The Gross and Minute Anatomy of the Heart of the Lizard, Leiolopisma grande (Gray)

Transactions and Proceedings of the Royal Society of New Zealand, Volume 84, 1956-57, Page 103

The Gross and Minute Anatomy of the Heart of the Lizard, Leiolopisma grande (Gray) By J. G. Buchanan Anatomy Department, University of Otago, Dunedin, New Zealand [Received by the Editor, June 23, 1955.] Abstract The structure of the heart of Leiolopisma grande has been described in detail. No Purkinje fibres or nodal fibres similar to those in birds and mammals are present anywhere in the heart, although there is direct muscular continuity without specialisation of the fibres at the S-A and A-V junctions. The glycogen content of the musculature of the various cardiac chambers varies, but no fibres are particularly conspicuous in glycogen content. It has been suggested that the conducting systems of birds and mammals are remnants of more extensive tissues of similar structure in lower vertebrates (Keith and Flack, 1907). My findings, however, accord with those of Davies and Francis, that lower vertebrates do not possess widespread collections of cells similar in structure to those of the conducting systems of the higher vertebrates, for in the heart of Leiolopisma no such tissue is demonstrable. Introduction Although the general structure of the lacertilian heart as revealed by gross dissection has been long known (Owen, 1866; Huxley, 1871; Wiedersheim, 1886; and more recently Meinertz, 1952) the study of its detailed anatomy as disclosed by histological methods and serial sectioning is of comparatively recent date, and, furthermore, investigations of this kind have been carried out on relatively few species and genera, such as Lacerta viridis and muralis (Külbs and Lange, 1911), Lacerta agilis (Krause, 1922), Tiliqua scincoides (Rau, 1924), Uromastix hardwickii (Bhatia, 1929), Hemidactylus flaviviridis (Mahendra, 1942) and Varanus monitor (Mathur, 1944). Much of the interest of these investigations and the discussion concerning them centres around the subdivision of the ventricular cavity and the precise positions of the various orifices, especially the arterial ones. Here Mahendra's work (1942) is of particular interest because his subdivision of the ventricle provides a convenient basis for its description in the Sauria, and thus more readily facilitates comparison between the various species.1Only when this paper had been prepared for publication did I have access to the paper by Prakash on the heart of the Indian spiny-tailed lizard, Uromastix hardwickii (Proc. Rajasthan Acad. Sc., 4, 1–12, 1953). In this paper, Prakash has also followed Mahendra's method of description and his conclusions in regard to the gross structure are in substantial agreement with mine. He has not, however, considered the coronary circulation nor the conducting system. It is well-known that in most reptiles (except crocodiles, which have a complete interventricular septum) the ventricular cavity is incompletely divided by a muscular ridge projecting in from the ventral wall. This projection (called “Muskelleiste” by German workers, and “interventricular septum” or “septum ventriculorum” by most English workers) is probably not equivalent to the entire interventricular septum of higher forms, and therefore Mahendra (1942) has preferred to call it the “muscular ridge”. This name, however, has been rather an unfortunate choice, for Rau (1924) in one of his figures, had previously labelled as “muscular ridge” an entirely different projection from the apical wall of the ventricle in some reptiles, at the place of separation of arterial and venous blood.

While the essential structure of the lacertilian heart is now fairly well established, and further study can only hope to reveal differences in detail as between different lizards, the position in regard to the conducting system of the heart is rather different. Although a conducting system of specialised muscle fibres is certainly present in mammalian and avian hearts (in spite of the contention that it is fictitious in certain species—Glomset and Glomset, 1940a, 1940b), there is considerable uncertainty about lower vertebrates (fish, amphibia, reptiles). Some believe that the muscular fibres connecting the various chambers are specialised, that is to say, that they may be clearly distinguished histologically from the rest of the myocardium; and that, in the evolution of the mammalian heart, the nodal tissue at the atrio-ventricular junction, like that at the sinu-atrial junction, has undergone reduction and concentration (Keith and Flack, 1907; Keith and Mackenzie, 1910; Mackenzie, 1913). Others, while agreeing that there is muscular continuity between the chambers of the heart, deny that this muscle shows any structural specialisation, so that “the S-A node, A-V node and A-V bundle of the hearts of mammals cannot be considered as remnants of more extensive tissues of similar structure in a lowly, generalised vertebrate heart like that of the spotted salamander” (Davies and Francis, 1941a, p. 125). According to this view, the conducting systems of birds and mammals are parallel neomorphic developments to be correlated with the more rapid rate of contraction of the hearts in these homoiothermal vertebrates than is the case in poikilothermal vertebrates (fish, amphibia, reptiles). Because birds and mammals have evolved from the great Sauropsidan and Theropsidan branches of reptiles (Goodrich, 1916), the reptiles obviously occupy a vital position in this controversy; for, if the conducting systems of birds and mammals are neomorphic developments with no counterpart in lower forms, they must then have arisen independently in those phyla; but if, on the other hand, they are remnants of a more extensive system in lower forms, we should surely expect to find these specialised tissues particularly well-developed in the reptilian heart. Since Davies and Francis (1946) have fully reviewed the extensive literature on this subject, I need refer here only to that which is relevant to the reptilian heart. Gaskell (1882, 1883) first claimed that in the frog and tortoise, the muscle of the S-A and A-V connexions, in contrast to ordinary cardiac muscle, is of an embryonic character, with a less distinct transverse striation, more sarcoplasm, and a slower rate of conduction to explain the interval between the contractions of the various chambers. This view was later taken up and elaborated by Keith and Mackenzie (1910) and Mackenzie (1913) after the discovery of the mammalian S-A node (Keith and Flack, 1907). They described, in the lizard and tortoise, an almost complete ring of “nodal tissue” around the junction of the sinus and the atrium: a ring of similar tissue was said to occur at the A-V junction. Since then, however, others have had great difficulty confirming these observations—many have found no obvious differences in the histological structure of the muscle at the S-A junction (Külbs and Lange, 1911; Külbs, 1912; Laurens, 1915; Swett, 1923); and at the A-V junction too, although generally workers have been more rectant to deny the presence of specialised muscle here, most of them regard it as much less distinctive than Keith and Mackenzie imply (Külbs and Lange, 1911; Külbs, 1912; Laurens, 1913, 1915). Swett (1923) goes even further and says categorically that in the alligator the A-V connexion is of ordinary cardiac musculature. More recently, Davies et al. (1952) have also been quite unequivocal on this matter, and have stated that there is no specialised nodal tissue or Purkinje fibres in any part of the heart of the alligator or crocodile—the conducting system, they say, appears as a structural entity only in birds and mammals. Nevertheless, Robb (1953) has recently questioned this by claiming that, in the turtle, “Purkinje-like tissue extends in scattered strands from the sinus venosus throughout the atria, forms a considerable portion of the A-V funnel and is distributed widely (deeply as well as subendocardially) within the ventricle” (p. 13).

Fig. 15.—A sagittal section (X 80) of the A-V junction, just to the right of the intriatrial septum. The section is viewed from the left. The atrial musculature is invaginated into the base of the ventricle as the A-V funnel. The epicardial tissue (Ep) accompanies this invagination for some distance, completely separating the atrial from the ventricular musculature. More caudally, however, these are seen to become directly continuous. This section also shows the right septal cusp (R. S. C.) of the bell-shaped valve with its concavity directed towards the ventricle and its caudal margin attached to the ventricular wall both ventrally and dorsally. Fig. 16.—Ventral view of the hear of Letolopisma grande in situ. Note the sac-like diverticulars (S. D.) of the right atrium projecting between the diverging carotid arteries. Fig. 17.—A wax-plate reconstruction of the sinu-atrial junctional region, from the ventral aspect. The ventral wall of the sinus venosus (S. V.) and the dorsal wall of the right atrium (R. A.) can be seen. The rostral (R. S. V.) and caudal (C. S. V.) cusps of the S. A. valve project into the right atrium. At its left-hand end the rostral cusp continues on to the dorsal part of the interatrial septum (I. S.). A white arrow passes through the opening of the pulmonary vein (P. V.) into the left atrium.

Figs. 18 21 are thick hand-cut transverse sections of the ventricle seen from above. The sections proceed in a caudal direction. × 16. Fig. 18.—Shows the muscular ridge separating the large cavum dorsale from the small cavum pulmonale (into which a hair has been inserted). Fig. 19.—This section includes a portion of both the muscular and apical ridges. The muscular ridge again separates the cavum pulmonale ventrally and to the right (with a hair inserted in it) from the cavum dorsale dorsally. The apical ridge can be seen running dorso-ventrally across the cavum dorsale, dividing it into the cavum venosum on the right, and cavum arteriosum on the left. Fig. 20.—The muscular ridge is not present at this level, but the apical ridge is prominent. Fig. 21.—Apex of the ventricle. Here muscular trabeculae divide the ventricular cavity into a number of regular spaces into which hairs have been inserted. Fig. 22.—Photomicrograph of ventricular muscle to show the characteristic appearance of the bulged portions of the fibres when these are cut in transverse section Azan × 600. Fig. 23.—Photomicrograph of a nest of epitheloid cells situated between the right and left systemic arteries in the adventitia of the truncus. The actual paraganglion cells may be distinguished by their distinct nucleoli (as indicated by arrow). Three fat cells are seen in the cell-nest in this section H. P. F. × 615. Fig. 24.—Subepicardial vein communicating with intertrabecula space H. P. F. × 290.

Fig. 25.—Photomicrograph of ventricular muscle showing the bulgings of the fibres around certain nuclei. It can be seen that these bulgings are quite localised and are not present along the entire length of the fibre. Azan. × 600. Fig. 26.—Photomicrograph of atrial muscle also showing occasional bulging of the fibres. Azan. × 650. Fig. 27.—Photomicrograph of ventricular muscle showing small masses of glycogen (arrows) within the individual fibres. Haematoxylin and Best's carmine × 720.

In an attempt to clarify this issue, I have made a detailed study, both gross and microscopic, of the heart of Leiolopisma grande, in which I have also studied the distribution of the coronary arteries, and the cardiac paraganglion cells which Trinci (1912) and Palme (1934) have described in other reptiles in the adventitia of the arteries of the truncus. Material and Methods Numerous specimens of Leiolopisma grande were used. This “Rock Lizard” as it is commonly known, is a member of the family Scincidae found in the southern part of the South Island of New Zealand. The average length of the adult is about 8 inches. The specimens were procured for me through the kindness of Mr. L. O. Simpson, of the Zoology Department, University of Otago. The hearts were studied both by gross dissection and histologically. In the former case, fresh or formalin-fixed material was dissected under the binocular microscope. For study of the internal anatomy, particularly of the ventricle, the heart was first washed out with 1% potassium chloride which causes increasing relaxation of the heart and finally its arrest in complete diastole. This greatly aids dissection of the ventricle, for when contracted, owing to the very spongy nature of its walls, its cavity is largely obliterated. In a heart thus relaxed, the general arrangement of the various septa projecting into the ventricular cavity can be easily determined by making about four thick transverse sections of the ventricle. For histological study, the hearts (except where used to show glycogen) were prepared as follows: The animals having been killed with ether, the heart (still beating) was perfused with Ringer-Locke solution through the postcaval vein, after which Bouin's or Zenker's fluid was injected, causing the heart to stop beating almost immediately. The heart and part of the trachea and lungs (to preserve the sinus venosus and the great veins), was then carefully removed and placed in fixative. After embedding in paraffin, three hearts were sectioned serially, one each in the transverse, frontal and sagittal planes. The sagittal sections were cut at 7μ, the others at 10μ; all were stained with iron haematoxylin and picro-fuchsin. A fourth was cut transversely at 10μ and the sections stained either with haematoxylin and eosin, or azan. For studying the distribution of glycogen, the procedure was varied so as to avoid its loss. The animal was quickly decapitated with a pair of serrated-edged scissors. The heart was then rapidly removed and placed, still beating, into ice-cold Rossman's fluid; it was then embedded in paraffin as usual and serially sectioned at 10μ in a frontal plane. The sections were stained by Best's carmine method, controlled by the saliva test and from liver sections stained at the same time. The distribution of glycogen in the musculature of the various cardiac chambers was then roughly estimated by the graphical method of Davies and Francis (1941b). To clarify the complex anatomy of the S-A orifice and valves, a wax-plate reconstruction of this region (X 75) was made from the first transverse series. Results Gross Anatomy In Leiolopisma grande the heart, in the pericardium, lies well forward in the pleuro-peritoneal cavity, almost in the midline. It is about 9 mm long and 7 mm wide with the atria distended. The sinus venosus. The sinus venosus as usual is a distinct thin-walled muscular chamber, dorsal to the right atrium and just in front of the ventricle. It is constricted at its middle into a larger right part, which receives the right precaval and postcaval veins, and a smaller left portion, receiving the left precaval vein at the level of the left extremity of the S-A aperture. The constriction is due to the invagination dorsally

Fig. 1.—A transverse section through sinus venosus and atria at the level of the pulmonary opening. (Figs. 1–8 and 11–13 have been drawn from projections of sections selected from the transverse and frontal series. Each of the various tissues is indicated in the same way throughout, according to the scheme in this figure). Fig. 2.—A transverse section through the heart, just caudal to the S-A orifice. Fig. 3.—A transverse section through the rostral part of the atria. Note the sac-like diverticulum (S.D.) of the right atrium, and dividing pulmonary artery.

and posteriorly of the sinus-musculature (which is thickened here) by a mass of connective tissue, which continues to the base of the ventricle as the so-called “dorsal ligament” or “sinu-ventricular fold”; in Leiolopisma this contains the coronary vein and numerous nerve-cells and fibres, but no muscle fibres. The coronary vein is single and opens posteriorly into the ventral wall of the sinus opposite the constriction and below the S-A aperture. The sinu-atrial orifice, somewhat crescentic and almost as wide as the sinus, is disposed rather obliquely (more transversely than longitudinally), with its right extremity the more rostral. It has two obvious but thin muscular cusps forming the sinu-atrial valve; whereas at the left these cusps are separate, at the right they join a small muscular mass (“suspending ligament” of Mathur, 1944) which attaches them to the dorsal wall of the right atrium. The caudal (or right) cusp is bilaminar, containing reflexions of both sinus and atrial musculature, which, although clearly separated at the base of the cusp by loose fibrous connective tissue, become directly continuous nearer the free margin (Fig. 1); thus here the sinu-atrial connexion is at the free border of the valve. The rostral (or left) cusp on the other hand, is almost entirely of atrial myocardium since the sinus and atrial musculature join at its base; at the very left, this cusp is directly continuous with the dorsal part of the interatrial septum and here separated from the dorsal atrial wall by a small fibrous mass containing some nerve cells (Fig. 2). The atria. The thin-walled atrial portion of the heart is concave ventrally and lodges the roots of the great arteries as they emerge from the ventricle (Fig. 1). Of the two atria, which are of course completely separated by the interatrial septum, the right is slightly more ventral. Medially, both atria are invaginated funnel-wise into the ventricle, so that the true A-V orifices actually lie within the ventricle. Near this funnel at the postero-medial part of each atrium, the atrial wall becomes much thicker and, unlike the rest of the atrium, is devoid of trabeculae. The interatrial septum is also quite smooth. The right atrium, the larger, is quite typical except for a small but constant sac-like diverticulum (Fig. 3) which often projects well forwards between the diverging bases of the carotid arteries; it was very prominent when the atria were fully distended with blood (Pl. 17, Fig. 16). The left atrium receives the common pulmonary vein, which opens well up the dorsal wall of the atrium, beside the interatrial septum (Fig. 1). Although there is no well-marked valve here, there is a very slight, membranous outgrowth at the caudal lip of the opening where the ventral wall of the vein joins the dorsal wall of the atrium; furthermore, the atrial musculature is slightly thickened around the orifice, which may help to close it during auricular systole. The interatrial septum is thin and muscular. In section it is usually markedly folded, but these folds disappear when the atria are turgid with blood (Mathur, List of Abbreviations in Figures and Plates. A.P.S., Aortico-pulmonary septum. Ao.V., Aortic vestibule. C.A., Cavum arteriosum. C.D., Cavum dorsale. C.P., Cavum pulmonale. C.S.V., Caudal cusp of S-A valve. C. V., Cavum venosum. Ca. P., Carotis primaria. Co.A., Coronary artery. D.L., Dorsal cusp of left systemic valve. D.P., Dorsal cusp of pulmonary valve. D.R., Dorsal cusp of right systemic valve. Ep., Epicardium. I.S., Interatrial septum. L.A, Left atrium. L.B.S., Left bulbo-ventricular sulcus. L.M.C., Marginal cusp of left A-V valve. L.P.A, Left pulmonary artery. L.P.C., Left precaval vein. L.S., Left systemic artery. L.S.C., Septal cusp of left A-V valve. M.R., Muscular ridge. P.A., Pulmonary artery. P.C.V., Postcaval vein. P.T.A., Post-truncal artery. P.V., Pulmonary vein. R.A., Right atrium. R.B.S., Right bulbo-ventricular sulcus. R.M.C., Marginal cusp of right A-V valve. R.P.A., Right pulmonary artery. R.S., Right systemic artery. R.S.C., Septal cusp of right A-V valve. R.S.V., Rostral cusp of S-A valve. S.D., Sac-like diverticulum of right atrium. S.V, Sinus venosus TR., Trachea. V.Co.A., Ventral coronary artery. V.L., Ventral cusp of left systemic valve. V.P., Ventral cusp of pulmonary valve. V.R., Ventral cusp of right systemic valve. X, Vagus nerve.

1944). Rostrally the septum is particularly thin, while caudally it is much thicker, and continues here into the A-V funnel, where it has attached to it the septal cusps of the A-V valves. The atrio-ventricular junction, as stated, is complicated by the funnel-like invagination of both atria into the ventricle. Epicardial tissue accompanies this invagination for a considerable distance, to separate the atrial and ventricular musculature (Pl. 17, Fig. 15). The muscle fibres of the A-V funnel run largely circularly, particularly at the base of the funnel where the atrial wall is thickened and there is an abrupt change from the trabecular arrangement found elsewhere in the atria, this forms what is generally known as the A-V ring. At the caudal end of the funnel, the atrial and ventricular musculatures are freely continuous, although ventrally and to the right, near the origins of the great vessels, they remain separated by fibrous tissue. The interatrial septum and the attached valves divide the A-V funnel into two channels, one from each atrium—the “atrium dextrum intraventriculare” and the “atrium sinistrum intraventriculare” (Mathur, 1944). The atrio-ventricular valves guard the right and left A-V openings. In each there is a large medial (septal) cusp and a very much smaller lateral (marginal) cusp, neither of which present chordae tendineae. The two septal cusps are entirely membranous; together they form a kind of bell with its concavity facing the ventricle. Above, this “bell” is attached to the interatrial septum, while dorsally and ventrally it is joined to the walls of the A-V canal by thin bands of connective tissue, which are themselves continuous rostrally with the interatrial septum. Thus, the “bell” and these two connective tissue bands divide the A-V canal into right and left parts. In transverse section this bell-shaped valve appears as a continuous ring of fibrous tissue which may be arbitrarily divided into two cusps by its dorsal and ventral attachments to the wall of the A-V canal (Fig. 11). More caudally, it becomes flattened dorso-ventrally, and so is more oval in section. The caudal margin of the “bell” although free on each side, is attached quite extensively, ventrally and especially dorsally, to the corresponding wall of the ventricle (Fig. 8). The lateral cusps of the A-V valves, which arise from the respective lateral walls of the A-V canal, are also entirely membranous but are considerably smaller than the septal cusps (Fig. 11). Fig. 4.—A frontal section of the heart through the origin of the pulmonary (P.A.) and left systemic (L.S.) trunks Note the carots primaria (Ca. P.) given off from the right systemic artery (R.S.) Note also the ventral coronary (V. Co. A.) and post-truncal (P. T. A.) arteries lying in the right and left bulbo-ventricular sulci respectively (R.B.S., L.B.S.).

The ventricle. The ventricle, as usual, is conical with its apex caudally. The region at the base from which the three arterial trunks emerge, is derived from the absorbed bulbus cordis, and may be called the bulbar region of the ventricle (Benninghoff, 1933). This region is partially marked off from the rest of the ventricle by two well-defined sulci: the left bulbo-ventricular sulcus (“interventricular sulcus”) which runs down ventrally for a certain distance from the bulbo-auricular infolding; and the right bulbo-ventricular sulcus which commences above on the right of the arterial trunks, and is much shorter than the left (Fig. 4). Following Mahendra (1942), the ventricle may be divided descriptively into four regions which pass gradually and insensibly into one another: (a), the apical region; (b), the region of the muscular ridge; (c), the region of the atrial apertures; and (d), the region of origin of the aortic trunks. The apical region presents a distinct muscular ridge, running dorso-ventrally, which I propose to call the “apical ridge”. In transverse sections near the apex, it is not distinguishable from the numerous other muscular trabeculae which intersect the ventricular cavity here and divide it into a series of irregular spaces (Fig. 5); but, as thick hand-cut sections readily show (Pl. 18, Figs. 18–21) it projects far more rostrally than these trabeculae—indeed about half-way up the ventricular cavity—and thus clearly divides the ventricle in the region of the apex into right and left Fig. 5.—A transverse section through the ventricle near its apex The ventricular cavity is divided by muscular trabeculae into a number of irregular spaces. Fig. 6.—A transverse section through the “apical region” of the ventricle, rostral to Fig. 5. The apical ridge (A.R.) divides the ventricular cavity in this region into the cavum venosum (C.V.) on the right, and the cavum arteriosum (C.A.) on the left. Fig. 7.—A transverse section of the ventricle through the more caudal part of the muscular ridge (M. R.) which separates the cavum pulmonale (C.P.) ventrally and on the right, from the cavum dorsale (C. D.). Fig. 8.—A transverse section of the ventricle through the more rostral part of the muscular ridge (M.R.). This ridge arises in relation with the left bulbo-ventricular (interventricular) sulcus (L.B.S.) and has its free border towards the right. The cavum dorsale, dorsal to the ridge is divided by the septal cusps of the A-V valves into the cavum arteriosum (C.A.) and cavum venosum (C.V.). The cavum venosum and cavum pulmonale (C.P.) communicate around the free border of the ridge.

parts (Fig. 6), freely continuous above its rostral free border. The plane of this ridge is in that of the interatrial septum, so that it projects towards the concavity of the bell-shaped valve. The region of the muscular ridge. On the right of the apical ridge, but in Leiolopisma entirely rostral to it, there springs from the wall of the ventricle a second ridge, which, running obliquely, subdivides the ventricle towards the base in a rather different fashion—for it separates a small ventro-lateral chamber on the right, the cavum pulmonale, from a very much larger compartment, the cavum dorsale, comprising the rest of the ventricular cavity (Figs. 7, 8). The muscular ridge may be said to have an attached and a free border. The attached border first follows along the left bulbo-ventricular (“interventricular”) Fig. 9.—A diagrammatic representation of the muscular ridge. The attached border of the ridge is represented by the broken line. Most of the ventral wall of the cavum pulmonale has been removed to show the muscular ridge which becomes continuous rostrally with the fibrous aortico-pulmonary septum. The black arrow is passing through the opening between cavum venosum and cavum pulmonale. This opening is bounded rostrally by the aortico-pulmonary septum, to the right by the right wall of the ventricle, and caudally and on the left by the free margin of the muscular ridge. The attachment of the dorsal cusp of the pulmonary valve to the lower free border of the aortico-pulmonary septum and rostral-most part of the muscular ridge is indicated by a dotted line. Arrows have been inserted in the pulmonary and left systemic arteries. Fig. 10.—Diagrammatic sagittal sections of the ventricle through the muscular ridge, seen from the right. Section A is lateral to (i.e., to the right of) Section B.

sulcus from the base of the heart, then curves obliquely around the right side of the ventricle, about midway between the base and the apex, finally to turn slightly upwards (Fig. 9). The free border is similarly curved; it faces towards the right and thus forms the left and caudal margin of an opening through which the cavum pulmonale communicates with the rest of the ventricular cavity (cavum dorsale). At the base of the heart the muscular ridge ends abruptly in a fibrous septum (aortico-pulmonary septum) which lies between the pulmonary artery and the “aortic vestibule”, and which extends right across the narrowing ventricular cavity to its right wall. Continuous on the left with the muscular ridge, on the right this septum has a caudal free border forming the rostral boundary of the opening, just described, between the cavum pulmonale and the cavum dorsale (Fig. 9). Thus, the cavum pulmonale is quite a small chamber which is limited below by the attachment of the muscular ridge; it does not therefore reach further than half-way down the ventricle. The cavum dorsale, on the contrary, is very extensive; it is divided, anteriorly, by the septal cusps of the A-V valve and, posteriorly, by the apical ridge, into a cavum venosum on the right and a cavum arteriosum on the left. These communicate with each other over the free border of the apical ridge, and the cavum venosum in turn communicates with the cavum pulmonale around the free border of the muscular ridge (Figs. 8, 9). The region of the atrial orifices is entirely dorso-lateral to the muscular ridge—i.e., in relationship with the cavum dorsale, the right “atrium intraventriculare” opening of the two systemic arteries (Fig. 9, 11). The pulmonary valve has a ventral opening into the cavum venosum and the left into the cavum arteriosum. The region of origin of the aortic trunks is at the base of the ventricle—the so-called “bulbar region”. The trunks—the pulmonary, left systemic and right systemic (systemico-carotid) arteries, are all bound together, where they leave the ventricle, in a common external covering—the epicardial sheath. Each trunk has at its origin a pair of semilunar valves; but these are not all at the same level, for the pulmonary valves are the most posterior, being just caudal to those of the left systemic trunk, while the right systemic semilunars are the most anterior. The pulmonary trunk arises separately from and caudal to the two systemic trunks, as a forward prolongation of the cavum pulmonale. Its dorsal wall is formed at first by fibrous tissue which is continuous, as stated, with the muscular ridge and also passes right across to the right wall of the bulbar portion of the ventricle, to form a fibrous septum (aortico-pulmonary septum) separating the origin of the pulmonary artery from what we may call the “aortic vestibule” leading to the common cusp (more strictly, ventromedial) attached to the ventral wall of the cavum pulmonale (Fig. 8), and a dorsal (more strictly, dorsolateral) cusp attached both to the free margin of the aortico-pulmonary septum, and to the muscular ridge (Fig. 10). Although the margins of the two cusps are at the same level, the attachment of the ventral one extends much further down in the cavum pulmonale than that of the dorsal. The two systemic trunks spring from a common “aortic vestibule”, which is a rostral extension of the cavum venosum (Fig. 11), the left systemic trunk arising somewhat ventrally and to the right of the right one. Like the pulmonary artery, each has at its origin paired semilunar valves, with a right dorsal and a left ventral cusp. The two ventral cusps are continuous with each other (Fig. 11); that of the right systemic artery is attached to the rostral part of the muscular ridge, while that of the left systemic is attached to the lower free border of the aortico-pulmonary septum (Fig. 10). The two dorsal cusps on the contrary are quite separate, and each attaches to the dorsal wall of the aortic vestibule, at the origins of the respective trunks. On leaving the base of the ventricle, the right systemic artery lies dorsal to and between the others, with the left systemic to the right of the pulmonary. The three arteries then wind spirally in the customary manner (Fig. 13), so that the left

Fig. 11.—A transverse section through the ventricle near the base. Note the invagination of the atria into the ventricle as the A-V funnel, and the division of this into atrium dextrum intraventriculare and atrium sinistrum intraventriculare by the bell-shaped valve and the two connective tissue bands. The fibrous aortico-pulmonary septum (A.P.S.) separates the pulmonary artery (P.A.) from the aortic vestibule (Ao. V.). Fig. 12.—A transverse section of the heart at the base of the ventricle. The postcaval vein (P.C.V.) is expanding into the sinus venosus. Note the ventral coronary artery (V.Co.A.) and the two terminal branches of the dorsal coronary artery. Note also the small mass of fibrous tissue which separates the interatrial septum (I.S.) from the dorsal atrial wall. Fig. 13.—Three transverse sections through the truncus to show the relative change in position of the three great arteries The sections are viewed from above; A. is the most rostral, C. the most caudal. systemic trunk is at first ventral then to the left, while the right ends up on the right; the pulmonary trunk eventually lies dorsal to and between the two systemic arteries, where it divides (Fig. 3). The right systemic artery as it arches laterally and dorsally gives origin to a very short, stout “carotis primaria” (Fig. 4) which divides almost at once into right and left common carotid arteries. The Coronary Vessels The Coronary Arteries. Leiolopisma has only one main coronary artery, which arises dorsally from the right systemic artery soon after its origin, and continues at first dorsal to it, running caudally right in the tunica externa (Fig. 2). At the base of the ventricle, where it

is now slightly more to the right, it divides into a ventral and a dorsal branch of approximately equal size. The ventral (coronary) artery runs beneath the epicardium in the right bulbo-ventricular sulcus at first under cover of the ventral part of the right atrium (Fig. 11). It gives off numerous small branches to the ventral surface of the ventricle on its right side, but soon becomes very small and cannot be traced further than half way down the ventricle. The dorsal (coronary) artery, after giving off on the right side, a small branch to the ventricle, divides into two small branches, one of which runs dorsally in the right A-V sulcus and ends by ramifying on the dorsal surface of the ventricle. The other branch forms the so-called “post-truncal branch” (Grant and Regnier, 1926), which passes across to the left between the right systemic artery and the atria, giving off as it does so branches to the wall of the A-V funnel; it then enters the left bulbo-ventricular sulcus, in which it runs caudally, giving off small branches to the bulbus musculature and the base of the ventricle on the left side (Fig. 4). As in other lizards, the ventricular myocardium of Leiolopisma is predominantly spongy (“spongiosa”) with a relatively thin layer of compact muscle superficially (“corticalis”). The blood in the ventricle is thus able to penetrate the wall for a considerable distance in the intertrabecular spaces so that, here, the coronary arteries and their branches lie right on the surface of the heart. However, in the bulbar region, and in the A-V ring where the myocardium is much thicker and more compact, small arteries actually lie within the myocardium. There are no branches of the coronary artery to the atria, apart from the region of the A-V funnel, where the thicker wall receives a small branch from the post-truncal artery. Elsewhere the atrial walls are extremely thin, and probably entirely nourished by the contained blood. The Coronary Veins. Numerous very small veins in the epicardium drain into larger vessels which ascend to the base of the ventricle, where they finally unite to form a single vein which opens into the sinus venosus. The coronary and inter-trabecular systems are not entirely independent, for fine prolongations of the inter-trabecular spaces may pass right through the compact muscle to join small veins on the surface of the heart (Pl. 18, Fig. 24). These communications correspond, of course, to the The besian veins of mammals (Grant and Regnier, 1926). In some reptiles an apical artery and vein pass to the heart in an apical ligament (“gubernaculum cordis”) which connects the apex of the ventricle to the parietal pericardium (Grant and Regnier, 1926)—in Leiolopisma no such apical ligament is present, and hence there are no apical vessels. Histological Results The muscle-fibres of the chambers of the heart are all transversely-striated, although the prominence of this striation is rather variable. The diameters of the fibres and their staining reactions also vary to some extent. Not only are there slight differences between the fibres of the different chambers but there may be a marked variation among the individual fibres of the same chamber, particularly of the ventricle. The nuclei, which usually lie axially within the fibres, can vary greatly in size and shape. The sarcoplasm is always relatively abundant, more so than in mammalian cardiac muscle; so that in transverse section the fibres appear as rounded or irregular areas of various sizes in which the myofibrils are quite obvious as brightly-staining dots. There was no sign of intercalated discs anywhere. In both atria and ventricle, fibres running in the same direction are usually grouped in small bundles, generally without any fibrous tissue around them. No smooth muscle occurs in any part of the heart; nor any cartilage, as is seen in the muscular ridge of Tiliqua (Rau, 1924).

The sinus musculature. Here the fibres, which lie in a relatively large amount of fibrous connective tissue, are thinner than elsewhere, with a fine but distinct transverse striation. They have very numerous, spherical or oval nuclei which, although slightly smaller than those of the atria, are quite large compared with the size of the fibres. The striated sinus-musculature extends a short way along the caval veins, and some distance on the inner aspect of the S-A valves where it eventually becomes directly continuous with the atrial musculature. The atrial musculature (Pl. 19, Fig. 26) is of course markedly trabeculated. Its fibres are rather thicker than those of the sinus, and have a somewhat coarser transverse striation; the oval nuclei may occasionally be so large as to bulge the surface of the fibre. In the interatrial septum the fibres run mainly dorso-ventrally and are embedded in a fair amount of fibrous connective tissue. In the A-V ring where the muscle is more compact, they intervene between the general musculature of the atria and that of the rest of the A-V funnel, where the muscle fibres are still circularly arranged; but, apart from the fact that they may have a slightly finer transverse striation, the fibres here are identical with those elsewhere in the atria. The ventricular musculature (Pl. 19, Fig. 25) is made up of interlacing bundles of anastomosing fibres. Except for a little in the bulbar region, fibrous tissue is negligible in the ventricular myocardium. The fibres—indeed, even different parts of the same fibre—may show wide variations in staining reaction (particularly with azan), ranging from quite pale to deep-red, through all intermediate shades. There is also considerable variation in fibre-diameter, myofibrillar content, coarseness of striation, and size of nuclei. In fibre-diameter and coarseness of striation, the fibres in general resemble the atrial fibres, but their nuclei are much longer and many of them are so large that they produce a striking local bulging of the fibre. These bulgings may indeed be so pronounced as to give to the fibre in regions where they occur, a very characteristic, in fact almost “specialised” appearance: a single large nucleus, or even two in succession, surrounded by an area of clear sarcoplasm, with the myofibrils restricted entirely to the periphery of the fibre. These enlargements are invariably lighter-staining than the rest of the fibre; when cut transversely the nucleus often appears to lie in a clear “hole”, an appearance which may be exaggerated by shrinkage (Pl. 18, Fig. 22). Although not present around all nuclei, these enlargements are common and widespread in the ventricular muscle (Pl. 19, Fig. 25). If this appearance, which here is quite localised and intermittent, were present continuously along the entire length of a fibre, one would have no hesitation in calling such a fibre specialised; but since it is confined purely to the region of the bulgings, and elsewhere the fibres have the typical structure of cylindrical ventricular fibres with distinct transverse striation right across the fibre, they are clearly not to be regarded as specialised in the accepted sense of the term. Similar enlargements also occur, but much less frequently, in atrial muscle fibres and in the A-V funnel. The sinu-atrial and atrio-ventricular junctions. At both junctions there appears to be free and direct muscular continuity between the chambers. Neither at the S-A nor at the A-V junction were any outstanding histological differences evident, between the muscle fibres here and those of the adjacent chambers. Such minor differences as were observed did not at all conform with those which we would expect to find if the fibres concerned were specialised in the way in which they are in mammals and birds. Furthermore, when the heart was examined for the glycogen content of its fibres, which in mammals and birds appears to be greater in those of the Purkinje system (Davies and Francis, 1941b), it was found that there was no comparable distinction in the heart of Leiolopisma In no part of the heart did the fibres show a particularly high glycogen content. It is true, however, that the average percentage of glycogen, as crudely determined by the graphical method, did vary in the different chambers of the heart. It was least in the sinus, then appeared increasingly in the atria, bulbus and ventricle in that order (Pl. 19, Fig. 27). As regards the junctional regions, the glycogen content of the two components of the

S-A valve was the same as that of the corresponding chambers; while in the A-V funnel the glycogen content resembled that of the general atrial musculature. Thus in spite of the gross limitations of this method for determining glycogen, these results support those obtained by ordinary histological methods. At the S-A junction the sinus and atrial musculature are freely continuous with each other, in the case of the caudal (right) S-A valve at its free border, and in the case of the rostral (left) S-A valve towards its base. At the A-V junction also, the fibres of the A-V funnel appear to be directly continuous with those of the ventricle (Pl. 17, Fig. 15). In the A-V funnel, although the muscle fibres largely form a fairly uniform layer running circularly around the funnel, as we approach the actual orifice this regularity is lost, so that on the left, the region of continuity between the atrial and the ventricular muscle may be only one fibre thick, and on the right, in the region of the bulbus, muscle fibres entirely disappear—here fibrous tissue completely separates the atrial from the bulbus musculature. Elsewhere at the A-V orifice, the muscular continuity is four or five cells thick. The Cardiac Paraganglion. In Leiolopisma, as in other lizards, the cardiac paraganglion is not an independent, discrete organ, but is formed of scattered groups or nests of epithelioid cells. These cell nests are irregular in outline, very numerous, and distributed widely in the adventitia along the entire length of the truncus. They occur particularly in the longitudinal sulci between the three arteries of the truncus, just beneath the visceral layer of the pericardium (Pl. 18, Fig. 23), as well as axially, in the adventitia connecting the three vessels. These groups of cells are completely devoid of any capsule of their own, and are often closely related to bundles of nerve-fibres, nerve-cells and fat cells in the adventitia, in which, occasionally, solitary paraganglion cells may occur. The cells of each nest lie in a delicate connective tissue stroma. Numerous nuclei of smaller cells and often fat cells and capillaries lie in the stroma between the cells. The paraganglion cells themselves are usually somewhat polygonal in shape, with pale cytoplasm (“chromophobe”), spherical nucleus, and very distinct nucleolus. They were never seen in the tunica media of the arteries, and no one group of cells appeared to be sufficiently large or constant in position to deserve description as the principal group. Discussion The heart of Leiolopisma is quite typical in structure, and as is usual in the Lacertilia, shows only slight variations from that of other lizards. For instance, S-A valves are present and well-developed in Leiolopisma, although apparently not in all lizards. In Uromastix hardwickii, for example, Bhatia (1929) has observed that the S-A aperture “is not guarded by any valve, but its lips are thick and muscular and are thus kept closed, except when the blood is to be forced through it into the right auricle” (p. 282); and in Hemidactylus flaviviridis, Mahendra (1942) has found a similar condition. However, these appear to be rather exceptional, for there is ample evidence that it is more usual to have distinct muscular valves guarding the S-A orifice in lizards (Külbs and Lange, 1911; Rau, 1924; Mathur, 1944; Meinertz, 1952). A constant feature in Leiolopisma is the presence of a distinctive diverticulum from the right atrium. This has also been described in Sphenodon (O'Donoghue, 1921), in Uromastix (Bhatia, 1929) and in a snake, Ptyas mucosus, in which, however, it arises from the left atrium (Ray, 1934). It is not present in Hemidactylus flaviviridis (Mahendra, 1942) nor in Varanus monitor (Mathur, 1944). It is difficult to see what function it could have. There is no distinct flap or valve at the opening of the pulmonary vein into the left atrium of Leiolopisma, except for a very slight membranous outgrowth at its caudal margin. This accords with the prevailing view that the opening of the pulmonary vein in reptiles is devoid of valves, but this may not be universal, for in

Uromastix hardwickii, apparently, a valve is present, “formed by a forward extension of the dorsal wall of the auricle” (Bhatia, 1929); in Varanus monitor, too, “the crypt containing the pulmonary opening is fairly long and it possesses a flap-like outgrowth on one side” (Mathur, 1944); and in Placovaranus komodoënsis “the ventrolateral wall of the pulmonary vein stands out in the auricle as a sharp fold that presumably functions as a flap-valve during systole of the auricle” (Meinertz, 1952). In Leiolopisma each A-V valve has two cusps, a larger medial (septal) one, and a much smaller lateral (marginal) one. In this respect it resembles Varanus monitor (Mathur, 1944) in which each A-V orifice also has two distinct cusps. However, many accounts of the heart in different lizards describe only a single medial (septal) cusp; but it is probable that there is considerable variation in the size of the lateral cusp in different lizards, and that when very small it has been overlooked. As regards the bell-shaped valve formed by the union of the two septal cusps, it is of interest to note that the distinction into right and left cusps can only be made rather arbitrarily in sections of the heart; Krause's error (1922) in describing dorsal and ventral cusps is therefore perhaps excusable. The differentiation and naming of the various ridges or septa within the ventricle has also been rather arbitrary, with the result that the same names have often been applied to quite different structures, and this, together with vague descriptions, has caused considerable confusion. For instance, in Tiliqua scincoides, Rau (1924) has indicated as a “muscular ridge”, a ridge which seldom receives particular mention. This ridge is also prominent in Leiolopisma, but I have preferred to call it the “apical ridge”, since the term “muscular ridge” has since been used by Mahendra (1942) specifically for the other muscular projection in the ventricle. Although the latter has been also called the “interventricular septum”, “septum ventriculorum” or “Muskelleiste”, I have preferred to use Mahendra's more non-committal term “muscular ridge”. Mathur (1944) has unfortunately taken what I have called the apical ridge (i.e., the “muscular ridge” of Rau) to be the same as the “muscular ridge” of Mahendra, which could further add to the confusion. There has been even more controversy and confusion over the precise relation of the muscular ridge (“interventricular septum”) to the three aortic trunks arising from the ventricle. Goodrich (1916) at first regarded this septum as dividing the ventricle in all Reptilia into a right chamber, giving off both the pulmonary and left systemic arteries; and a left chamber, giving off the right systemic alone. Later, in response to a criticism by O'Donoghue (1918), he modified this view slightly by admitting that, in the Lacertilia and Ophidia, the “ostium” of the left systemic artery “comes to lie close to that of the right arch and dorsally to the free edge of the septum” (Goodrich, 1919, p. 301). At the same time, he was quite emphatic that this septum is “essentially always a ventral septum developed in relation to the sulcus interventricularis” (i.e., the left bulbo-ventricular sulcus here) (1919, p. 301). Fig. 14.—Schema of the relation of the “interventricular septum” to the aortic orifice: in the Lacertilia and Ophidia. A, As I interpret O'Donoghue's remarks (1918) B, According to Goodrich (1919).

O'Donoghue's criticism (1918) of Goodrich's earlier view had been that, in the Lacertilia and Ophidia, the arterial trunks do not come off as suggested, but that (although possessing the usual positions relative to each other) the pulmonary artery comes off alone from the left side, while the two systemic trunks come from the right. He based this radically different arrangement on the fact that, contrary to what has been described by all other observers, the interventricular septum runs from the left dorso-lateral wall of the ventricle towards the right ventro-lateral wall. If this were so, the septum must be (as he apparently realised) in a plane approximately at right angles to that of the interventricular septum as described by Goodrich and others (see Fig. 14). Of those who have particularly studied this point since then, Rau (1924) from studies of Tiliqua and Eunectes believes that O'Donoghue's account of the septum is quite incorrect, and that it is substantially as Goodrich described it; from Mathur's account of the heart of Varanus monitor, it is quite clear that in this lizard, too, the septum is not as O'Donoghue (1918) describes it in Varanus salvator (p. 477). Clearly, in Leiolopisma also, the muscular ridge (“interventricular septum” of the above authors) is quite different in its orientation from that which O'Donoghue has described as typical of the lizards. It does in fact arise just as Goodrich found—i.e., from the ventral wall of the ventricle in relation to the left bulbo-ventricular sulcus. Traced toward the base of the ventricle, the muscular ridge gives place to a fibrous band, which is in fact an aortico-pulmonary septum, distinctly separating the pulmonary artery (arising from the cavum pulmonale ventrally and to the right of the muscular ridge) from the systemic trunks (arising from the cavum venosum, dorsal to the muscular ridge). These two cavities (i.e., the cavum venosum and cavum pulmonale) communicate with each other around the free border of the muscular ridge, which faces towards the right wall of the ventricle. Thus it seems probable, that in ventricular systole this opening will be closed by the approximation of the right wall of the ventricle (the “Bulbuslamelle” of Benninghoff, 1933) to the free margin of the muscular ridge. When this happens the cavum pulmonale must be entirely shut off from the rest of the ventricular cavity, and can lead out into one vessel and one vessel only—the pulmonary artery. A careful study of the structure of the cardiac muscle fibres in the heart of Leiolopisma by a variety of histological procedures, did not disclose any evidence of specialised conduction pathways. Numerous fibres occur in the atria and ventricles which, because they show frequent localised bulgings in which there is an increase of sarcoplasm, may be mistaken for specialised fibres such as are found in the mammalian conducting system; but, as stated, there is no reason to believe that these are not merely ordinary cardiac muscle fibres. It would seem that the Purkinje-like tissue which Robb (1953) has described and figured (her Figs. 6, 7, p. 11) in the heart of the turtle, may in fact be these very fibres, for she says: “Apart from refusing the stain, the fibers of this tissue are wider than other muscle fibers, the walls are not parallel, but bulge…. The nuclei are large, conspicuous, sometimes multiple, and often appear to lie in a hole. There are faint striations at the periphery of the fiber” (p. 8); these are all features which characterise the local bulgings of the muscle fibres of Leiolopisma, and which occur quite widely and randomly throughout the heart, including the A-V funnel, where, however, they are certainly no more numerous than elsewhere. There is thus no reason to believe that in Leiolopisma they could be exclusively concerned with conduction. That there are no “typical” Purkinje fibres (in the sense of Blair and Davies, 1935, p. 321) amongst the cardiac musculature does not necessarily mean, of course, that no conducting tissue is present. Even in birds and mammals considerable variation exists in the conducting tissues as between different species (Davies and Francis, 1952) and in some mammals “typical” Purkinje fibres are absent in certain parts of the conducting system (e.g., human A-V bundle, Blair and Davies, 1935).

The search for specialised fibres in the lizard is further complicated—particularly in the ventricle—by the marked variation which the fibres show in breadth, myo-fibrillar content and depth of staining. However, fibres of all intermediate grades can always be found so that the fact that a particular fibre happens to be slightly thicker and possesses fewer myofibrillae is not necessarily an adequate reason for considering it specialised for the purpose of conduction; again, variation in depth of staining is probably due rather to the physiological state of the fibre at the time of fixation, or to the thickness of it included in the section, than to any constant histoligical difference. As the fibres of the conducting system may not necessarily show marked histological variation from the ordinary cardiac fibres, and because there is normally so much variation in the ordinary fibres anyway, it would be a decided advantage to have some specific method for the demonstration of conducting tissue. Taussig (1934) considered Best's carmine stain to be the best criterion of specialised tissue at that time, and no better technique appears to have been developed since. By this method, if successful, “the Purkinje cells stand out as brilliant red against the yellow tone of cardiac muscle”. As we have seen, when this technique was applied to the heart of Leiolopisma, the results were quite negative, in that no conspicuous fibres or groups of fibres were to be seen. The fact that the glycogen content of the musculature of the different chambers increases as one proceeds from the sinus to the ventricle, may possibly be correlated with the energy demands of each chamber for the work of contraction, rather than with any specific conduction properties of the muscle (Davies and Francis, 1941b). This result confirms that of Külbs and Lange (1911) who were also unable, in Lacerta, to find fibres with a particularly characteristic glycogen content, although they were able to demonstrate glycogen in all parts of the heart. Thus it appears that although there are slight differences in fibre breadth, coarseness of striation and depth of staining in the fibres of the various chambers, all the muscle fibres in the heart of Leiolopisma have the same general histological structure. Nowhere in the heart were there any fibres at all resembling the Purkinje fibres or nodal fibres of birds and mammals. There is, however, extensive muscular continuity between the sinus and the atria, and the atria and the ventricle, which obviously must provide a pathway whereby the impulse may pass from one chamber to the next. The muscle fibres at these junctions do not show any histological specialisation at all, to distinguish them from the remainder of the myocardium. The well-known delay in the passage of the impulse from the atria to the ventricle may be due to the fact that the fibres of the A-V ring and funnel are arranged circularly and not longitudinally (Davies and Francis, 1941a). Furthermore, the evidence in regard to glycogen content does not support the presence of any specialised connexions between the chambers, or indeed, specialised fibres in the chambers. My findings therefore permit me to say that in Leiolopisma there are no fibres similar in structure to those described in the conducting systems in the birds and mammals, either isolated or in the form of an extensive system. While this can be regarded as support for the contention of Davies and Francis that lower vertebrates do not possess widespread collections of cells similar in structure to those of the conducting systems of the higher vertebrates, it still does not exclude the possibility that conducting tissue of a histologically “unspecialised” nature (i.e, identical histologically with the ordinary cardiac muscle), may be present. Acknowledgments I am deeply indebted to Professor W. E. Adams both for the opportunity of carrying out this work and for his constant help and guidance during its progress I also wish to thank Mr. J. G. Howard for his valuable technical assistance, and Miss Margaret Ogilvie, who kindly labelled the diagrams for me.

References Benninghoff, A., 1933. “Das Herz.” Bolk, Göppert, Kallius and Lubosch's Handbuch der vergleichenden Anatomie der Wirbeltiere Berlin. Urban und Schwarzenberg. 6: 467–556. Bhatia, M. L., 1929. On the arterial system of the lizard, Uromastix hardwickii Gray. J. Morph, 48: 281–316. Blair, D. M. and Davies, F., 1935. Observations on the conducting system of the heart. J. Anat., Lond 69: 303–325. Davies, F. and Francis, E. T. B, 1941a. The heart of the salamander (Salamandra salamandra L.), with special reference to the conducting (connecting) system and its bearing on the phylogeny of the conducting systems of mammalian and avian hearts. Philos. Trans. B. 231: 99–130. —— 1941b. The glycogen content of the frog's heart. J. Physiol. 100: 329–336. —— 1946. The conducting system of the vertebrate heart. Biol. Rev 21: 173–188. —— 1952. The conduction of the impulse for cardiac contraction. J. Anat., Lond. 86: 302–309. Davies, F., and Francis, E. T. B. and King, T. S., 1952. The conducting (connecting) system of the crocodilian heart. J. Anat., Lond. 86: 152–161. Gaskell, W. H., 1882. Observations on the innervation of the heart. On the sequence of the contractions of the different portions of the heart. Bnt. med. J. 2: 572–573. —— 1883. On the innervation of the heart, with especial reference to the heart of the tortoise. J. Physiol. 4: 43–127. Glomset, D. J. and Glomset, A. T. A., 1940a. A morphologic study of the cardiac conducting system in ungulates, dog, and man. Part I: The Sinoatrial Node. Amer. Heart J. 20: 389–398. —— 1940b. A morphologic study of the cardiac conduction system in ungulates, dog, and man. Part II: The Purkinje System. Amer. Heart. J. 20: 677–701. Goodrich, E. S., 1916. On the classification of the Reptilia. Proc. Roy. Soc. B. 89: 261–276. —— 1919. Note on the reptilian heart. J. Anat., Lond. 53: 298–304. Grant, R. T. and Regnier, M., 1926. The comparative anatomy of the cardiac coronary vessels. Heart 13: 285–317. Huxley, T. H., 1871. A manual of the anatomy of vertebrated animals. J. & A. Churchill, London. 431 pp. Keith, A and Flack, M., 1907. The form and nature of the muscular connections between the primary divisions of the vertebrate heart. J. Anat., Lond. 41: 172–189. Keith, A. and Mackenzie, I., 1910. Recent researches on the anatomy of the heart. Lancet 1: 101–103. Krause, R., 1922. Mikroskopische Anatomie der Wirbeltiere. Walter de Gruyter & Co., Berlin. II, Vögel und Reptilien. 434–439. Külbs, F., 1912. Ueber das Reizleitungssystem bei Amphibien, Reptilien und Vögeln. Z. exp. Path. Ther. 11: 51–68. —— and Lange, W., 1911. Anatomische und experimentelle Untersuchungen über das Reizleitungssystem im Eidechsenherzen. Z. exp. Path. Ther. 8: 313–322. Laurens, H., 1913. The atrio-ventricular connection in the reptiles. Anat. Rec. 7: 273–285. —— 1915. The connecting systems of the reptile heart. Anat. Rec. 9: 427–446. Mackenzie, I., 1913. The excitatory and connecting muscular system of the heart (a study in comparative anatomy). Trans. Int. Congr. Med. Lond. Sect. 1, 121–150. Mahendra, B. C., 1942. Contributions to the bionomics, anatomy, reproduction and development of the Indian house-gecko, Hemidactylus flaviviridis Rüppel. Part III. The heart and the venous system. Proc. Indian Acad. Sci. B. 15: 231–252. Mathur, P. N., 1944. The anatomy of the reptilian heart. Part I. Varanus monitor (Linne). Proc. Indian Acad. Sci. B. 20: 1–29. Meinertz, T., 1952. Das Herz und die grossen Blutgefässe bei der Komodoeidechse, Placovaranus komodoënsis Ouw. Z. Anat. Entw. Gesch. 116: 315–325. O'Donoghue, C. H., 1918. The heart of the leathery turtle, Dermochelys (Sphargis) coriacea with a note on the septum ventriculorum in the Reptilia. J. Anat., Lond. 52 467–480. —— 1921. The blood vascular system of the tuatara, Sphenodon punctatus. Philos. Trans. B. 210: 175–252. Owen, R. On the anatomy of vertebrates. London, Longmans, Green and Co. 1866. xlii + 650 pp. Vol. 1 Fishes and Reptiles. Palme, F., 1934. Die Paraganglien über dem Herzen und im Endigungsgebiet des Nervus depressor. Z. mikr.-anat. Forsch. 36: 391–420. Rau, A. S., 1924. Observations on the anatomy of the heart of Tiliqua scincoides and Eunectes murinus. J. Anat., Lond. 59: 60–71. Ray, H. C., 1934. On the arterial system of the common Indian rat-snake, Ptyas mucosus (Linn). J. Morph. 56: 533–576.

Robb, J. S., 1953. Specialized (conducting) tissue in the turtle heart. Amer. J. Physiol. 172: 7–13. Swett, F. H., 1923. The connecting systems of the reptile heart—alligator. Anat. Rec. 26: 129–140. Taussig, H. B., 1934. On the technique for the demonstration of the specialized tissue in the heart. J. tech. Meth. 13: 85–87. Trinci, G., 1912. Il sistema cromaffine cardiaco-cervicale nei Sauri. Arch. ital. Anat. Embriol. 10: 197–260. Wiedersheim, R., 1886. Elements of the comparative anatomy of vertebrates. English ed. Macmillan and Co., London. 345 pp. Mr. J. G. Buchanan, 70 Seacliffe Avenue, Belmont, Takapuna, Auckland.