm. The technic has clearly preser\'ed such delicate structures as sarcolenuna, differences between contracted and relaxed portions of the fiber, nuclear wall and reticulum, and cytoplasmic (sarcoplasmic) reticulum and granules.

But in face of these facts no definite indication of a cell membrane appears. The perinuclear sarcoplasm shades away into the peripheral fibrillar and finely granular sarcoplasm without sharp line of demarcation except such as is simulated at certain levels of focus by adjacent myofibrillae. ^Moreover, one perinuclear cone of sarcoplasm connects with adjacent cones through narrow strands of structurally similar sarcoplasm. In sections of such trabeculae a slight obliquity of cut would give an entirely false impression. This observation of a continuity of axial undifferentiated sarcoplasm, swelling at the levels of the nuclei, can be made onl>' Aery rarely in sectioned tissue. However, figure 2 shows a similar condition in a section of ventricle of adult white mouse. In cross-sections of cardiac nmscle of various manunals such a central core is almost invariably discernible, though frequently so frail as to escape notice unless specially looked for.

The undifferentiated sarcoplasm fonns an axial granular covo: the telophragmata extend thiough it, though frequently somewhat irregularly, as can be clearly observed in tissue deejily stained with iron-hematoxylin. There is no evidence warranting a distinction between the perinuclear plasm as cytoplasm, and the myofibrillai- portion as sarcoplasm. No clear evidence appears of a cell membrane in macerated tissue; in fixed tissue an adjacent myofibril, not showing a clear alter


MAMMALIAN CARDIAC MUSCLE 427

nation of light and dark discs, or a condensation (fixation artefact) of peripheral protoplasm, may simulate a cell membrane. Such apparent 'cell membranes' can frequently be traced at some distance into an undoubted fibril. Xor is there any indication of the complicated investment of sarcolemma with respect to myofibrils as conceived by Baldwin. The main observations, however, arguing against Apathy's original conception are the continuity of the axial strand of granular undifferentiated sarcoplasm, and the continuity of the telophragmata throughout the extranuclear portion of the muscle.

But granted that fusiform heart muscle cells actually do exist as illustrated by Baldwin: such cells should then appear isolated in properly macerated material. On the contrary one finds only such short fragments as illustrated in figure 3. Fractures occur in the macerating fluid along the telophragmata, frequently at the levels of the intercalated discs. Such fragments suggest a close structural association between the granular perinuclear sarcoplasm and the non-granular sarcoplasm among the fibrils, most proba))ly by virtue of the telophragmata as first described by MacCallum (9). "VMien the maceration has progressed further only naked nuclei appear (fig. 4), with small adherent masses of sarcoplasm. Occasionalh' such a structure as illusrated in figure 5 appears. Here a i^erijiheral coarse fiber-structure simulates a portion of a cell membrane. But usually its striped character reveals its myofibril nature.

That the technic does actually isolate cells when present is shown by abundant spindle shaped cells (fig. 6) from the endomysium. If the nuiscle nuclei and spindle shaped areas of enveloping sarcoplasm actually' constituted spindle shaped colls surrounded by a membrane, the same technic which isolated such structures from the connective tissue of the same material would also be expected to isolate them from the muscle comjilex. The hypothetical spindle shaped cell of cardiac muscle is obviously structiu'ally not closely similar to the fusiform cells of the endomysium, or of smooth iiuisclo.

Smooth muscle from tlie intestine of the cat subjected to an identical technic vields the usual fusifonn cell, enclosed in a




Fig. 3 Fragment of fiber from macerated cat's heart, drawn as if it were a transparent object. The nucleus is surrounded by granuhir sarcoplasm. The breaks follow tclophragmata, without any relation to hypothetical cell membranes. The manner of fracture strongly indicates that the central granular and more peripheral non-granular sarcoplasm are continuous, probably by reason of the meso- anrl tclophragmata. and that the former is not invested bj^ a cell membrane. X 1.500. •

Fig. 4 Naked nucleus from similar ])rci)arati()n. with adherent clumps of sarcoplasm.

Fig. Nucleus with an adherent mass of graiudar sarcoi)lasm from Ihc same preparation. The sarcoplasm is delimiteil at the right bj' a stout fibril which simulates a membrane, but a faint segmentation reveals its true myofibril nature.

Fig. 6 I.solated fusiform connective tissue cell of the endomysium from the same preparation.

Fig. 7 Isolatcfl large fusiform smooth muscle cell from cat's intestine. The nucleus is surrounded by coarsely granular sarcoplasm as in cardiac muscle, which is continuf)Us similarly with the more peripheral non-granular or finely granular sarcoi)lasm, though myofibrils may simulate, as in heart muscle, a cell membrane. X 1500; reduced one-half in reproduction.

Fig. 8 Oblique transverse section of smooth muscle cell from nuiscularis mucosae of esophagus of cat. The perinuclear sarcoplasm has contracted away from the nucleus leaving a clear space, peripherally limited by a sharp line, a fixation artefact, simulating a cell membrane. The light-blue-staining sarcoplasm however is peripheral to this 'membrane' Zenker's fixation; hematoxylineosin stain. X 1500.

42S


MAMMALIAN CARDIAC MUSCLE 429

distinct nieiii})ran(\ The centrally placed elongate nucleus is enveloped by undifferentiated granular sarcoplasin in a like manner, and of apparently identical structure, even to a delicate peripheral 'membrane,' as in cardiac muscle trabecular Further mechanical treatment (teasing) separates a similar essentially bare nucleus. Fixed smooth muscle stained with the hematox\'lin-eosin combination shows the central perinuclear mass of sarcoplasm stained a faint blue, in contrast to the deep blue of the nucleus and the bright red of the cytoplasm. Certain cells show contraction artefacts. In these cases a space, empty, except for occasional very delicate strands, appears between nucleus and contracted cytoplasm. The inner surface of the latter exhibits a sharp contour, simulating a delicate membrane.

In cross-sections one finds appearances like figure 8 (oblique section of smooth muscle fiber of muscularis mucosae of esophagus of cat). Occasional strands spanning the space might be interpreted as spongioplasm; but the space is colorless, while the light-blue-staining sarcoplasm is without but closely applied (indicating contraction) to the peripheral border or 'membrane' of the space.

If cardiac muscle can be appropriately interpreted in terms of fusiform cells and extracellular masses of myofibrils and sarcoplasm, smooth muscle should be similarly interpreted, since apparently exactly the same cytologic conditions as regards nuclear relation to sarcoplasm prevails in both, irrespective of course of telophragmata. But the histogenesis of smooth nmscle renders such interpretation very improbable. ^Moreover, maceration separates the genetic units, not secondary structures. Shnilarly in the case of cardiac muscle: genetically we start with a syncytium in which myofibrillae are dejiosited: maceration separates irregular fragments, and ultimately yields naked nuclei and masses of myofibrillae imbedded in sarcoplasm. Since the intercalated discs, locations wIumv fragmentation frequently takes place, can not be regarded as cell boundaries, the cardiac muscle must be conceived to persist in its original syncytial condition.


430 H. E. JORDAN

The results of this study of cardiac muscle by the dissociation method, and comparative observations of snnilarly treated endomysium and smooth muscle tissue, yield no evidence in favor of the cellular conception of heart muscle suggested by Ai:)athy and supported by Baldwin. On the contrary heart muscle appears to l)e a true syncytimn, the anastomosing nmscle trabeculae consisting of axial strands of undifferentiated coarsely granular sarcoplasm containing nuclei, and peripheral layers of ai^parently non-granular or finely granular sarcoplasm differentiated in that it contains myofibrillae marked by alternating dark and light discs and intercalated discs, the telophragmata being continuous throughout the extranuclear muscle complex.

LITERATURE CITED

(1) ZiMMEK.MANX, K. W. 1910 t'ber den Bau tier Herzmuskulatur. Arch. f.

mikr. Anat. u. Entwickl., Bd. 75, no. 1.

(2) Jordan, H. E. 1911 The structure of heart muscle of the humming bird,

with special reference to the intercalated discs. Anat. Rec, vol. .5, no. 11.

(3) Jordan*, H. E., and Steele, K. B. 1912 a A comparative microscopic

study of the intercalated discs of vertebrate heart muscle. Am. Jour. Anat., vol. 13, no. 2.

(4) .JoRDA.N, H. E. 1912 The intercalated discs of hypertrophied heart muscle.

Anat. Rec, vol. 6, no. 9.

1012 The intercalated discs of atrophied heart muscle. Proc. Soc. Exp. Biol, and .Med., vol. 10, no. 2. (.5) Jordan, H. E., and Bardin, J. 1913 The relation of the intercalated di.scs to the so-called segmentation and fragmentation of heart muscle. Anat. Anz., Bd. 43, pp. 23 and 24.

(6) Lewis, F. T. 1913 Textbook of histology. Philadelphia; p. 128.

(7) Baldwi.v, VV. M. 1912 The relation of muscle cell to muscle fiber in volun tary striped muscle. Zcits. f. .\llgem. Physiologic, Bd. 14. 1912 b .Muscle fibers and muscle cells of the adult white mouse heart. Anat. Anz., Bd. 42, 7 and 8. <9) SzY.MONOwicz, L., and MacCallu.m. J. B. 1902 Textbook of histology. Philadelphia; p. 86.


THE THYROID GLAXD OF THE OPOSSOI

R. R. BEXSLEY From the Hull Zoological Laboratory, University of Chicago

THREE FIGURES

In the thyroid glands of several of the opossums obtained in the autumn of 1912 from New Jersey several features of interest were noted, which may be briefly stated as follows : The th\Toid epithelium, instead of being uniform, as is usual in vertebrates, contained, in addition to the usual tj-pe of ei)ithelial cells, ovoid cells, parietal in position with reference to the foUicles. filled with fine eosinophile granules which gave them a striking resemblance to the oxyphile cells of the anterior lobe of the h^-pophysis. The epithelium of the thyroid follicles contained large needle-shaped crystals. The th\Toid glands of those animals which were kept in the laborator}' for two weeks or longer showed a high degree of hyperplasia, associated with the disappearance of the contents of the follicles and of the crystals and, in those kept for a longer period, the appearance of a new secretion antecedent in the form of granules along the free border of the cell.

Shice it was possible, considering the time of year at which the collections were made, that the characters and changes in question were associated in some way with the phenomena of hibernation. I made preparations to obtain a larger number of animals, during the past winter, in small groups caught at different parts of the season, and immediately shipped to the laboratory. I have been able to secure these, through the aid of Mr. Ell^ert Clark, from ^^ aldo, -Vrkansas. From each of these series, received on October 21, 1913, December 5, 1913, December 22, and January 19. one or more animals were examined immediately, while others were kept for varying lengths of time in the laborat<M'v. and then examined, or were usetl for experiments as indicated later.

431


432 H. U. HENSLEY

Tlie tliyroids of all animals examined immediately after their arri\al at the laboratory showed the normal type of gland. There were minor variations in the size of the vesicles and in the shape of ej)ithelial cells. ])ut, in all, the follicles were well formed and spherical and filled with a deeply staining colloid. In all the animals also the ovoid cells mentioned above were present in large numbers, and large crystals occurred in the cells of the follicular epithelium.

The thyroids of the animals examined two or more weeks after their arri\al at the laboratory showed a high degree of hyperplasia and cell overgrowth. That hibernation had nothing to do with these changes is indicated b}' the fact that no change of this sort was apparent in animals taken in midwinter and examined immediately, and that in an animal examined on June first there were but slight evidences of involution of the hyperplastic gland.

The thyroid gland of the opossum consists of two ovoidal lobes situated one on either side of the trachea at the posterior end of the larynx. Usually no isthmus is present but remains of it are seen in the form of minute lobes attached to the posterior end of the main lobe. In some cases this isthmus lobe reaches a considerable size, and in one case a complete isthmus was found. In one case also an accessory thyroid gland was found in a mesial position low down in the neck.

Figure 1 represents a section from a thyroid gland of an animal sacrificed immediateh' after its arrival at the laboratory. The vesicles in many were larger than in the illustration, the amount of colloid greater and the epithelium flatter. It will be noted that this gland differs from the familiar type of vertebrate thyroid in two features, namely, the presence of a second sort of cell and the presence of large crj'stalloids in the regular epithelial cells.

^riie crystals are invarialjly present in the cells of the thyroid gland examined immediately after capture of the animal, though as will ai)pear later they may be absent or nearly so in the glands during the period of active hyperplasia. They are exclusively intracellular in the epithelium of the vesicles, and never occur in the parietally placed ovoidal cells. Their solubilities have not been accurately determined. Their prf)toin nature is indicated by


THYROID GLAND OF THE OPOSSUM


433



Figure 1


the strong Millon's and xanthoproteic reaction^; which the>- give in sections. They are visible in the fresh tissue examined in salt solution, but disappear with crossed Xicols. ^^^len the animal is injected with an oxazime dye known conmiercially as "new methylene blue GO," the crystals stain a deep Ulac color. The cholesterin reaction, and the reaction for phosphates are negative. The staining with new methylene blue GG in the fresh gland indicates their permeability to this dye. and suggests that notwithstanding their crystalline form they are permeable to oth(M" substances in solution.

The ovoid cells resemble anterior lobe cells of the hypophysis. Thoy contain a multitude of tiny granules, easily visible in the fresh cell, though of low refractive index, and staining readily in the living cell with the dye mentioned in the foregoing paragraph. The nucleus is large, oval in outline, located nearer one end of the cell, and richer in chromatin than the nuclei of the regular thyroid


434 H- H. BENSLEY

epitliolium. In stained preparations, among the granules in the pole of the cell which contains the larger amount of protoplasm, may be seen the tlelicate network of canals which Holmgren regards as of trophospongial origin but which I consider the homologue of the vacuolar system of plant cells. In these canals is a substance which stains faintly pink in sections of formalin Zenker material stained in ]Mallory's phosphotungstic acid hematoxylin. Sometimes these canals are expanded locally to oval, fusifomn, or spherical vacuoles containing the same substance.

These cells are always peripheral in position, and never extend to the lumen of the follicle. They are, however, in immediate contact with the follicular epithelium, and no reticulum extends between the two type of cells. They are distinguished from tissue mast-cells by the fact that the granules of the latter stain pink intra vitam with new methylene blue GG, while those of the ovoid cells stain blue. From Unna's plasma cells and from fibroblastic cells they are distinguished by the discreteness of their granulations, l)y their size and by the fixation properties. The best mode of demonstrating them is fixation in formalin zenker, staining in IMallory's phos])hotungstic hematoxylin, in which the granules stain deeply blue. In preparations stained with hematoxjdin and eosin the granules stain red, and a similar distribution of the acid and basic dyes follows stainhig with toluidin blue and acid fuchsin. In the preparations so stained, blue stained floccules may be seen distributed through the cell protoplasm, in additif)n to the small oxyphile granules.

The distriliution of the ovoid cells in the thyroid of the opossum is irregular. In three glands serially sectioned they were more abundant in the anterior three-fourths of the gland. The posterior fourth contained few and the isthmus and one accessory thyroid none. That the cells in question are special internal secreting cells there can be little doubt but what their homologiies in other vertebrates may be, remains wholly obscure. The possibility that they may represent a dispersed parathyroid has been considered but no proof of this has been obtained. Indeed they resemble the usual cells of the parathyroid glands as little as they do those of the thyroid though some resemblance to the eosin


THYROID GLAND OF THE OPOSSUM 435

ophile cells described in the human thyroid by Welsh and others may be perceived.

As indicated above, the thyroids of all animals examined after two or more weeks in the laboratory show a high degree of hyperplasia and cell overgrowth. In the first month of this process numerous mitoses may be seen in the cells of the th>Toid gland, and the latter increase considerably in size. This increase in size is associated with a proportional increase in the quantity of mitochondria and with an increase in size of the individual filaments. Though the process is invariable in the animals kept in captivity there is some variation in the rate at which it proceeds.

Figure 2 represents a follicle from a gland taken from an animal kept in the laboratory for a period of six months. The hj^perplasia though high is not materially greater than that observed in two animals from the same group killed during the first month of captivity. Indeed, the results of the obserxations on all the groups indicate that the hyperplasia proceeds very rapidly at first, then slows down, though mitoses may be found even after six months even in glands which are reverting to normal type after iodine administration. It will be noted in figure 2 that the follicle has expanded to a large complex mass of cells in which the original lumen is still visible though it no longer constitutes a secretion space. Instead, secondar}' secretion spaces, suggesting an attempt to reconstruct the thyroid by breaking the hyperjilastic cell-mass into new independent alveoli, are to be seen. These secondary secretion spaces, not much larger than a red blood corpuscle, contain a small gl()l)ule of colloid, wliich stains blue in Jones' modification of ^Nlallory's anilin blue method. The i)orders of the cells next the lumen contain a few granules, apjiarently a secretion antecedent. These granules differ from intracellular colloid in their properties inasmuch as they occur always at the extreme border of the cell while the colloid may be deep in the protoplasm of the cells or even external to the nuchnis itig. 3). They are with diliiculty preserved and then only in the most peripheral part of the piece of tissue, indicil iiig a different solubility from that of the intracellular colloid, ^\'hen fixed by formalin Zenker or by acetic osmic bichromat(^ thev stnin readilv with neutnil gentian and are


436


H. J{. HKNSLKV



1^ ^ P^




Figure


stained blue by Mallory's phospliotuiigstic hainatoxylin. They resemble certain granules which I have found in the hyperplastic thyroid of man but the granules are smaller than in man. Whether these granules represent an incompletely elaborated colloid, or another normal secretion which is exaggerated in the hyperplastic gland or a new secretion, I cannot fully determine. The facts so far, however, are opposed to the first assumption for as will appear later, in the gland which is reverting to normal as the result of iodine administration, and in which colloid secretion is proceeding


K


THYROID GLAXD OF THE OPOSSUM 437

at a rapid rate, the latter makes its appearance in the cells in the form of droplets which have unquestionably the characters of the colloid as seen inside the follicle, and this resumption of normal activity is associated with a disappearance of the granules described above. I am inclined therefore to the view that the granules represent a new secretory product or a nonnal secretorj* product different from thyreoglobulin, which is not present in the normally secreting gland in sufficient quantities for microscopic detection. In either event the secretory condition of the h\'perplastic gland would represent a perverted secretion indicated either by the introduction of new secretory products or the disturbed equilibrium of normal products. In these hyperplastic glands, here and there, but very rarely, small droplets of colloid may be seen in the protoplasm of the cells. They are usually more deeply placed in the cell than the granules discussed above and show the characteristic staining pro])erties of intrafollicular colloid.

The ovoid cells apparently share in the hyperplasia for in the hyperplastic glands they are much more numerous than in the normal glands. In some cases large groups of cells differing in some respects from both types, but which I take to be the result of hyperplasia of the ovoid ty^^e, are found. These cells stain deeply blue in ^lallorj-'s phosphotungstic acid hematoxyhn but the proplasm is much reduced in comparison with the normal oval cells, and the definite granulation is not seen. In these groups also mitoses may be seen.

The degree and rate of hyperplasia is subject, in different animals, to some variation, but some degree of hj-perplasia is invariable in the animals kept in captivity. In some animals the degree of hyperplasia is such that the whole thyroid gland is converted into a continuous complex of branching and anastomosing epithelial cords, which giv(> it a superficial re.«;emblance to the parathyroid gland.

In the intermediate stages of this hyi>erplasia some glands show a complete absence of colloid, most of them show a great reduction of colloid and. in the earlier phases, few of the border granules descrilied above. Tlir crystals also disappear or beconie greatly


43S


K. H. BENSLEY



Figure 3


redurod in size and in niiml)or. Since colloid originally present in large amount rapidh' disappears from the gland and since the evidences of normal colloid production are almost wholly lacking Tnote the absence of intracellular colloid as well as intrafoUicular colloid) and since the resumption of activity after the administration of iodine is marked by the ap]iearance of colloid both in the cells and in the follicles, in my opinion the conclusion, which is also probal)le on general biological grounds, that this gland in the phase of acti\e hyperplasia and cell ()\ergrowth has a low secretory rate and potential, is justified. On the other hand, after the period of most active hyperplasia is past the gland resumes activity though of an abnormal sort, marked by a low rate of colloid i^roduction and of crystal i)roduction and the appearance of a new secretory antecedent in the cells.

Since I wished to establish the independence of the hyperi)lasia of captivity with reference to the phenomena of hibernation the


THYROID GLAND OF THE OPOSSUM 439

number of animals available for experiment has been small. I hope to return to this aspect of the question next year for the readiness with which this animal's thyroid undergoes overgrowth in captivity suggests the possibility of controlling both hji^erplasia and reversion experimentally. The experiments on feeding gave no definite results though the border granules were more abundant in an animal kept for two months on meat and egg diet exclusively, as compared with several animals kept for a similar length of time on a mixed diet of meat, bread, and apples.

With regard to the effect of iodine administration one experiment made in the autumn suggests the possibility that there is a refractory period, in regard to iodine, at the height of hjT^erplastic activity. This animal, a female received October 21, was made the subject of a hemilobectomy on November 18. The gland removed showed a high degree of h^-perplasia. For two weeks the animal received a dose of 5 drops of s>Tup of iodide of iron daily, and, at the end of the time, the other lobe was removed and examined. No substantial change was noted in the second lobe.

On the contrary, experiments made in the spring on animals which had been present in the laboratory all winter and in which as the controls showed there was still a high degree of hyperplasia and but little tendency to reversion gave a prompt and characteristic reaction to iodine. On ^lay 8 four animals which remained in my collection were set aside for iodine experiments. Of these two were retained for control and to each of the others was administered daily 5 drops of syrup of iodide of iron. One of the controls died and was iuiavailal)le for examination. One of the iodide animals was killed after 17 days, the other after 24 daj's. The remaining control animal was thyrodectomised on the same day as the last iodine animal and the thyroids hxod for histological examination.

Both of the animals which received iodide showed an advanced degree of colloid involution of the thyroid gland, and this was more advanced in tlio twenty-four-day annual than in the seventeen-day one. In lioth practically every cell in the thyroid gland contained one or more droplets of colloid, and in each the re-formation of t)i(^ follicles had advanced to a degree which was pnv

THE ANATOMICAL KKCOHll. VOL. S. NO. 1>


440 li. K. BENSLEY

portioiial to the duration of tlic experiment. The control gland differed in no respect from the hyperplastic glands examined earlier in the j^ear, that is, showed practically no reversion. Figure 3 shows a section of the thjToid from the twenty-fourdays iodide animal. In this gland it is to be noted that there was no increase in the number of intercellular crystals as compared with the control gland. The resumption of colloid activity was a common i)roperty of the thyroid epithelimn and was not associated with the formation of the so-called colloid cells of Langendorff. The border granules disappeared from the cells with the resumption of colloid activity.

These observations on the opossum establish in my opinion by studj' of both th^ hyperplasia and involution the high degree of labilit}' and rapidity of reaction in the thyroid gland, on which ^larine and his co-workers have long insisted. They furnish an opportunity to control and to analyze the factors involved in thyroid hyperplasia. They confirm Marine's conclusion of a low rate of colloid production in the hyperplastic gland. In addition, by demonstrating a new tj^De of cell in the thyroid gland of the opossum and a new secretory product in the cells, they furnish objective evidence of that polyvalenc}' of thyroid secretion which has been so often postulated in the discussion of morbid conditions of the gland.


COVERS FOR DISSECTING TABLES

T. WIXGATE TODD From the Analominal La')oralory of Western Reserve University, Cleveland, Ohio

THREE FIGURES

In order that the border-line areas between the different 'parts' of a cadaver may be studied efficiently by the student, it is essential that dissection of the whole subject be carried out before dismembennent. In this way alone is it possible to obtain a correct impression of the vagus system, of the relations and distril)ution of the limb plexuses, of such muscles as the psoas and the obturators and of many other points in human anatomy. But the greatest technical difficulty in attempting this lies in the fact that the cadavers drv out so easily in spite of every care. For the student to be able to revise his work from time to time, it is essential that the cadaver be kept hi the best possible condition for six months or even more, during which time, of course, dissection is proceeding ever}- day. The utmost care leaves much to be desired in this respect because it is impossible, even were it desirable, to move the cadaver into a tank or store each day when dissection is finished, and no subject kept continuously on the dissecting-room table can long remahi in fresh condition even with the precautionary- measures of plenty of cloths and waterjiroof covers in addition to repeated moistening with the various fluids in vogue for such a i)uri">ose.

Our final attempt at Western Pu'ser\-e University to overcome the difhculty of adecjuately preser\-ing cadavera in the di.<;secting-room has met with such success that it may perhaps be a useful suggestion to other laboratories where the same difficulties are to he met.

Each table is mounted on castors so that it may be easily moved, and is jirovided with a galvanized iron cover, which can be dra\\ni to the ceiling when work is in j)rogre.ss. The cover is made of numl^er 20 galvanized siieet iron. Its dimensions are: length. 5 feet 10 niches:' breadth. 23 inches; depth. 14 inches. It is made watertight. Its lower margin is flanged round a steel wire to give it greater rigidity and finish. Its top is strengthened by three strips of iron i inch in tiiickness and 1\ inches in breacHh. These are riveted to the top, as showni in figures 1 to 3, and at eacii intersection an iron ring (2 inches in diam 1 The length had to be o feet, 10 inches, to fit our tables. A 6-foot length is desirable, but we find that with our routine measure of keeping the feet at right angles to the legs when the cadavera are embalmed, there is no difficulty in getting the cover to fit over all ordinary subjects.

441


442


T. W INC ATE TODD




3C


n


^



^



N





A B



c


I T!



' «L> .



m


■ 3.i'


Figs. 1 to 3 Plans for the making of the covers; scale one-half inch to one foot. Figure 1 represents the plan of the roof of the cover; figures 2 and 3 correspond to side and end elevations respectively.

eter) is bolted to the cover. The cover fits accurately on to the table, and can easily be raised or lowered at will ]\v means of cords antl pulleys fixed to the ceiling. Parallel with the suspension jiuUeys, but about 3 feet distant from them, two lOO-candle-power Mazda lamps are suspended, so that when the cover has been raised the table can be dra^^^l from a position directly under the cover to one beneath the electric lights. At first it was thought that when the cover was raised any contained condensed moisture might drip from the inside, but it has been found that there is not moisture enough to cause any trouble. The covers are j^ainted white both within and without, and thus add to the clean and tidy ajipearance of the room, while they do not interfere with the lighting when they are raised.

In addition to the efficient preservation of the cadavera, the covers keep the tables free from dust and therefore prevent any soiling of dissections placed on the tables. Moreover, they add to the neatness and well-being of the dissecting-room, and transform it, so far as the members of other departinents of the University are concerned, from a gruesome, somewliat rejiulsive apartment into a clean and pleasing lai)()r;itory. ('ort.ainly tlie appearance of covered tables is much more agreeable than that of tables on which the outlines of the cadavera are plainly suggested under the folds of the dark-colored waterproof covering.


COVERS FOR DISSECTING TABLES 443

The practical application of the idea was carried out by Prosector Leonhart, and the students were ciuick to recognize the advantages of the covers so that no 'regulations' are necessary for their use.

One last consideration may be mentioned in favor of such covers as are herein described. Should they become obsolete for their present purpose, their size, their watertight build and their strengthened top allow them to be turned upside down and used as tanks for the preservation of material. In case of such use the removal of the iron rings leaves the tank with two holes in the bottom into which can be fitted ordinary corks, and which provide for drainage of fluid and efficient cleaning.


A TANK von THE PRESEHN ATIOX OF ANATOMICAL

MATERIAL

r. WINGATE TODD From the Atinfotnicnl LnhoniU.ri/ of Western fiexerre Univerxity, Cleveland, Ohio

THKEE FIGURES

TIk' a(U'(iuate i^rpsen'iition of gross anatomical material requires some form of rccc^ptacle or tank, and if large (juantities of material or dissections eitlier of human or mammalian anatomy are to be carefully kept in good condition, it is necessary that the recej^tacles he so cheap as to he readily multiplied with the growing needs of the dejiartmiMit. In Western Heser\'e University we have, besides the dissecting-room, a museum, material for which accumulates faster than it can be mounted for exhibition, under present circumstances. In addition, arrangements with the various hospitals and with the city administration result in the ac(iuisition of nuich fetal material and the bodies of all animals from tiie Zoological (iardens. For these reasons it has been necessary to provide such accommodations as shall be at once cheaj) and ser\^iceable.

The form of tank descrilx'd below has fulfilled these recjuirements, and is therefore now being used in other laboratories. Hence it seemed advisable to make a record of it as one more laboratory furnishing .suitable for anatomical departments.

The tank, the plans for the manufacture of wiuch are also submitted with this connnunication, is made of galvanized iron, numlxM- 20 thickness. It is watcrtigiit, and ))rovided with a flange running round its upper margiii, tlie flange being \\ inches broad. The lid is sim|)ly a sheet of the same metal sligjitly scored diagonally from comer to comer so that a somewhat concave surface is i)resented to the contents and the flow of the condensed fluid which accumulates on the under surface of the lid directed to the comers. In order to seal the tank hermetically, the flange is thickly smeared with vaseline.

The thickness of the iron is found to be sufficient to prevent undue bending of the fUmge, but of course the n\sult of any accident to tlaiige or lid can readily be re|>aired witli tjie iiammer. The vaseline method of sealing has proved eciually efficient with the metliod wiierebv the lid is made to fit into a channel filled with glycerine round tiie top side.= of the tank, and is much more convenient than any other .scheme of tank lid. The inner surfaces of tank and lid are coated with asphaltum, which is renewed from time to time. The tank is cheap, tight, portable, does not get out of order, and is ver>' easily opened, closed or cleaned.

444


TANK FOR ANATOMICAL MATERIAL


445



i 1


3


Figs. 1 to 3. Plan of tank measuring 48 X 11 X 1-1 inches; scale one-half inch to one foot. Figure 1 shows the plan of the tank, with the lid in dotted lines. Figures 2 and 3 represent side and end elevations.

It has proved much easier to use and more convenient than the usual form of tank made of slate, stoneware, wood or lead-hned wood. It can always l)e made locall}', and the stock can be increased at verAshort notice. It can he made of any size iiji to one which will hold the larger Mammalia. But it is well to have a jihio; in the floor of the bigger tanks so that they may be emptied of fluid and cleaned more readily. The idea originated in a somewhat similar tank in use for the preparation of color specimens by the Kaiserling method in the Pathological Laboratories of the University of Manchester.

It is convenient for storing purjioses to have .-standard sizes, and the dimensions which have been found most useful by Mr. Leonhart, Prosector to the Department, arc the following:


Kaiserling prci)ar!itiun

Brain storage

Limbs or pelves

Torsos


DIMENSIONS I.V INCHES


I.oncth


Breadth


Depth


•J 4


11


s


4S


U


s


4S


11


14


Is


•V)


14


446 T, \vix(;ate todd

It is obvious that any size may l)e made; only those which are most generally useful have been detailed. If it is desired to suspend the brains in fiuitl from rods i)laced across the tank, plaster slabs 1 inch in thickness may be made to fit the length of the tank. Grooves may then be gouged out of the upper margin to accommodate the rods. Finally, the plaster slabs may l)e rendered hard by lioiling them in oil. one of the l^rain tanks being used temporarily Un tliis purpose.


A SIMPLE ELECTRICAL HEATING DEVICE FOR INCUBATORS, ETC.

A. O. WEESE Deparlmenl of Biology, The University of New Mexico

FOUR FIGURES

The advantages of electricity as a means of heating incubators, paraffine baths, etc., are, I believe, everywhere recognized. Gas, \\ith all its uncertainty and inconvenience, is still employed in man}' laboratories on account of the prohibitive cost of the various high-priced electric incubators on the market. Many excellent devices have been designed by laboraton,' workers, and described in this and other journals, for the utilization of electricity for heating incubators originally designed for gas, and the most of these work \er\ satisfactorily where direct current is available in the laboratory-. The fact that there may be other laboratories confronted with the same problems that we have here has prompted me to offer the suggestions contained in this paper. Our laboratories are at present supplied with 110 volt alternating current only, and a lack of mechanical facilities suggested the modification of the old forms of gas regulators for use with electricity.

The heating element made use of was made of nichrome wire. About 40 feet of number 18 nichrome wire was wound, on a lathe, into coils of 5-inch inside diameter, and the entire coil .stretched between pegs arranged in a 'transite' base the size of the incubator. A wooden frame, lined also with transite board, was constructed so as to support the incubator about } inch above the coils. The heating chamber thus formed was completely insulated on all sides with transite. Copper leads connected the ends of the heating coils with ])inding posts on the outside.

For thermo-regulators I have ]>een able to modify several forms of gas regulators so as to operate a "make and break" device. Two examples will suffice. The mercur>--gla.- rises to the point (/) a cont-act is made, and when the mercur>' falls, due to cooling, the cont^'ict is broken.

The Roux bimetallic regulator may be utilized as follows: The hole at {x, fig. 2) should be reamed out and the upper part (c. d, c), insulated from the arm (//) by means of washers of mica or other non-conducting

447


448


A. O. WEESE



1


liij. 1 A modified Reiclicrt regulator; d, platinum terminal sealed into inside tube; d, terminal fastened to regulating screw;/, point of contact between platinum terminal and mercury.

Fig. 2' A modified Roux regulator; a, regulating screw; b, lock nut with terminal attached; e, terminal soldered to gas tube; .r, insulate! joiiif ; //, rcmilntor arm; z, point f)f contact.

inatorial. Wires are then attached at (b) and (c), tlic circuit Ix'ing mado and broken at (z). In this case the circuit is made when the himetaUic part, of the regulator is cooled below the desired teni})erature. Therefore, with this regulator, the connections siiould be arranged so that the current will flow through the heating clement when tlie cir


' Those figures have been reproduceil by courtesy of liausch and Lomb Optical Co.


HEA'I'IN'G DEVICE FOR INCUBATORS


449


cuit through the regulator is closed. When the mercury-glass form is used the opposite should be the case. Many of the other forms of gas regulator may be modified in a similar manner, at a ver>' small cost and with very little work.

These devices may be used with either direct or alternating current, but the arrangement will be somewhat different in the two cases. With either type of current the actual making and breaking of the heating current is accomplished by means of an ordinary telegraphic relay, to prevent arcing at the jjoints of contact in the regulators, through which only a very small current can be allowed to pass. \\'hen direct current is used, the regulator and electromagnet of the rela}' together with a large resistance {R) are connected in parallel with the heating element (fig. 3). The resistance is very large so as to allow just enough current to pass through the regulator to actuate the relay, and no more.



Fig. 3 Diagram of connections for direct current; .V, lu'ating elcineni ; L, electric line; t, regulator; r, relay; R, resistance.

Fig. 4 Diagram of connections for alternating current; B, battery; otherwise lettered the same as figure 3.


When alternating current is used, otlier means must i^e employed to operate the relay. In this case the manner of connecting is as shown in figure 4. To supply tlie regulating current hero 1 use a batter>- of three Columbia dry cells, with a "i.lO Ohm. tolograpiiic relay. As used here the cells require renewing about once in four months. Within that time there is absolutely no danger of tiie relay failing to operate, in fact in some cases the same i)atterv lias been used for a much longer time. With any of the devices mentioned temperature regulation is much more accurate than woukl be supposed. The variations, on a scale extending from room tcmiierature to a pohit 70'^(\ above room temperature, is always less than one degree, which is accurate enougii for all routine zo logical work, and nnich more accurate than tlie ordhiary gas regulator, t'sjiecially if the gas pressure is somewhat variable.


NOTICE

The next annual meeting; of the American Association of Anatomists "vvill be held in St. LouLs during December. The Association goes to St. Louis as the guest of "Washington University. The exact dates of the scientific sessions will be announced later.