Talk:Book - Contributions to Embryology Carnegie Institution No.32
CONTRIBUTIONS TO EMBRYOLOGY, No. 32.
THE SINO-YENTRICULAR BUNDLE : A FUNCTIONAL INTERPRETATION OF MORPHOLOGICAL FINDINGS.
By Robert Retzer,
Professor of Anatomy in the University of Pittsburgh.
With one plate.
In spite of the large amount of seientifif work that has been done on the subject, tlie problem of how the contraction wave passes from the venous to the arterial end of the heart still remains unsolved. When a muscular connection between atria and ventricles was discovered and it was found that the severance of this connection brought about arhythmia, many adherents of the myogenic theory felt that the burden of proof lay with the neurogenists. It was, however, not long before it was establi.shed that there were nerve fibers accompan^-ing this muscular connection and the problem stood as it did before the discoveries of His and Kent. I have devoted the last fifteen years to anatomical studies bearing on this subject and during tliis time my views have undergone considerable change.* In the present papers I have undertaken to outline the possible interpretations of the facts as we now know them.
It seems that there is a lack of vmderstanding between the anatomist and the phy.siologist, because neither seems to ajipreciate the value of the facts described by the other. For instance, it should not be left out of account that the heart of Linuihis (whicli is unquestionably neurogenic) has more resemblance to voluntary than to cardiac musculature, considering its morphology. Again, the peculiarity of the histologic structure of the muscular connection between the atria and ventricles and of the Purkinje fibers is left almost entirely out of consideration by those who hold that the myogenic theory is the only correct one. It is this phase of the question which is the subject of this paper.
The Uterature has been so thoroughly reviewed recently by others that I propose to discuss only those articles that have a direct bearing no the subject at hand. The term "atrio-ventricular bundle" was given by His, jr. (1893) to a strand of cardiac muscle which arises in the posterior wall of the right atrium, near the atrial septum, is sui)erimposed on the superior edge of the ventricular septum with numerous interchange of fibers, and (passing forward) divides near the aorta into a right and left branch, the latter ending in the aortic leaflet of the mitral valve. It was later foinid that this bundle does not begin nor end in the manner described by His, but that the region of origin represents the primitive sinus region of the heart, and that it terminates in a complex network of fibers long known as "Purkinje fibers." Imbued with the idea that this connection conducts impulses from atria to ventricles, Tawara (1906) gave it the name of "Reizleitungs-system." This term is still used by manj' authors, but feeling that a morphological one must be used as long as the question of function is still in doubt, in 1908 I employed the name "sino-ventrioular l^undle." By this term we definitely commit ourselves only to the fact that the bundle forms an anatomical connection between the sinus region and the ventricles. The terms "conductive system" and "atrio-ventricular bundle" still have their place, as I shall indicate later.
Although a large part of this work was done during my association with Professor Mall. I should like also to record my indebtedness to Professor Keibel of Freiburg, and Professor Felix, of Zurich, for the courtesies extended to me while a guest in their laboratories; and to Professor Bensley for his kindly interest during my residence in Chicago.
It may be noted that most of our anatonaical text-books take little cognizance of the change of our knowledge on this subject and still persist in using the term "bundle of His." It has already been pointed out that Ilis's description is far from correct. I maj' add here that considerable skepticism as to the correctness of the observations of His existed in the mind of liis father and of other mature anatomists when I made my first observations under Professor Spalteholz, in 1903, and it was for this reason that the problem was undertaken again. Kent's (1893) original description is far more correct in the Ught of our modern investigations than was that of His, and the bundle has been named the "bundle of Kent" by some authors; but we should recognize the fact that it was His, who by exi^erimental methods, was the first to attempt to prove that all impulses from atria to ventricles pass bj' way of the bundle.
In my notes, made in 1903, when I began my studies on the subject, I find the follo\\dng review of Kent's article:
"In his figure on p. 244, it seems to me that he mistakes the Purkinje fibers for fibers of the mj'ocardiuln. His description of the cells standing between muscle and connective tissue reminds me also of these (Purkinje fibers)."
This note was made five years before the appearance of Tawara's monograph, which showed indisputably that the Purkinje fibers represented the end-ramifica- tion of the atrio-ventricular bundle. Recently I read the Proceedings of the Physiological' Society of November 12, 1892, which I had previoush^ overlooked, and I was much interested in the report of Kent's paper before the society:
"Between the auricle and ventricle and lying in the cimnective-tissue ring are modified muscle cells, usually spindle-shaped, nucleated, granular, becoming extremely narrow in parts and then swelling out again, transversely striated, Ijranched, and usually connected into a network. These cells are somewhat rudimentary in the case of the rat and are verj' well developed in the monkey. In the latter animal they exist as a complete network, permeating the fibrous connective tissue of the groove and extending through from auricle to ventricle. IJpcm approaching the groove the normal cardiac fillers split up ijito similar fibers and become connected with the network of cells lying in the fibrous tissue."
This shows even more conclusively that Kent did recognize the difference in appearance between the bundle-fibers and the heart-muscle fibers and also saw the ventricular transitions between them, although he did not recognize that the end- ramifications of the bundle really represent the Purkinje fibers. His descrip tion, however, lacked the dogmatism which seems so essential for the general acceptance of a new idea. Thus he states in a summary:
"It would appear* then, that the fact of two masses of muscle being joined together by fibrous tissue is in itself no argument agamst the muscular continuity of such masses,
The italics are mine (R. R.)
the fibrous nature of the intervening tissue by no means excluding the possibility of mus- cular fibers running through it and preserving the muscular connection. .\nd in the mammalian heart such a connection appears to exist."
Kent (1913) finds more than one connection in some hearts he examined and all careful researches will confirm tliis observation. In the pig I have seen bundles of muscle-tissue streaming down along the outer posterior wall connecting the muscular tissue in the region of the coronary sinus with the ventricular musculature.
Since then innumerable papers on the subject have appeared and some authors have claimed priority in the discovery of the "bundle," either for themselves or for others. Thus priority was claimed bj^ Paladino (1914) and some authors contend that Kiirschner described the bundle in Wagner's Handbuch. These authors did describe muscular connections, but unlike His, they did not limit these to a restricted area of the septum. Until the monograph of Tawara appeared all work done was in the main confirmatory of His. According to those authors, the muscular con- nection consisted of ordinary cardiac musculature and contained but a negligible amount of nerve-fibers. Tawara was the first to show the connection between the bundle and the Purkinje fibers, although (as I have stated above) Kent had appre- ciated that there was a difference in the histological structure of the bundle and the rest of the cardiac musculature.*
Tawara called this connection between atria and ventricles and the widely ramifying Purkinje fibers the "Reizleitungssystem." He described nerve-fibers accompanying the system but paid little attention to them. Indeed, his mono- graph was used extensively bj^ the adherents of the myogenic theorj^ as further evidence in their favor. In spite of the careful work of Erlanger and others, it has never been determined definitely' whether it is the nerve-fibers accompanj'ing the bundle or the bundle itself that is the conducting element. The term "con- ductive system" is, therefore, only applicable to the nerve-muscle complex. It must be divided anatomically and physiologically into its component parts— the nervous mechanism and the muscular mechanism. This is essentially the problem of to-daj'. Its solution will definitely settle the myogenic or neurogenic theory of the mammalian heart-beat.
Authors using the term "bundle of His" usually mean the muscular connec- tion. They use the term "Tawara's node" also in the sense that it is a muscular node — as described by Tawara. In view of the fact that the bundle is in aU instances accompanied by nerve-fibers, we must be more accurate in our ter- nainology. Therefore, it has been the author's custom to use the term "conductive system" not in Tawara's sense, but as a designation for the neuro-vmscular complex extending from an indefinite area in the wall of the right atrium to the ventricles including the Purkinje fibers. His's bundle or "atrio-ventricular bundle" I have termed the muscular connection between atria and ventricles, not including the Pur- kinje fibers. The term "sino-ventricular bundle" is used to define the muscular connection extending from the region of the coronary sinus and the great veins to the ventricles, inckiding the Purkinje fibers. The nervous mechanism is without a name.
Although Tawara's work has been confirmed in this respect by all who have published on the subject since then, we find in the 1912 edition of Quain's Anatomy, Schaefer's Text-book of Microscopic Anatomy, the foUowinE statements, p. 200: "In man and most animals, distinct Purkinje fibers do not exist, but the cardiac fibers are rather larger near the ven- tricular endocardium than elsewhere." And further on: "In animals which have fibers of Purkinje. the ventricular end of the bundle resembles these in r'ructure." Statements like these in text-lxioks which we have learned to look upon as more or less authoritative should be corrected.
The general gross anatomy of the muscular connections has been so ably described and so well illustrated by Aagaard and Hall (1912) that it would seem unnecessary to describe it again were it not that slight differences of opinion concerning the extent of these connections necessitate a definite statement upon which I base my later discussion. We generally find the descriptions and figures begin with the node of Tawara and end with the ramification of the Purkinje fibers. Such descriptions are quite correct as far as one can recognize these structures in the gross. Injection methods which give most satisfactory results for demon- strating the system are useless beyond these areas, and it is doubtful whether any other gross method will reveal anything further that can be considered conclusive. The use of the Zeiss-Greenough binocular in dissecting out the sinus portion will show fibers leading from the node to the coronary sinus and inferior vena cava, but whether these fibers are continuous with the bundle or not is a question that needs close histological examination. Both with the gross and with the micro- scopic investigation there seems no difficulty, as long as the bundle is invested by fibrous sheath. It is this fibrous sheath which is injected, and it is this sheath also wliich stands out most prominently when stained with dyes that differentiate between musculature and fibrous tissue. As soon as we reach the node of Tawara, when tracing the bundle to its origin, the fibrous sheath disappears, and from here on the tracing is difficult, if not impossible, by any gross methods. Sim- ilarlj'-, at the other end of the bundle, we can not demonstrate in the gross the transition between it and the cardiac musculature.
The description of the gross, therefore, will closely- correspond to the descrip- tion given by most writers of recent years, but to make it comjilete we must add to it what the microscope reveals. The bundle takes its origin in the musculature of the right side of the interatrial septum immediately in front of the coronary sinus or of the left superior vena cava in those mammals that lack a coronary sinus — that is, in those that have a persistent left suioerior vena cava. It passes forward and downward* in the septum and comes to he beneath the insertion of the septal leaflet of the tricuspid valve. This part of the sin()-\-(>ntricular bundle is so delicate and diffuse that it can not always be distinguished from the sur- rounding atrial musculature, even after the removal of the endocardium. Here the diffuse strands are collected into a bundle which can readily be seen as a light- colored strand having a diameter of 2 to 3 mm., running perpendicularly to the direction of the fibers of the ventricular septum and between them and the pars membranacea septi. At a i)oint that lies in a line drawn from the apex of the; left ventricle to the junction of the right and of the i)osterior cusps of the aortic valve, it divides into a right and a left branch that straddle the muscular interventricular septum. These two main branches pass slightly forward and downward toward the base of the pupillary muscles. In the left ventricle, there being two papillary muscles, the bundle must necessarily chvide, while in the right ventricle it remains single, reacliing the large; papillary muscle by way of the moderator band if there is one, or in relatively the same position if it is lacking. In the left ventricle the main branches are visible without dissection, appearing as light strands under- neath the endocardium, sweeping downwards towards the base of the papillary muscles making an inverted V. In the right ventricle the bundle is frequently beneath a millimeter or more of musculature, l:)ut becomes vi.sible at the base of the papillary muscle. The end-ramifications or the Purkinje fibers are quite visible in the fresh heart, but in many of the preserved hearts they can not be defined. They spread out underneath the endocardium in a complex network, usually bridging over the recesses between the trabeculae carneae.
•The directions refer to tlic excised liCiirt with Imau upwani ami septum aiilxTo-po.sterior,
This description is apphcable to all mammalian hearts. In the human heart as well as in the dog's, it should be noted that the strands are exceedingly' delicate and tear easily when chssected out. In the Herbivora, however, the entire system is ensheathed by collagenic fibers that give the bundle a very much fighter color than the surrounding musculature and also make dissection easy. In these species, sheep especially, the Purkinje fibers are accompanied by fat-cells that stand out gray on the dark background. In these hearts, furthermore, the right branch Ues immediately beneath the endocardium, while in the human and dog it must be dissected out of a depth of a milHmcter or more of musculature.
The microscopic anatomy is far more complex than the gross. It is due to inadecjuate liistological work that our present conception of the nature and function of the sino-ventricular bundle is somewhat erroneous. The microscope must settle two questions — one, the origin of the bundle where we lose trace of it by dissection, and the other, the nature of the material that constitutes the bundle. In reference to the first question I must confess that we have not advanced far. As previously stated, it is my opinion that the bundle does not originate in Tawara's node but in the musculature that represents the primitive sinus region of the heart. I [base this opinion upon the examination of the embryonic heart, where the musculature of the various parts presents greater cytological differ- ences than it does in the adult. In the latter the problem is similar to tracing finer nerve tracts in the medulla without the use of pathological material which singles out these tracts by degeneration. Nerves, connective tissues of various kinds, and poorly striated musculature interwoven in a most complex manner present almost insurmountable difficulties. In reference to the second question we reafize that our histological knowledge is based almost entirely upon Tawara's work. His illustrations are more or less diagrammatic and for the sake of clearness such structures as seemed to him unessential were omitted. He showed that there were histological differences between various mammals and in various parts of the same heart. But as to why we have these differences he does not even spec- ulate. I think that we can show the reason of these differences and what their significance is.
As that material was at one time most available, I used the pig heart for study. For comparison, many other hearts were studied, including the human. Various methods were used, such as the maceration, the injection, the dissection, and the use of other reagents to bring about a clear jiicture of the system in gross, but, as stated before, it meant merely a repetition of what is already known. I was, therefore, forced to depend entirely upon histological technique. At first I was content with the use of such fixatives as Zenker's, formaldehj'de, mercuric bichloride, and other combinations, and with such stains as would diiferentiate between mus- cular tissue and connective tis.sue. It was soon seen that these methods would not lead me bej'ond the confirmation of previous work. As stated before, as long as we have an investment of the bundle there is no difficulty in tracing it, but I can state from broad experience that no one who is not thoroughly conversant with the finer structures of carchac musculature can distinguish between a few cells of the bundle and those of ordinary cardiac musculature. This is especiallj' true of the human heart.
With Zenker's fixation and hematoxylin and eosin stain, or hematoxylin and van Gieson, there is one characteristic common to the entire system, and that is a clear perinuclear space. The nature of this perinuclear space will be discussed later.
Let us look more closely into the liistology and cytolog}^ of the conductive system in the pig's heart. For descriptive purpose we divide the system into five parts: the atrial part, the junctional part, the ventricular stems, the preterminal part, and the terminal part or Purkinje fibers. There is a gradual transition from one part to the other, but the difference between the first and last is verj^ great.
The first part begins in the region of the great veins and ends in the node of Tawara. It is most difficult to tell where this musculature begins and where the atrial musculature ends, but if one looks underneath the endocardium or underneath the intima of the great veins one will see the clearlj^ defined cells with a perinuclear space. The fibrils are not as numerous nor is the striation of them as marked as in the atrial musculature.
It is a matter of opinion whether these cells around the great veins belong to the sino-ventricular bundle or to the system belonging to the Keith-Flack node. It makes but little difference, because one system is connected with the other and it becomes more a question as to where one ends and the other begins. It is my opinion that we can not make any anatomical distinction between the Keith-Flack system on the one hand and the sino-ventricular on the other. Physiologists may find an easier explanation of cardiac phenomena by considering these systems separate and distinct, but we have no anatomical ])r()of that such a condition exists.
The second or junctional part is what is known as Tawara's node. In most animals this node is situated in a restricted area between the trigonum fibrosum and the endocardium in the right side of the interatrial sej^tum immediately above the interventricular septum. In most mammals (including man) the trigone above mentioned contains either bone or cartilage. It is owing to this fact that thin coronal sections of this region are most (lillimlt to ol)taiii. The necessity of using an acid fixative in order to decalcify and soften the tissue prohibits the use of finer cytologic stains. It is therefore very fortunate that in the pig this node is not restricted to this limited area but streams down into the ventricles, accompanying and gradually merging into the third part.
In figure 1 this area is shown. The section is taken parallel to the ventricular septum and lies a little below the atrial junction. At a is shown the ventricular musculature with many nuclei in the same fiber; betw^een a and b is loose connective tissue which forms Curran's (1910) bursa and which can readily be injected; b shows the third jjurt and c the second part. It will be noted that it consists of an intricate interlacement of fibrils that show but Httle cross striation. These fibrils constitute a syncytium that incloses clear areas containing nuclei. It will be noted that there are generally two nuclei in the same area. In several places, where but one nucleus is drawn, another one can be found in a lower level of the section. Again the raembrane of one nucleus is directly continuous with that of another. Unquestionably we are dealing here with amitotic nuclear division.
The transition from the second to the third part of the sj'stem is greater than between any other two parts. The third part or ventricular stem (6) shows an enormous increase in the number of fibrils. It is difficult to determine whether this increase has been brought about by a splitting of the fibrils or whether it has developed from a differentiation of the cytoplasm. Each nucleus is now surrounded by its own perinuclear space. The bundle is here divided up into a number of strands by Jhe invasion of delicate coUagenic fibers and j^resumably nerve-fibers. It seems that we are dealing here also with a sj-ncj'tium, as far as the fibrils are concerned. Nevertheless, there is a definite condensation of fibrils midway between the nuclei. This gives it an appearance that would suggest cellular deUm- itations. The question whether the fibrils would then be considered as exoplasm or as extracellular deposits hes without the scope of this paper.
In the fourth or preterminal part (see fig. 2) we find the cell area increased in size and two nuclei are found to be situated in the same perinuclear space, which again indicates a multiplication of nuclei, as it does in the second part. The fibrils are less numerous in a given area than thej'^ are in the second part of the bundle. We find a considerable increase in the amount of collagenic fibers, which we recognize in the gross as the sheath of the bundle.
In the fifth or terminal part (see fig. 3), which constitutes the Purkinje fibers proper, we find that the nuclei have moved to the border of the cell area. Each one is surrounded by its own perinuclear space. In the pig we find as a rule four nuclei for one cell area. The striation of the fibrils is a little more marked than it is in the other parts of the bundle, but not as much as in cardiac muscle proper. The collagenic fibers are restricted to the border. A few delicate strands can be seen which delimit the cell areas. In various parts (shown at d) of this Purkinje system we see transition to the cardiac musculature; tliis transition consists in an increase and a more longitudinal arrangement of the fibrils and a further multiplication of the number of nuclei. In the pig tliis nuclear multiplication brings about rows of nuclei, as many as eight, that are situated in the central core of the fiber.
The description just given applies only to the pig's heart. Similar conditions will be found in other Ungulata, but in the heart of man and of Carnivora one would have difficulty in recognizing the cellular multiphcations just described. The first thing wc have to consider is that in the Carnivora there are but two nuclei to a hoart-muscle cell, wliile in th(; pig there are eight. In the production of an adult human carchac muscle-fiber we would see but one-fourth as many nuclear divisions as we do in the pig. These nuclear cUvisions seem to be restricted to the first and second part of the bundle and to the Purkinje fibers. The appearance in the human is, therefore, not as striking as it is in the Herbivora.
The cytologic details of the various parts of the bundle do not differ as markedly as one might expect from the low-power examination. As in cardiac muscle, we must distinguish three parts of the cell: the fibrils, the sarcoplasm around the fibrils or sarcostyle, and the perinuclear sarcoplasms.
It has been previously pointed out that it is not difficult to trace the bundle from Tawara's node to the Purkinje fibers in sections that have been stained with hematoxj'lin and van Gieson stain, or with Mallory's connective-tissue stain, because these stains bring out the fibrous investment of the bundle very clearly. One of these stains has been used by every pathologist who has reported on the histological findings of Stokes-Adams disease. Many will admit that they would be absolutely at a loss if they were to use a general stain, such as hematoxylin and eosin. Further, it is evident that thick sections studied with the low powers of the microscope make tracing of the bundle easier than thin sections with oil immersion. The reason Hes in the fact that the differences, in the human heart especially, between the bundle cells and normal cardiac muscle cells are so slight that they become evident only when seen en masse. The optical cUfference is primarilj- due to a proportionately larger amount of sarcoplasm in the bundle fibers than there is in the musculature. Sarcoplasm does not stain as heavily as do fibrils, with the result that the bundle always stands out in thick sections as a hght-colored area. It is obvious that the thinner the section the more do these color differences disappear.
I have endeavored to analyze the chemical nature and the cytologic details of the sarcoplasm because I felt that an analysis of tliis cell element would give us a clue as to the nature of the bundle. I attempted to determine the glycogen, lipoid, and protein content and the mitochondria. I regret to say that I was only partly successful. As regards the glycogen content, we confirm the work of Aschoff (1908), who finds but very small amounts in the pig's heart. It should be noted, however, that in the heart of beef and sheep the glycogen content far exceeds that of the ventricular musculature. Indeed, it is not difficult to trace the course of the bundle by means of the glycogen reaction. The reason the pig's heart has so little is due to the fact that the nodal tissue which is restricted to the second part of the bundle extends far into the ventricles. The sarcoplasm is much reduced. The results in the heart of the rat were negative, but as only two hearts were examined I do not consider them conclusive. The preponderance of evidence (see results and discussions by Engel, 1910, and Aschoff, 190S) hes in favor of the view that the bundle-fibers contain a larger amount of glycogen than do cardiac fibers. But the ratio seems to be in the same proportion as the amount of sarcoplasm.
A study of the lipoid, protein, and mitochondria content can not be dissociated. Mitochondria or chondriosomes are morphological constituents of the cell, of whose chemical nature we know but httle. The lipoid elements may appear in the form of granules, hposomes, or may be attached to or a part of the mitochon- dria. As to the protein, myogen, and myosin, we may say that it can be either in soluble form and optically inactive or attached to a part of some of the morphologi- cal constituents which have been lumped under the term interstitial granules. We must make a distinction, however, between the granules which are situated between the fibrils or sarcostyles and those that are found in the perinuclear space. My work has led me to a closer analysis of these structures, but I can not say that the results are conclusive. I therefore make the following statements with some reservation :
Some interfibrillar granules seem to bear the same relation to the fibrils, in regard to both size and structure, in the bundle fibers as they do in cardiac muscle. I am inclined to consider them as contraction jjhases in Holmgren's (1910) sense. There are others which have been called mitochondria, but as they do not give all the mitochondrial reactions I am doubtful of their nature. ]\lironesco (1898) has found, however, that a large amount of mitochondria are present in the Purkinje fibers; he concludes that the latter represent a reserve material for cardiac mus- culature. Duesberg (1910) has showii (and so far his work has not been disproved) that myofibrils arise from mitochondria or chondriosomes. The perinuclear gran- ules, however, are both larger and more numerous in the bundle, especially in the terminal portion, than they are in cardiac muscle. Some of these granules give the glycogen reaction, while others show that they must be Upoid in nature.
The object of this paper is to give a physiological interpretation of the facts that are brought forth by careful anatomical observations. We realize that to a great extent the study of function belongs to the physiological laboratory; but when we reahze that in former years the physiologists were also the histologists, and that furthermore much of our early physiological knowledge was derived purely from microscopical stucUes, it may not be amiss to agtiin use our anatomical knowledge for the explanation of physiological phenomena. In doing so we fully reaUze that it is speculative, but it has its justification if it can prove of value in directing the efforts of the experimentalist.
McCallum (1898) stated that the ultimate number of muscle-fibers in volun- tai-y muscles is laid down at bu-th. Increase in the amount of musculature after tliis time is due to the increase in the number of fibrils. The same was held to be true for the muscles of the heart. It seems inconceivable to me that muscle-cells should differ fundamentally from all other cells which have even a less constant activity. W^e know that gland cells are replaced. Why should not muscle-fibers be replaced? This question seems especially pertinent in reference to the heart.
Can this organ, which has a continuous activity, contract every minute of the day and night from several months before birth until maturity and death, without its constituent elements being replaced or rejuvenated? We find nothing in the protoplasm of musculature that seems fundamentally different from the protoplasm of any other cells, nor does chemistry reveal any reason why muscle- cells should live far beyond the life period of other cells. It is true that we have not observed tlie constant degeneration and regeneration that is so evident in many other tissues, but I take the view that they are nevertheless there and can be seen and their presence demonstrated if proper cytologic methods can be found.
In many organs of lesser activity we have found evidence of such replacement, and no one to-day can fail to see them in epithehal structures. In these cases we have been led to the interpretation of cell-replacement by the fact that we fre- quently recognize that one type resembles closely the cells that predominate in the embryo. In certain instances we may see distinct evidence of the degeneration of the superficial cell layer or of individual cells, and then again these uiSLy become evident onlj' in pathological conditions. In the heart we likewise have two or more types of cells, one of which, the cells of the atrioventricular bundle, resembles the embrj'onic heart-muscle cell. To argue that Purkinje fibers are already differen- tiated in the embryo and do not present in the adult morphological features identical with those found in the embryonic state is puerile. We have no difficulty in recog- nizing as distinctive the various layers of embryonic squamous epithehum, and no one would mistake them for adult; yet no one doubts that the superficial layer in the adult is continually being replaced by the cells beneath — which, though not embryonic in appearance, yet divide and multiply as the embryonic cells. The same structures which in the embryo grow and multiply by a process of karyo- kinesis, in the adult absolutely fail to show such evidences of multiplication. Simi- larly, as Bensley (1911) has shown in the case of the pancreas, the duct-cells repre- sent the multipotential elements, which are embryonic in everj' sense but their appearance, yet nevertheless produce both islet and acinus cells. In other words, if we argue that Purkinje fibers can not be embryonic in type because they are morphologically unlike the embryonic cells, we must Ukewise concede that because the deeper layers of stratified epithelium are not embryonic in api^earance they can not replace the superficial, and that duct-cells can not rej^lace islet cells and acinus cells because they arc unlike them.
As regards the degeneration process, I must state that 1 lia\e no evidence to present. I have looked for proof of my assertion, but the same difficulty presents itself that is found in studying the origin of myofibrils. After cmplojdng methods such as arc used for fixing and staining mitochondria, I find in the embryonic mus- culature rods that have a definite color reaction. Suddenly we have fibrils that have a slightlj^ different tinge. If these fibrils are derived from these rods a certain chemical or ph)\sical change must have taken place that brings about a change in color reaction. Likewise we would expect a similar change of reaction when the myofibrils break down. A single degenerated muscle-fiber is difficult to distinguish. W'g know that in certain tyi)es of degeneration the first sign of degeneracy is the disappearance of isotropic and anisotropic bands and a reduction in the amount of sarcoplasm. The nuclei appear shrunken. The appearance is not at all unhke a strand of collagenic tissue and undoubtedly many such degenerated muscle-cells have been mistaken for connective tissue.
It is therefore our opinion that the sino-ventricular bundle is to cardiac muscle what the deep layer is to the superficial in transitional epithelium. It represents the center of growth of heart muscle. It grows from the sinus region, the region which is the most primitive of the heart, and by a series of cell divisions reaches the ventricles of the heart to replace the worn-out heart-muscle cells. That is its essential function. When we study the development of the heart (cf. Mall, 1915) we readily see that the course of the bundle is predetermined by developmental changes taking place in the primitive heart tube. In other words, the bundle takes the shortest course possible between the sinus region and the ventricles of the heart.
It has been pointed out that nerve fibers always accompany the bundle, and thus far it has not been determined whether this accompaniment is due to phj^si- ological association or is merely incidental. In the earlj- embrj^os gangUon cells ai'e distinguishable in the atrial septum in the immediate \dcinity of the bundle, and it is but natural that their processes should take the same course as do the bundle-fibers, because during development nerves always take the shortest course to their sites of innervation. It is, however, significant that Wilson (1909), Morrison (1912), and others have shown nerve terminations on the bundle. Basing it upon their observations, there are some who are therefore inclined to argue that the bundle is a specialized neuromuscular mechanism; but as the natm-e of these ter- minations is not known and as there are likewise similar nerve terminations to be found in other parts of the heart, it seems unwarranted to assume that there exists a physiological difference between them. It is our opinion that these nerve terminations have their origin in postganglionic neurones and most Ukelj^ form a synapse with the vagus terminations directly or through the intervention of another neurone.
Most experimentalists agree that after section of the conductive sj^stem stimulation of the vagus -ndU not inhibit the ventricular beat. This bears out the anatomical observation, namely, that the bundle is sinus tissue and naturallj' in its growth down to the ventricles brings with it the nerve that is associated with it— viz., the vagus. I do not mean to imply that in these experiments vagus fibers were necessarily cut. If the conduction passes by way of the sino-ventricular bundle the very origin of the latter will explain the reason of the lack of response to vagus stimulation upon the section of the area in question. The accelerator fibers follow the course of the coronary vessels and are not anatomically associated with the bundle.
The work of Burrows (1912) dispels any doubt that may have existed that embryonic mammaUan heart-muscle has the power of rhythmic contraction. Em- bryonic tissues have, however, greater potentiality than adult structures and it is quite conceivable that in the adult the muscle-tissue has become so speciahzed that it retains its power of contraction but has lost its power of conduction. If this is the case, conduction takes place by nervous pathways, but it should be remembered that the sino-vcntricular bundle is not adult muscle. It should be noted in this connection that as long as there is a complete muscular continuity between the atria and ventricles in the embryo, gangUon cells are not visible in the heart; but ganglion cells appear with the invasion of the atrioventricular ring by connective tissue and the formation of the valves. In other words, before the definitive physi- ologic reactions take place there are important morphologic changes, namely, the appearance of cross-striations, of ganglion cells, and of nerves.
Attention should again be called to the accessory muscular connections that are found between the right atrium and ventricles. No exi)erimentalist should leave these out of consideration when interpreting the results of his findings. It belongs to future anatomical researches to establish whether these connections also have their origin in the sinus region of the heart or whether they are truly atrioventricular fibers.
To sum up, therefore, we have in the sino-ventricular bundle primarily a growth center for the replacement of cardiac musculature. The cells begin their growth in the sinus region of the heart and by a series of direct nuclear division eventually reach the subendocardial musculature where they jjass into cardiac- muscle cells. Being embryonic in type they may retain their embryonic bipo- tentiality of contractihty and conductivity. The nerves which accompany the bundle are associated with the vagus nerve as is the sinus tissue which it innervates.
Aagaard, O. C. and H. C. Hall, 1915. (Iber Injcktionen des "ReizleitungSisystem" und dor Lymphgffasse des Saugetierherzcns. Anat. Heftc, 1 Abt., Bd. 51, p. 357-427.
AscHOFF, L., 1908. tlber den Glycogengchalt des Reizlcitungssystems des Saugetierherzcns. Vcr- handl. d. deutsch. pathol. Gesellsoh., p. 1.50-153.
Bensbly, R. R., 1911. Studies on the pancreiis of guinea, pigs. Amer. Jour. Anat., vol. 12, p. 297-388.
Burrows, M. T., 1912. Rythmical activity of isolated heart muscle cells in vitro. Science, vol. 36, p. 90-92.
Cdhran, E. J., 1910. \ constant bursa in relation with the bundle of His; with studies of auricular con- nections of the biuidlc. Anat. \nv.., Hd. 35, p. 89.
Di;e8BERG, J., 1910. liCS chondriosomes des cellules cmbryonnaire du poulet, et le\a r61e dans la gen6.se des myofibrils, avec quelques observa- tions sur Ic d(5veloppement des fibres musculaires strides. Arch. f. Zellf., Bd. 4, p. 602-671.
Enoel, Im(;ard, 1910. Hcitrilge zvu' nonnalcn uud patlio- logisrhen Histologic des Atriov(ntri(ularl>\aiil(ls. Zieglcr's Heitr. z. path. Anat., H.l. 4s, |) I'.l'.t 525.
His, Wiliiklm, .In., 1S93. Die Tatigkcit des ernhry- onalen Herzcns und diren licdcutung fiu" die Ivchre von der Uerzbewt.'gung beim Erwachsencn. Arbeiten ausder med. Klinik zu Leipzig, p. 14-49
HoLL, M., 1912. Makroskopischc Darstellung des atrio- ventrikularcn Verbindungsbiindels am men- schlichcn unci tierischcn Herzen. Arch. f. Anat. u. I'hys., Anat. Abt., p. 02-93.
HoLMGRF.N, Iv, 1910. Untcrsuchuugcn iiber die mor- phologisch nachweisbaren stofllich<>n llniset- zuDgcD der quergestreiftcn MuHkclfasem. Arch, f. mikr. Anat., Bd. 75, p. 240-337.
Kent, A. F. S., 1893. Researches on the structure and function of the mammalian heart. Jour. Physiol., vol. 14, p. 233-254.
1913. Observations on the auriculo-ventricular junction of the mammaUan heart. Quart. J. Expcr. Physiol., vol. 7, p. 193-195.
MacCallum, J. ii., 1898. On the histogenesis of the striated muscle fibre and the growth of the human sartorius muscle. Johns Hopkins Hosp. Bull, No. 9, p. 208-215.
Mall, F. P., 1915. Development of the heart. Refer- ence Handbook of Med. Sciences, (Wood & Co., 3 ed.), vol. 5, p. ,58-64.
MiRONESco, Theodore, 1912. IjC chrondriome du r6seau de Purkinjc du ccxnir. C. rend. Soc. dc Biol., t. 72, p. 30-31.
Morrison, Alexander, 1912. On the innervation of the sino-aiu-icular bundle (Kent-His). Jour. Anat. and Physiol, vol. 7, p. 319-327.
Paladino, Giov., 1914. .Vncora per una que.stione di prioritii a proposito del fascio atrio-ventricuhirc del cuore. Anat. Anz., Hd. 46, p. 90-94.
Retzer, Rohert, 1908. Some results of recent investi- gations on the manmialian heart. Anat. Rcc., vol. 2, p. 149-1.55.
Tawara, S., 1906. Das Reizleitungssystem des Siiu- gethierherzens. Jena, G. Fischer.
Wilson, J. Gohdon, 1909. The nerves of the atrio- ventricular bundle. I'roc. Royal Soc, scr. B, vol 81, p. 151.
Fig. 1. This shows the second (r) and third (b) parts of the sinovcntricular bundle invading the ventricular mus- culature (a). The upper part of the figure points toward the apex of the heart. Between the cardiac muscle (a) and the bundle (6, c) is seen Curran's sheath.
Fig. 2. The preterminal part of the bundle is here shown, lying in a dense mass of fibrous tissue. The size of the cells should be compared with the ventricular musculature.
Fig. 3 This illustrates the terminal part of the sino-ventricular bundle or the Purkmje fibers (6) and a transition to the cardiac muscl- (a) shown at (rf). C is endocardium.
The drawings were all made from the same series. The sections are from an adult pig s septum fi-xcd with leUysniczky's fluid and stained with hematoxylin and eosin. Enlargement 210 X.