t ' m§,i HARVARD UNIVERSITY ^ Library of the Museum of Comparative Zoology • BULLETIN OF THE MUSEUM OF COMPAEATIVE ZOOLOGY AT HARVARD COLLEGE, IN CAMBRIDGE VOL. 117 CAMBRIDGE, MASS., U. S. A. 1957 The Cosmos Press, Inc. Cambridge, Mass., U.S.A. J CONTENTS PAGE No. 1. — A Study op the Sharks of the Suborder Squa- LOIDEA. By Henry B. Bigelow and William C. Schroeder. (4 plates) Augrust, 1957 ... 1 No. 2. — Check List of the Reptiles and Amphibians OF East Africa (Uganda ; Kenya ; Tanganyika; Zanzibar). By Arthur Loveridge. August, 1957 151 No. 3. — The Spider Genera Crustulina and Steatoda IN North America, Central America, and the West Indies (Araneae, Theridiidae). By Herb- ert W. Levi. August, 1957 . . . ' . .365 No. 4. — Contribution to a Revision of the Earthworm Family Ocnerodrilidae. The Genus Nemato- GENiA. By G. E. Gates. September, 1957 . . 425 No. 5. — Triassic Reptile Footprint Faunules from Mil- ford, New Jersey. By Donald Baird. (4 plates) November, 1957 447 BuUetin of the Museum of Comparative Zoology AT HARVARD COLLEGE Vol. 117, No. 1 A STUDY OF THE SHARKS OF THE SUBORDER SQUALOIDEA By Henry B. Biqelow and William C. Scheoeder With Four Plates CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM August 1957 Publications Issued by or in Connection WITH THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE Bulletin (octavo) 1863 — The current volume is Vol. 117. Breviora (octavo) 1952 — No. 80 is current. Memoirs (quarto) 1864-1938 — Publication was terminated with Vol. 55. JoHNSONiA (quarto) 1941 — A publication of the Department of Mollusks. Vol. 3, no. 35 is current. Occasional Papers of the Department of Mollusks (octavo) 1945 — Vol. 2, no. 21 is current. Proceedings of the New England Zoological Club (octavo) 1899- 1948 — Published in connection with the Museum. Publication terminated with Vol. 24. The continuing publications are issued at irregular intervals in numbers which may be purchased separately. Prices and lists may be obtained on application to the Director of the Museum of Comparative Zoology, Cambridge 38, Massachusetts. Of the Peters "Check List of Birds of the World," volumes 1-3 are out of print; volumes 4 and 6 may be obtained from the Harvard University Press; volumes 5 and 7 are sold by the Museum, and future volumes will be published under Museum auspices. Bulletin of the Musetun of Comparative Zoology AT HARVARD COLLEGE Vol. 117, No. 1 A STUDY OF THE SHARKS OF THE SUBORDER SQUALOIDEA By Henry B. Bigelow and William C. Schroeder With Four Plates CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM August 1957 No. 1 — A Study of the Sharks of the Suborder Squaloidea ' By Henry B. Bigelow and William C. Sohroeder TABLE OF CONTENTS PAGE Foreword 5 Acknowledgments 5 Classification of the Squaloidea 7 Key to Families of Squaloidea 13 Family Oxynotidae 13 Genus Oxynotus 14 Key to species of Oxynotus 17 Family Squalidae 17 Key to Subfamilies of Squalidae 18 Subfamily Squalinae 18 Synopsis of Genera of Squalinae 24 Genus Squulu^ 26 Genus CirrMgaleus 37 Genus Centroscyllium 38 Provisional Key to Species of Centroscyllium 47 Genus Etmopterus 47 E. h ullisi n. sp 50 Key to Species of Etmopterus 60 Genus Centropliorus 63 Nomenclatural status of the type species 73 C. foliaceus Giinther 1877 78 Key to species of Centropliorus 83 1 Contribution 848, Woods Hole Oceanographic Institution. 4 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Genus Centroscymnus 84 Key to species of Centroscymnus 95 Genus Scymnodon 96 Key to species of Scymnodon 101 Genus Deania 101 Key to species Deania 107 Subfamily Dalatiinae 109 Key to Genera of Dalatiinae 112 Genus Dalatias 113 Genus Somniosus 115 Key to species of Somniosus in the Northern Hemisphere 122 Genus Isistius 123 Genus Scymnodalatias 124 Genus Euprotomicrus 126 Genus Squuliolus 128 Genus Keteroscymnoides 132 Subfamily Echinorhininae 134 Genus Echinorhinus 134 References 136 BIQELOW AND SCHROEDER : SQUALOID SHARKS 5 FOREWORD The sharks that lack an anal fin but do have snouts of the ordinary "shark" form, and pectoral fins that are wholly pos- terior to the gill openings (grouped together here as the sub- order Squaloidea), have been surveyed of late for Philippine waters and adjacent seas by Fowler (1941) and for the western North Atlantic by Bigelow and Schroeder (1948). But the most recent comprehensive accounts of the group for the oceans as a whole, by Regan (1908) and by Garman (1913, pp. 188-244), now lie many years in the past. And so much additional in- formation has come to hand during these years, from various sources, that the time seems ripe for a fresh survey of the group. Many taxonomic questions, however, still await for their solution the critical examination of large series of specimens. We attempt little more in the following pages than to summarize the present state of knowledge of the group. Anything as ambitious as would be implied by the word "revision" is a task for the future. It will save repetition if we state here (once and for all) that the names of the species we have seen are preceded by an asterisk Also, we should remind the reader, who may not be familiar with the use of the trawl in deep water, that the depths of cap- ture reported by this method, under the several genera, are in reality those at which the trawl was dragging on the bottom. This is not necessarily the depth at which the shark in question was living, for it is always possible that any particular specimen (even of the species that hold closest to the bottom) may have been picked up when the trawl was being lowered, or hauled up again through the water. The only precise data in this regard are for specimens caught on hook and line. And we have very few definite records of depths of capture for line-caught sharks. ACKNOWLEDGMENTS We welcome the opportunity of thanking our colleagues in different parts of the world who have assisted us in various ways : especially Dr. Tokiharu Abe, Tokyo, for contributing a specimen of Somniosus long us; the Australian Museum, through Dr. G. P. Whitley, for a specimen of Oxynotus hruniensis; Dr. G. Bel- loc of the Museum, Monaco, for the gift of an excellent specimen 6 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY of Oxynotus from Italy; Dr. James Bohlke, of the Academy of Natural Sciences for making available to us for study the type specimen of Heteroscymnoides marleyi, from the Natal Coast, southern Africa; the British Museum for the loan of the speci- mens of C entroscymnus crepidater and of Scymnodon ringens, on which Giinther (1870) based his descriptions of these species, and of a specimen of Oxynotus paradoxus ; Dr. Paul Budker, of the Paris Museum, for denticles from the type specimen of Scymnodon ohscurus, and for drawings of its teeth; Dr. Fer- nando Frade for re-examining for us the type specimen of Oxy- notus paradoxus; Dr. Arni Fridriksson for specimens of Centrophorus sqiiamosus and of Somniosus microcephalus from Iceland; Dr. J. A. F. Garrick for kindly allowing us to quote from his manuscripts in advance of publication, for a specimen of a New Zealand shark apparently not separable from Centro- scymnus crepidater of the eastern North Atlantic and for a para- type of Etmopterus abernathyi; Dr. Finnur Gudmundsson, of the Museum of Natural History, Reykjavik, for the loan of an excellent series of C entroscymnus crepidater (see p. 91) ; Col. John K. Howard for specimens of Squalus megalops from New South Wales; Mr. N. B. Marshall for notes on Euprotomicrus (p. 127) ; Dr. George S. Myers, of the Natural History Museum, Stanford University, for loaning us a paratype of Centroscyllmm ruscosum; Mr. G. Palmer, of the British Museum, for contribut- ing tracings of pectoral fins of Richardson's young specimen of Acanthias vulgaris, renamed Squalus megalops by Regan; Dr. J. R. Pfaff, University Zoological Museum, Copenhagen, for the loan of specimens of Centrophorus squamosus, Deania calcea, and Centroscymnus crepidater, from Icelandic waters, and for the gift of a specimen of C. crepidater; Dr. A. M. Ramalho for many kindnesses, including the contribution of a series of Oxynotus centrina from off Portugal and of an excellent specimen of Echi- norhinus hrucus from northwest Africa, also for photographs of a specimen (now in the Bocage Museum in Lisbon) that had been described and pictured in 1870 by Capello under the name Lae- margus rostratus Risso 1826; Dr. Leonard Schultz, for making available to us the extensive collections of the U. S. National Museum, as well as for assistance in various other ways; Dr. J. L. B. Smith, Rhodes University, Grahamstown, for photographs of the second dorsal fin spine of the shark described by Gilchrist BIGELOW AND SCHROEDER : SQUALOID SHARKS 7 1922 as Atractophorus armatus; Dr. Ethelwyn Trewavas of the British Museum for contributing a drawing of a lower tooth of the type specimen of Centroscymnus macracanthiis as well as for constant cooperation; Mr. Denys W. Tucker of the British Mu- seum for drawings of the first and second dorsal fins of the type specimen of Centroscymnus macr acanthus ; and Dr. Gilbert P. Whitley, of the Australian Museum, for making available to us a lower tooth from the type specimen of Centrophorus harrisonii and for contributing a tracing of the pectoral fins of the type specimen of Squalus megalops. Drawings are by Eugene N. Fischer, Jessie H. Sawyer and Henry B. Bigelow, except as otherwise noted. Photographs were contributed by the Woods Hole Oceanographie Institution. CLASSIFICATION OF THE SQUALOIDEA While the sharks in question are an extremely homogeneous assemblage differing conspicuously in their general makeup from the other groups of sharks with which they share the lack of an anal fin (pristiophoroids and squatinoids), the taxonomic rank- ing assigned to them by recent authors has reflected divergent viewpoints. Thus they have been grouped with the pristiophor- oids by Garman (1913, pp. 12, 13) as the "group of families" Squaloidei; by Berg (1940; Eng. Trans. 1947, p. 381) as the suborder Squaloidei; by White (1937, pp. 100, 101) as the sub- order Squalida; and by Bertin (1939, pp. 17-19) as the suborder Squaliformes. Fowler (1941, pp. 222, 279) has placed the squaloids in a separate order (Cyclospondyli), joining the pristiophoroids with the squatinoids as the order Squatinae. Bigelow and Schroeder (1948, p. 77) classed the squaloids as one suborder (Squaloidea), the pristiophoroids as a second sub- order (Pristiophoroidea), and the squatinoids as a third sub- order (Squatinoidea) of the order Selachii. Still more recently Berg, 1955, while similarly classifying the squaloids and the squatinoids as suborders, has set the pristiophoroids apart as a separate order. And this arrangement has much to recommend it for the pristiophoroids are distinguished not only by their saw- like snout, but l3y the "presence of a separate antorbital bar, from which the upper jaw is suspended by a broad ligament, in addition to the articulation to the cranium" (Bigelow and Schroeder, 1948, p. 532). 8 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Characters. Sharks without an anal fin, but with the snout of the ordinary shape (i.e. not beak-like and without lateral teeth or cirri), and subcylindrical in general form, the anterior mar- gins of the pectoral fins not expanded forward past the first pair of gill openings. Two dorsal fins, with or without fin-spines. Only 5 gill openings, all of them anterior to the pectorals. Inner margins of pelvics entirely separate posterior to cloaca. Nostrils entirely separate from mouth. Spiracles present and conspic- uous. Eyes without nictitating fold or membrane. (For internal characters, see Bigelow and Schroeder, 1948, p. 449.) Families. Views have differed as to whether a subdivision of the Squaloidea into separate families is justified on any basis that has yet been proposed. Garman (1913, p. 13) distributed them among three families: Squalidae for those with spines in the dorsal fins, Scymnorhinidae and Echinorhinidae for those without spines, separating these last two, one from the other, by differences in their teeth. And this same scheme was accepted by Bigelow and Schroeder (1948, p. 450), except that they re- placed the name Scymnorhinidae by Dalatiidae because the gen- eric name ScymnorJiinus Bonaparte 1846 is synonymous with Dalatias Rafinesque 1810. Bertin (1939), however, recognized only two families among the sharks in question — Squalidae for those with fin spines, and Scymnorhinidae for those without spines — while Berg (1940) has united them all in the single family Squalidae. And Hubbs and McHugh (1951, pp. 163, 164), for the reasons stated on p. 9, follow Berg in their suggestion "that the Dalatiidae be fused with Squalidae" and that the Echinorhinidae "should likewise be synonymized with Squali- dae," though they retain the Dalatiinae and Echinorhininae as subfamilies, pending future study. These differences do not stem from any recent increase in factual knowledge, for the external features of the squaloid sharks were about as well known structurally a hundred years ago as they are today. What the different systems do chiefly mirror is the opinions held by the several authors as to the de- grees of consanguinity among the animals in question. The ex- pression, however, of views in this regard is but the secondary purpose of animal classification; the primary purpose, as so trenchantly worded by Simpson (1945, p. 13), is to "provide a convenient practical means by which zoologists may know what BIQELOW AND SCHROEDER : SQUALOID SHARKS 9 they are talking about and others may find out." In the ease of any suborder as homogeneous structurally as the squaloid sharks are, and for which the fossil record throws little light on their evolutionary history, the decision, whether any subdivision into different families is warranted depends primarily (in our opinion) on whether it will help anyone from the practical standpoint. The scheme adopted here is based on this premise. At the same time it also emphasizes the strong probability that one of the genera with which we are here concerned (Oxynotus) represents a separate line of descent, for it is set apart from all the others not only by the arrangement of its upper teeth, by the nature of its dermal denticles, and by its body form (p. 14), but also because it was differentiated at least as early as the Miocene. Oxynotus was made the basis of a separate family (Oxynoti- dae) by Jordan (1923, p. 103), followed by Key (1928, p. 467).' And this course is followed here, as better indicative of the ap- parent degrees of genetic relationship among squaloid genera than was our former inclusion of it in the family Squalidae (Bigelow and Schroeder, 1948, p. 450). The Linnean genus Squalus, which includes the common spiny dogfishes, has been regarded as what may be termed the "key genus" among the squaloids that remain after the subtraction of Oxynotus. This implies that the ancestors of the modern squaloids had spines in their dorsal fins. And this view, recently urged by Hubbs and McHugh (1951), is at least consistent with the fossil record, for Protospinax Woodward 1919, from the Upper Jurassic of Bavaria, with a spine of the squaloid type in each of its dorsal fins, but with a small anal fin (if Wood- ward's interpretation of the specimen be correct), and with teeth but little differentiated in nature, may in some measure bridge the gap between the squaloid sharks and the galeoids. Hubbs and McHugh further argue that the absence of a spine in the second dorsal fin in such of the sharks in question as lack one (also of a spine in the first dorsal fin in all but one of the genera in question) probably does not point to any close relation- ship among them, but more likely represents the end result of convergent evolution (i.e. loss of fin spines) along sep- arate phyletic lines leading back to as many separate spiny- finned ancestors. As one illustration, they point out that "com- 2 Regarded as a subfamily by Fowler, 1941, p. 223. 10 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY plete obsolescence of the spines would render Centroscymnus scarcely separable from Somniosiis." This leads them to con- clude that the presence or absence of spines in the dorsal fins has been given undue weight in current classifications of the squaloid sharks (as by Bigelow and Schroeder, 1948) in which it is in- voked as the basis for the separation of families. Here the fossil record does not help us at all, for this reaches as far back for the genera Dalatias and Isistius, which lack fin spines, as for the spiny-finned genera Squalus, Etmopterus and Centrophorus, i.e. to the Upper Cretaceous — always provided that identification of the fossils in question with these modern genera was correct. For that matter, a shark with a rostral blade (but lacking rostral teeth) that may reasonably be regarded as ancestral to the pristiophoroids (Woodward 1932, p. 476, PI. 18; see also Jaekel, 1890, for other early pristiophorid fossils) is similarly known from the Upper Cretaceous ; Sqnatina is even known from as far back as the Jurassic. Thus both the squaloids with spiny fins and those lacking spines, the progenitors of the pristiophoroids with their bizarre snouts, and the squatinoids (ray-like in ap- pearance though true sharks) had completed their major evolu- tionary history by the Upper Cretaceous period, which (the geologists tell us) lies something like 75 million years in the past. For all that is known to the contrary, this may equally apply to the aberrant squaloid genera Oxynotus and Echinorhinus, though the known record for these runs back only to the Miocene. (For the fossil record of modern shark genera, see Romer 1945, pp. 576-577.) Were one approaching the problem anew, the shape of the caudal fin, combined with the degree of elevation of the caudal axis, would seem about as appropriate as would the presence or absence of fin spines for a primary alternative among the genera of squaloids. Thus the one character is about as susceptible of precise determination as the other, while, in each case, the ex- treme states are bridged by intermediates so that division is not sharp-cut on either basis. Also the use of either one would result in the same grouping of genera as would the use of the other, with one exception. A third character that is alternative between the groups of genera under discussion here, though it seems not to have been emphasized previously in this particular connection, is the con- BIGELOW AND SCHROEDER : SQUALOID SHARKS 11 i^sSSfeii -(iSsJi^^S^ Fig. 1. Corner of mouth, of different squaloids, to show preoral clefts and preoial pouches, if any; the latter outlined by the broken curves. A, Oxynotus centrina, male 590 mm. long, Mus. Comp. Zool. No. 3!;»576, x about 3. B, Centrophorus squamosiis, female 1230 mm. long, Iceland, Mus. Comp. Zool. No. 37825, x about 0.5. C, Centroscymnus coelolepis, female, 1035 mm. long, offing of Delaware Bay, Mus. Comp. Zool. No. 38295, x about 2. D, Centroscymnus crepidatcr, female, 784 mm. long, Faroes, Mus. Comp. Zool., No. 39377, with preoral cleft and upper labial furrow pinned open, x about 1. E, Dalatias Helm, male, 1114 mm. long, Japan, Mus. Comp. Zool. No. 1116, X about 0.5. F, Isistius hrasiliensis, female, 501 mm. long, Japan, Mus. Comp. Zool. No. 1245, x about 2. 12 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY formation of the more or less voluminous inpocketing of the lower surface of the head, that bounds each angle of the jaw on the outer side. In all the known genera of Squaloidea that have a spine in the second dorsal fin as well as in the first (in- cluding Oxynotus, Fig. lA), this oral pocket extends inward- forward across the lower surface of the head, anterior to the upper jaw, as an open and conspicuous preoral cleft, that is simple in some (e.g. Oxynotus, Fig. lA; Centroscymnus crepi- dater, Fig. ID), but that expands inwardly in most to form a more or less voluminous subdermal pouch (e.g. in Centroscymnus coelolepis, Fig. IC). On the other hand, the majority of the gen- era that lack a spine in the second dorsal fin equally lack the open preoral cleft, although the preoral pouch extending forward from each of the oral pockets is voluminous (Fig. IE, F). Two, how- ever, of the genera of this group, Scymnodalatias and Echino- rhinus do have short preoral clefts. And, in any ease, the group difference in this respect, while a convenient aid in the definition and identification of genera, is obviously of the kind rated by White (1937, pp. 51-53) as of ''physiological importance," rather than as an indication of lines of ancestral descent. There seems, in short, to be "no securely objective basis," as pointed out by Hubbs and McHugh (1951, p. 163) "on which to choose between the characters to be emphasized" in the classifi- cation of the Squaloidea. But the cause of stability in nomenclature seems best served by using the presence or absence of fin spines as the primary alternative, since this accords with precedent, whereas to use the shape of the caudal fin, as have Hubbs and McHugh (1951, p. 164) in their analytical key to the dalatiine sharks, or to emphasize the presence or absence of the preoral clefts, would re- sult in a novel alignment which we could not support by any cogent reasoning. But we wish it understood that this choice does not carry any phylogenetic implication but is purely a mat- ter of convenience. In our earlier synopsis of the Squaloidea (Bigelow and Schroeder 1948, p. 450), we recognize two families for such of the genera as lack a fin spine in the second dorsal fin : Dalatiidae for those in which the teeth have only one cusp, but are of con- spicuously different shapes in the two jaws, contrasted with BIGELOW AND SCHROEDER : SQUALOID SHARKS 33 Ecliinorhinidae (for Ecliinorhinus), in which the teeth are simi- lar in the two jaws, with several cusps. But it seems more logical. in view of the uncertainties just discussed, to follow Hubbs and McHugh (1951) in reducing these categories to the rank of sub- families of the family Squalidae, the structural contrast between them obviously being much narrower than that between either of them and the genus Oxynotus (p. 14). Key to Families of Squaloidea Only two or three upper teeth in the hrst row, but with the number of teeth greater by one in each successive row, so that the rows of func- tional teeth are successively longer from front to rear; trunk sub- triangular in cross section with a prominent ridge low down on either side anterior to the pelvic fins Family Oxynotidae, p. 13 About as many upper teeth in the anterior row as in the succeeding rows, so that any one functional row is about as long as any other; trunk subcylindrieal in cross section in most; longitudinal ridges, if any, confined to the tail sector posterior to the pelvic fins Family Squalidae, p. 17 Family OXYNOTIDAE Family cliaracters (based on only known genus, Oxynotus). Squaloidea with the upper teeth occupying a triangular area on the roof of the mouth, in quincuucial arrangement, the functional rows successively longer from front to rear, the first row con- sisting of three teeth only (on specimen studied, two, according to Eey 1928, p. 469), the number of teeth greater by one in each successive row, and with about six rows functional; lower teeth in a single functional row, blade-like, the median tooth sym- metrically triangular, but the successive teeth increasingly oblique, outward (PI. 1) ; the uppers smooth-edged, but the lowers finely serrate. The trunk subtriangular in cross section, with a prominent longitudinal ridge low down on either side an- terior to the pelvic fins; preoral clefts well developed and very conspicuous, extending forward-inward so far that they are sep- arated by only a narrow isthmus in the midline of the snout ( Fig. lA) ; lips thick, spongy, with a complex series of cross folds (Fig. 14 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY lA) ; first dorsal fin spine slanting either rearward or forward;' the first dorsal fin either with or without supporting radial cartilages posterior to the imbedded portion of the spine ; both fin spines conspicuously stout, but projecting only slightly beyond the skin. Dermal denticles so large and prominent that the skin is very rough. On young specimens they are tridentate, with the median point much the longest. But they are much more com- plex in form on adults, mostly with three subsidiary points, the basal point with a varying number of radial furrows (for excel- lent illustrations of the denticles, see Rey 1928, p. 471, Fig. 157). Genera. Only one genus is known : Oxynotus Rafinesque 1810. Remarks. The pattern in which its upper teeth are arranged sets Oxynotus apart from all other squaloid sharks — from all other sharks for that matter — which of itself seems to us war- rant enough for referring it to a separate family. Genus OxYNOTUS Rafinesque 1810 Oxynotus Eafinesque, 1810, pp. 45, 60, type species *Squnlus oentrina Lin- naeus 1758, p. 233, type locality, Mediterranean, For generic synonyms, see Garman, 1913, p. 190. Generic characters. Those of the family (p. 13). Maximum length probably not far from one meter. Depth range. The type species (centrina) has attracted scien- tific attention at least since the time of Belon (1515, p, 46 bis), and is well known to Mediterranean fishermen, in moderately deep water. But the only definite statements we have found as to depth of capture are Roule's (1919, p. 123) report of a Mediterranean specimen taken at 60 meters, and Poll's (1951, p. 57) report of three, from 100-180 meters off equatorial West Africa. Depth records for the newly discovered *0. paradoxus (p. 16) have ranged between 265 and 530 meters (Tucker and Palmer 1949, p, 930). All we have found for *hruniensis of the Australian-New Zealand region is that a specimen at hand (p. 5) was trawled at about 180 meters. 3 This seems an appropriate place to remind the reader that it is only the first dorsal fin spine that is inclined forward in any Ijnown species of Oxynotus, not both of the fin spines, as is erroneously stated in our earlier Icey to the fenera of Squalidae (Bigelow and Schroeder 1948, p. 450, alternative lA) ; also hat in constructing that lS'. acanthias (p. 30) occurs in numbers anywhere' between the surface and the 160-180 meter level. *8. fernandinus ranges from close to the surface (reported at 16-20 meters by Poll, 1951, p. 62) down to the maximum depth reported above for its genus, while the center of abundance lies considerably deeper for it than for * acanthias, to judge from Poll's (1951, p. 61) report of 608 specimens ti-awled at 220 meters off equatorial West Africa. And it seems that neither *cuhensis (reported from 137-384 meters) nor *megalops rise, normally, as near to the surface as does *acanthias or * fernandinus. Species. All known representatives of the genus Squalus from various parts of the world (under whatever names thej^ may have been reported) seem referable to one or other of three divisions as defined by the shape of the pectoral fins; by the position (farther forward or rearward) of the first dorsal fin; by the presence or absence of a small secondary lobelet on the outer edge of the flap-like expansion of the anterior margin of the nostril ; BIQELOW AND SCHROEDER : SQUALOID SHARKS 27 and by the color pattern. In the following discussion each of these divisions is referred to under the name of its earliest de- scribed representative. The differences in the shape of the pectorals, as outlined be- low, are so obvious that no further discussion of them is called for just here. Similarh', the position of the first dorsal spine, relative to that of the origin of the pectoral fins, has proved definitely alternative for the specimens yet measured, though with the reservation that measurements of a larger number may eventually narrow the gap in this respect. And the shape of the narial flap seems equally diagnostic for the majority of specimens, though we have seen one individual of the fcrnandinus division where one of the nostrils shows the secondary lobelet, but the other does not — a situation analogous to that of the shape of the inner corner of the pectoral fins in Centrophorus (p. 67). But our earlier selection (Bigelow and Schroeder 1948, p. 455; Bigelow, Schroeder and Springer 1953, p. 221) of the midpoint of the bases of the pelvics relative to the positions of the dorsals as a species-criterion was a less happy one, for additional in- formation has shown that in this respect the extremes intergrade. And while the presence of white spots is diagnostic for the speci- mens that show them, the absence of white markings is not diag- nostic, for it has long been known that in the type species of the genus {acanthias) these spots tend to fade with growth, while some large specimens show no ti'ace of them. Synopsis of divisions of Squalus Acanthias division. Inner corner of pectoral fins well rounded ; the distal margin not deeply concave; point of emergence from skin of 1st dorsal spine is posterior to 5th gill openings by a dis- tance as long as from anterior margin of eye to 5th gill openings or slightly longer ; midpoint of base of pelvics about midway be- tween a perpendicular at point of emergence of 2nd dorsal spine and one at free rear tip of 1st dorsal on most specimens, but nearer to tip of 1st dorsal than to 2nd dorsal spine on one large male from Massachusetts (Mus. Comp. Zool. No. 35863) that we have seen. Anterior narial flap simple, without secondary lobelet on its outer edge. Sides of trunk marked with white spots on small and middle sized specimens, more faintly so on larger ones. 28 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Fig. 3. Dermal denticles from side of body below first dorsal fin. A, Squalus f.ernandirius, female, 914 mm. long, island of Juan Fernandez, Mus. Comp. Zool. No. 841, x about 50. B, S. fernandirm^, male, 397 mm. long, off South Carolina, Mus. Comp. Zool. No. 37064, x about 50. C, S. megalops, same specimen as in Fig. 4A, x about 70. D, Squalus acnnthias, Massachusetts, X about 34, after Bigelow and Schroeder 1948, Fig. 88C. E, Etmopterus spinax, female, 250 mm. long, Norway, Mus. Comp. Zool. No. 1022, x about 15. F, Etmopterus princeps, female, 165 mm. long, offing of Nantucket, Mus. Comp. Zool. No. 37403, x about 12. G, Etmopterus princeps, female, 610 mm. long, offing of southern edge of Georges Bank, Mus. Comp. Zool. No. 37445, X about 10. BIGELOW AND SCHROEDEB: SQUALOID SHARKS 29 while some adults lack these spots altogether. Dermal denticles on sides below 1st dorsal fin of the shape shown in Figure 3D. Fernandinus division. Inner corner of pectorals about a right angle, more or less blunted, the distal margin weakly concave ; point of emergence from skin of 1st dorsal spine is posterior to 5th gill openings by a distance at least not longer than from an- terior edge of eye to 2nd gill openings, and slightly shorter than that in most ; midpoint of base of pelvics at least as near to a per- pendicular at rear tip of 1st dorsal as to one at emergence from A '^ Fig. 4. Outlines of fins of Squalus megalops and of Squalus cubensis, superimposed to show the differences in shape. A, caudal fins of S. mega- lops, female 526 mm. long, New South Wales, Mus. Comp. Zool. No. 38619 (dotted line), and of -S. cubensis, female, 672 mm. long, Cuba, Mus. Comp. Zool. No. 1461 (solid line), adjusted to equal lengths along upper margin. B, pectoral fins of S. megalops (type specimen), from drawing kindly con- tributed by Dr. G. P. Whitley (dotted line), and of S. cubensis, same specimen as in A (solid line), adjusted to equal lengths along outer margin. skin of 2nd dorsal spine, and nearer on most (but see above, under acanthias division) ; anterior narial flap with a small sec- ondary lobelet on its outer edge (Bigelow and Schroeder 1948, Fig. 87F), though in occasional specimens this lobelet may be lacking on the one nostril or on the other (see above) ; dermal denticles on sides below first dorsal fin simpler in form than in 30 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY acanthias division, tridentate, the marginal points longer on the denticles of small specimens than on those of larger (Fig. 3, A, B). Trunk without white markings at any stage in growth. Megalops division. Inner corner of pectoral fin pointed, or only very slightly blunted, the distal margin deeply concave (Fig. 4B) ; relative positions of fins, shape of narial flap, and color (without spots) essentially as in fernandinus division (see above) ; dermal denticles on sides below first dorsal fin differing from those both of the axanthias division and of the fernandinus division in a way easier presented pictorially (Fig. 3C) than verbally. Remarks. There is no danger of confusing any representative of the megalops division with an}'- of the acanthias division or any of the fernandinus division, so sharply diagnostic is the' shape of the pectoral fins. But the discontinuity as regards fin characters is so small between the acanthias and the fernandinus divisions that identification as the one or the other of large plain-colored specimens may call for close scrutiny of the shape of the narial flap and of the denticles. Acanthias division. *Squalus acanthias Linnaeus 1758, type species of the genus® and locally the most numerous of sharks, represents this division in the North Atlantic, where its regular range extends in the east from Iceland, Norway and the Murman coast to Morocco (including the Mediterranean and the Black Sea), with reports from the Canaries and Senegal; and in the west from southeastern Labrador (straying to southwest Green- land) to the southern part of the North Carolina coast in abun- dance, with a few ranging to southern Florida and to Cuba.' The common spiny dogfish of the coast of the northwestern Pacific was originally described as a separate species, *suckleyi Girard 1854, and it has most often been reported under that name. But our own comparison of Californian specimens with others from both sides of the North Atlantic has not revealed any consistent differences that might be regarded as specific, whether in the position of the first dorsal fin relative to the pectorals (the character the most often invoked as alternative), or in any other respect (Bigelow and Schroeder 1948, p. 453). 6 For synonyms see Bigelow and Schroeder 1948, p. 467. " For details, see Bigelow and Schroeder 194S, pp. 463-464. BIGELOW AND SCHROEDER : SQUALOID SHARKS 31 Spiny dogfishes of the acanthias division have been reported from northeastern Asia both as *acanthias Linnaeus 1758 and as *sucMeyi Girard 1854 ; from Japanese waters under the name vulgaris Risso 1826, repeatedly as *suckleyi Girard 1854, as mitsukurii Jordan and Fowler 1903 (their illustration, but not their description, which was based on an example of the fer- nandinus division, see below, p. 33), and as wakiyae Tanaka 1918, a name proposed by Tanaka for the shark that had been pictured by Jordan and Fowler (1903, p. 630, fig. 3) as mitsu- kurii; from Korea, China and Formosa as suckleyi. There is nothing, however, in available information to suggest that these populations differ specifically, one from another, or from the typical * acanthias of the North Atlantic. Schmidt (1931, p. 7) indeed, has long since expressed a doubt as to whether the spiny dogfishes of the acanthias division of Japan are separable from those of the Black Sea. For North Pacific references to Sqvalus of the acanthias division, see Bigelow and Schroeder 1948, p. 472. A representative of the acanthias division has also been cred- ited (by name only) to Hawaii (Fowler 1930, p. 495, as S. suckleyi). But an earlier report of mitsitkurii from Hawaiian waters proves actually to have been based on *S. fernandinus (p. 33). The acanthias division has not been reliably reported for the equatorial-subequatorial belt, whether of the Atlantic or of the Indo-Pacific. But seemingly it is as widespread in mid-latitudes of the southern hemisphere as it is of the northern. Thus sharks apparently of this division have been reported from southern Africa as *acanthias and as vulgaris ; from Uruguay and north- ern Argentina as ^acanthias; from the Magellanic region as "^acanthias and as lehruni Vaillant 1888^ ; from Chile as ^suckleyi (by Fowler 1930, p. 495) ; from New Zealand as * fernandinus (by Waite 1909, p. 142, PL 16, fig. 1) but identity as acanthias- like established by rearward position of first dorsal fin and by white spotted sides, and as kirki Phillipps 1931 ; from Australia as vulgaris (repeatedly) and as whitleyi Phillipps 1931 (see also \Vhitley 1940, p. 193); and from the island of Reunion in the southern Indian Ocean as vulgaris. No one of these southern populations, however, has yet been studied in detail, nor has any- thing developed within the past few years to contradict our 32 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY earlier statement (Bigelow and Schroeder 1948, p. 454) that "the relationsliip of these southern hemisphere forms to one another and to the northern acanthias is uncertain." Fernandinus division. The earliest named representative of this division is Squalus fernandinus Molina 1782, originally described from the island of Juan Fernandez off the coast of Chile. And our recent comparison of a female, 520 mm. long (Mus. Comp, Zool. No. 446) from Nice, France, with a * fernan- dinus of 914 mm. from the type locality, as well as with others from the Gulf of Mexico and from the offing of South Carolina, has confirmed the conclusion earlier arrived at by Garman (1913, p. 195), by Poll (1951, p. 59), and by Bigelow, Schroeder and Springer (1953, p. 222), that the shark described from the Medi- terranean in 1826 by Risso under the species-name hlainvillii (Risso 1826, p. 133, PI. 3, fig. 6) is identical with fernandimis. This joint species (most often reported as ^^Nainvillii^^) occurs widespread in the Mediterranean, in the Black Sea, and off the open Atlantic coast from Portugal southward to equatorial waters (reported from 6° 31' S., see Poll 1951, p. 59). And it is to be expected all along the West African seaboard to the south- ward, for spiny dogfishes apparently identical with * fernandinus have been reported in abundance in southern African waters, both as hlainvilUi (by Bleeker, 1860, pp. 50, 58, 80, and by Giin- ther 1870, p. 419) : as acutipinnis Regan 1908 (by Regan 1908\ p. 248, and by Barnard 1925, p. 48), and as * fernandinus (by Gilchrist 1922, p. 48, by von Bonde 1924, p. 5, and by Smith 1949, p. 60). *;S'. fernandinus is known in the western side of the At- lantic also, both from the offing of South Carolina and in the Gulf of Mexico in the north (Bigelow, Schroeder and Springer 1953, p. 220), and off the coast of Argentina in the south (for locality records from the western South Atlantic, see Bigelow and Schroe- der 1948, p. 480). But the number of specimens so far reported from the western side of the Atlantic, north or south, has not been large enough to suggest that *fer7iandinus is as plentiful there, or as generally distributed, as it is in the eastern side, and in the Mediterranean. The reports that have come to hand for *fer7iandinus in the Indo-Pacific are numerous enough, and are distributed widely enough to show that it is as widespread there as it is in the At- lantic, not only in the temperate zones both north and south but BIGELOW AND SCHROEDER : SQUALOID SHARKS 33 in tropical-subtropical latitudes as well. Thus *fernandinus, or a form so closely allied to it that no significant differences appear from the published accounts, is widespread in Japanese waters, whence it has been reported as mitsukurii Jordan and Fowler 1903 (their description, but not their illustration, which was based on a specimen of the acanthias division, see above p. 31^) ; repeatedly also as japonicus Tanaka 1917 (for an excellent de- scription and illustration, see Tanaka 1917, p. 467, PI, 30, figs. 366, 367). It has also been described and pictured from southern China by Fang and Wang (1932, pp. 249, 250, fig. 16), not only under the name mitsukurii Jordan and Fowler 1903, but also as brevirostris Tanaka 1917, though this last name had actually been based on the Japanese representatives of the meg- alops division, not on one of the fernandinus division (see below, p. 36). The specific identity of the spiny dogfishes that had been re- ported by Snyder (1904, p. 515), by Jordan and Evermann (1905, p. 45), by Gilbert (1905, p. 580), and by Giinther (1910, p. 490), from the Hawaiian Islands under the name mitsukurii Jordan and Fowler 1903, had remained uncertain, for the brief descriptions seemed equally applicable to the fernandinus di- vision of the genus or to the megalops division. But our recent examination of five of Jordan and Evermann 's and Gilbert's specimens has shown nothing to separate them specifically from the specimens of ^fernandinus which we have seen from the offing of South Carolina, from the Gulf of Mexico, from the Mediter- ranean, and from Juan Fernandez (type locality), whether in relative positions of fins, in bodily proportions, or in the shape of the denticles on the sides of the body. *S. fernandinus has also been reported from the Philippines under its own name (with brief description) by Herre (1923, p. 73; 1934, p. 12) ; also as philippinus Smith and Radcliffe (1912, p. 677, PI. 51, fig. 1), a supposedly new species based, however, on a specimen that is not separable from fernandinus either in fin characters or in its denticles, as Mr. Stewart Springer informs us from a recent examination of it in the U. S. National Museum. We should note in passing that Whitley (1931, p. 310) had pro- posed a new name, montalbani, to replace philippinus Smith and Radcliffe, the latter name having long been preoccupied within 8 On this point, see also Jordan and Hubbs 1925, pp. 105-106. 34 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY a S3 o BIOELOW AND SCHROEDER : SQUALOID SHARKS 35 the genus Squalus by philippimis Shaw 1804 (p. 341), proposed for an Australian heterodontid. For details see Whitley. *S. fernandinus, to continue, has been reported from Mauritius, both as hlainvillei (by Sauvage 1891, p. 11) and as acutipinnis (by Regan 1908, p. 47). It has long been known also in Aus- tralian waters, where it seems first to have been reported, with at least presumptive evidence of identity, by Giinther (1870, p. 419) as 'blainvillei, and subsequently by that same name but with- out description by MacLeay (1881), by Ogilby (1889, p. 185), and by Lucas (1890, p. 44). A confusing feature here is that the only Australian report of a spiny dogfish under the name *fer- nandinus, the identification of which was supported by a de- scription or by an illustration (Waite 1921, p. 23, fig. 30), seems not to have been based on that species, but on the local repre- sentative of the megalops division of the genus. Eegan's (1908, p. 46) identification, however, of a spiny dog- fish from Tasmania as * fernandinus was no doubt correct, for we have found nothing to separate the type specimen of *tasmanien- sis Rivero 1936 (now in the Museum of Comparative Zoology) from our specimen of *fernandinus from the type locality of the species (Juan Fernandez). For an extensive list of Indo-Pacific records for fernandinus see Herre 1953, p. 31. With the evidence of widespread occurrence so conclusive for fernandinus for the Australian region in general, it seems astonishing that there has been no report yet for New Zealand waters of a spiny dogfish, the accompanying account of which places it definitely in *' fernandinus. In fact, the only report under that name of a New Zealand Squalus that includes any clear evi- Fig. 5, A, Etmopterus bnllisi, new species, juvenile male, 230 mm. long (U. S. Nat. Mus. No. 158186), off northeastern Florida. B, upper teeth of same, x about 9. C, lower teeth of same x about 12. D, dermal denticles of same, from side below first dorsal fin, x about 30. E, Etmopterus polli, immature male about 226 mm. long, off south equatorial West Africa, Mus. Comp. Zool. No. 38000, lower surface of anterior part of head, x about 1.4. F, Etmopterus spvnax, Norway, to show black pattern on posterior part of trunk. G, corresponding drawings of E. virens, northern part of Gulf of Mexico. H, corresponding drawings of E. hillianus, north coast of Cuba; F-H, X about 0.4, after Bigelow, Schroeder and Springer 1953, Fig. 6. I, corresponding drawing of E. ahernathyi, type specimen, x about 0.4, after Garrick 1957, Fig. 3C. 36 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY dence of specific identity (Waite 1909, p. 142, PI. 16, fig. 1; Phillipps 1928, p. 223, fig. 3) was evidently based on a member of the acanthias division. Thus the accompanying illustration shows its sides as white spotted, its first dorsal fin as originating posterior to the rear inner corner of the pectorals when the lat- ter are laid back, and the mid-point of the bases of its pelvics as nearer to the origin of the second dorsal than to the rear end of the base of the first dorsal. It is astonishing, also, that while *fernandinus must be com- mon around Juan Fernandez, to judge from the number of speci- mens that have found their way thence to various museums (Del- fin 1901, p. 21), and is reported from northern Argentine waters, on the conclusive evidence of a small one found in the stomach of an albatross (Lahille 1928, p. 327, fig. 17), we have found no positive report of its presence in continental waters along the coast of Chile. Thus Philippi's (1887, p. 27) inclusion of it (under the name firmandezianus) in his account of the sharks of Chile seeminglj^ harked back to Molina's original report of it from Juan Fernandez. Neither does Delfin (1901) mention any Chilean localities for it other than Juan Fernandez, which is some 240 miles ofl^ the Chilean coastline ; nor does Fowler's (1930, p. 495) nominal reference of it to "Chile" include any con- tinental locality. Neither has it been reported from Peru, nor from the Pacific coast of Central or North America, for that matter. Megalops division. The earliest adequate account (with illus- tration) of a representative of this was by McCulloch (1927, p. 9, PI. 2, fig. 25a; illustration later copied by "Whitley 1940, p. 138, fig. 147) of an Australian shark that was identified by him as *megalo2)s MacLeay 1881. And Dr. Whitley writes us that he has verified this identification, by comparison with the type specimen of *megalops, which he has had the kindness to ex- amine at our request. The accompanying illustration (Fig. 4B) reproduces a tracing of its pectoral fin which Dr. Whitley has kindly sent us. We are also fortunate enough to have at hand three excellent specimens of *megalops, 526-550 mm. long, from New South Wales, for which the Museum is indebted to Col. John K. Howard. The Japanese brevirostris, described and beautifully pictured by Tanaka (1917, p. 464, PI. 129, figs. 362, 363; PI. 130, fig. 364) BIQELOW AND SCHROEDER : SQUALOID SHARKS 37 resembles *m€lagops so closely that we have no doubt a com- parison of specimens from the two regions would lead to their union as ^megalops, this being the older of the two names. The *megalops division is represented in the Atlantic by *cuhensis Rivero 1936, which is known from Cuban waters (type region), from the northwestern part of the Gulf of Mexico, (specimen recently taken by "Oregon" and referred to us by Mr. Stewart Springer), probably Trinidad and from Rio de Jan- eiro, whence it was reported by Ribeiro (1907, p. 168) under the name hlainvillei (Bigelow and Schroeder 1948, p. 478, footnote 63). *S. cuhensis closelj^ resembles *;S^. megalops in proportional dimensions, and in the shape of the denticles on the sides of its body (Fig. 3), but differs in a relatively longer first dorsal fin spine, in somewhat differently shaped caudal and 'pectoral fins (Fig. 4), and in the location (relative to the dorsal fins) of the pelvic fins, the midpoint of the bases of which are about midway between the rear end of the base of the first dorsal and the origin of the second dorsal in * cuhensis (a little nearer to the rear end of the base of the first dorsal than to the origin of the second dorsal in * megalops). No representative of this division of the genus has yet been reported from the eastern side of the Atlantic, or from the Mediterranean. Genus CiRRHIGALEUS Tanaka 1912 Cirrhigaleus Tanaka 1912, type species C. harbifer Tanaka 1912b, p. 151, PI. 41, figs. 156-162; Japan. Generic synonym : Phaenopogon Herre 1935, type species P. harhulifer Herre 1935, p. 123, fig. 1; Japan. Generic characters. In general as in Squalus (p. 26), but without precaudal pit either above or below; posterior part of trunk without longitudinal ridges; and anterior margin of nos- tril, near inner end, with a well developed and very conspicuous fleshy barbel tapering to a slender tip (a feature unique among squaloid sharks) ; dermal denticles on sides of body with tri- dentate rear margins, their outer surface with three correspond- ing ridges. Also, in the only species that has yet been seen, the back is more highly arched than it is in any known Squalus, and 38 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY the body is subtriangular in cross section. Maximum recorded length 855 mm. Remarks. Garman (1913, p. 457), followed by Fowler (1941, pp. 255, 275) reduced Cirrhigaleus to a subgenus of Squalus in rank. But the presence of the nasal barbel, combined with the lack of precaudal pits and of longitudinal-lateral ridges on the caudal peduncle seem fully to justify its retention as a distinct genus. Phaenopogon Herre, proposed without mention of Tan- aka's (1912) earlier account of Cirrhigaleus, is clearly equivalent to the latter, to the synonymy of which Herre himself (1936, p. 59) has already relegated it. Species. One species only is known, C. harhifer Tanaka 1912, which was based on a single male, 855 mm. long, from the Tokyo market. A second specimen (a female of 555 mm.) from Misaki, Japan, has more recently been described and pictured by Herre (1935, p. 123, fig. 1) as Phaenopogon barhulifer. Genus CeNTEOSCYLLIUM Miiller and Henle 1841 Centroscyllium Miiller and Henle 1841, p. 191. Type species *Spinax fabricii Eeinhardt 1825, West Greenland. Generic synonym: Paracentroscyllium Alcock 1889, p. 379, type species P. ornatum Alcock, Bay of Bengal. Generic characters. Squalinae with upper teeth similar to lowers, erect and symmetrical, or nearly so, all along the jaw ; the teeth in both jaws with 3-5 cusps, the median cusp much the largest; dorsal fin spines well developed, with two longi- tudinal grooves on either side ; lower posterior margin of caudal fin with subterminal notch (but see discussion, p. 46) ; caudal peduncle without precaudal notch either above or below, and without lateral-longitudinal ridges ; inner corner of pectoral fins rounded in known species; snout in front of mouth not more than one-half as long as head to origin of pectoral fins; pre- oral clefts present, slightly expanded inwardly; dermal denti- cles thorn-like, on stellate bases; newborn specimens naked, the denticles developing first along the back, spreading later to sides, and subsequently to the lower surface (see further, p. 41) ; some species, at least, with thickenings of the skin that presum- ably are luminous. Maximum recorded length 829 mm., for C. fahricii. BIQELOW AND SCHROEDER : SQUALOID SHARKS 39 Depth range. The only member of this genus for which we have more than scattered records {fabricii, p. 41) seldom (if ever) has been taken shoaler than 275 meters; its center of abundance lies deeper than about 500 meters. It is known to occur down at least to 1100 meters, and there is a somewhat questionable report of it from about 1495 meters (for details see Bigelow and Schroeder, 1948, p. 485). Depth records for other members of the genus have been from between 448 and 1024 meters. Remarks. The few known species that fall in Centroscyllium as generally defined are set apart from all other known mem- bers of the subfamily Squalinae by their dentition, the teeth having 3-5 cusps in both jaws. In this respect they are ap- proached the most nearly by Etmopterus, in which the uppers have 3-7 cusps, but the lowers have only one. In all other known members of the subfamily the teeth (uppers as well as lowers) have only one cusp. The genus Paracentroscyllium was proposed by Alcock (1889, p. 379) for three juvenile sharks, 5i/4 inches long, from the Bay of Bengal, allied to Centroscyllium except that they had "mono- cuspid teeth" and an "absolutely smooth integument." But the nakedness of the skin was clearly a juvenile character (this accords with conditions in *fahricii, see p. 41) ; and the lateral cusps on the teeth seem merely to have escaped notice, for Al- cock's (1896, p. 308) subsequent study of four additional speci- mens up to nearly one foot long from the Arabian Sea led him to emend his original diagnosis of ornatum to read "minute tri- cuspid teeth in both jaws" and body "covered with minute placoid deciduous scales." Accordingly, in 1896 (pp. 308, 310) he united Paracevitroscyllium with Centroscyllium. Hence it was as a Centroscyllium that he pictured ornatum in 1900 (PI. 35, figs. 1-lB, including a figure of one of its denticles). And Fowler's (1941, p. 252) revival of Paracentroscyllium, as a sub- genus of Centroscyllium, for species in which the "caudal ends in point posteriorly, lower edge of fin without distinct posterior notch," (harking back, no doubt, to Regan's [1908, p. 40] key to the species of Centroscyllium), runs counter to Burkhardt's (1900, p. 567, fig. 8) illustration of one of Alcock 's original specimens, for this shows the caudal as blunt-tipped and with a distinct (though shallow) subterminal notch. 40 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY CO OS o BIGELOW AND SCHROEDER : SQUALOID SHARKS 41 Species. The following supposedly separate species fall in Centroscyllium : *fabricii Reinhardt 1825 (type species), both sides of the northern Atlantic ; grayiulatum Giinther 1887, Falk- land Islands ; ornatum Alcock 1889, Bay of Bengal and Arabian Gulf; *nigrum Garman 1899 (inel. *ruscosiim Gilbert 1905), eastern Pacific off Panama (vicinity of the Galapagos, vicinity of Cocos Island), also Hawaiian region; and *'ritteri Jordan and Fowler 1903, Japan. *C. fahricii, the common ''black dogfish" of English-speaking North Atlantic fishermen, has been described by several authors. Even so, examination of large catches made recently off the Nova Scotian coast (Bigelow and Schroeder 1954, p. 40) enables us to expand our earlier account (Bigelow and Schroeder 1948, p. 482) in the following respects. The teeth are already tricuspid in both jaws on specimens not more than 165 mm. long (suggest- ing that this is already the case at the time of birth), and the skin, then, is wholly naked above as well as below, just as Alcock (1889, p. 379) described it for the original (133 mm.) speci- mens of his ornatum, from the Bay of Bengal. But a * fahricii of 171 mm. already shows a dense band of denticles along the back, with a few on the upper part of the sides below the second dor- sal fin. On one of 177 mm., the back as a whole, and the sides above the mid-level are rough with denticles, though the lower surface still is smooth. And the situation is much the same on one of 273 mm. But on specimens of 300 mm. and larger, the belly, like the upper parts, is strewn with denticles. The color pattern, too, undergoes an interesting alteration with growth. The smallest specimen we have seen (165 mm. long) is ink-black below and slightly paler along the back and on the basal parts of the fins, with the two dorsals and the pectorals margined and the pelvics tipped with white, and with Fig. 6. A, Etmopterus lucifer, female 337 mm. long, Japan, Mus. Comp. Zool. No. 1109. B, outlines of juvenile male, 267 mm. long, Japan, Mus. Comp. Zool. No. 1123, to show the distribution of the black, presumably luminescent, markings. C, denticles from side of body below first dorsal fin of specimen shown in A, x about 35. D, Etmopterus brachyurus, male, 300 mm. long, Japan, after Smitii and Radcliffe with the black markings added from the type specimen in the U. S. National Museum. E, ventral view of posterior part of trunk of same specimen. 42 BULLETIN : MUSEUM OF COMPAEATIVE ZOOLOGY white fin spines. The development of the denticles (see above) is accompanied by a paling to deep chocolate brown, first along the mid-zone of the back, then along the upper part of the sides, as illustrated by specimens of 171-177mm. (see above). The belly, however, continues black and the fins continue white- edged up to a length of 200-225 mm. But it is only the extreme tips of the dorsal fins that show any trace of white by the time a length of 280-330 mm. is reached, and all the fins on larger specimens are uniformly blackish right out to the margins. Meantime, the back and the upper part of the sides darken again after the dermal denticles have spread to the lower surface, so that half-grown specimens, and larger, are black above as well as below when taken from the water, unless the skin has been rubbed off, as often happens in the trawl. We have seen no ex- ception to this rule among the many specimens we have handled during the cruises of "Caryn" and of "Capn. Bill II" during 1949, 1952 and 1953 (Bigelow and Schroeder 1954, p. 41). Our earlier account of the color of adult *fahricii as "deep chocolate brown, darkest (almost black) below and on the fins generally" (Bigelow and Schroeder 1948, p. 485) was based on a specimen that had reposed for several years in alcohol. We think it evident, from the foregoing, that changes with growth must be taken into account in any discussion of the pos- sible taxonomic significance of fin-markings and of the regional distribution of dermal denticles in this particular genus. The North Atlantic, Japanese, and Falkland Island repre- sentatives of the genus all seem to deserve continued recognition as separate species, though they resemble one another closely in general appearance. Thus the distance from the upper origin of the caudal fin to the rear end of the base of the second dorsal fin is as long as from the eye to the level of the 4th-5th gill openings in *ritteri of which we have an excellent specimen at hand, about 425 mm. long (Fig. 7A), whereas in *fahricii of about that same size the caudal peduncle (measured as above) is only about as long as from the eye to the first gill openings. This contrast, indeed, was invoked as alternative between the Japan- ese and North Atlantic populations in our earlier key to the species of Centroscylliuni (Bigelow and Schroeder 1948, p. 482). The denticles, too, on specimens of about equal sizes are some- what sparser on *ritteri than on *fahricii; as an illustration, BIGELOW AND SCHROEDER : SQUALOID SHARKS 43 7 or 8 denticles on the side, below the first dorsal fin, counted along a nearly straight line, occupy a length of 1 cm. on our *ritteri of 425 mm., but 8 to 12 denticles occupy this space on the specimens of *fahricii of about 400 mm. The general coloration of the body seems also to be distinctive here, for all the *fahricii we have handled, large enough for the denticles to have spread to the lower surface (p. 41), have been plain above as well as below, not only when fresh but after a few years' stay in alcohol. But the two original specimens of *ritteri, taken three years previous, are described as "uniform dark grayish brown, blackish below in front" (Jordan and Fowler, 1903, p. 635). And this applies equally to the alcoholic specimen of ^ritteri we have at hand, the lower surface of which is densely- denticulate, while it also shows a definite black flank mark on either side of the rearward part of the trunk (Fig. 7A). Judging from Burkhardt's (1900, p. 567, fig. 7) outline draw- ing of the type specimen (270 mm. long), C. granulatum Giinther 1887, from the Falkland Islands, agrees more closely with *ritteri than with *fabricii in the length of the caudal peduncle relative to the dimensions of the head. But Giinther 's (1887, p. 7) characterization of the dermal denticles of granu- latum as "much coarser" than those of '*fahricii "and in the form of granulations" sets it apart equally from *ritteri, the thorn-like denticles of which (Fig. 7C) more nearly recall those of *fabricii (Bigelow and Schroeder 1948, fig. 91), except that thev are relativelv somewhat longer and more slender. With *riiteri clearly separable from *fahricii, and granulatum seemingly so, the same might have been expected to hold good for ^nigrum (including *ruscosum Gilbert 1905) ; or at the least, a closer relationship might have been expected with *ritteri than with *fahricii on geographic grounds. Actually, however, *ni- grum is pictured both by Garman (1899, PI. 1, fig. 2) and by Beebe and Tee Van (1941, p. 120, fig. 32), as falling with *fah- ricii in the shortness of its caudal peduncle relative to the di- mensions of the head, which w^e can corroborate from our exam- ination of one of Gilbert's original Hawaiian specimens of his *ruscosum, about 400 mm. long (the co-type), loaned us by the Stanford University Museum through the kind offices of Dr. 44 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY H O I— I BIGELOW AND SCHROEDER : SQUALOID SHARKS 45 George Myers,® and of the type specimen in the U. S. National Museum. Nor can we find anything in the dermal denticles to separate *nignim from *fahricii. Also, Garman's (1913, p. 230) use of the number of cusps on the lower teeth as alternative between * nigrum (with five) and *fahricii (with three) has failed to withstand the test of time, for most of the lowers midway along either side of the jaw actually have five cusps in *fahricii (Bigelow and Schroeder 1948, p. 483, fig. 91E). And Beebe and Tee Van (1941, p. 121) have found that on their Cocos Island specimens of *nigrum the outermost pair of cusps were not as prominent as pictured by Garman (1899, PI. 4, fig. 6). *C. nigrum, however, maturing at a length of only about 400 mm. (Gilbert 1905, p. 581) or perhaps smaller still (Beebe and Tee Van 1941, p. 120) is a much smaller shark than *fabricii, the males of which seemingly do not mature until at least 600 mm. long. Also, the pectoral and dorsal fins continue white tipped to maturity in *nigrum (Gilbert 1905, p. 581, fig. 230; Beebe and Tee Van 1941, p. 120, fig. 32) though with the white areas somew^hat reduced, whereas half-grown specimens of */a&- ricii, and larger, show no trace of the white fin-markings that are so conspicuous on small specimens (p. 42). It may also prove (when more specimens have been examined) that it is characteristic for the pectorals to extend somewhat farther rear- ward in *nigrum than in *fabricii, for they are pictured both by 9 The larger of Garman's original two specimens of *nigrum, used as the basis for his account of the anatomy of the genus GentroscylUum, is now in a verj- fragmentary condition. Fig. 7. A, Centroscyllium ritteri, female about 425 mm. long, Mus. Comp. Zool. No. 1370. B, second dorsal fin spine of same drawn to larger scale to show the longitudinal grooving. C, dermal denticles from side below first dorsal fin, x about 6 to 8. D, upper and lower teeth from central part of jaw, X about 4. E, upper and lower teeth from midway along side of jaw, X about 4. F, Centrophorus foUaceus, immature male, 351 mm. long, in U. S. Nat, Mus., from Japan. G, upper and lower teeth of same at center of mouth, X about 4. H, denticles of same from side below first dorsal fin, X about 13. I, lateral view of denticle, x about 18. J, Centrophorus armatus, southern Africa, second dorsal spine of specimen 330 mm, long, X about 3,5, from a photograph kindly contributed by Dr. J, L. B, Smith (see p. 6). 46 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Garman (1899, PI. 1, fig. 2) and by Beebe and Tee Van (1941, p. 120, fig. 32) as reaching nearly or quite to a perpendicular at the point of emergence (from the skin) of the first dorsal spine, which is true also of the Hawaiian specimen we have seen (men- tioned above). It therefore seems justifiable to accept ^nigrum as a distinct species, at least until a larger number of specimens from the Pacific populations have been studied with a critical eye. C. ornatum of the northern Indian Ocean falls with the fah- ricii-nigrum-r Uteri group in its thorn-like denticles (Alcock 1900, PI. 35, fig. 1). It agrees more nearly in proportional di- mensions with *nigrum of low latitudes in the eastern Pacific than with *ritteri of the northwestern Pacific, the distance from the rear end of the base of its second dorsal fin to the origin of the upper side of its caudal fin being only about as long as from the eye to the first pair of gill openings, while its pectorals reach rearward about to a perpendicular at the point of emergence of the first dorsal fin spine. But Alcock (1896, p. 308) describes the fins as well as the trunk as "uniform jet black" on speci- mens nearly one foot long. Neither does his account of the type specimen, 133 mm. long, nor his illustration of it (Alcock 1894, PI. 8, fig. 2) suggest any white fin-markings, such as are con- spicuous on nigrum to maturity. While Regan (1908, pp. 40-41) characterized ornatum as without a distinct posterior notch on the caudal fin, seemingly following Alcock 's (1894, PI. 8, fig. 2) original illustration, Burkhardt's (1900, p. 567, fig. 6) drawing of one of the original series in the British Museum clearly indi- cates a subterminal notch. And Alcock 's (1900, PI. 35, fig. 1) second representation of the caudal of a larger specimen seems more fanciful than realistic, so widely does it differ from his earlier illustrations. With our knowledge of ornatum so far from adequate, and with the specimens of it in the collection of the Zoological Survey of India in a bad state of preservation (the late Dr. S. L. Hora has so written us in a recent letter), a final decision as to its status obviously is a matter for some future student. Meanwhile we believe a tentative key to the species of Centroscyllium, using the length of the caudal peduncle as the first alternative, seems likely to be more helpful for identifica- tion than our earlier key, in which the primary alternative is the location of the rear margin of the pectoral fin (when laid back) relative to the location of the first dorsal fin spine. BIGELOW AND SCHROEDER : SQUALOID SHARKS 47 Provisional Key to Species of C entroscylUum 1. Distance from rear end of base of second dorsal fin to origin of upper side of caudal fin nearly or quite as long as from eye to level of origin of pectoral fins *ritteri Jordan and Fowler 1903 Japan (Fig. 7 A,B,C). p. 42 Distance from rear end of base of second dorsal fin to origin of upper side of caudal fin only about as long as from eye to first-second gill openings 2 2. Dermal denticles on anterior part of trunk "in the form of granula- tions" (Giinther 1887, p. 7) fjranuJatum Giinther 1887 Falkland Islands, p. 43 Dermal denticles in the form of spines or thorns on stellate bases .... 3 3. Margin of pectorals, when laid back, falls short of a perpendicular at point of emergence from skin of first dorsal spine by a distance at least 25 per cent as long as eye *fabricii Eeinhardt 1825. North Atlantic, p. 41 Margin of pectorals, when laid back, reaches nearly or quite to below point of emergence from skin of first dorsal spine 4 4. Dorsal, pectoral and pelvic fins white edged at all stages in growth . *nigrum Garman 1899 (incl. ruscosum Gilbert 1905), east- ern tropical Pacific off Central America, also Hawaiian region, p. 43. Fins, as well as body, black at all stages in growth ornatum Alcock 1889 Bay of Bengal and Arabian Gulf. p. 46 Genus EtmOPTEEUS Eafinesque 1810 Etmoptents Eafinesque 1810, p. 14; type species E. aculeatus Eafinesque, Mediterranean, equals *Squalus spinax Linnaeus 1758. Generic syn- onyms, see Bigelow and Schroeder 1948, p. 487. Generic characters. Essentially as in CentroscylUum (p. 38), except with lower teeth much as in Squalus, i.e., much broader than the uppers and with only one cusp, the base of each tooth overlapping that of the next outward so that each row forms a continuous dental band, and with the cusps directed so strongly outward all along the jaw that their successive inner margins act as a nearly unbroken cutting edge (Bigelow, Schroeder and Springer 1953, figs. 7-10) ; dermal denticles on sides of body 48 BULLETIN : MUSEUM OF COMPAEATIVE ZOOLOGY ranging from thorn-like to bristle-like and to truncate, on quad- rate bases. Size. At maturity the various species of Etmopterus range in length from slightly more than 200 mm. {*virens, perhaps also *villosus) to a recorded maximum of 728 mm. in the case of *princeps (Collett 1904, p. 4). Depth range. The members of this genus are creatures of at least moderately deep water, where (as a group) their depth range is a wide one. The shoalest captures we find recorded for any of them are 73-91 meters for *spi7iax in the northernmost parts of its range (Smitt 1895, p. 1165) and 183 meters for ^abernathyi, off New Zealand (Garrick 1957, p. 181) ; the deep- est are 2074 meters for E. princeps (see Bigelow, Schroeder and Springer 1953, p. 251; Bigelow and Schroeder 1954, p. 47; and Grey 1956, p. 95 for details as to * princeps). Most of the records for one species or another in various parts of the world have been from about 300 meters to about 900-1000 meters. Remarks. These small, black-bellied deep-water sharks, of which some 16 recognizably distinct species are known, form a very compact unit. Characters that have proved the most use- ful for diagnosis of species within the genus are : the sizes and relative positions of the fins; the color pattern, whether uni- formly as dark above as below or paler above with conspicuous dark flank markings of characteristic shape, whether or not with complex patterns of longitudinal bands of black dashes and dots (presumably luminescent) and whether with or not with white fin markings; also, for half-grown specimens and larger, the shapes of the dermal denticles, and their arrangement. But in using the denticles for diagnosis allowance must be made for the size of the specimens in question, for we have found that in one species {"^'princeps) where the denticles are bristle-like on very young specimens, they are succeeded by thorn-like forms on half- grown individuals and larger (Fig. 3, F, G). For accounts of the luminescent organs (for the species that have them) and of luminosity in this genus, see Johann (1899) and Oshima (1911). Whitley (1939, p. 266) has revived the genus Acanthidium Lowe 1839, type species *A. pusillum Lowe 1839 {=Centrina nigra Lowe 1834, preoccupied), designated by Jordan and Ever- mann (1896, p. 55), as distinct from Etmopterus Rafinesque BIGELOW AND SCHROEDER : SQUALOID SHARKS 49 1810, type species E. aculeatus Rafinesque, {=*Squalus spinax Linnaeus 1758). And Fowler (1941, p. 251) has recognized two corresponding subgenera within the genus Etmopterus, namely Acanthidivm Lowe 1839, with "second dorsal origin be- hind ventral base ; first dorsal origin midway between orbit and second dorsal origin," and Etmopterus Rafinesque 1810 with "second dorsal origin over or a little before ventral base; ventral origin slightly nearer subcaudal origin than pectoral origin." But these differences do not seem to us wide enough for generic — or even subgenerie — separation (p. 22; Bigelow, Schroeder and Springer 1953, p. 238). Acanthidium as employed by Gar- man (1913, p. 215) in a different sense, is discussed on p. 21. Species. The genus Etmopterus is especially interesting from the taxonomic standpoint, because it is already known to include a larger number of recognizable species than we have any reason to suppose is included in any other genus of squaloid sharks. These species, it is true, resemble one another so closely in gen- eral aspect that any one of them might easily be mistaken for any other, on cursory inspection. But closer examination has shown that they are sharply separated by differences both in the relative positions of the fins and in the nature and arrangement of the dermal denticles, and also in the color pattern which has proved a much more convenient species-character here than it is usually thought to be among other sharks. The following seven species have been known from the North and Equatorial Atlantic, including the Mediterranean and the Gulf of Mexico. 1. *spinax Linnaeus 1758, type species of the genus (many of the reports of it have been as niger Cloquet 1820, p. 93) ; eastern North Atlantic (including the Mediterranean) from the Cape Verde Islands, Morocco and the Azores to Norway. 2. *pusillus Lowe 1839 {=Centrina nigra Lowe 1834, but nig7'a is preoccupied in Etmopterus by niger Cloquet 1820, pro- posed in substitution for spinax Linnaeus 1758 and employed by many subsequent authors) ; tropical-subtropical Atlantic, from tropical West Africa to the Canaries, Madeira, and the Azores in the east (including the Mediterranean) and Gulf of Mexico in the west, where it has been found recently (Bigelow, Schroeder and Springer 1955, p. 2). The Japanese shark de- scribed and pictured under the name pusillus by Tanaka (1912, 50 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY PI. 22; 1912^, p. 88) was based on a clearly separable form (see below, p. 58). 3. *hillianus Poey 1861, Cuban-West Indian region, north- ward to the ofiEing of Chesapeake Bay. 4. *princeps Collett 1904, both sides of the North Atlantic, P'aroes-Hebrides region and offing of Gibraltar in the east; con- tinental slope in the west where it is common from the offing of southern Nova Scotia to that of southern New England (Bigelow and Schroeder 1954, p. 46). 5. *polli Bigelow, Schroeder and Springer 1953, equatorial West Africa. 6. *schultzi Bigelow, Schroeder and Springer 1953, northern part of the Gulf of Mexico. 7. *'virens Bigelow, Schroeder and Springer 1953, northern part of the Gulf of Mexico. For the diagnostic characters of the members of this group, see Bigelow, Schroeder and Springer 1953, p. 237. An eighth North Atlantic Etmopterus, from the east coast of Florida, is described here as a new species, *E. hnllisi. Etmopterus bullisi n. sp. Figure 5 A-D Plate 2 Type specimen. Female, 196 mm. long; "Pelican" Sta. 42, off northeast coast of Florida, Lat. 30°02' N, Long. 80°05' W; 205 fathoms (U. S. Nat. Mus. No. 158186). Additional material : female of 205 mm. and immature male of 230 mm. from this same locality; also immature male of 212 mm., "Pelican" Sta. 51, off eastern Florida, Lat. 29°48' N, Long. 80°09' W, about 200 fathoms. Description. Proportional dimensions in per cent of total length of type specimen, and of a male 230 mm. long, to nearest 0.1 per cent. Snout length in front of: outer nostrils 2.5, 2.8; mouth 11.7, 12.2. Eye: horizontal diameter 4.6, 4.6. Mouth: breadth 7.1, 7.0. Nostrils: distance between inner ends 2.8, 2.8. Gill openings: lengths first 2.3, 1.7; third 2.3, 1.7; fifth 2.0, 1.5. BIGELOW AND SCHROEDER : SQUALOID SHARKS 51 First dorsal fin: vertical height 3.1, 2.8; length of base 4.6, 4.4. Second dorsal fin: vertical height 4.6, 4.4; length of base 7.4, 7.0 Caudal fin: upper margin 24.0, 23.5. Pectoral fin: outer margin 11.0, 10.0; inner margin 4.8, 5.6, width 4.6, 5.2. Distance from snout to: 1st dorsal 33.7, 33.0; 2nd dorsal 56.6, 57.9; upper caudal 76.0, 76.5; pectorals 25.0, 24.4; pelvics 51.5, 50.0. Interspace hctween : First and second dorsals 18.3, 18.7 ; 2nd dorsal and caudal 12.7, 13.9; base of pelvics and caudal 18.9, 20.4. 20 18 Teeth : — , — 27, 31 Description of type specimen. Trunk noticeably slender, its height at pectorals (where highest) about 12 per cent as great as its length to upper origin of caudal fin, its greatest thickness about equal to its greatest height ; body narrowing rearward with caudal peduncle nearly as thick as high. Head flattened above, its length to origin of pectorals occupying about 33 per cent of trunk to origin of caudal fin. Snout obtusely wedge-shaped anteriorly, its length in front of mouth about 50 per cent as great as length of head to origin of pectorals, its length in front of eyes about 28 per cent. Eye about 1.4 times as long as high, its horizontal diameter about 40 per cent as long as snout to mouth. Nostril with outer end close to edge of snout; anterior narial flap triangular, with pointed tip, reaching across narial opening. Pores on lower surface of snout so small that they are visible only on close scrutiny. Mouth low-arched, occupying about 80 per cent of breadth of head. Preoral clefts short, extending about 33 per cent the distance from corners of mouth toward nostrils. Fur- rows from corners of mouth reaching rearward half way toward first gill openings. First gill openings a little shorter than distance between nos- trils ; fifth openings a little shorter than first to fourth ; an- terior margins only slightly concave, but tips of filaments ex- posed in first to third. Teeth smooth edged as in other mem- bers of this genus ; uppers with only 3 cusps, the axial much the largest. Lowers with cutting edge nearly parallel with genera] trend of the jaw. Cusp of the median lower tooth, identified as such by its base overlapping that of the next tooth on either side, directed strongly outward, to the one side or the other, on all specimens seen. Two or three rows functional in upper jaw hut one row only in lower jaw. 52 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Dermal denticles low, eonical-thorn-like, erect or nearly so and so sharp that the skin feels very rough. Denticles on lower surface straight, or nearly so, in random arrangement, but those on sides and back slightly curved rearward and arranged in single series in regular longitudinal rows (PI. 2) ; the rows on the back anterior to the first dorsal fin converging rearward, but the rows on the sides paralleling in general the main axis of the trunk as far rearward as abreast the second dorsal, whence those above the mid-level trend downward-rearward along the tail sector and out along the caudal axis. The lower surface of the trunk, like sides and back, is closely set with denticles, except for the chin, the lips, a small area in the midline of the snout close in front of the mouth, and the margins of the nostrils, which are naked. Fins as a whole naked. Base of first dorsal fin about as long as eye, its origin pos- terior to the origin of the pectorals by a distance about as long as from tip of snout to mouth. Rear end of base of first dorsal about equidistant between perpendiculars at axil of pectorals and at origin of pelvics. Distance from point of emergence from skin of first dorsal spine to point of emergence of second dorsal spine about as long as head to origin of pectorals. Second dorsal spine about 1.8 times as long as first, each measured from point of emergence from the skin, its tip reaching about 75 per cent toward the upper corner of the fin. Distance from rear end of base of second dorsal fin to origin of upper side of caudal fin about 54 per cent as long as between points of emergence from skin of first and second spines, and about 47 per cent as long as length of head to pectorals; upper side of caudal fin about as long as head to origin of pectorals, its shape as pictured in Fig- ure 5A. Lower anterior margin of caudal 43 per cent as long as upper margin. Distance from origin of lower side of caudal fin to rear end of bases of pelvics about 75 per cent as long as head to origin of pectorals. Rear end of bases of pelvics a little anterior to point of emergence, from skin, of second dorsal spine. Pec- torals truncate, with rounded corners, the distal margin, when laid back, falling about abreast of point of emergence from skin of first dorsal spine. Tn their present state the edges of the pectorals, dorsals, and pelvics, and the terminal sector of the caudal, are much frayed out on all the specimens. But close examination makes it evident BIGELOW AND SCHROEDER : SQUALOID SHARKS 53 that, normally, tliey are entire, not fringed as they are on *E. schultzi (Bigelow, Schroeder, and Springer, 1953, p. 254, Fig. 9D). Claspers of mature males not yet seen. Color. Upper parts in general verj^ dark sooty gray, the mid- line of the back paler (as a definite band on one of the speci- mens), with a vaguely outlined pale yellowish spot on the top of the head betw^een the eyes. Lower surface of head and body sectors black; the black extending rearward in a narrow band along the lower surface of the tail sector, nearly or quite as far as the origin of the caudal ; also upward on each side close be- hind the pelvics, to continue forward as a flank mark of the shape shown on Figure 5A. The pattern of coloration, however, is evident only on close examination, the sides, as a whole, being very nearly as dark as the belly. Distal parts of the dorsal, pec- toral and pelvic fins, and lower edge of caudal pale gray after preservation, perhaps whitish in life. The (presumablj^) luminescent system of markings to be seen on various other members of Etmopterus is evident here only as a row of black dots along the midline of the back from nape to base of caudal. If similar markings are present elsewhere they are masked by the dark hue and dense pigmentation of the sides. Size. The size at maturity is not yet known. Geographic range. So far known only off the northeast coast of Florida, at the localities listed above (p. 50). Remarks. The linear arrangement of the denticles on its back and sides marks *huUisi off from all the other species of Etmop- terus that were known previously from the North Atlantic and Gulf of Mexico. It differs further from *spinax in its stouter, thorn-like denticles, and in that the interspace between its first and second dorsal fins is only about as long as from tip of snout to first gill openings (as long as from tip of snout to origin of pectorals in *spinax) ; from *hillianus and from *virens in its noticeably shorter caudal peduncle, relative to the length of the head; from *pusillus in its thorn-shaped denticles (truncate with concave crown in ^pusillus) ; from *schultzi both in the na- ture of the margins of its fins (p. 52) and in a relatively shorter caudal fin (upper margin only as long as from tip of snout to origin of pectorals in *bullisi, but as long as from snout to rear edge of pectorals in * schultzi) ; from *princeps in a noticeably 54 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY longer interspace between its second dorsal and the origin of the upper side of its caudal (to mention only the most conspicuous difference) ; and from *polU in its tricuspid upper teeth (mostly with 5 cusps in *polU), in the roughness of the lower side of its snout, and in the minuteness of the pores there. Among species known from other seas, it agrees with *lucAfer (originally described from Japan) in the linear arrangement of the denticles on the sides of the head and body, and in their thorn-like nature, also with *villosus of Hawaiian waters; with ^hrachyurns, originally described from the Philippines (p. 59) ; and with *aher7iathyi, from New Zealand. But there is no dan- ger of confusing it with either one of these for the following reasons: the distance from the tip of its snout to its first dorsal fin spine is shorter relatively than in *villosus but its caudal fin is considerably longer; the upper margin of its caudal is much longer relatively than in *brachyurus ; the interspace between its two dorsal fins (not longer than from tip of snout to first gill openings) is noticeably shorter than in either *lucifer or *aher- 7iathyi (at least as long as from tip of snout to origin of pec- torals), and its flank marks are much less conspicuous (cf. Fig. 5 A with Fig. 6A) than in either of these last two. The Mediterranean shark, also, that was named Squalus in- fernus by Blainville (1825, p. 59), a century and a quarter ago, is referable to Etmopterus by Blainville 's account of its upper teeth as three-cusped, but of its lowers as one-cusped, with hori- zontal cutting edge." And there is nothing in the dimensions given for it by Blainville to set it apart from *E. spinax, as ap- pears from the accompanying comparative table. Our earlier statement (Bigelow, Schroeder and Springer 1953, p. 224) that its proportional dimensions differ from those of other Atlantic species was based on a misreading of the measurements given for it by Blainville. Proportional dimensions in per cent of total length (A) for the type specimen, 255 mm. ("dix pouces") long, of inf emits Blainville 1825; and (B) for a female * spinax, about 285 mm. long, from the Mediterranean, in the Mu- seum of Comparative Zoology. 10 Notwithstanding its tooth characters, Garman (1913, p. 197) classed it as a synonym of *CLntrophorus uyato Raflnesque 1810. We may also point out that Blainville's (1825, p. 59) reference for it to his "PI. 14, Fig. 2" was incorrect; iictually this figure represents a "Lmnna nnsua. BIGELOW AND SCHROEDER : SQUALOID SHARKS 55 A R Snout to 5th gill openings 19 21 Snout to cloaca 57 58 Snout to 1st darsal fin 31 32 Snout to 2nd dorsal fin C3 63 Cloaca to lower origin of caudal fin 23 19 *S. spinax has been reported from southern Africa as well as from the North Atlantic and Mediterranean, on the evidence of a shark that was taken off Cape Point by the "Pickle" (Gil- christ 1922, pp. 42, 49 ; Barnard 1925, p. 51 ; Smith 1949, p. 59), with two others taken in the same general region by the "Dis- covery" (Norman 1935, p. 37). And the "Pickle" specimen certainly falls with *spinax in the relative positions of its fins, as well as in the bristle-like nature and irregular distribution of its dermal denticles, to judge from what little information is available. But we still await a detailed description of it (or of the specimens taken by "Discovery") ; nor are the outlines of the black belly area and of the flank-markings shown on Smith's illustration. Final decision, in short, as to the specific relation- ship that the spinax-like form of South African waters bears to the typical *spinax of the northeastern Atlantic must await a comparison of specimens from the two ocean areas. None of the other Atlantic species has been reported from farther southward in the southern hemisphere that latitude 6°08'S {S. polli Big- elow, Schroeder and Springer 1953, p. 241). Other named sharks that can be definitely placed in Etmop- terus are the following. E. granulosus (Giinther) 1880, originally reported from the southwest coast of South America, and subsequently from Ar- gentina (Lahille 1921, p. 16, by name only) as well as from southern Africa (Gilchrist 1922, p. 41; Barnard 1925, p. 49; Smith 1949, p. 58). Barnard's reference of it to the Hawaiian Islands, without supporting date, may actually have referred to *E. villosus. The chief diagnostic features of granulosus are: distance from tip of snout to emergence from skin of first dorsal spine is at least as long as from emergence of first dorsal spine to rear end of base of second dorsal (Giinther 1880, PL 2, fig. C; Barnard 1927, PI. 2, fig. 8; Smith 1949, p. 58, fig. 50) ; dis- tance from rear end of base of pelvics to origin of lower side of caudal is shorter than from tip of snout to first gill openings; 56 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY margin of pectorals (when laid back) reaches nearly to abreast of base of first dorsal spine; dermal denticles on sides of body- described as granular, without median spine and in irregular pattern, but those on tail in the form of minute spines and ar- ranged in longitudinal bands (Giinther 1880, p. 19) ; body uni- formly dark brown to blackish, but the fins edged with white (Giinther 1880, p. 19), a feature not mentioned either by Barnard or by Smith. Incidentally, the upper side of the caudal fin is shown as about as long as from the tip of the snout to the origin of the pectoral fins both by Giinther (1880, PI. 2, fig. C) and by Barnard (1927, PI. 2, fig. 8), but only as long as from snout to first gill openings by Smith (1949, p. 58, fig. 50), a disagreement we cannot resolve. *E. lucifer Jordan and Snyder 1902, Japan (Fig. 6 A-C). The chief diagnostic features, as appear both from the original ac- count and from our own examination of 5 specimens (two in the Museum of Comparative Zoology, three in the U. S. National Museum ) are : distance from tip of snout to point of emergence from skin of first dorsal spine is about as long as from emergence of first dorsal spine to midpoint of base of second dorsal ; upper side of caudal and distance from rear end of base of pelvics to origin of lower side of caudal both are about as long as from tip of snout to origin of pectorals; rear end of base of first dor- sal is about midway between perpendiculars at origin of pelvics and at axil of pectorals ; dermal denticles slender, thorn-like, weakly curved, sloping only slightly, and close set; those on the sides arranged in general in longitudinal rows and directed downward-rearward (a conspicuous feature), but those on the black areas of the lower surface in random pattern, di- rected obliquely downward-rearward on either side of the body toward the midline, where a single row is directed rearward. Upper surface of head, anterior to spiracles, with extensive naked areas. Back and upper part of sides chocolate brown; lower surface black, with a distinctive black marking on either flank (Fig. 6A; Bigelow, Schroeder and Springer 1953, fig. 6B), the line of transition from black to brown marked by a pale band as shown on Figure 6 A. Imposed on the sides, there is also a complex pattern of black dash-like markings, presumably luminescent (Fig. 6B ; Oshima 1911, pp. 1-3), in longitudinal linear arrangement. BIGELOW AND SCHROEDER : SQUALOID SHARKS 57 Originally described from Japan, Hucifer has since been found in Sonth African waters (Gilchrist 1922, p. 49; Barnard 1925, p. 50; Norman 1935, p. 37; Smith 1949, p. 59, fig. 52). It has also been reported by name from the Philippines (Smith and Radcliffe 1912, p. 679, 21 specimens; Fowler 1941, p. 248). and from Flores Straits in the East Indies (Weber 1913, p. 597). But we think it more likely on geographic grounds that these reports from equatorial waters actually were based on *E. hrachyunis (p. 58). *E. villosus Gilbert 1905, from the Hawaiian Islands. The combination of slender, spine-like denticles (those on back and sides in linear series) with a first dorsal fin standing so far rear- ward that its point of origin is about midway between the tip of the snout and the origin of the upper side of the caudal fin (corroborated by our examination of the type specimen in the U. S. National Museum), sets ^villosus apart from all other members of the genus yet known. Other features diagnostic of *villosus are : upper side of caudal about as long as from tip of snout to first-second gill openings; distance from rear end of pelvics to origin of lower side of caudal fin only a very little longer than from snout to level of spiracles; rear end of base of first dorsal about midway between origin of pelvics and rear margin of pectorals when the latter are laid back ; abdomen and lower side of head blue-black, the sides brown, no black flank marks being suggested by Gilbert's (1905, PI. 66) illustration or by our recent examination of the type specimen; dorsals, pectorals and pelvic fins broadly edged rearward with white ; lower surface in general rough with denticles, the skin being naked only on the lips and nostrils, close around the eyes and spiracles, and close behind the dorsal, pectoral and pelvic fins, which are partially so; the denticles along the back and on the tail arranged in lengthwise series. On the type specimen, the denticles along the midline of the lower surface are smaller than they are pictured ; the first dorsal spine is only about 65 per cent as long as the second ; the upper teeth have only three cusps ; and the cusps of the low^er teeth are so strongly oblique that their inner margins are nearly parallel with the jaw. *Yillosus is so far known only from the type specimen, 170 mm. long, taken by the "Albatross" off the south coast of Molokai (Gilbert 1905, p. 580, PI. 66). 58 BULLETIN: MUSEUM OP COMPARATIVE ZOOLOGY E. frontiniaculatus Pietschmann 1907, from Japan. In gen- eral, this species resembles *E. lucifer in color, with brown back and sides contrasting with ink-black belly and black flank marks. But the latter differ in shape from those of Hucifer, as Oshima (1911, p. 2, fig. 1) has pointed out, while neither Pietschmann 's account nor his illustration (1908, p. 657, PI. 1, fig. 2) suggest the presence of the linear pattern of luminous organs to be seen on the sides of the body in Hucifer (p. 56). E. frontimaculatus is set apart still more sharply from *lucifer by its denticles, which (as pictured by Pietschmann 1908, p. 657, figs. 1, 2) are truncate, with concave crowns (slender, thorn-like in *lucifer, p. 56), while those on the sides are arranged irregularly (in linear series in "^'lucifer). Also, according to Pietschmann (1908, pp. 656-657) the skin of frontimaculatus is naked on the black areas of the lower surface and of the tail, which in Hucifer (p. 56) are as densely denticulate as the paler sides are and the back. E. frontimaculatus resembles *E. pusillus of the At- lantic in the nature of its denticles; Tanaka, in fact (1912, PI. 22, 1912^, p. 88) pictured and described a specimen from Misaki under that name. But it differs sharply from *pusillus in the nakedness of the black areas on its lower surface and on its tail ; in a somewhat more rearward position of its first dorsal fin rela- tive to the pectorals, and — more conspicuously — in its color pattern, for the black belly and flank marks contrast with paler brown sides and back which is not the case in * pusillus (com- pare Pietschmann 1908, PI. 1, fig. 2 with Bigelow, Schroeder and Springer 1955, p. 3, fig. 1). E. frontimaculatus is so far known only from Japan. *E. hrachyurvs Smith and Radcliffe 1912 from the Philip- pines. This shark (Fig. 6 D, E) falls with Hucifer in the rela- tive position of its pectoral, dorsal and pelvic fins ; in the thorn- like nature of its dermal denticles (these roughen the lower sur- face of its body as well as the upper) ; and in the arrangement in longitudinal bauds of those on the upper surface, on the caudal peduncle, and on the bases of the fins. It falls further with Hucifer in its general color pattern, with sharply outlined black flank markings, which are not shown on Smith and Radcliffe's original illustration of it. But it differs from *lucifer in a rela- tively much shorter caudal fin, the upper margin of which is only about 66 per cent as long as from the tip of the snout to BIQELOW AND SCHROEDER : SQUALOID SHARKS 59 the origin of the pectorals, as we can verify from recent examin- ation of the type specimen in the U. S. National Museum (about as long as from snout to pectorals in *lucifer). And there are five cusps on its upper teeth (three in Hucifer), with the laterals longer, relative to the median member, than in Hucifer. *E. hrachyunis was originally described from an adult male, 227 mm. long, taken off Jolo Island in the Philippines by "Alba- tross" in 1909, from 481 meters (263 fathoms). But its range extends to southern Australia, provided our union with it of mol- leri Whitley 1939 be correct (see below). E. molleri (Whitley) 1939, from Australia. This shark de- scribed by Whitley as Acanthidiuni molleri. seems not to differ from E. brachyurus in any significant respect, to judge from Whitley's description or illustration (Whitley 1939, p. 265, fig. 1; 1940, pp. 147-148, fig. 163). E. haxteri Garrick 1957, from New Zealand. This newly de- scribed Etmoptcrus, known from a female 742 mm. long, caught at 914 meters (500 fathoms) seven miles south of Kaikoura, N. Z., finds a close counterpart in E. princeps of the North At- lantic. But it differs so sharply from princeps in the respects summarized in the accompanying key (p. 63, alternatives 14-15), that the two clearly represent distinct species. E. ahernathyi Clarrick 1957, from New Zealand. This well de- fined species, described from an immature male 338 mm. long and from a female 278 mm. long, caught at 183-366 meters off' Kaikoura, N. Z., falls in the subdivision of the genus that is characterized by slender thorn-like denticles, with those on the upper part of the sides arranged in unilinear, longitudinal rows, and by a conspicuous pattern of black markings. Among the other members of this division it comes the closest to brachyurus. But it is separated from the latter by a much longer caudal fin and by a more forward position of the first dorsal fin, to men- tion only the most conspicuous differences (see Key, pp. 61, 62, alternatives 7-10). For comparison Avith other members of its geuus, we refer the reader to Garrick 's detailed account of aher- nathyi and to his excellent illustrations of it. The U. S. National Museum also contains a female Etmopterus, about 270 mm. long (U.S.N.M. No. 91576) collected among the Galapagos by the "'Albatross" in 1888, from 717 meters (392 fathoms), which agrees very closely with the West African 60 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY *polli, except that such of its upper teeth as are still intact have only three cusps, contrasting with five cusps in *polli. Un- fortunately the specimen is now in such poor condition that we dare do no more than call attention to the presence in Gala- pagan waters of a species of Etniopterus, as yet undescribed. Finally, we face the case of E. paessleri Lonnberg (1907, p. 5, fig. 2) from the Straits of Magellan. Lonnberg 's reference of this shark to Etniopterus has been accepted both by Regan (1908, p. 43) and by Garman (1913, p. 229). And while Lonn- berg states that lateral cusps were not visible ("nicht sichtbar") on its upper teeth, the type (and only known) specimen was so small (about 160 mm. long) that these structures may have been overlooked. Until this point can be settled its generic status must remain open. Meantime, we are content to let this species rest provisionally in Etniopterus. Whatever the final outcome in this regard, paessleri is set apart from all other members of Etniop- terus by the nature of its dermal denticles, which are described (they have not been pictured) as having a large central spine, surrounded by several smaller spines ("einen grosseren Stachel und mehere kleinere Nebenstacheln, " Lonnberg 1907, p. 5). Also, its second dorsal fin-spine is described as a little smaller than the first, just the reverse of the usual relationship. Key to species of Etmopterus 1. Dermal denticles with larger central spine surrounded by several smaller spines; second dorsal fin spine smaller than first dorsal spine; lateral cusps not easily visible on upper teeth of small specimens paessleri Lonnberg 1907. Straits of Magellan, p. 60 Dermal denticles citiier truncate or with a central spine only, the latter bristle-like or thorn-like ; second dorsal fin spine longer than first dorsal spine; lateral cusps on upper teeth clearly visible, even on very small specimens . . 2 2. Upper margin of caudal fin nearly as long as from tip of snout to rear edge of pectorals when these are laid back *schult2i Bigelow, Schroeder and Springer 1953. Northern part of Gulf of Mexico, p. 50 Upper margin of caudal fin very little (if any) longer than from tip of snout to origin of pectorals and considerably shorter in most 3 3. Dermal denticles on sides of body truncate, without central spine ... .4 Dermal denticles on sides of body with a central spine, either conical. BIGELOW AND SCHROEDER : SQUALOID SHARKS 61 thorn-like, or bristle-like 6 Back and upper part of sides dark brown, contrasting sharply with a l)lack marking on either Hank, and with lower surface of head and body; the black areas partially naked frontimaculattts Pietschmann 1907. Japan, p. 58 Back and upper part of sides black or sooty like lower surface; no conspicuous flank-markings ; lower surface, like upper, rough with denticles 5 Denticles on sides of tail not arranged in regular longitudinal rows, and of same shape as those on sides of body ; interspace between first and second dorsal fins about as long as from tip of snout to axil of pectorals *pusillus Lowe 1839. p. 49 Eastern Atlantic from equatorial West Africa to the coast of Portu- gal, including the Cape Verde Islands, the Canaries, Azores and Madeira; also Gulf of Mexico, p. 49. Dermal denticles on tail arranged in longitudinal rows, and more spine-like than those on the body; interspace between first and second dorsal fins shorter than from snout to first gill openings granulosus Giinther 1880. Coast of Chile, also reported from Natal coast of southern Africa and from Argentina, p. 55. Denticles on sides in definite uniserial longitudinal rows 7 Denticles on sides in random arrangement, not in longitudinal rows . . 11 Upper side of caudal fin at least not longer than from tip of snout to second gill openings 8 Upper side of caudal fin about as long as from tip of snout to origin of pectorals 9 Distance from tip of snout to first dorsal spine is about as long as from base of first dorsal spine to origin of upper side of caudal; interspace between first and second dorsal fins is only about as long as from tip of snout to level of spiracles *villosus Gilbert 1905. Hawaiian Islands, p. 57. Distance from tip of snout to first dorsal spine is little longer than from first dorsal spine to rear end of base of second dorsal fin; interspace between first and second dorsal fins is at least as long as from tip of snout to second gill openings *bradhyurus Smith and Eadcliffe 1912. Philippines; apparently in- cluding E. molleri (Whitley) 1939, southern Australia, p. 58. 62 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 9. Interspace between first and second dorsal fins is much shorter than from snout to first gill openings ; black flank markings not con- spicuous *hullisi, new species, off north- east coast of Florida, p. 50 Interspace between first and second dorsal fins is as long as from tip of snout to origin of pectorals, or longer; black flank markings conspicuous .... 10 10. Distance from base of pelvics to origin of lower side of caudal is only about as long as from tip of snout to first gill openings *al)ernathyi Garrick 1957. New Zealand, p. 5 Distance from base of pelvics to origin of lower side of caudal is as long as from tip of snout to origin of pectorals *lucifer Jordan and Snyder 1902. Japan; Atlantic off southern Africa; also reported by name from Argentina, the Philippines and the East In- dies, p. 56. For list of Philippine, East Indian, and South African records, see Herre 1953, p. 30. 11. Distance from rear end of base of pelvics to origin of lower side of caudal is at least as long as from axils of pectorals to origin of pelvics 12 Distance from rear end of base of pelvics to origin of lower side of caudal is only about V2 as long as from axil of pectorals to origin of pelvics 14 12. Distance from origin of pectorals to origin of pelvics is not longer than from tip of snout to origin of pectorals *polli Bigelow, Schroeder and Springer 1953. trop- ical West Africa, p. 50 Distance from origin of pectorals to origin of pelvics is longer than from tip of snout to origin of pectorals 13 13. Denticles on sides below first dorsal fin are slender, bristle-like (Bige low, Schroeder 1948, Fig. 92B) ; skin on lower surface of snout and in region of gill openings is rough with denticles; snout in front of spiracles about as long as from spiracles to level of axil of pectorals ; black flank markings as in Figure 5H *}iillianus (Poey) 1861. Cuban and West Indian region, and north- ward along the American slope to the offing of Chesapeake Bay. p. 50. Denticles on sides below first dorsal fin low, conical (Bigelow, Schroe- der and Springer 1953, Fig. 10, D, E) ; skin on lower surface of snout BIGELOW AND SCHROEDER : SQUALOID SHARKS 63 and in region of gill openings naked; snout in front of spiracles only about as long as from spiracles to level of origin of pectorals ; black flank markings as in figure 5G *viren.'i Bigelow, Schroeder and Springer 1953. Northern part of Gulf of Mexico, p. 50 14. Interspace between first and second dorsal fins is longer than from tip of snout to axil of pectorals; distance from rear end of second dorsal to origin of upper side of caudal is only about 35 per cent as long as interspace between first and second dorsals baxteri Garrick 1957. New Zealand, p. 59 Interspace between first and second dorsal fins is only about as long as from tip of snout to origin of pectorals; distance from rear end of second dorsal to origin of upper side of caudal is about 50 per cent as long as interspace between first and second dorsals 15 1."). Dermal denticles on sides slender, bristle-like, and closely spaced (Fig. 3E) ; breadth of head about as great as length of snout to mouth; longest gill openings only 14-33 per cent as long as eye; black flank marks conspicuous, of shape shown in Figure 5F.. . .*spinax (Linnaeus) 1758. Eastern Atlantic, including Mediterranean, from the Cape Verde Islands, Morocco and the Azores to Norway, p. 49 ; also reported from southern Africa, p. 56. Dermal denticles on sides low, conical to thorn-like (more slender on juveniles), and more loosely spaced (Fig. 3, F, G) ; breadth of head 1.2-1.4 times as great as length of snout (to mouth) ; longest gill openings 33-50 per cent as long as eye; flank markings not conspicu- ous, if visible at all *pnnceps Collett 1904. Faroes- Hebrides region and offing of Gibraltar in eastern Atlantic ; offing of southern Nova Scotia to offing of southern New England in west- ern, p. 50. Genus CexteOPHORUS Miiller and Henle 1837 Centrophorus Muller and Henle 1837, p. 398; 1937A, p. 115; 1841, p. 89; type species *C. granulosus Miiller and Henle, Mediterranean, con- sidered by them and by subsequent authors in general as the same as Squalus granulosus Bloch and Schneider 1801, no locality given. Generic Synonyms: Squalus in part, Bonnaterre, 1788, p. 12, for *S. squnmosus Bonnaterre, 64 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY type locality not known; Bloch and Schneider 1801, p. 135 for *S. granulosus Bloch and Schneider, no stated locality. Acanihorlnnus in part, Blainville 1816, p. 121, for *A. squamosals Blain- ville, equals Sqnalns squamosiis Bonnaterre 1788, and A. granulosus Blainville, equals *Sqtialus granulosus Bloch and Schneider 1801. Ccntrina in part, Cuvier 1817, p. 130, for C. squamosals Cuvier, equals *Sqnalm squamosus Bonnaterre 1788. Lcpidorhinus Bonaparte 1838, type and only included species, *Squ/ilus squamosus Bonnaterre 1788. Acanthias in part, Miiller and Henle 1841, p. 85, for A. uyatus Miiller and Ilenle, equals *Sqiialus uyato Eafincsque 1810, Mediterranean. Spinax in part, Bonaparte 1841, PI. 57, for Spinax uyatus Bonaparte, equals ^Squalus uyato Rafinesque 1810. MaohephUus Johnson 1867, p. 713, type and only included species M. duvi,eriU Johnson, Madeira. Entoxychirus Gill 1862, p. 498, type and only included species E. uyatus Gill, equals *Sqnalus uyato Rafinesque 1810. Gaboa (subgenus of Centrophorus) Whitley 1940, p. 145, type and only included species, *Centrophorus harrisonii McCuUoch 1915, Victoria, Australia. Som-nispinax (subgenus of Centrophorus) Whitley 1940, p. 146, type and only included species, Centrophorus nilsoni Thompson 1930, New Zea- land. Probable Generic Synonym : Atractophorus Gilchrist 1922, p. 48, type and only included species A. armatus Gilchrist, off Natal, southern Africa. For discussion, see p. 82. Not Squalus Linnaeus 1758, type species S. acanthias Linnaeus. Not Spinax Cuvier 1817, p. 129, type species Squalus acanthias Linnaeus 1758. Not Acanthias Risso 1826, p. 131, type species A. vulgaris Risso, Mediter- ranean, equals Squalus acanthias Linnaeus 1758. Generic characters. Squalinae with the fin spines laterally grooved and well exposed in most (but see p. 81 under C. nil- soni), the second longer than the first; lower margin of caudal with subterminal notch ; teeth one-cusped in both jaws, the lowers wider than the uppers, with cusp directed strongly out- ward all along each side of jaw, and with the bases of successive teeth overlapping outwardly ; margins of upper teeth smooth ; margins of low-er teeth finely serrate in some species but smooth in others; the cusp of the median lower tooth either erect and symmetrical or oblique ; upper teeth subtriangular, with the BIQELOW AND SCHBOEDER : SQUALOID SHARKS 65 Fig. 8. Pectoral fin in diiferent species of Centropliorus, adjusted to ap- proximately equal lengths along outer margin, to show shape of inner posterior corner. A, foliaceus, same specimen as in Fig. 7F. B, squamosus, female, 1230 mm. long, southwest of Iceland. C, stclndachneri, type speci- men, after Pietschmann 1908, PI. 1, Fig. 2B. D, acws, type specimen, male 820 mm. long, Japan, Mus. Comp. Zool. No. 1049. E, uyato, male, 429 mm. long, northern part of Gulf of Mexico, Mus. Comp. Zool. No. 38165. F, granulosus, female, 922 mm. long, northern part of Gulf of Mexico, U. S. Nat. Mus. No. 157833. G, scalpratus, Victoria, Australia, after McCuUoch. 66 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY bases of successive teeth in contact along row, or slightly over- lapping, the cusp erect and nearly symmetrical near center of mouth in most species, but nearly or quite as oblique as the lowers all along each side of jaw in a few species (p. 72) ; median upper tooth either symmetrical or oblique ; inner corner of pectoral fins at least quadrate and angular, and more or less extended in most species (Fig. 8) ; snout to mouth in all known species, not longer than from mouth to level of origin of pec- toral fins ; nostrils without barbel ; preoral clefts present, more or less expanded inwardlj^ (Fig. IB). Caudal peduncle without preeaudal pits or lateral ridges ; dermal denticles on sides of body low ; block-like, scale-like, or conical in different species (Bigelow, Sehroeder and Springer 1953, p. 270, fig. 3). Maximum recorded lengths, 1500 mm. for *C. granulosus and 1420 mm. for *C. squamosus (Bocage and Capello 1866, pp. 26, 27). Depth range. The only definite records of depths of capture that we have found are 631 meters for *foliaceus, off Japan (Giinther 1887, p. 5), and 1756 meters in Philippine waters (Smith and Radcliffe 1912, p. 679) ; 366-375 meters for * gran- ulosus (specimen in U. S. Nat. Mus. from Gulf of Mexico) ; 50- 400 meters for *uyaio (Poll 1951, and specimens we have seen from Gulf of Mexico) ; "common between about 400 and 1500 meters" for * squamosus (Grey 1956, p. 96) with one record from 1875 meters; 400 meters for machiquensis (Maul 1955, p. 5 ; and 128 to 439 meters for scaipratus (Whitley 1940, p. 145). While no definite information is available for any one of the group of additional species described from Japan {^acus, *atromarginatus, *tesselatus, and steindachneri) ,_ it seems that their normal ranges are restricted to depths greater than 100- 200 meters. Remarks. The most distinctive single character of the genus (though probabl}^ not one of any great phylogenetic importance) is the angular and usually more or less extended inner corner of the pectoral fins. The range in this respect (Fig. 8) may be illustrated by *C. foliaceus Giinther 1877 of Japan, in which the pectoral inner corner is only veiy slightly extended or even rectangular, through conditions in ^squamosus Bonnaterre 1788, in *acus Garman 1906 (incl. steindachneri Pietschmann 1907), in ^'uyato Rafinesque 1810, and in harrisonii McCulloch 1915, BIGELOW AND SCHROEDER : SQUALOID SHARKS 67 where it is extended somewhat more, and in *granulosus Bloch and Schneider 1801, where it is still further prolonged, as it also is in scalpratus McCulloch 1915. The gradation, in short, in the shape of the pectoral, is so nearly unbroken that the revival of Lepidorhinus Bonaparte 1838 V Garman (1913, p. 211) and of Entoxychirus Gill 1862 by Fowler (1941, p. 242), based on differences in this regard, serves no useful purpose, whether from the practical standpoint, or as representative of probable phylogenetic relationship. In the type specimens, in fact, of C. nilsoni from New Zealand (Thompson 1930, p. 277), and of *C. tesselatus from Japan (now in the Museum of Comparative Zoology; Fig. 10 C, D), the inner corner of one of the pectoral fins is extended, but not that of the other pectoral (apparently undamaged). Teeth. The cutting edge of the lower teeth of *C. granulosus is described and pictured both by Miiller and Henle (1841, p. 89, PI. 33) and by Bocage and Capello (1866, p. 26, PI. 1, fig. 3D) as finely serrate. Similarly, the reserve rows (5 in number) of our eastern Atlantic specimen of this species are finely but irregularly serrate in some places (Fig. 9A), though not in others. The lowers in the more anterior of the two rows that are visible from outside the mouth (no longer in function) also show faint traces of serration here and there. But the lowers in the func- tional row are slightly jagged in places, at most. And the situa- tion is essentially similar in the case of a Gulf of Mexico specimen of *granulosus, on which some of the low^ers show microscopically fine serration, but others are merely somewhat irregular in outline, probably as the result of wear. (For further details, see Bigelow, Schroeder and Springer 1955, p. 6, Fig. 2C, p. 8). Bocage and Capello (1866, PI. 1, Fig. 3B) pic- ture the upper teeth of *granulosus as finely serrate, like the lowers. But this seems to have been an error on the part of the artist, for the uppers are smooth on both of the specimens of *granulosus that we have examined — also on all the speci- mens of other species of the genus that we have studied, for that matter. In *G. squamosus, as represented by a female about 1250 mm. long from Iceland (Fig. lOA), and by jaws of a somewhat larger individual, presumably from near the Faroes (Fig. IIB), the lowers in the outermost of the two rows that are visible from 68 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Figure 9 BIGELOW AND SCHROEDER : SQUALOID SHARKS 69 outside the mouth are finely serrate along the central part of the cutting edge, those in the reserve rows less clearly so. The lowers in the first reserve row (the edges of the functional rows are badly worn) are also unmistakably though faintly serrate in places (Fig. 9C) on the type specimen of *tesselatus. On the type specimen of *atromarginatus (on which the teeth are in better condition), the lowers (Fig. 9D) are irregularly but un- mistakably serrate. They are described as "serrated" in scalpra- tus (McCulloch 1915, p. 98) from Australia also. And the two rows of lower teeth that are visible from outside the mouth in the type specimen of the Japanese *acus Garman 1906 show faint and irregular traces of microscopically fine serration here and there, though not so distinctly as these appear on Garman 's (1913, PI. 12, fig. 6) illustration when the latter is viewed under a lens. In the Australian harrisonii, also, the edges of the lower teeth are irregularly serrate (Fig. 9E). On the other hand, we have not detected any trace of marginal serration either on the lower teeth of a juvenile male foliaceus, 351 mm. long, from Japan (Fig. 7G), or on those of uyato, whether on a 429 mm. male from the Gulf of Mexico (Bigelow, Schroeder and Springer 1953, p. 271, fig. 4) or on a 480 mm. male from the Mediterranean. Thus the presence or absence of serrations on the cutting edge of the lower teeth seems not to be a useful basis for generic separation here, though it has long been so regarded among the carcharhinid sharks. Similarly, the degree of obliquity of the upper teeth ranges Fig. 9. Lower teeth in different species of Centrophorus to show the degree of marginal serration. A, graiiulosus, male, about 855 mm. long, eastern Atlantic, Mus. Comp. Zool. No. 662, 2nd and 3rd reserve rows, x about 9. B, sq^iamosus, third tooth, outermost row, from jaw of specimen taken near the Faroes, Mus. Comp. Zool. No. 39571, x about 5.5. C, tess.e- latus, male, 887 mm. long, type specimen, Mus. Comp. Zool. No. 1031, Japan, first row visible from outside the mouth, x about 15. D, atromarginatus, female about 870 mm. long, type specimen, Japan, Mus. Comp. Zool. No. 1455, first two rows visible from outside the mouth, x about 8. E, harrisonii, Australia, teeth from one of McCulloch 's original specimens, made available to us through the kindness of Dr. G. P. Whitley, x about 10. F, squamosus, lower teeth at center of jaw of same specimen as in B, to show the obliquity of the median tooth, x about 2.4. 70 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1 sT PI <» o -1-1 s c3 f3 3 ^■i e ft OJ o cS ;-< CQ « l-j o a o CO T-T a u CO a> «H cc o 0) a •1— t ^ iH ^ bO *; ■+J o fl O , a> o rSi O ^ ft O ^ o O a 03 03 o S so o a 03 09 'o o tS] si a o o >-< cS ft w 13 a O '4-1 o 5 O 3 ^ a o ^ TS § p =0 c3 o !-i Oj s ^ -t-2 C «i5 2 CH o a o 13 Q .r-t +i ;-( g 0) oS o ^ ft to ft tc t-< ft ^ r-4 Oi o V C3 ft o o •C >. ^ fl -tJ -4^ •+J 5ii O) ;3 > M 03 O +1 O •d r-" ^ o 1— 1 OO 13 «H CO -|J ft o 13 > oJ -£3 32 ^ g ;::r .5 BIGELOW AND SCHROEDER : SQUALOID SHARKS 71 Fig. 11. A, Ccntrophorus tessclatus, same specimen as in Fig. 9C, upper and lower teeth, left hand side of mouth, x about 3. B, Centrophorus squamosus, female, about 1100 mm. long, from southwest of the Faroes, upper and lower teeth, left hand side of mouth, x about 1.3. C, Centro- phorus squamosus, same specimen as in B, denticles from side below first dorsal fin, x about 5.5. D, Centrophorns acus, male, 810 mm. long, type specimen, Mus. Comp. Zool. No. 1049, Japan, denticles from side below first dorsal fin, x about 15. E, Centrophorus atromargmatus, female, about 850 mm. long, type specimen, Mus. Comp. Zool. No. 1455, denticles from side below first dorsal fin, x about 17. F, Centrophorus tesselatus, same speci- men as in A, denticles from side, x about 17. 72 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY from the state in * granulosus and in *atromarginatus (Garman 1913, PI. 13, fig. 2), where those along the central sector of the jaw are erect and symmetrical, or nearly so, to the state in '*uyato (Bigelow, Schroeder and Springer 1953, p. 271, fig. 4) and in harrisonii (McCulloch 1915, p. 99, PI. 14, fig. 3), where this is true only of the first few teeth next to the midpoint of the jaw, the teeth farther out along the latter being nearly or quite as strongly oblique as the lower teeth are. As remarked above (p. 21), the specimen on which Johnson (1867) based his neAv genus and species MachepMlus dumerili, but which Vaillant (1888, p. 70), Garman (1913, p. 212), Key (1928, p. 440), and Bigelow, Schroeder and Springer (1953, p. 224) have considered as identical with *Centrophorus squamosus (Bonnaterre) 1788, had a symmetrical tooth at the center of the lower jaw. Accordingly Vaillant (1888, p. 70), followed by Rey (1928, p. 443), included the presence or absence of a symmetrical median lower tooth as alternative between two varieties of *squamosus. Conditions in this respect among specimens of Centrophorus that we have examined are as follows : *granulosus (male about 855 mm. long, Europe, and female 922 mm.. Gulf of Mexico) ; * squamosus (2 females about 1250 and 1280 mm. long from Iceland and jaws of 4 somewhat larger specimens, probably from west of the Faroes, Fig. IIB) ; *acus (type speci- men, male about 820 mm., Japan) ; *atromarginatus (type speci- men, 857 mm., Japan) ; and *foliaceus (male, 351 mm. long, Japan), the cusp of the median lower tooth is about as oblique as the cusps of the successive teeth outward, in each of the rows that are visible from outside the mouth. On the other hand, the cusp of the median lower tooth is erect and symmetrical in both these rows on the type specimen of *C. tesselatus from Japan (Fig. IIA). And the two specimens we have seen of *C. uyaio are intermediate between these two extremes, the cusp of the median lower tooth being strongly oblique on a male of 480 mm. from Nice, France, but only very weakly so, as well as smaller than that of the succeeding teeth on another male of 445 mm. from the Gulf of Mexico (Bigelow, Schroeder and Springer 1953, p. 271, fig. 4). Evidently the degree of obliquity of the cusp of the median lower tooth is not an appropriate basis for generic separation here. Dermal denticles. In some members of the genus the denticles BIGELOW AND SCHROEDER : SQUALOID SHARKS 73 on the sides of the trunk are scale-like and overlapping, the rear edges even or tridentate, and with the ridges on the outer surface either parallel or diverging slightly rearward. In other species the denticles are more block-like or conical-thorn-like, not over- lapping, more or less pointed, with the ridges converging rear- ward. And Key's (1928, p. 239, fig. 143) account of the denticles of *C. granulosus as more pointed on juvenile specimens than on adults seems actually to have been based on a specimen of *C. uyato (Rey 1928, PI. 3, fig. 2), a species for which pointed denticles are diagnostic. The shape of the denticles, indeed, is the most obvious character among those likely to serve as alterna- tive in this genus as between groups of species. But to subdivide Ccntrophorus generieally on this basis, or by the other minor characters on which Whitley (1940, pp. 145, 146) based his sub- genera Gaboa (type Cenfrophorus harrisonii McCulloch 1915) and Somnispinax (type Centrophorus nilsoni Thompson 1930), seems not to serve any useful purpose, the total number of species concerned being so small, and their resemblance, in general, being so close one to another. Nomenclatural status of the type species. The original descrip- tion by Bloch and Schneider (1801, p. 135) of their Squalus granulosus reads as follows: "Sq. bruneus cute granulosa, acu- leis 2 robustis compressis, postice parum ineu ruis, intus con- cavis, primo in initio pinnae dorsi, secundo in initio mesurae, pinnis brevibus, dorsali pectoralibus propinquiorae quam ven- tralibus remotissimis, mesura caudale vicina, foraminibus tem- porum remotis, lunatis." Obviously this is not enough to place Bloch and Schneider's five-foot specimen, whether as to genus among the Squalinae, or as to species, for it does not mention either the shape of the pectoral fins or the nature of the teeth. It is therefore fortunate for the sake of nomenclatural stability that while Miiller and Henle's (1841, p. 89, PI. 33) description and illustrations of their * granulosus were based on an alcoholic specimen from near Sicily, they did have, for comparison, a dried specimen in Berlin from Bloch 's collection, probably the one on which Bloch and Schneider's 1801 account was based, for they mention only the one. Miiller and Henle's identification of their ''granulosus" with the "granulosus" of Bloch and Schneider 1801 is therefore to be accepted — unless re-examination of the 74 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY type specimen (if still in existence) should eventually prove otherwise. This, in turn, implies that Miiller and Henle's very excellent description and illustrations take precedence over the several accounts by subsequent authors, as the underlying basis for the comparison of species within the genus Centrophorus. Species. The following named species, listed in chronological order, fall in Centrophorus as defined above (p. 64) : *Squalus squamosus Bonnaterre 1788 eastern North Atlantic (Figs. IB; SB; 9B; lOA, B; IIB, C), including Machephilus dumerili Johnson 1867 ; *S. granulosus Bloch and Schneider 1801 includ- ing Centrophorus hraganzae Regan 1906), type species, eastern North Atlantic (including the Mediterranean) and Gulf of Mexico; *8qualus uyato Rafinesque 1810, eastern Atlantic (in- cluding the Mediterranean) and Gulf of Mexico (Fig. 8E) ; Centrophorus moluccensis Bleeker 1860, Amboina ; Centrophorus lusitanicus Bocage and Capello 1864, coast of Portugal ; *Centro- phorus foliaceus Giinther 1877, Japan (Fig. 7F-I) ; *C. acus Garman 1906, Japan (Fig. IID) ; *C. tesselatus Garman 1906, Japan (Figs. 9C, IOC, llA) ; C. steindachneri Pietschmann 1907, Japan (Fig. 8C) ; *C atromarginatus Garman 1913, Japan (Figs. 9D, HE) ; C. harrisonii McCulloch 1915, Australia (Fig. 9E) ; C. scalpratus McCulloch 1915, Australia (Fig. 8G) ; C. nilsoni Thompson 1930, New Zealand ; and C. machiquensis Maul 1955, Madeira. The specimen from southeastern Africa, described by Gil- christ (1922, p. 48) as Atractophorus armatus, is also provision- ally referred here to Centrophorus (for discussion, see p. 82). The members of the genus may be segregated by the shape of the denticles on the sides of the body, in three groups : A, denticles block-like, close set but not overlapping; B, denticles scale-like and overlapping more or less ; C, denticles spur-like or conical-thorn-like and more widely spaced. In the Atlantic, group A is represented by *C. granulosus, by lusitanicus if this finally proves a distinct species (see below), and by the newly described C. machiquensis, which maj^ prove to be identical with '^granulosus when a larger number of speci- mens of it have been examined; group B is represented by * squamosus; group C by * uyato. The distinctive features of *granidosus, ^squamosus and *uyato — all of which are fairly BIGELOW AND SCHBOEDER : SQUALOID SHARKS 75 well known now — are discussed elsewhere (Bigelow, Schroeder and Springer 1953, pp. 224-225; 1955, p. 7). The status of lusitanicus remains uncertain. Originally de- scribed as distinct from * granulosus by Bocage and Capello in 1864 but definitely classed by them in 1866 with *granulosus, lusitanicus was revived as a separate species by Giinther (1870, pp. 420-421),' and was so regarded both by Regan (1908) and by Garman (1913, p. 199). Again relegated to the synonymy of * granulosus by Rey (1928, p. 436), it has been revived by Nobre (1935, p. 449). And Mr. G. E. Maul informs us that the first dorsal fin of a specimen which he has examined in the British Museum is larger than in ^granulosus, as appears from the following tabulation. Proportional dimensions, in per cent of total length, of (A) Centropliorus lusitanicus, male, 742 mm. long, coast of Portugal, based on data furnished to us by Mr. G. E. Maul from a specimen in the British Museum, and (B) * Centropliorus granulosus, female, 922 mm. long, Gulf of Mexico (U. S. Nat. Mus., No. 157833). Snout length in front of movth First dorsal fin : vertical height Base, from anterior origin of spine Free rear margin of fin Second dorsal fin: vertical height Base, from anterior origin of spine Free rear margin of fin Caudal fin: length of upper margin Distance from snout to: origin of first dorsal spine origin of second dorsal spine upper origin of caudal fin origin of pectoral fins origin of pelvic fins Interspace between : origin of first dorsal spine and second dorsal spine base of second dorsal and lower origin of caudal Distance from: origin of pectorals to origin of pelvics origin of pelvics to lower origin of caudal Until a larger number of specimens have been measured to determine how wide the range of variation may be, it must A B 9.4 9.1 7.4 5.4 11.7 8.,5 7.3 6.9 6.1 5.0 6.6 5.7 4.0 4.5 19.7 20.3 32.1 34.8 67.0 69.2 80.3 79.7 22.2 22.2 60.7 58.7 34.9 34.4 6.6 6.0 38.4 37.6 15.8 18.1 76 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY remain an open question whether lusitanicus deserves recogni- tion as a species separate from granulosus. The situation is not clear for the species of Centrophorus that are known from the northern Pacific (all of them from Japan). Among these, group A (with block-like denticles closely spaced in quincunx arrangement) is represented by *tesselatus and by '*atroniarginatus. The demarcation between these two seems clear cut, from our examination of the type specimens in the Museum of Comparative Zoology. Thus the interspace be- tween the rear end of the base of the first dorsal fin and the point of emergence from the skin of the second dorsal spine is as long as from tip of snout to axils of pectorals in *atromarginatus, but is only as long as from the snout to origin of pectorals in *tesselatus. Also, the inner corner of the pectoral fins is ex- tended somewhat farther, the second dorsal spine is noticeably longer, and the cusps of the lower teeth are not only more strongly oblique but their cutting edge less deeply concave in * atromarginatus than in *tesselatus. Furthermore, the cusp of the median lower tooth, which is strongly oblique on the type specimen of * airomarginatus, is erect and symmetrical on the type specimen of *tesselatus, as noted above (p. 72). But this last difference may be a sexual one. It remains an open question whether *afromarginatus is sep- arable from * granulosus of the Atlantic. At least, our compari- son of the type specimen of the former (a female of 850 mm.) with a male ^granulosus of 860 mm. from the eastern Atlantic, with a female of that species, 922 mm. long from the Gulf of Mexico (Bigelow, Schroeder and Springer 1955, fig. 2), and with Miiller and Henle's (1841, PI. 33) excellent picture of a ^granulosus from the Mediterranean, has not revealed any dif- ferences that might seem of specific rank, unless it be that the gill openings are not as long in the Japanese form (third gill is about 64 per cent as long as distance between nostrils) as in our Gulf of Mexico specimen (third gill about 80 per cent as long as distance between nostrils). And while Garman (1913, p. 197) desscribed the teeth of * airomarginatus as "not ser- rate," the lowers of the type specimen do show traces of fine serrations along the cutting edge (Fig. 9D) much as in the case of *granulosus (Fig. 9A), in *tesselatus also, for that matter (Fig. 9C). Pietschmann (1908, p. 663), in fact, has reported under BIGELOW AND SCHROEDER : SQUALOID SHARKS 77 the name granidosns two Japanese specimens of Cenfrophorus, respectively 536 mm. and 908 mm. long, that he had compared with two granulosus of corresponding sizes from the Mediter- ranean. But the actual identity of his Japanese specimens calls for verification (if they are still in existence) for their denticles are described (Pietsehmann 1908, pp. 663-667) as differing in shape from those of ^granulosus of the Atlantic. Pietsehmann also united both *tesselatus and *acus Garman with *granulosus. But as implied in the foregoing discussion, *tesselatus differs from * atromarginatus. And the denticles on the sides of the body of *acus (no figures of which had appeared when Pietseh- mann wrote) differ widely from those of * granulosus, for they are scale-like, with tridentate rear margins (Fig. IID). The *granulosus-*tesselatus-* atromarginatus group of the Northern Hemisphere has a close ally in the Southern Hemisphere of the Pacific in scalpratus of Australian waters, except that McCulloch's (1915, PI. 13, figs. 2-7) illustrations show the denti- cles on its sides as more pointed than those of * granulosus of the Atlantic and Gulf of Mexico (Bigelow, Schroeder and Springer 1953, p. 270, fig. 3A; 1955, fig. 2D), or of the Japanese representative of the latter (p. 76). Other differences sug- gested by McCulloch's illustrations of scalpratus may be due to dift'erent methods of measurement, etc. The embryo from Amboina that was described by Bleeker in 1860 as C. moluccensis appears also to fall in this same group, for Regan (1908, p. 51), who saw the type specimen, located it in the subsection of the genus with "posterior angle of pectoral fin considerably pro- duced and acutely pointed." And the little else that is known of its fins and bodily dimensions is consistent with this. But the shape of its denticles is not known. Group B, with scale-like denticles (p. 74), is represented in the northwestern Pacific by *foliaceus, originally described from Japan, and reported by name (with very brief comment) from deep water (960 fathoms) among the Philippines (Smith and Radcliffe 1912, p. 679) ; by *acus from Japan; and by steindach- neri, also from Japan. Among these, steindachneri, with tri- dentate flank denticles and with the inner corner of the pectorals only slightly extended (Fig. 80; Pietsehmann 1908, p. 676, text fig. 3; PI. 1, fig. 1) agrees so closely with the type specimen of •C. acus as to make it reasonably certain that the two represent 78 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY but a single species, to be named aciis on the ground of priority. ^'Acus, in turn, is allied to *foliaceiis. But it differs from the latter in the evident extension of the inner corner of its pectoral fins, for these are not appreciably extended in *foliaceus, either as they are represented by Giinther (1887, p. 5, PI. 2, fig. A), or as they appear on two immature males of *foHaceus, respectively 325 and 350 mm. long, from Japan (U. S. Nat. Mus., Nos. 161517 and 161518) that we have examined (Fig. 7F). The dermal denticles, also, on the sides of the type specimen of *acus differ conspicuously from those of *foUa,ceus, for while they are tri- dentate in both cases, they are not only broader, relatively, with shorter marginal teeth and less regularly overlapping in *acus (Fig. IID) than in *foliaceus (Fig. 7H), but so much smaller relatively that the one species is separable from the other at a glance by this character alone. *C'. foUaceits, indeed, is by far the roughest-skinned species of Centrophorns that we have seen. *Acus is also firmer-bodied than *foliaccus, at least in the pre- served state, and it is of a paler hue below than above, whereas the specimens of *foliaceus we have seen are of as dark a choco- late brown below as above. The species *acus may, therefore, be retained as distinct, at least until someone re-examines the type specimen of *foliaceus, now in the British Museum. (For an account of *acus, with beautiful illustrations, see Garman 1913, p. 199, PI. 12, figs. 5-8.) The accounts of *fol\aceus by Giinther (1887, p. 5) and by Jordan and Fowler (1903, p. 631), do not especially emphasize the characters now known to be diagnostic of species within the genus Centrophorns. A more detailed account from this point of view, with measurements, is therefore included here (p. 78). Centrophorus foliaceus Giinther 1877 Figure 7F-I Study material. Immature male, 351 mm. long, U. S. Nat. Mus. No. 161518, Okinosa, Japan. Proportional dimensions of specimen listed above, in per cent of total length. Snout, lengtli in front of: outer ends of nostrils 3.4; mouth 10.8. Eye : length, 4.8. Mouth: breadth, 7.6. BIGELOW AND SCHROEDER : SQUALOID SHARKS 79 Xostrils: distance between inner ends, 4.0. GUI opening lengths: First, 2.1; second, 2.4; third, 2.4; fourth, 2.4; fifth 2.0. First dorsal fin: vertical height, 5.0; length from imbedded base of spine to rear end of base of fin, 9.0. Second dorsal fin: vertical height, 5.0; length from imbedded base of spine to rear end of base of fin, 8.0. Caudal fin: upper margin, 21.1; lower anterior margin, 15.7. Pectoral fin: outer margin, 10.8; inner margin, 7.7; width, 7.3. Distance from snout to: imbedded base of first dorsal spine, 36.2; imbedded base of second dorsal spine, 65.5; upper origin of caudal, 78.9; origin of pectorals, 25.4. Interspace between: rear end of base of first dorsal and imbedded base of second dorsal spine, 22.5; rear end of base of second dorsal and upper origin of caudal, 5.7; base of pelvics and lower origin of caudal, 12.4. Distance from origin to origin of: pectorals and pelvics, 33.7; pelvics and caudal, 17.6. 16-1-16 Teeth : 14-14 Head, to orig-in of pectorals, about % of trunk to upper origin of caudal fin; snout in front of mouth about 42 per cent of head to pectorals; eye about 19 per cent as long as head; distance from tip of snout to front of eye about 29 per cent of head; distance between inner ends of nostrils about 37 per cent as long as snout to mouth; longest gill openings (2nd, 3rd and 4th) about half as long as eye; mouth transverse, scarcely arched; distance between inner ends of preoral clefts about 1.75 times as great as between nostrils. Upper teeth short, triangular, the median tooth erect, sym- metrical, but the subsequent teeth increasingly oblique outward along the jaw, with the inner edges of those toward the corners of the mouth nearly horizontal; the lowers wider than the up- pers, the cusps directed so sharply outward as to form a nearly unbroken cutting edge paralleling either side of the jaw. Cut- ting edge of lowers perfectly smooth, both in the reserve rows and in the outermost of the two rows visible from outside the mouth ; the 2nd row slightly irregular, but not definitely serrate. The condition of the specimen does not allow us to determine whether or not there was a symmetrical median tooth in the lower jaw. 80 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Dermal denticles scale-like, close set and largely overlapping, those on upper and lower sides of snout ovate, but those on body as a whole tridentate, lanceolate in general form, with the median tooth much the longest, on short pedicle, rising steeply from the skin and with high median crest; those on back, an- terior to first dorsal spine, the largest. Length of first dorsal fin, measured from anterior side of imbedded origin of spine about 2.5 times as long as distance between nostrils; interspace between rear end of base of first dorsal fin and anterior edge of imbedded base of second dorsal spine about as long as from tip of snout to second gill openings. Second dorsal fin similar in shape to first dorsal, about 90 per cent as long, basally, as first, similarly measured (see above). Dorsal spines both well exposed; exposed portion of second about 1.5 times as long as that of first ; distance between first and second spines, at points of emergence from the skin, about 1.5 times as long as from tip of snout to first gill openings. Interspace between rear end of base of second dorsal fin and origin of upper side of caudal about as long as from point of emergence from skin of second dorsal spine to rear end of base of second dorsal fin. Caudal fin about 20 per cent of total length, of the shape shown in Figure 7F. Distance from origin of lower side of caudal to rear end of bases of pelvics about as long as from rear edge of eye to origin of pectoral fins. Rear tips of pelvics (when laid back) about in a line with point of emergence from skin of second dorsal spine ; origin of pelvics anterior to imbedded base of second dorsal spine by a distance about as long as from rear end of eye to first gill openings. Pectorals with outer margin about as long as from eye to fourth gill openings ; outer corner broadly rounded ; distal margin weakly sinuous, reaching, when laid back, about even with imbedded base of first dorsal spine ; inner corner angular, with a few of the horny rays slightly ex- tended (Fig. 8A). Color. Plain dark brownish gray above and below; the rear margins of pectorals and tip of dorsals pale edged, after many years in alcohol. Remarks. *C. foliaceus is recognizable at a glance, so widely does its dermal armature, with close-set, strongly tridentate, high-crested denticles, rising steeply from the skin, differ from that of every other Centrophorus that we have seen, or that has BIGELOW AND SCHROEDER : SQUALOID SHARKS 81 been pictured. Other distinctive characters for it (^Yithin the genus Ceniropkorus) are: the shape of the inner corner of its pectoral fins (p. 66), the lack of any abrupt transition from the general contour of its back to the rising slope of the first dorsal fin, and the pronounced obliquity of its teeth, uppers as well as lowers, as described above (p. 79). In all these respects the specimen here described agrees with Giinther's (1887, p. 5, PL 2, fig. A) account of a 470 mm. (ISVo inch) male, except that its inner pectoral corner is very slightly extended ("rectangular, not produced" according to Giinther), and that it is only the median upper tooth that is erect in the upper jaw, the lateral upper teeth being strongly oblique (see above), whereas Giinther characterized "the anterior teeth of the upper jaw" as erect, the laterals as "slightly oblique." These slight differences do not seem to call for recognition in nomenclature on the basis of so few specimens. The Japanese *foliaceus resembles *squamosus of the eastern North Atlantic in its bodily proportions in general, in the rela- tive locations of its fins, and in the scale-like nature of the denticles on the sides of its trunk. But it differs from *squamo- sus in the quadrate outline of the inner corner of its pectoral fins (definitely though only slightly extended in *squamosus cf. Fig. 8A with Fig. 8B), and in the narrower, more regularly tri- dentate and more steeply elevated denticles (cf. Fig. 7H with Fig. IIC), so that the two species clearly are separable, the one from the other. The Northern Hemisphere species with scale-like denticles are represented in the Southern Hemisphere by nilsoni of New Zealand (Thompson 1930, p. 277, PI. 44). Nilsoni falls close to *acus in the shapes of the inner corner of its pectoral fins and of the denticles on the sides of its trunk. But its fin-spines are described by Thompson (1930, p. 277) as "low and largely covered," a character which (if normal) sets it apart. The North Atlantic *uyato, with conical-thorn-like denticles on its sides and with the inner pectoral corner moderately ex- tended, is similarly replaced in the South Pacific by harrisonii of Australian waters. Going back a step further in shark his- tory, the embryo reported from Port Jackson, Australia, many years ago by Steiudachner (1867, p. 333) under the name Acanthias uyato Rafinesque, may have been the first specimen 82 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY of harrisonii to come under scientific observation. On the other hand, Whitley (1934, p. 199) has suggested that it may have been a Squalus, a view supported by Steindachner's description of it as marked witli large white spots on its back. However this may be, the cutting edge of the lower teeth, which are smooth in the Atlantic *uyaio (p. 69), are irregularly serrate (Fig. 9E) in Jiarrisonii (described as imperfectly so by McCulloch 1915, p. 99). And other differences between the Atlantic and Indo-Pacific populations may come to light when adequate series of specimens from the two ocean areas have been compared. Finally, we face the case of Atractophorus armatus from the Natal coast of southern Africa, the type specimen of which seems to have been a typical Centrophorus in all respects, except that the tip of its second dorsal fin spine was described and pictured as "barbed, like an arrowhead" (Gilchrist 1922, p. 48, PI. 7, fig. 3). Unfortunately, the type specimen of armatus is no longer to be found. ^^ But the collection of the Government Marine Survey of South Africa does contain a juvenile shark, 330 mm. long, apparently an armatus, the sec- ond dorsal fin spine of which (while not two-barbed) bears a single, wing-like expansion on its rear side, close to its tip (Fig. 7J). And any conformation of the spine tip, other than tapering to a point (if normal) makes armatus unique among known sharks, whether modern or fossil. We may either accept this feature as justifying the generic separation of Atractophorus from its parent genus Centrophorus, or we may transfer armatus to Centrophorus, as the genus in which it would fall on the basis of ancestral relationship, were it not for the single character in question. We choose this last alternative, partly because we see no useful purpose to be served in acting otherwise; partly be- cause this choice seems the more likely of the two to focus attention on armatus when additional specimens come to hand ; and partly because of the chance that the peculiar conformation of the second dorsal spine tip on the two specimens that have been examined may have represented an abnormality, or may have been the result of accidental damage. If armatus is, in fact, a Centrophorus, its closest relationship is with harrisonii, with which — and with *uyato of the North 11 We are so informed by Prof. J. L. B. Smith, who has made a special search for it in the collection of the Government Marine Survey of South Africn. BIGELOW AND SCHROEDER : SQUALOID SHARKS 83 Atlantic — it shares pectoral fins with extended inner corner (but shorter from origin to rear tip than from tip of snout to level of first gill openings), and denticles on sides of body that are described as acuminate, with many sharp ridges (Gilchrist 1922, p. 49). Key to species of Centrophorus 1. Denticles on sides of trunk scale-like, overlapping, the ridges on their outer surface (if more than one) parallel or diverging slightly rear- ward 2 Denticles on sides of trunk block-like or conieal-thorn-like, not over- lapping; the ridges on their outer surface converging rearvsrard, if at all 5 2. Denticles on sides of body not definitely tridentate (Fig. IIC) *squamosus Bonnaterre 1788. Northeastern Atlantic, p. 74 Denticles on sides of body regularly tridentate (Figs. 7H, IID) ... .3 3. Inner corner of pectoral fms quadrate, not appreciably extended (Fig. 8A) ; dermal denticles on sides of body as in Figure 7H *foliaceus Giinther 1877. Japan ; also reported by name from the Philippines (Smith and Radcliffe 1912, p. 679), and from New Zea- land (Richardson 1956, p. 7). p. 78 Inner corner of pectoral fins definitely extended, though not greatly so (Fig. 8D) ; denticles on sides of body as in Figure IID) 4 4. Both of the dorsal fin spines prominent and well exposed aciis Garman 1906, Japan, p. 77 Both of the dorsal fin spines low and largely covered . nilsoni Thompson 1930, New Zealand; p. 81 5. Pectoral fin, from origin to inner rear corner, about as long as from tip of snout to level of fourth gill openings scalpratus McCulloch 1915, Australia, p. 74 Pectoral fin from origin to inner rear corner shorter than from tip of snout to level of first gill openings 6 6. Denticles on sides of trunk block-like, c^uadrate, closely spaced in quincunx arrangement; upper teeth erect and symmetrical, or nearly so, all along central part of jaw 7 Denticles on sides of trunk conical-thorn shaped, loosely spaced in random arrangement; upper teeth strongly oblique, except for the first few next to the mid-point of the jaw, which are erect 8 7. Distance from rear end of base of first dorsal fin to point of emergence 84 BULLETIN : MUSEUM OF COMPAEATIVE ZOOLOGY from the skin of second dorsal spine is as long as from tip of snout to axil of pectorals ; exposed part of second dorsal fin spine is about 75 per cent as long as distance between nostrils * granulosus Bloch and Schneider 1801 (including lusitanicus Bocage and Capello 1864) ; Mediterranean with neighboring eastern Atlantic, Gulf of Mexico, p. 73 ; and re- ported by name from tropical West Africa (Cadenat 1950, p. 101); also machiqucnsis Maul 1955, Ma- deira (p. 74) ; and atromargina- tus Garman 1913, Japan (see dis- cussion, pp. 76-77). Distance from rear end of base of first dorsal fin to point of emergence from the skin of second dorsal fin spine is only about as long as from tip of snout to origin of pectorals; exposed part of second dorsal spine only about 40 per cent as long as distance between nostrils *tcsselatus Garman 1906, Japan, p. 74 8. Cutting edges of lower teeth smooth *uyato Eafinesque 1810. Mediterranean, eastern tropical At- lantic along the African coast from Senegambia (Rochebrune 1883- 1885, p. 24) to Lat. about 20° S (Poll 1951, p. 65); also Gulf of Mexico, p. 74 Cutting edges of lower teeth irregularly serrate (Fig. 9E) . Jiarrisonii McCulloch 1915. Australia, p. 74 Position uncertain : moluccensis Bleeker 1860, Amboina ; also armatus Gilchrist 1922, southeastern Africa (p. 82). Genus CeNTROSCYMNUS BocaRe and Capello 1864 Centroscymnus Bocage and Capello 1864, p. 263; 1866, p. 29; type species *C. coelolepis Bocage and Capello, off Portugal. Generic Synonyms : Centrophorus in part, Giinther 1870, p. 423, for C. coelolepis Giinther, equals * Centroscymnus coelolepis Bocage and Capello 1864; Thompson 1930, p. 277, for C. tvaitei Thompson, New Zealand. Centroselachus Garman 1913, p. 206, type species *Centrophorus crepidater Bocage and Capello 1864, off Portugal. Proscymnodon (subgenus) Fowler 1934, p. 239, type and only included species Centrophorus pluiiketi Waite 1909, New Zealand. BIGELOW AND SCHROEDER : SQUALOID SHARKS 85 Scymnodon in part, Fowler 1941, p. 225, for Centrophorus plunlceti Waite 1909, New Zealand. Not Scymnodon Bigelow, Sehroeder and Springer 1953, p. 230, for Scymno- don melas Bigelow, Sehroeder and Springer 1953, p. 233, fig. 5, which is a young stage of Centroscymnus coelolcpis (p. 100-). Generic characters. In fjeneral as in Centrophorus (p. 64), except with inner corner of pectoral fins broadly rounded (this is the most conspicuous difference between the two genera) ; the upper teeth more slender than the loAvers, lancet shaped, those along central sector of mouth nearly as long as those midway out along either side of upper jaw; the successive teeth in the functional row not in contact basally, or hardly so ; lower teeth smooth edged, the successive cusps directed so strongly outward as to form a nearly continuous cutting edge ; median lower tooth not symmetrical ; second dorsal fin spine, like first, either ex- posed more or less at tip, or entirely enclosed in the skin ; dermal denticles on sides of body low, scale-like, with tridentate margins on young specimens in some species, and probably in all, also on adults in some species, but replaced later in growth in other species by denticles with evenly rounded margins and concave crowns ; preoral clefts short in some species but so long in others as to leave only a narrow gap between their inner ends (see dis- cussion, p. 87), considerably expanded inwardly in species where they are short (Fig. IC) but not appreciably so in those where they are longest (Fig. ID). Maximum recorded length 1140 cm. for C. coelolepis (Bocage and Capello 1866, p. 31) and 1414 for C. plunketi (Waite 1910, p. 384). Depth range. Centroscymnus coelolepis (p. 88), the shoalest record for which has been 329 meters, has its center of distribu- tion between 400 meters and 2000 meters according to locality, and has been taken down to about 2700 meters (maximum yet recorded for anj- shark ) . It ma}- fairly be classed as the deepest living of modern sharks. (For further details see Bigelow and Sehroeder 1948, p. 498, and Grey 1956, pp. 97-98.) Depth records for the other members of this genus are 270-920 meters for crepidater off Iceland (Saemundsson 1922, pp. 192, 196, as Centrophorus jonsonii) ; 220 meters for plunketi (Waite 1910, p. 384) ; 780 meters for rossi (Alcoek 1898, p. 144) ; and 512 meters for fiiscus (Gilchrist and von Bonde 1924, p. 3). 86 BULLETIN : MUSEUM OF COMPAEATIVE ZOOLOGY Remarks. Teeth. The only characters mentioned by Bocage and Capello (1864, p. 263;' 1866, pp. 19-20, 24, 29) in their successive generic diagnoses of Centroscymnus that contrast with their diagnosis for Centrophorus (1866, pp. 19, 22) were that the upper teeth of the former are awl-shaped and loosely spaced ("subuliformes ecartes") and that its first dorsal fin spine is almost entirely enveloped by the skin. But a much more con- spicuous difference than either of these is the rounded inner corner of the pectorals of Centroscymnus, as contrasted with their angular inner corner in Ceritrophorus (p. 66). Similarlj^ the onlj^ alternative character in their original diagnosis of Centroscymnus, as contrasted with Scymnoclon, is the shape of the lower teeth, which Bocage and Capello characterized for the former as not unlike those of CentropJiorus (i.e. as oblique, without symmetrical median tooth), but for Scymnodon as with symmetrical median tooth, the first tooth outward from it on either side erect, the succeeding teeth as increasingly oblique toward corners of mouth ("dens medianus impar, post eum dantes premum erecti, deinde versus angulum oris magis ac magis decumbentes, " Bocage and Capello 1864, p. 263). But a second difference between the type species of Centroscymnus (*coelo- lepis Bocage and Capello 1864) and of Scymnodon {* ring ens Bocage and Capello 1864), equally diagnostic and much more conspicuous, is that while the upper teeth along the central sector of the mouth are nearly as long as those midway out along each side of the jaw in Centroscymnus, they are much shorter than those midway along the jaw in Scymnodon. The value of this feature as a diagnostic character seems first to liave been realized by Key (1928, p. 486, fig. 152). Dermal denticles. In the adults of some species {plunketi Waite 1910, New Zealand; *crepidatcr Bocage and Capello 1864, eastern Atlantic) that are referable to Centroscymnus by their dentition, the denticles on the sides of the trunk are tridentate in outline, the outer surface sculptured with longitudinal ridges. In the adults of others {*coclolepis Bocage and Capello 1864, both sides of the North Atlantic; cryptacanfhus Regan 1906, Madeira; *owstoni Garman 1913, Japan), they are rounded in outline, with smooth, concave crowns; also in fiiscus Gilchrist and von Boude 1924, southern Africa, so far as the concavity BIGELOW AND SCHROEDER : SQUALOID SHARKS 87 of the crowns is concerned/" But it has been found by Tortonese (1952, p. 386, corroborated by Bigelow and Schroeder 1954, p. 47) that the rounded denticles of the adult *coelolepis, and ap- parentW those of *owstoni as well (Bigelow and Schroeder 1954, p. 50), are preceded at early stages in growth (as proven by embryos of known parentage) by trideutate forms, the replace- ment of denticles of the juvenile shape by others of the adult shape taking place when a given specimen is half to two-thirds grown. For further details we refer the reader to our earlier account (Bigelow and Schroeder 1954, p. 47, fig. 2). It follows from this that our earlier acceptance of the shape of the denticles (Bigelow and Schroeder 1948, p. 451, following Garman 1913, p. 189) as alternative between Centroscymnus, where they are rounded on the type species at maturity {*coelolepis Bocage and Capello 1864) and Scymnodon, where they are tridentate at maturity, is no longer tenable. It seems appropriate here to recall that we owe to Steen- strup (1861) our knowledge that the dermal denticles of sharks do not persist throughout the life of the fish, growing as the latter grows, as the scales of bony fish do, but that they are constantly being shed, to be replaced by other, larger denticles. Thus several generations of denticles succeed one another dur- ing the life-span of any individual shark. Fill spines. It seems worth emphasizing that the imbedded basal part of the fin spines is easily felt, in species in which their tips are concealed by the skin (p. 89). Preoral clefts. The preoral clefts extend for about 40-50 per cent of the distance from the corner of the mouth toward the midline of the snout in *coelolepis (incl. cryptacanthus) of the North Atlantic, in '*owstoni of Japan, and in macracantJnis of the Magellanic region ; about 60 per cent of that distance in waitei from New Zealand;^" but for about 90 per cent of that distance in *crepidater (incl. *jonsonii) of the North Atlantic (Fig. ID) and in rossi Alcock 1898 of the Indian Ocean (Alcock 1899, PI. 26, fig. 3). Both *crepidater and rossi fall, however, with the type species of Centroscymnus {* coelolepis) in dental characters, in shape of pectoral fins, and in shortness of fin- 12 The contour of the margins of the denticles is not known for fuseus. 13 Taken from a drawing made available to us by Dr. J. A. F. Garrick before publication. 88 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY spines. Hence the difference in the length of the clefts does not call for generic recognition, especially where the total number of species concerned is so small. Shape of caudal fin. Conditions in Centroscymnus argue against the presence or absence of a subterminal notch on the loAver margin of the caudal as a subfamily alternative among the Squalidae, for while this notch is well marked in most of the members of the genus, it is indistinct in waitei (Garrick 1955, p. 233, fig. lA). This seems the appropriate place for us to point out that the caudal of *coelolepis is not correctly pictured in our earlier illustration (Bigelow and Schroeder 1948, fig. 94). Actually it is of the shape shown here on Figure 12 H. Species. The following named species, listed in chronological order, are referable to Centroscymnus as defined here : *Centro- scynmus coelolepis Bocage and Capello 1864, type species, North Atlantic, east and west; *Centrophorus crepidater (Bocage and Capello 1864), off Portugal; Centrophorus rossi Alcock 1898, Travancore Coast, India; Centroscymnus cryptacanthus Regan 1906, Madeira ; Centroscymnus macracayithus Regan 1906, Magel- lanic region; * Centroscymnus owstoni Garman 1906, Japan; CeyiiropJwrus plunketi (Waite) 1910, New Zealand (referred by Garman 1913, p. 210 and by Fowler 1941, p. 228 to Scymnodon because of its tridentate denticles) ; * Centrophorus jonsonii (Saemundsson) 1922, Iceland; Centroscymnus fuscus Gilchrist and von Bonde 1924, southern Africa ; Centrophorus waitei (Thompson) 1930, New Zealand." Detailed descriptions of the type species, *coelolepis, with illustrations, have been given by Garman (1913, p. 204, PI. 17, figs. 5-8), by Rey (1928, p. 451) and by Bigelow and Schroeder (1948, p. 494; 1954, p. 47). Coelolepis is known in the eastern Atlantic from the offing of Cape Verde and of Morocco to the Faroe Bank and to Iceland, including the western part of the Mediterranean ; from the offing of Delaware Bay to the Banks of Newfoundland in the western. According to Regan's (1906, p. 437) account, his cryptacanthus from Madeira differs from the specimens of * coelolepis that we have seen (including the one M Our inclusion of lonitei in CentroscymnuB follows Garrick's (1955) study of the type specimen, ttie result of which was contributed to us by Dr. Garricli in advance of his publication. BIGELOW AND SCHROEDER : SQUALOID SHARKS 89 on which Garman's illustration was based) in a slightly longer snout; in an interspace between the two dorsal fins about 6 times as long as the base of the first dorsal (4-5 times in coelo- Ic-pis) ; in the extension rearward of the tips of the pelvic fins to abreast of the tip of the second dorsal fin; and in complete enclosure of the fin spines by the skin. Also, the denticles on the trunk rearward from abreast the first dorsal fin are described (Regan 1906, p. 437) as already rounded, and with concave crowns on a cryptacanthus only 700 mm. long (the type speci- men), whereas various transitional stages are represented on a *coelolepis of about that same size (690 mm., Bigelow and Schroeder 1954, p. 48, fig. 2). But with both *coelolepis and cryptacanthus recorded from Madeira, the desirability is self- evident of comparison of the type of cryptacanthus (in the British Museum) with northern specimens of *coelolepis. *C. owstoni from Japan (the type specimen of which is in the Museum of Comparative Zoology) differs from *coelolepis of the Atlantic in a considerably longer snout, a somewhat larger eye, a considerably larger second dorsal fin relative to the size of the first dorsal, with the rear tip reaching more nearly to the upper origin of the caudal fin, and in relatively smaller — consequently more numerous — dermal denticles on the sides of the trunk. The New Zealand shark that was described by Waite (1910) as Centrophorus plunketi, but which has been referred by Gar- man (1913 p. 210) to Scymnodon, and by Fowler (1933, p. 239; 1941, p. 228) to his subgenus Proscyinnodon, is set apart from the coelolepis-owstoni pair hy the tridentate shape of the denti- cles on the sides of the trunk on large specimens (type specimen a female 1414 mm. long) — presumably on small specimens also. And its denticles are both described and pictured (Waite 1910, p. 386, text fig.) as having as many as eight separate roots. The tips, also, of its pelvic fins are pictured (Waite 1909, PI. 37) as reaching back only about as far as a perpendicular at the point of emergence of the second dorsal fin spine (they reach to a perpendicular at about the midpoint of the free lower margin of the second dorsal fin both in *coelolepis and in * owstoni), while the second dorsal fin is pictured as considerably larger in plunketi than it is either in *coeloIepis or in * owstoni. Dr. J. A. F. Garrick's study of the type specimen (a juvenile 90 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY male, 318 mm. long) of the shark described by Thompson (1930) as Scymnodon waitei places it definitely in the genus Centro- scymnus as here defined, while his detailed account of it, with excellent illustrations, makes possible, for the first time, a satis- factory vicAv of such of its characteristics as are the most evidently diagnostic of the species. Briefly, these are : snout in front of mouth shorter than distance from mouth to level of first gill openings; head, to origin of pectorals, about 20 per cent of total length ; distance between inner ends of preoral clefts nearly or quite as long as between nostrils; base of second dorsal about twice (1.9) as long as base of first dorsal; dorsal fin spines with only the tips exposed; caudal fin without subterminal notch ; dermal denticles on sides of body strongly tridentate, on four-angled base, and rising steeply from the skin ; the outer exposed surface concave, M'ithout axial ridge ; the denticles on the lower side of the snout ovate, with pointed tip and with prominent axial ridge. (For further details, including teeth, see Garrick 1955, pp. 233-238, figs. 1, 2.) Thompson (1930, p. 278) followed by Richardson and Garrick (1953, p. 35) has suggested that waitei may prove to represent a stage in the growth of Centroscymnus j)lunketi, intermediate between the adult female that was the basis of Waite's original (1910) account of the latter, and the embryo, 165 mm. long, that was pictured subsequently by him (Waite 1914, p. 127, PI. 3), from a litter of 36 taken from a female measuring 1398 mm. Solution of this question is, however, a matter for the future. On the Magellanic shark named C. macracanthus by Regan (1906, p. 436), the dorsal fin spines (first as well as second, Fig. 12 E, F) project farther from the skin than they do in any of the other representatives of the genus discussed so far, though not as much as is suggested by Regan's (1906, p. 437) character- ization of them as "strongly projecting." Its two dorsal fins also — especially the first — are larger relatively than in the specimens of *coelolepis and of *owsioni that we have seen, more nearly paralleling plunketi in this respect.^^ And the distance between the inner ends of its preoral clefts is nearly as long as 15 We are indebted to Dr. Ethelwyn Trewavas and to Mr. Denys W. Tucker of the British Museum (Natural History) for tracings of the first and second dorsal tins, and to Dr. Ethelwyn Trewavas for the accompanying drawinir of a lower tooth (Fig. 12 G), and for other details of the type specimen. BIGELOW AND SCHROEDER : SQUALOID SHARKS 91 between the inner ends of the nostrils, to judge from Regan's (1906, p. 436) statement "length of anterior labial fold about equal to its distance from the symphysis." The flank denticles are tridentate on the type (and only known) specimen, but the latter is so small (640 mm. long, Regan 1906, p. 437) that this may represent a transitory state ; those on mature specimens may be of a different shape, as they are in *C. coelolepis (p. 87). In the species of Centroscymnus so far mentioned, the gap between the inner ends of the preoral clefts is at least % as long as the distance between the nostrils. In the three named species yet to be considered {*crepidater and *jonsonii of the northeast- ern Atlantic; rossi of the Indian Ocean) the clefts are so much longer that the space between their inner ends is only about one- third as long as between the nostrils (Figs. ID, 12B). Other than this, the most sharply diagnostic features of *crepidater are that its flank-denticles are tridentate on large specimens (Fig. 12D), as they also are in C. plmiketi (p. 89), and that its first dorsal fin spine projects from the skin nearly as far as in C. macracanthus (cf. Fig. 12A Avith 12E, F). The original (and only detailed) account of *G. jonsonii (Saemundsson 1922, p. 192, PI. 5) does not include any com- parison with *crepidaier, nor does it suggest any definite ground for considering it as distinct from the latter. Comparison of seven specimens (male and female, 587-796 mm. long) from south of Iceland and southwest of the Faroes ^"^ with the 787 mm. female from Madeira (Fig. 12A), on which Giinther (1870, p. 421) based his first-hand account of *crepidater,^' has revealed no significant differences, whether in teeth, in the shape of the denticles on large specimens, in bodily proportions ( p. 93), or in the shapes, the sizes or the relative locations of fins. Con- sequently we have no hesitation in classing * jonsonii as a syn- onym of *crepidater, the known range of which extends (on this basis) from Madeira and the coast of Portugal northward to the Faroes and southern Iceland. 16 Kindly made available to us by the Natural History Museum, Reykjavik, and by the University Zoological Museum, Copenhagen. 1" We owe to the British Museum the opportunity to examine this specimen. Its identification as *crepidater has been checlsed through correspondence with Dr. A. M. Kamalho, who has contributed measurements and photographs of the teeth of the type specimen which he has examined in Lisbon on our behalf. 92 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY r-l O BIGELOW AND SCHROEDER : SQUALOID SHARKS 93 Dr. J. A. F. Garrick informs us, also/^ that the Canterbury Museum contains two specimens of a shark taken on a long line set at 914 meters (500 fathoms) off Keikoura, New Zealand, and two others presumably from New Zealand but without definite locality, that differ from *crepidaier (including *jon- sonii) of the eastern North Atlantic only in such minor partic- ulars that it is doubtful Avhether specific separation is warranted. And this we can corroborate, for we find no significant differ- ences in bodily proportions, fins, teeth or denticles between one of these New Zealand specimens (kindly sent us by Dr. Garrick) and the Atlantic specimens that we have seen of *crepidater (pp. 93-94). Proportional dimensions in per cent of total length of *Cen- troscymnus crepidater. A, female, 787 mm., Madeira, from the British Museum (Natural History) ; B, four females, 746 to 796 mm. long and C, male, 587 mm. long, off Iceland and the Faroes, from the University Zoological Museum, Copenhagen ; D, female, 822 mm. long, presumably from New Zealand (see above), Mus. Comp. Zool. No. 39570.' Fig. 1:2. A, Centroscymnus ci-epidater, female, 787 mm. long, from Madeira, in British Museum. B, lower surface of head of same, to show the great length of the preoral clefts. C, upper and lower teeth of same, x about 2.3. D, denticles of same, from side below first dorsal fin, x about 6.5. E, Centroscymnus macracanihiis, outline of first dorsal fin of a female, 640 mm. long, type specimen, in British Museum, from Magellanic region. F, outline of second dorsal fin of same; E and F, after tracings of the. fins kindly contributed by Mr. Denys W. Tucker. G, third right hand lower tooth of same, x about 7, from drawing kindly contributed by Dr. Ethelwyn Trewavas. H, Centroscymnus coelolepis, caudal fin of female, 1035 mm. long, from offing of Delaware Bay, Mus. Comp. Zool. No. 38295. 18 Information contributed by Dr. Garrick in advance of publication. 94 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY A B C D Trunk at origin of pectoral : breadth 8.9 8.3-9.3 8.7 9.0 height 7.4 6.7-8.8 8.2 7.1 Snotit in front of : eye 8.6 8.2-9.3 8,3 8.9 mouth 11.4 11.4-12.7 11.7 11.8 Eye: horizontal dianietpr 4.4 4.4-4.6 4.6 4.5 Mouth: breadth 6.1 6.1-7.5 7.5 7.1 Xostrilft: distaiiee between inner ends 3.3 3.1-3.6 3.2 3.6 Preoral clefts: distance between inner ends 1.1 1.1-1.3 1.0 0.7 (iill opening lengths: first 1.5 1.3-1.5 1.4 1.3 third 1.5 1.3-1.5 1.4 1.3 fifth 1.5 1.3-1.8 1.6 1.6 First dorsal fin: vertical height 5.2 5.1-5.5 5.4 4.6 length of base from origin of spine 7.5 5.9-6.7 7.0 6.6 Second dorsal fin: vertical height 4.8 4.3-5.9 5.4 5.0 length of base from origin of spine 7.1 6.3-6.8 7.0 7.3 Caudal fin : upper margin oo o 19.0-20.7 20.8 18.9 lower anterior margin 13.2 10.7-13.5 13.0 13.4 Pectoral fin : width 5.3 5.2-6.8 4.2 5.2 length of anterior margin 13.1 11.8-13.1 11.4 12.4 Distance from snovt to : first dorsal 28.9 28.1-30.5 30.7 29.2 second dorsal 60.2 61.0 04.2 60.3 61.7 upper caudal 77.8 79.3-81.5 79.2 81.1 pectoral 24.4 23.7-24.1 24.7 23.3 pelvics 60.6 61.0-65.0 .16.3 62.2 Interspace between: origins of first and second dorsal spines 30.9 28.0-31. .-; 32.0 29.5 second dorsal and caudal G.l 6.9-8.0 7.0 7.7 pelvics and caudal 8.0 7.4-8.8 lO.O 7.9 Distance from origin to origin of : pectorals and pelvics 36.1 36.6-40.7 31.7 40.5 pelvics and caudal 15.8 15.2-17.9 17.9 15.7 Teeth, number of series : uppers 43 39-48 51 36-f^'> lowers 36 32 36 32 364-1=' This adds to the list of cases among sharks and skates where a species, or pair of very closely allied species, is known from temperate latitudes in the two hemispheres, hut not from the tropical-subtropical belt that intervenes. C. rossi from the Indian Ocean, if represented correctly in Alcock's (1899, PI. 26, Fig. 3) illustrations, differs from *crep- idater in a much longer head, and in detail as to the vinpaired fins, especially the caudal which appears to lack the subterminal 19 Several teeth have been lost at the corners of the mouth. BIGELOW AND SCHROEDER : SQUALOID SHARKS 95 notch ; perhaps also in the shapes of the teeth, though the illus- trations of the latter by Alcoek appear to be someAvhat diagram- matic. The denticles on the sides of the trunk are stronglv tridentate on juveniles (represented by the type specimen 254 ram. long), much as they are in *C. coelolepis (Bigelow and Schroeder 1954, p. 48, Fig. 2B) ; the shape is not known for adults. The relationship of C. fnscus of southern African waters to other members of its genus remains doubtful, for while it falls with *coelolepis in length of snout, in the sizes and relative posi- tions of fins, and in the concave crowns of the denticles on the sides of its body when mature (type and only known specimen 1,100 mm. long),^° the length of its preoral clefts is not known. And the characterization of it in the original description (Gil- christ and von Bonde 1924, p. 2, our only source of information) as with orbits 1.7 times as long as the snout (''orbit .6 in snout") seems an obvious error, for this would credit it with eyes far larger than those of any related shark. Key to Species of Centroscyninus 1. Distaiue between inner ends oi preoral clefts at least 7.j per cent as long as distance between nostrils 2 Distance between inner ends of preoral clefts is not more than abont 30 per cent as long as distance between nostrils C -. Exposed part of lirst dorsal fin spine, measured along its anterior margin, is very nearlj- 50 per cent as long as distance from its point of emergence from tlie skin to rear end of base of the fin (Fifj. 1'2F) . . macracanthus Began 1906. Magellanic region, p. 90 Dorsal tin spines exposed only at their extreme tips, if at all ii 3. Both of the dorsal fin spines are wholly concealed by the skin cryptacantlius Regan 1906. Madeira, p. 88 The tips of the dorsal fin spines protrude more or less from the skin . .4 4. First dorsal fin, from point of emergence of spine to rear end of base, is about as long as second dorsal fin, similarly measured . ^coelolepis Bocage and Capello 1864. North Atlantic, east and west. p. 88 First dorsal fin, from point of emergence of spine to rear end of base. is shorter than second dorsal fin, similarly measured r> 20 .Apparently the tj'pe specimen has been lost (Smith 1949, p. 58). 96 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 5. Snout in front of mouth is at least as long as from mouth to level of first gill openings *owstoni Garman 1906. Japan, p. 89 Snout in front of mouth is noticeably shorter than from mouth to level of first gill openings plunlceti Waite 1910. New Zealand, p. 89 ; also waitH Thompson 1930, New Zealand, per- haps a young stage of plunheti (see discussion, p. 90-). 6. Head, to level of origin of pectoral fins, is only about 24-25 per cent of total length ; caudal fin with distinct subterminal notch *crepidater Bocage and Capello 1864, including *jonsonii Sae- mundsson 1922. Eastern North Atlantic, p. 91 ; also New Zea- land, or represented there by a close ally (see discussion, p. 93). Head to level of origin of pectoral fins about 31 per cent of total length, caudal fin without distinct subterminal notch rossi Alcock 1898. Indian Ocean, p. 94 Systematic position in the genus doubtful fvscus Gilchrist and von Bonde 1924. Southern Africa, see discus- sion, p. 95. Genus SCYMNODON Boeage and Capello 1864 Seymnodon Bocage and Capello 1864, p. 263; 1866, p. 31, type species *S. ringens Bocage and Capello 1864, p. 263; 1866, p. 32, PI. 1, fig. 1; PI. 3, fig. 2A ; type locality off Portugal. Generic Synonyms: Ceniroplwrus in part, Giinther 1870, p. 423, for *Scymno(lnn ringens Bocage and Capello 1864. Zameus Jordan and Fowler 1903, type species Centrophorus squamulosu.t Giinther 1877, Japan. Not Ce7itrophorus Miiller and Henle 1837, type species *Sgualus granulosus Bloch and Schneider 1801. Generic characters. As in Centroscymnus (p. 85), except with the upper teeth much longer midway along each side of jaw than either toward the center of the mouth or toward its outer corners (Fig. 13C). Lower teeth triangular, highest and symmetrical along median sector of jaw, shorter but only weakly BIGELOW AND SCHROEDER : SQUALOID SHARKS 97 oblique toward outer corners of mouth ; first and second dorsal fin spines scarcely projecting beyond the skin ; dermal denticles on sides of trunk scale-like, tridentate, the outer surface with longitudinal ridges alone in some species, including the type but also with cross ridges (Fig. 13, F, G) in one species {ohscurus Vaillant 1888) that seems referable to Scyninodon by its denti- tion. Maximum recorded length 1100 mm. for ring ens (Bocage and Capello 1866, p. 32). Depth range. One of the four known representatives of this genus has been reported from a depth as great as 1400-1435 meters (one specimen, Yaillant 1888, p. 68) ; a second, sqiiamu- losus (Giinther 1887, p. 6), from 631 meters. A young specimen, however (the only one known) of a third was found on the beach at North Brighton, New Zealand (Archey 1921, p. 195). No information in this respect is at hand for the type species of the genus, so far known only from Portugal and from the northern coast of Spain (Rey 1928, p. 457). Remarks. Giinther (1870, p. 420) thought that "the passage . . . from reclining to erect teeth in the lower ' ' jaw is too gradual to justify the separation of Scyninodon from C entroscymnus ; or the separation of either of these from CentropJiorus, for that matter. But the difference in dentition between the species re- ferred here to Scymnodo7i and those referred to C entroscymnus (p. 88) is in fact so conspicuous that a glance at the mouth is enough to place a given specimen in the one category or in the other (provided it is properly referable to either), which makes the retention of Scymnodon as a separate genus a matter of convenience, even if nothing more. The contrast is also interesting between the pointed or tri- dentate denticles, with longitudinal ridges, on the sides of the trunk on mature specimens of Scymnodo7i and the replacement of denticles of this type by evenly rounded ones with concave blades, in some of the species of Centroscymnus (p. 86). But this difference can no longer be regarded as alternative between the two genera, for the denticles are tridentate in the adults of two of the species {*crepidater Bocage and Capello 1864, and plunketi Waite 1909), the teeth of w^hich place them in Centro- scymnus (p. 88). In any case choice of the denticles rather than of the dentition as the primary distinction between Seym- 98 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY CO w O BIGELOW AND SCHROEDER : SQUALOID SHARKS 99 nodon and Ceniroscymnus, as by Garman (1913), bv Fowler (1941), and by Bigelow and Schroeder (1948) would provide no basis on which to refer a given specimen to the one genus rather than to the other, unless it chanced to be full grown or nearly so. AVhile the only illustration of one of the three known species of Seym nodon {squamulosns Giinther 1877) show^s no sign of fin spines, the artist seems to have overlooked them, for Gunther (1887, p. 6, PI. 2, Fig. B) wrote of them as "very small, scarcely projecting beyond the skin." Species: Named species referable to Scymnodon as defined here are : *S. ringens Bocage and Capello 1864, type species, offing of Portugal; Centrophonis squamnlosus Giinther 1877, Japan; and Ceniroscymnus ohscurus Vaillant 1888, off the coast of north- western Africa,-^ in a haul from 1400-1435 meters. The New Zealand shark, also, that was described by Archey (1921) as Scymnodon sherwoodi has been retained in that genus by the several authors (Whitley 1934, 1940; Fowler 1941; Big-elow, Schroeder and Springer 1953 ; and Richardson and Garrick 1953), who have had occasion to refer to it. But we now learn from Dr. J. A. F. Garrick" that the type (and only known) specimen, which he has recently examined, shows no trace of a spine in either of its dorsal fins, and that it is therefore one Fig. 13. A, Scymnodon ringens, female, 893 mm., long, coast of Portugal, ill British Museum. B, ventral view of anterior part of head of same. C, upper and lower teeth of same, left hand side of mouth, x about 1, midpoint of jaws marked bj' the dotted line. D, dermal denticles of same, from side below first dorsal fin, x about 7. E, dermal denticles of same from side of caudal peduncle x about 7. F, Scymnodon ohscuriis, dermal denticle of type specimen, female, 590 mm. long, off northwest Africa (p. 100), from side below first dorsal fin, x about 18, made available to us through kindness of Dr. P. Budker. G, dermal denticle of same, after Vaillant, 1888, PI. 2, Fig. 2. H, lower teeth of same, from drawing con- tributed by Dr. P. Budker, x about 4. I, Scymnodon squamulosus, type specimen, female, 658 mm. long, Japan, in British Museum, outline of caudal fin from drawing contributed by Mr. Denys W. Tucker. 21 The locality as given by Vaillant (1888, p. 68), is "sur les c6tes du Soudan," a designation referring to the general region from southern Morocco to Senegal (Folin, 1887. pp. 287-300). 22 Information contributed by Dr. Garrick in advance of publication. 100 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY of the Dalatiinae not of the Squalinae, as the sub-families are defined here (p. 18). We have pointed out elsewhere (Bigelow and Schroeder 1954, p. 51) that the juvenile shark from the northwestern Atlantic, on which Bigelow, Schroeder and Springer (1953, p. 233) based their new species *Scymnodon melas, actually represented a juvenile stage in the growth of *Centroscymnus coelolepis Bocage and Capello 1864. The dentition of ohscurus closely resembles that of *ringens, as Dr. Paul Budker has kindly verified for us from examination of the type specimen of ohscurus in the Paris Museum. Indeed, Garman (1913, p. 208) has definitely rated ohscurus as a synonym of *ringens. But a more abruptly truncate caudal fin, with more definite subterminal notch and smaller pectorals in oh- scurus (as pictured by Yaillant 1888, PI. 2, Fig. 2) than in *ringens, and flank denticles with a double series of transverse ridges between the more prominent longitudinal ridges (Fig. 13, F, G; see also Vaillant 1888, PL 2, Fig. 2C) argue against this union, hence ohscurus is retained here as a distinct species. It is interesting, also, that in ohscurus some of the denticles on the sides of the trunk of a given individual may be tridentate, while others are not (Fig. 13, F, G). The Japanese squamulosus (Giinther 1887, p. 5, PI. 2, Fig. B) agrees with * ring ens in the shape of its denticles. But an exam- ination of the type specimen in the British Museum by Mr. Denys W. Tucker (to whom we are greatly indebted), shows its gill openings as only about 50 per cent as long, relatively, as those of *ringens; its snout as longer relatively and more pointed (length from tip of snout to mouth about as great as distance between inner ends of preoral clefts) ; and both its sec- ond dorsal, its pectoral and ventral fins, also its caudal (Fig. 13, I), as differing so conspicuously from those of *ringens, that the two species are clearly distinct. In all these respects, in- deed, squamulosus so nearly resembles ohscurus (as pictured by Vaillant 1888, PI. 2, Fig. 2) that these two species might easily be confused. But the denticles on the sides of the trunk of squamulosus (as pictured by Giinther 1887, PI. 2, fig. B) show no trace of the transverse ridges that mark the denticles of ohscurus (Fig. 13, F, G). BIQELOW AND SCHROEDER : SQUALOID SHARKS 101 Key to Species of Scymnodon 1. Lower outline of caudal fin nearly continuously rounded, -with sub- terminal notch only faintly marked, and without definite lower-anterior lobe (Fig. 13 A) ; pectoral fin, when laid back, reaching very nearly to a perpendicular at base of first dorsal fin spine *ringens Bocage and Capello 1864, eastern North Atlantic, p. 99 Lower outline of caudal fin with well-marked subterminal notch and prominent lower anterior lobe (Fig. 13, I) ; pectoral fin, when laid back, falls short of a perpendicular at base of first dorsal fin spine by a distance at least 50 per cent as long as snout anterior to eyes 2 2. Outer surface of dermal denticles on sides of trunk with weak trans- verse ridges in addition to the three longitudinal ridges (Fig. 13, F, G) ohscurus Vaillant 1888, off coast of Northwestern Africa, p. 100 Outer surface of dermal denticles on sides of trunk without trans- verse ridges squamulosus Giinther 1877, Japan, p. 100 Genus DeaNIA Jordan and Snyder 1902 Deania Jordan and Snyder 1902, p. 80, type species, *D. eglantina Jordan and Snyder, Japan. Generic Synonyms: Acanthidium in part, Lowe, 1839, p. 92, for *A. calceum Lowe, Madeira. Centrophorus in part, Lowe, 1843, p. 93, and subsequent authors, for C. calceus Lowe 1843, equals * Acanthidium calceum Lowe 1839. Centrophorus in part, Bocage and Capello 1864, 1866, for C. crepidalius Bocage and Capello, evidently equals *Acanthidiuvi calceum Lowe 1839. Scymnodon in part, Goode and Bean 1895, by error, their PI. 4, Fig. 12, labelled ' ' S. ringens Bocage and Capello ' ' in the explanation of plates being a copy of the illustration by Bocage and Capello (1866, PI. 2, Fig. 1) of Centrophorus crepidalhus Bocage and Capello 1864, equals Acanthidium calceum Lowe 1839. Nasisqualus Smith and Eadcliffe 1912, p. 681, type species *N . profundorum Smith and Eadcliffe, Philippines. Acnnthidium Garman 1913, p. 215; type designated as *A. calceum Lowe 1839; but incorrectly so, because *A. pusillum Lowe 1839 (which is a species of Etmopterus Eafinesque 1810, see below) had previously been designated as the type of Acanthidium by Jordan and Evermann (1896, p. 55). Daeniops Whitley 1932, p. 36, type species Acanthidium, quadrispinosum McCulloch 1915, Australia. 102 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Not Centrophonis Miiller and Henle 1837, type species *Sqiuilus granulosus Bloeh and Schneider 1801. Not Acanthidium Lowe 1839, type species *A. pu^illum Lowe as designated by Jordan and Everraann 1896, p. 55, and by Jordan 1919, p. 195, making Acanthidium a synonym of Etmopterus Eafinesque 1810 (sec above). Not Scymnodon Bocage and Capello 1864, 1866, type species *S. ringenn Bocage and Capello 1864. Generic characters. Inner corner of pectoral fins rounded in most species, perhaps rectangular in some, but not extended (p. 103) ; snout in front of mouth longer than from mouth to level of origin of pectoral fins; preoral clefts only very slightly expanded inwardly ; dermal denticles on sides of trunk high, pitchfork shaped ; teeth smooth-edged in all known species ; other characters as in CentropJiorus (p. 64). Maximum recorded lengths, 1138 mm. {qnadrispinosa Mc- (.^ulloch 1915, p. 100) ; 1070 mm. {kaikourae Thompson 1930, as "calceus") ; and 1060 mm. {calcea, Vaillant 1888, p. 71). Depth range. The depths of capture so far recorded for Deaiiia (mostly, at least, in the trawl, hence subject to the un- certainty emphasized on p. 5) are: 605-1431 meters for calcea (Roule 1919, p. 119) ; 715-1785 meters for *profundorum (Smith and Radcliffe 1912, p. 683) ; 190-366 meters for *eglantina (Smith 1949, p. 58) ; and 238-823 meters for qnadrispinosa (McCulloeh 1915, p. 100; Whitley 1940, p. 147). While definite information as to depth is wanting for kaikourae, it was reported from "deep water" (Thompson 1930, p. 276, as "calceus"). And the Jap- anese records for *acicidata (Garman) 1906, *rostrata Garman 1906, and hystricosa (Garman) 1906 which we here refer to *eglantina (p. 104) were from the deep line-fishery in Sagami Ba3^ Evidently Deania can fairly be termed a deep-water genus. Remarks. The chief respects in which the various species that have been grouped together as Deania (or as one or other of its synonyms) difiler from Centrophonis are: a much longer snout, and pitchfork-shaped denticles on the sides of the body. Most of the species of Deania are set further apart from Centrophonis by the rounded shape of the inner corner of their pectoral fins. But the two genera intergrade in this respect, the inner corner of the pectorals being quadrate, both in one member of Centro- phonis ifoliaceus Giinther 1877), as noted above (p. 66), and BIGELOW AND SCHROEDER : SQUALOID SHARKS 103 also in one member of Deania, if Gilchrist's (1922, PI. 7, Fig. 2) illustration of his D. natalense represents the latter correctly.^^ Deania differs from Centroscymnns and from Scymnodon in a much longer snout; in a longer exposed portion of the second dorsal fin spine ; and in broader upper teeth. It differs further from Scymnodon in that the teeth in the central part of its upper jaw are nearly or quite as long as those midway out along either side. It had appeared also until recently that its high, pitch- fork-shaped denticles were equally diagnostic of Deania as con- trasted with Centrophorus and with the C entroscymnus-Scymno- don group. But Garriek (1955, p. 234, Fig. 2) has found that the denticles on the sides of the trunk of young specimens of Centroscymnus ivaitei (Thompson) 1930 of New Zealand (p. 90) approach the pitchfork-like shape that was formerly thought to be diagnostic of the denticles of Deania. The cusps of the upper teeth of Deania are sharp-tipped, rather narrowly triangular, with concave margins, on broad bases ;'■* erect along the central part of the jaw, but increasingly oblique toward the outer corners of the mouth. The lower teeth are very strongly oblique all along either side, and the median tooth is not symmetrical in the female of any species for which the condition in this respect is known. In *calcea (Lowe) 1839 (the oldest species), the lowers are of the same shape in the males as in the females, as illustrated by a male of 848 mm. and a female of 1037 mm. that we have seen from southwest of the Faroes.'^ Similarly, the lowers are de- scribed and pictured as stronglj^ oblique in the male in *pro- fundorum (Smith and Radcliffe 1912, p. 681, PI. 53), which we can verify from our examination of the type specimen. But the lowers are described by McCulloch (1915, p. 101) as much less strongly oblique in the males of quadrispinosa, from Australia, than in the females, which is also the case in the Japanese species-complex that we unite here (p. 104) under the name *eglantina Jordan and Snyder 1902. Thus the lowers (including the median tooth) of the female pictured by Garman 23 The pectoral fins are not mentioned in Gilchrist's (1922, p. 49) description of his natalense. 2* Our re-examinatlon of the type specimen of *Deania eglantina (type of the genus) has shown that Jordan and Snyder's (1902, p. 81) description of the teeth as with "small basal cusp" is not correct. 25 We are indebted to Dr. J. R. Pfaff, of the University Zoological Museum, Copenhagen, for the opportunity of seeing these specimens. 104 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY (1913, PI. 11, Fig. 7) as Acanthidmm hystricosum^^ are shown as strongly oblique, whereas those of the adult male figured by him (Garman 1913, PI. 12, Fig. 2) as *Acanthidium acicula- tum (which we have examined) are erect in the front of the mouth, with the median tooth symmetrical. And the reserve rows of lower teeth of an immature male, 682 mm. long, in the Museum of Comparative Zoology, are similarly nearly erect along the center of the jaws (the functional row is badly worn). The situation in this respect is not known for natalense or for kaikourae. It is a matter of common knowledge that sexual dimorphism of this sort is common among batoid elasmobranchs, but it seems not to have been reported previously among sharks. Species: Named species referable to Deania: are * Acantkidium calceum Lowe, 1839, eastern Atlantic; * Deania eglantina Jordan and Snyder 1902, type species, Japan ; * Acanthidium, acicMlatum Garman 1906, A. hystricosum Garman 1906, and *A. rostratum Garman 1906, all from Japan ; *Nasisqualus profimdorum Smith and Radcliffe 1912, Philippines; Acanthidium quadrispinosum McCulloch 1915, Australia; Acanthidium natalense Gilchrist 1922, Natal, southeastern Africa; and Centrophorus kaikourae Whitley 1934, New Zealand, which was originally reported and pictured by Thompson (1930, p. 275, PI. 42 as Centrophorus calceus Lowe). Our re-examination of the type specimens of *eglantina (a female of 300 mm.), *aciculata (a male of 890 mm.) and *ro- strata (a female of 873 mm.) has revealed no differences between them in proportional dimensions (see below) greater than may reasonably be credited to the difference between them in stage of growth, or to the variability that has already been reported by Regan (1908, p. 52) for the allied *calcea of the Atlantic. Cases in point are the somewhat longer head and caudal fin, but shorter interdorsal space of the * eglantina. And while the type specimen of the fourth member of this group {hystricosa) is no longer to be found, Carman's (1913, PI. 11, Figs. 5-8) beautiful illustrations of it, by the well-known zoological artist E. N. Fischer, do not suggest anything to set it apart either. Hence, all deanias yet reported from Japan are referred here to the one species * eglantina Jordan and Snyder 1902. 26 This specimen is no longer to be found. BIGELOW AND SCHROEDER : SQUALOID SHARKS 105 Proportional dimensions in per cent of total length. A. *Deania eglantina, female, 300 mm., type, U.S.N.M. No. 49524; B. *D. rostrata, female, 873 mm., type, M.C.Z. No. 1047; C. *D. aciculata, male 890 mm., type, M.C.Z. No. 1128; all from Japan. A B C Trunk at origin of pectoral: breadth 9.3 7.5 6.7 height 7.7 8.0 6.7 Snout in front of: outer nostrils 5.3 5.3 4.0 mouth opening 15.3 13.7 12.1 Eye: horizontal diameter 5.3 3.6 4.7 Mouth: breadth 7.7 5.6 6.4 Nostrils: distance between inner ends 4.5 3.4 3.4 Preoral clefts: distance between inner ends 4.8 3.4 3.7 Gill opening lengths: first 2.0 — 1.7 third 2.0 2.5 1.7 fifth 2.0 2.1 1.8 First dorsal fin: vertical height 3.8 3.5 2.9 length of l)ase from spine origin 9.5 9.7 11.8 Second dorsal fin: vertical height 4.7 5.5 4.8 length of base from spine origin 9.2 10.0 10.0 Caudal fin: upper margin 20.0 16.0 16.8 lower anterior margin 11.6 10.8 9.8 Pectoral fin : width 6.2 6.1 6.1 length of anterior margin 9.3 9.0 8.9 Distance from snout to : 1st dorsal spine 42.3 39.9 40.2 2nd dorsal spine 68.2 70.5 70.5 upper caudal 80.0 84.0 83.2 pectoral 28.7 24.0 23.2 pelvics 61.6 64.7 62.6 Interspace hetiveen: origins of first and second dorsal spines, at base 24.6 30.5 30.7 second dorsal and caudal 3.8 3.2 3.4 pelvics and caudal 8.9 10.3 10.1 Distance from origin to origin of : pectorals and pelvics 33.3 40.8 39.3 pelvics and caudal 13.3 13.4 14.4 Teeth, ■number of series: uppers 28 34 29 lowers 13-1-14 ? 29 The Japanese *eglantina (inclnding Garman's supposedly dis- tinct species, as just noted), is so closely allied to *Deania calcea of the eastern Atlantic that Regan (1908, p. 51) has already relegated it unequivocally to the synonymy of * calcea, and 106 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY *aciculata doubtfully so. But it seems to us wiser to retain *eglantina as a separate species, at least until a large number of specimens of the Japanese and North Atlantic populations have been studied, for while the Japanese specimens listed above agree very closel}' indeed Avith the two *calcea we have seen, in fin characters and in bodily proportions in general, the lower teeth are as oblique in the males of *calcea as in the females, but are more erect in mature males than in females in *cglantina, as noted above (p. 103). Though this last difference would not be an aid to identification unless the species in hand chanced to be well-grown males, it deserves taxonomic recognition of some sort. And the ([uestion as to whether the North Atlantic and North Pacific populations of this general type have diverged significantly since they became isolated geographically, one from the other, is one too interesting to be answered offhand. The * calcca-* eglantina group of the Northern Hemisphere has close counterparts, in the Southern Hemisphere, in quadri- spinosa of Australia ; also in a shark reported and pictured from southern African waters by Smith (1949, p. 58, Fig. 49) as '*eglantina, which certainly resembles the type specimen of *eg- lantina very closely. But the eventual decision as to the identity of either or both of these with ^eglaniina of Japan must await the comparison of specimens from the different ocean areas. In the * calcea-* eglantina-quadrispinosa group (however many species this may eventually prove to represent), the pectoral fins, when laid back, fall considerably short of a vertical at the point of emergence from the skin of the first dorsal fin spine. But the pectorals of *profundorum from the Philippines are pictured as reaching rearward to abreast of the first dorsal spine (Smith and Radclift'e, 1912, PI. 53), and our own examination of the type specimen has shown this to be correct. As this is a rather conspicuous difference, * prof undornm may be accepted as a good species. The original illustration of natale7ise (Gilchrist 1922, PI. 7, Pig. 2) of southern African waters, shows the pectorals not only as reaching rearward beyond the first dorsal spine, but also with the inner corner quadrate, instead of rounded as they are in other known representatives of the genus Deania. But Smith's (1949, p. 58) allocation of it (with quadrispinosa) to the syn- onymy of *eglanihia — perhaps from examination of the type BIQELOW AND SCHROEDER : SQUALOID SHARKS 107 which Barnard (1925, p. 51) has reported as being in the collec- tion of the Government Marine Surve}- of South Africa — suggests that the artist's representation of natalensc may not have been correct. Hence its status remains in doubt, pending further information. Finally, kaikourae (falling with the * calcea-* eglantina group in length of pectoral fin) differs from all other known members of Deania in a first dorsal fin base about 1.6 times as long as the second (each measured from the point of emergence of the respective fin spine from the skin), contrasted with 0.9-1.2 times for the *calcea-*eglantina-quadrispinosa group, for *pro- fundorum, and for natalcnse. Thompson (1930, p. 276) indeed, has already called attention to this feature in his original ac- count of kaikourae (as "CentropJiorus calceus"). To sum up, the nine representatives of the genus Deania that have been named from various seas may represent as few as four species, *calcea in the North Atlantic, '*e.glantiiia in the North and South Pacific and southern Indian Ocean, kaikourae from New Zealand, and *profundorum in Philippine waters. Or they may represent as many as six, if quadrispinosa from Australia, and the South African form reported by Gilchrist 1922 as 7iai- alense, but by Smith (1949) as '*eglantina, should finally prove to be separable, not only from *cglantina of the North Pacific, but one from the other as well. Key to Species of Deania 1. Pectorals, when laid back, falling considerably short of a perpendicu lar at point of emergence of first dorsal fiu spine from the skin 2 Pectorals, when laid back, reach at least as far as a perpendicular at point of emergence of first dorsal fin spine 4 2. First dorsal fin, from point of emergence of spine to rear end of base is 1.5-1.6 times as long as second dorsal, similarly measured Icaikourae Whitley 1934, Australia, p. 107 First dorsal fin, from point of emergence of fin spine to rear end of base of fin, is only 0.9-1.2 times as long as second dorsal fin, similarly measured 3 3. Lower teeth as strongly oblique in males .is in females *calcea Lowe 1939. Northeastern Atlantic, including Mediterranean, p. 105 Lower teeth more erect in males than in females *eglantina 108 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Figure 14 BIGELOW AND SCHROEDER : SQUALOID SHARKS 109 Jordan and Snyder 1902 (incl. *aci- culata Garman 1906, *rostrata Gar- man 1906, and hystricosa Garman 1906), Japan, and reported from southern Africa (p. 106) ; also quadrispinosa Gilchrist 1922, Au- stralia, p. 106 4. Inner corner of pectoral fins rounded; first dorsal fin only about 70 per cent as high vertically as second dorsal fin *profundorwm Smith and Eadcliffe 1912, Philippines, p. 106 Inner corner of pectoral fins quadrate; first dorsal fin about as high vertically as second dorsal fin natalense Gilchrist 1922. Off Natal coast, southern Africa (but see discussion), p. 107 Subfamily DALATIINAE Subfamily characters. Squalidae lacking a second dorsal fin spine (lacking a first spine also in most cases) ; preoral clefts lacking in most but present in one of the known genera (Scym- iwdalatias, p. 124) ; dermal denticles either truncate with hollow crown, or with a single cusp or spine ; teeth with only one cusp, the uppers narrow, awl shaped or lancet shaped ; the lowers much broader, with quadrate base, each overlapping the next outward ; caudal axis raised in some but not in others ; the caudal fin varying widely in shape accordingly (Fig. 14), its tip more or less evidently truncate ; sides of body, anterior to cloaca, with- out longitudinal ridges. Fig. 14. Outlines of caudal fins of representative genera, adjusted to equal lengths along upper margin, to show relative breadth of fin above axis and below ; also degree to which caudal axis is raised. A, Dalatias UcJm, male, 1114 mm. long, Japan, Mus. Comp. Zool. No. 1116. B, Somnio- diis microcephalus, male, 1334 mm. long, Massachusetts Bay, Mus. Comp. Zool. No. 39609. C, Isistius brasilietisis, male, 383 mm. long, western North Atlantic north of the Bahamas, Mus. Comp. Zool. No. 36039. D, Squaliolus laticaudus, same specimen as in Fig. 16A. E, Echinorhinus bruous, male, 1650 mm. long, Mauritanian coast, northwest Africa, Mus. Comp. Zool. No. 39633. F, Lanina ditropis, type specimen, male, 2085 mm. long, off La Jolla, California, Mus. Comp. Zool. No. 36471, the vertebral axis indicated by the broken line. 110 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Figure 15 BTGELOW AND SCHROEDER : SQUALOID SHARKS 111 Genera. The shark species referable to this admittedly "arti- ficial" group have been distributed among the following named genera: (1) DaJatias Rafinesque 1810, type species D. sparo- phagus Rafinesque, equals *Squalus licha Bonnaterre 1788; (2) Somniosus Lesueur 1818, type species iS^. hrevipmna Lesueur, equals *Squalus microcephalus Bloch and Schneider 1801; (3) Euprotomicrus Gill 1864, type species Scymnus {Laemargiis) labordii Miiller and Henle 1841, equals *Scymnus hispinatus Quoy and Gaimard 1824; (4) Isis-tius Gill 1864, type species *Scymnus brasiliejisis Quoy and Gaimard 1824; (5) Heteroscyni- nus Tanaka 1912, type species *n. longus Tanaka; (6) Squalio- lus Smith and Radcliffe 1912, type species *S. longicaudus Smith and Radcliffe; (7) Heteroscymnoides Fowler 1934, type species *H. marleyi Fowler; (8) Pseudoscymmis Herre 1935, type species P. hoshuensis Herre; and (9) Scymnodalafias Garrick 1956, type species Scymnodon sherwoodi Archey 1921. There seems nothing, however, to set Pseudoscymnus apart from Dal- atias (Bigelow and Schroeder 1948, pp. 501, 502). And Hubbs and McHugh's (1951, p. 164, Footnote) suggestion that Hetero- scymnus should be reduced "perhaps to the synonymy of Somniosus," is supported by our examination of the teeth (PL 3) and denticles (Fig. 15B) of the one known species, *//. longus Tanaka 1912, from Japan. For further discussion of it see page 121. Fig. l.j. A, Somniosus longus, type specimen, female, 13G0 luni. luny, .lapan, al'ter Tanaka, the .shapes of the fins somewhat emended to accord with a Japanese male, 1010 mm. long, Miis. Comp. Zool. No. 396r)0, kindly (ontiibuted by Dr. Tokiharu Abe. B, dermal denticles of latter, from side below first dorsal fin, x about 25. C, side view of same. D, Somniosus rostratus, lower tooth of Mediten-anean female, 820 mm. long, after Canestrini, ISfi-l, PI. 2, Fig. 4. E, lower teeth near center of jaw, of Portuguese shark described by Capello 1870 under the name " rostratus'', traced from a photograph contributed by Dr. A. M. Eamalho. F, Hetero- scymnoides marleyi, type specimen (see Fig. 16E), upper and lower teeth at center of mouth, x about 12. G, dermal denticles of same, from side bv-low first dorsal fin, x 36. H, Squaliolus laticaudus, same specimen as in Figure IGA, denticles on side below first dorsal fin, x about 45. I, Dalatias licha, same specimen as in Figure 14 A, a sector of the exposed part of upper and lower lips, x about 4. 112 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY In short, the dozen or so known species of Dalatiinae that appear to deserve separate recognition seem referable to 6 or 7 genera as defined by the positions and relative sizes of the first and second dorsal fins, presence or absence of a spine on the first dorsal fin, nature of the teeth, shape of the dermal denticles on the sides of the body, and degree to which the portion of the caudal fin above the axis is developed relative to the portion below the axis. This last matter is one of especial interest, as already stressed by Hubbs and McHugh (1951, p. 164) in their key to genera, for in the genera in which the fin is the most widely expanded above the axis (Fig. 14), the latter is raised only very slightly above the main axis of the shark's body (if at all). In other words, one of the evolutionary trends of the group has been a tendency for the tail to lose the heterocercal nature so characteristic for it among sharks in general. Cau- dal fins even more lunate in form (Fig. 14F) are, it is true, well known characteristics of the isurid, basking, and whale sharks. But in all of these the caudal axis is steeply raised ; they are, in a word, as truly heterocercal as any other shark, more noticeably so, indeed, than many, such as the erectolobids, scyliorhinids and some of the triakids. Key to Genera of Dalatiinae 1. Lips with a complex series of fringed cross folds (Fig. 15, I) ; margins of lower teeth regularly serrate Dalaiias Eafinesque 1810. p. 113 Lips smooth, or nearly so; margins of lower teeth smooth, or only faintly and partially serrate at most 2 2. First dorsal fin with a spine, the tip either exposed or buried in the skin Squaliolus Smith and Radcliffe 1912. p. 128 First dorsal fin with no trace of a spine 3 3. Preoral clefts present ; upper margin of caudal fin about twice as long as lower anterior margin Scymnodalatias Garrick 1956. p. 124 No preoral clefts ; upper margin of caudal fin not more than 1.5-1.7 times as long as lower anterior margin, and relatively shorter than this in most 4 4. Eear end of base of first dorsal fin is over or posterior to origin of pelvic fins ; interspace between first and second dorsal fins is shorter than between second dorsal fin and origin of upper margin of caudal ; lower teeth erect, symmetrical Isistiiis Gill 1864. p. 123 BIGELOW AND SCHROEDER : SQUALOID SHARKS 113 Rear end of base of first dorsal fin is clearly anterior to origin of pelvics; interspace between first and second dorsal fins is at least as long as between second dorsal fin and origin of upper side of caudal ; lower teeth asymmetrical, the cusp more or less strongly oblique, outward 5 5. Origin of first dorsal fin is about over axil of pectorals Hetcroseymnoidcs Fowler 1934. p. 182 Origin of first dorsal fin is posterior to tips of pectorals when these are laid back 6 6. Base of first dorsal fin about as long as base of second dorsal ; eye only about 20-25 per cent as long as snout in front of mouth Somniosus Lesueur 1818. p. 115 Base of first dorsal fin only about 25-33 per cent as long as base of second dorsal ; eye more than nO per cent as long as snout in front of mouth Euprotomicrvs Gill 1864. p. 126 Genus DaLATIAS Rafinesque 1810 Dalatias Rafinesque 1810,^ p. 10; type species D. sparophagus Rafinesque, Mediterranean, equals *Squalus licha Bonnaterre 1788, designated by Jordan, Tanaka and Snyder 1913. For generic synonyms and refer- ences, see Bigelow and Schroedcr 1948, p. 501. Geyieric characters. Dalatiinae with blade-like lower teeth, the cusp erect, triangular, with serrate edges, the base quadrate with a conspicuous notch on the outer edge marking the transi- tion from cusp to base. Snout in front of mouth much shorter than from mouth to level of first gill openings ; longest gill open- ings about 8-11 per cent as long as head to origin of pectorals (specimens measured) ; lips only moderately expanded at corners of mouth, and without special cartilages, but conspicuously thick and fleshy inward along each jaw and close-set with a complex series of cross folds with finely fringed edges (Fig. 15, I) ; fur- row from corner of mouth reaching rearward about 30 per cent of distance towards first gill openings ; no preoral clefts, but with voluminous subdermal pouches extending forward from the oral pockets (Fig. IE) ; caudal fin of the shape shown in Fig. 14A, its axis rather steeply raised ; origin of first dorsal fin much nearer to rear edge of pectorals when these are laid back than to origin of pelvics ; second dorsal fin about 1.3 times as long as first dorsal at base ; no spine in either of the dorsal fins. Denticles on sides of body thick, quadrate, outer surface with three weak ridges united rearward in a point that varies 114 BULLETIN : MUSEUM OP COMPAEATIVE ZOOLOGY in length and in aeuteness from denticle to denticle; denticles on lower side of snout overlapping, ovate, without marginal teeth (Bigelow and Sehroeder 1948, Fig. 96, D, E). Greatest recorded length, 1820 mm. (Dumeril 1865, p. 452). Depth range. Definite depths of capture that we have found recorded for Dalatias in one part of the oceans or another have ranged from about 90 meters down to 600 meters. Probably it is more numerous in the sboaler part of its vertical range, else fishermen would hardly be as familiar with it as they have long been in the eastern side of the North Atlantic, from northwestern Africa and the Canaries to the Irish slope ; also in the Mediter- ranean. Remarks. The serrate lower teeth of Dalatias, and the shape of its caudal fin, with its conspicuously fleshy and complexly sculptured lips, set it so sharply apart that there is no danger of confusing it with any other genus of Dalatiinae. Species. It is probable that all the reports for Dalatias that have appeared in scientific literature during the past century and a half have been based on the single species *D. licha (Bonnaterre) 1788, type of the genus, which was originally de- scribed from "Cap Breton" on the coast of Brittany, by Brous- sonet (1780, p. 677), as "La Liche." In the eastern side of the Atlantic, Hicha is now known to be generally distributed from tropical West Africa (Rio de Oro), Morocco, the western Mediterranean, the Canaries, and Madeira to the Irish Atlantic slope and to the fishing grounds west of Ireland. Contrasting with this widespread occurrence, and with its local abundance in the Mediterranean and near Ireland, there have been onlv two pul_)lished records of its capture in the western Atlantic, the one for a female of about 5 feet (about 1550 mm.), taken on Georges Bank, August 19, 1937 (Nichols and Firth 1939, p. 85; Bigelow and Sehroeder 1948, p. 502) ; the other for an 845 mm. female from the northern part of the Giilf of Mexico (Big- elow, Sehroeder and Springer 1955, p. 10). But the recent cap- ture by "Oregon" of a female only about 380 mm. long (U. S. Nat. Mus. No. 157844) in the northern side of the Gulf, sug- gests that the latter may harbor a resident population. *D. licha has also been reported under its o-\vn name by Smith (1949, p. 56) from southern Africa, whence it had previoush been recorded as a separate species [D. hrevipinnis Smitli BIGELOW AND SCHROEDER : SQUALOID SHARKS 115 1936, p. 1); from New Zealand; from Australia; also from Japan both as JicJia and as americanus Gmelin 1789 (an obvious synonym for h'cha). Identification as licha in the case of the Japanese form is corroborated by our own comparison of a Japanese specimen with the specimens we have seen from Georges Bank and from the Gulf of Mexico. (For the list of New Zea- land, Australian and Japanese references for *licha, see Bigelow and Schroeder 1948, p. 508.) Nor do we find anything in Herre's (1935, p. 124) account of the Japanese shark for which he proposed the new generic and specific names Pseud oscymnus hohuensis to suggest that it differs in any way from *Z). licha. And this applies equally to McCulloch's (1914, p. 81, PI. 14) excellent account and illustration of the Australian shark that he described as "licha" but which Whitley (1931, p. 310; 1940, p. 151) rechristened phillippsi, although without including any supporting evidence to justify the new specific name. AVith the known range of *Z>. licha so extensive, it seems astonishing that the genus has not been reported as yet from the eastern side of the Pacific, south or north, or from the southern Atlantic, though it is known from Algoa Bay, south- eastern Africa. Genus SOMNIOSUS Lesueur 1818 Somniosiis Lesueur 1818, p. 222; type species S. brevipinna Lesueur, Massa- rhusetts, equals *Squalus TuicrocepJialus Bloch and Schneider 1801. For generic synonyms, see the list given by Bigelow and Schroeder 1948, p. 514, with the addition of Iletcroscymnus Tanaka 1912, type species 77. longus Tanaka, Japan (see p. 121). Generic characters. Dalatiinae, without fin spines ; teeth smooth edged, the lowers with the cusp directed so strongly outward in some species that the inner edges function as a nearly unbroken cutting edge parallel with the jaw, but more nearly erect in others (PI. 3) ; snout in front of mouth at least not longer than from mouth to level of first gill openings; longest gill openings about 9-11 per cent as long as head to origin of pectorals on specimens measured ; lips not conspicuously thick, their surface smooth ; furrows from corners of mouth reaching rearward only about 35-40 per cent of the distance toward first gill openings; no preoral clefts but with voluminous preoral pouches extend- 116 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY ing forward from oral pockets; caudal fin with upper division widely expanded, as well as the lower (Fig. 14B) ; upper margin about 1.5 times as long as lower anterior margin. Caudal axis moderately raised. Origin of first dorsal fin nearer to rear edge of pectorals (when these are laid back) than to origin of pelvics; second dorsal fin not larger than the first dorsal ; interspace between first and second dorsal fins 1.7-3 times as long as be- tween second dorsal and caudal ; denticles on sides of body coni- cal to thorn-like, curved rearward, differing widely in size from species to species, and in degree of elevation from the skin. Size. The different species of Somniosus differ widely in size. Thus rostratus is not known to grow longer than about one meter. But microcephalus averages 8-14 feet (2.4-4.3 meters) at maturity, with specimens of 16-18 feet (4.9-5.5 meters) taken occasionally, and one of 21 feet (6.4 meters) definitely recorded (see Bigelow and Schroeder 1948, p. 519 for further informa- tion). Depth range. The best known member of the genus {*micro- cephalus, p. 117) has long been known to occupy a wide depth range, coming right up to the ice in the northern part of its range (Greenland, Labrador) in winter, but in summer most often caught at 180-550 meters, with one recorded from 1207 meters. (For further details see Bigelow and Schroeder 1948, p. 521.) And the depth range is, presumably, about the same for its representative in the North Pacific {pacificus, p. 121), though precise information is lacking. The only definite depth record we find for S. rostratus of the eastern Atlantic, said by Tortonese (1937-1938, p. 75) to occur off Portugal in shoal water as well as deep, is about 1000 meters, near Madeira, by line fishermen's report (Maul, 1955, p. 7). And the only information at hand in this respect for *S. longus of Japan (p. 121) is that the s*pecimens so far seen Avere the product of the deep hook and line fishery in Sagami Bay, Tanaka 1912a, p. 104). Remarks. The combination of a very short snout with a caudal of the shape shown in Figure 14B, a first dorsal fin originating near the mid-length of the trunk, a second dorsal not larger than the first, and thorn-like denticles, with lower teeth more or less strongly oblique, and razor sharp, makes the recognition of BIGELOW AND SCHROEDER : SQUALOID SHARKS 117 a Somniosus easy, even as one is dumped on deck from the trawl. The caudal fin in Somniosus has a distinct subterminal notch in the specimens we have seen from the Gulf of Maine, from Iceland, and from Japan ; indeed it has been so pictured for rostratus by Burkhardt (1900, p. 564, Fig. 3) and by Helbing (1904, PI. 9, fig. 1) ; for *microcephalus by Helbing (1904, PI. 9, fig. 2, as "horealis") ; by Garman (1913, PI. 15, fig. 4), and by Bigelow and Schroeder (1948, Fig. 100) ; by Tanaka (1912a, PI. 26) for Hongiis. Since it is similarly notched on the type specimen of *>S^. pacificus Bigelow and Schroeder 1944, from Japan (pictured by Garman 1913, PI. 15, fig. 1, under the name "hrevipinna"), we suspect that Tanaka 's (1911, PI. 11, fig. 32) earlier illustration of it for that species (under the name microcephalus Bloch and Schneider 1801) as without a notch, represents the result either of wear, or of mutilation. More significant from the taxonomic standpoint is the wide expansion of the fin above the caudal axis, and its sublunate shape (Fig. 14B). One species {rostratus of the Mediterranean and adjacent Atlantic) is brilliantly luminescent. For a first hand account of its luminescence, and of its light-producing organs, see Burk- hardt 1900, p. 560. Species. The following named species (listed here in chrono- logical order) fall in Somniosus : *Squali(s microcephalus Bloch and Schneider 1801, boreal- subarctic latitudes in the eastern Atlantic ; *Somniosus brevi- pinna Lesueur 1818, corresponding zone in the western North Atlantic; Scymnus rostratus Risso 1826, Mediterranean and Madeira; *Hetero scymnus longus Tanaka 1912, Japan; Somnio- sus antarcticus Whitley 1939, Macquarie Island, Subantarctica, south of Australia; * Somniosus pacijlcus Bigelow and Schroe- der 1944, Japan, probably also northeastern Asiatic coast, and American coast from Bering Strait southward to Puget Sound, occasionally to southern California. Our own comparison of a 1334 mm. male, taken off the coast of Massachusetts, with a 1564 mm. male from Iceland corrobo- rates the view, now generally held, that the Greenland sharks of the subarctic on the two sides of the Atlantic (including the White Sea) and of the neighboring parts of the Arctic Ocean 118 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY belong to a single species. How closely the population of *S. microcephalus in the two sides of the North Atlantic agree in proportional dimensions appears from the following tabulation : A, for a male, 1334 mm. long taken near Boston Lightship in 32 fathoms (Mus. Comp. Zool. No. 39609) and B, for a second male, 1564 mm. long, taken northwest of Iceland, Lat. 66°48'N., Long. 25°10'W., in 180 fathoms, October 1952 (Mus. Comp. Zool. No. 37826). Proportional dimensions in per cent of total length Snout length in front of : outer nostrils eye mouth Eye: horizontal diameter Mouth: breadth Nostrii.s: di.stance between inner ends cm opening lengths: first second third fourth fifth First dorsal fin: vertical height length of base Second dorsal fin : vertical height length of base Caudal fin: upper margin lower anterior margin Pectoral fin: outer margin inner margin greatest width Distance from snout to : first dorsal second dorsal upper caudal pectorals pelvics interspace between: first and second dorsals second dorsal and upper caudal pelvics and lower caudal Distance between origins of: pectorals and pelvics pelvics and subcaudal Teeth : A 3.1 B 2.4 7.3 6.4 10.3 9.3 2.2 2.4 6.7 7.2 3.9 3.4 1.8 2.0 2.0 2.2 2.0 2.2 2.0 2.2 1.8 2.0 3.4 3.5 7.6 8.4 2.9 2.6 6.6 5.8 20.5 20.6 14.9 15.0 12.1 13.2 6.0 5.6 6.5 6.2 41.2 38.0 65.3 64.9 79.5 79.4 24.6 24.6 59.7 59.3 16.5 18.3 7.7 8.7 12.1 .11.6 35.2 34.8 17.1 17.0 48 45 48 51 BIGELOW AND SCHROEDEE : SQUALOID SHARKS 119 This joint species was originally described by Gunneriis in 1766 under the name Squalus carcharias, but by the rules of nomenclature, it must be called *microcephalus Bloch and Schnei- der 1801, "the name Squalus carcharias having been used pre- viously by Linnaeus 1758, for a very different shark" (Bigelow and Schroeder 1948, p. 515). The known range of *microcephalus extends from the White Sea, Bear Island, Spitzbergen and east and west Greenland, southward to the North Sea (to the mouth of the Seine as a straggler) in the east, to the Gulf of Maine in the west. A second North Atlantic species, rostratus Risso 1826, re- sembles *microcephalus so closely in general appearance that as recently as 1928 Key (1928, p. 480), following Garman (1913, p. 241), classed it as identical with the former, while — to fur- ther confuse the issue — recent illustrations of rostratus by Burkhardt (1900, p. 564, fig. 3) and by Maul (1955, PI. 3, fig. 17) differ as to the length of the upper side of the caudal peduncle relative to the length of the base of the second dorsal fin. Both, however, like Canestrini (1864, PL 2, fig. 2) and Ilelbing (1904, PL 9, fig. A) show the distance from the rear tip of the second dorsal to the origin of the upper side of the caudal as less than 70 per cent as long as the base of the second dorsal (about as long as the base of the second dorsal in *micro- cephalus). A more important difference is that the lower teeth are pictured as with the cusp more nearly erect in rostratus (Fig. 15D, see also Maul 1955, PL 3, fig. 19) than they are in *micro- cephalus (PL 3), and its denticles as rising less steeply from the skin (cf. Maul 1955, PL 3, fig. 20, with Bigelow and Schroe- der 1948, fig. lOlB). Rostratus is also much the smaller member of the pair, the greatest length we find definitely recorded for it being only 1000 mm. (Helbing 1904, p. 347) ; whereas adults of *micro- cephalus average 8-14 feet (2.4-4.3 meters) long, with a few growing considerably larger still (p. 116). A further difference already remarked upon by Burkhardt (1900, pp. 561-562, 565) is that while the luminescent organs of rostratus (p. 117) are surrounded and thus rendered conspicuous by denticles that differ in blunter cusps and more definitely stellate bases from those elsewhere on the sides, this is not the case in *niicro- cephalus, as we can verify from our own examination of a well 120 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY preserved male 1334 mm. long, taken off Massachusetts. For that matter, we have not been able to detect anything suggestive of luminescent organs on it, or on a 1564 mm. male from Iceland (p. 118). Helbing (1904), also, in his extensive treatise on the anatomy and systematic relationships of Somniosus (as "Laemargiis'^), has pointed out several skeletal differences between rostratus and * micro cepJialus (as "horealis") . Finally, rostratus is ovo- viviparous (see Helbing 1904, p. 358, for illustration of its embrj^os at successive stages in growth), as are all the other genera of Squaloidea for which the mode of development is known, whereas it is still an open question whether *micro- cephalus is ovoviviparous or oviparous (for a summary of avail- able evidence, see Bigelow and Schroeder 1948, p. 520). For a more detailed discussion of the differences in propor- tional dimensions between rostratus and *microcephalus, see Tortonese 1937-1938, p. 74. It has recently been suggested by Maul (1955, p. 9) that the Portuguese shark described and pictured under the name rostratus by Capello (1870, p. 148, PI. 9, tig. 2) actually repre- sents a third Atlantic species of Somniosus (as yet unnamed), for it is shown and described as differing from rostratus in a longer snout and head, and as with the position of the second dorsal fin different in relation to the pelvics, to which we may add a longer caudal peduncle. This appears from the following com- parative dimensions of (A) rostratus according to Maul (1955, measurements, p. 9, and fig. 17), and of (B) the Portuguese specimen reported under that name, from Capello 's measure- ments, and illustration, also measurements, contributed by Dr. A. M. Ramalho, who kindly re-examined the specimen now in the Bocage Museum in Lisbon. *p^ A B Length of head, to pectorals, in per cent of total length 20 28-29 Snout to mouth, in per cent of length of head 25 35 Distance from rear end of base of 2nd dorsal to upper origin of caudal, in per cent of total length 7 11 BIGELOW AND SCHROEDER : SQUALOID SHARKS 121 Also, the origin of the second dorsal, pictured as opposite the rear end of the bases of the pelvics by Maul, for his Madeira specimen, is about 10 mm. posterior to the pelvics on the Capello specimen, as we are informed by Dr. Ramalho, while excellent photographs of its mouth, supplied by him (Fig. 15E) show its lower teeth as somewhat more strongly oblique than they are pictured for the typical rostratus (Fig. 15D after Canestrini, 1864, PI. 2, fig. 4; see also Maul 1955, fig. 19), i.e., resembling, rather, those of microcephalus. Thus, present indications are that the Capello specimen may well represent an undescribed species more nearly related to * microcephalus than to rostratus. But it seems to us wiser to leave the matter open than to burden shark literature with a new name that might soon be relegated to oblivion. The early reports of Somniosus from the North Pacific, west and east, were either as ^'microcephalus" or as "hrevipinna," on the assumption that the Pacific representative of the genus was identical with the Atlantic ^microcephalus. And the Jap- anese specimen, on which we based our new species pacificus (described and pictured by Garman 1913, p. 240, PI. 15, figs. 1-3, as '"hrevipimia"), agrees with *microcephalus (p. 117) in lacking any evident dermal structures to which a luminescent nature might be ascribed, and in the shape of its lower teeth. But its first dorsal stands considerably farther rearward than on the Atlantic specimens that we have seen of *microcephalus, or than is pictured for rostratus; the distance from the tip of its second dorsal to the origin of the upper side of its caudal is shorter relative to the size of the second dorsal ; the upper pos- terior and lower anterior margins of its caudal are more convex ; and its upper teeth are broader. It was because of these differ- ences that Ave proposed the new species *pacificus for it (Bige- low and Schroeder 1944, p. 35). And it seems reasonably certain (though not yet proven) that the various reports of Som^iios us from northeastern Asia, Alaska, and thence southward along the Pacific Coast of America to California all were based on this .same species. Japanese waters also harbor a second Somniosus (Fig. 15A, B, C, PI. 3), the *Heteroscynmus lo7igus of Tanaka 1912a (p. 102, PI. 26, figs. 102-107), more nearly resembling roslratus than either * microcephalus or *pacificus in the shape of its lower 122 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY teeth, in the minuteness of its denticles, and in the length of the base of its first dorsal fin relative to the length of the head. So closely, indeed, does Tanaka's account of its type specimen, corroborated by our examination of a female of about 1010 mm. (in rather poor condition), received from Dr. Tokiharu Abe, agree with the published accounts of rostratus (we have not seen that species) that it is doubtful whether the Japanese popula- tion differs consistently enough in any respect from the At- lantic population to deserve a separate specific name — final decision is a matter for the future. Somniosus has not been reported from tropical-equatorial waters. But the micro cephalus-pacificus division of the genus is represented in suhantarctic latitudes by 8. antarcticus Whit- ley 1939, a name proposed for an ''8 foot, 2 inch" (2491 mm.) specimen that was found stranded on the beach at Macquarie Island in latitude about 54°49'S., south of Australia (Waite 1916, p. 51, fig. 10). The position of its first dorsal fin, as pictured by Waite, and the shape of its caudal suggests a closer resemblance to *micro- cephalus of the northern Atlantic than to *pacificus. But final decision as to its specific relationship must await a more de- tailed study of the Southern Hemisphere population. Key to Species of Somniosus in the Northern Hemisphere 1. Lower teeth only moderately oblique (Fig. 15D) ; adults with evident luminescent organs, at least in best known species (see p. 117) rostratus Eisso 1826. Mediterranean and Madeira; p. 119; also longus Tanaka 1912a, Japan, perhaps not distinguish- able from rostratus, see discussion, p. 121. Lower teeth so strongly oblique that the inner margins are almost parallel with the trend of the jaw, forming a nearly uninterrupted cutting edge (PI. 3) ; no evident luminescent organs 2 2. Head to pectorals 28-30 per cent of total length; lower teeth as in Fig. 15E rostratus Capello 1870 {not of Eisso 1826), off Portugal; see discussion, p. 120. Head to pectorals only about 24-26 per cent of total length 3 BIGELOW AND SCHROEDER : SQUALOID SHARKS 123 3. Origin of first dorsal much nearer to tip of snout than to tip of caudal ; interspace between first and second dorsals as long as from snout to lpt-2nd gill openings or longer *'microcephalus Bloch and Schneider 1801. Both sides of northern North Atlantic, p. 117. Origin of first dorsal almost as near to tip of caudal as to tip of snout; interspace between first and second dorsals only about 66 per cent as long as from tip of snout to second gill openings *pacificus Bigelow and Schroeder 1944. Type locality Japan, prob- ably also northeastern Asia and Pacific Coast of North America. from Bering Sea south to Puget Sound, occasionally to southern California, p. 121. Genus ISISTIUS Gill 1864 I.. Three new shark records from the Gulf of Mexico. Breviora, Mus. Comp. Zool., no. 49, 12 pp., 2 text figs. ni..\INVlI,LE, H. M. DUCROTY DE 1816. Prodrome d'une nouvelle distribution systematique du regne animal. Bull. Sci., Soc. Philomatique, Paris, 1816, pp. 105-124. 1825. Vertebres, classe G, Poissons. In Viellot, Desmarest, etc., Faunc Fran(,'aise . . ., vol. 1, part 3, 96 pp., 120 pis. (As advertised, l)ut 22 plates only, were published.) 138 BULLETIN : MUSEUM OF COMPAKATIVE ZOOLOGY Bleekee, Pietek 1860. Elfde Bidjdrage tot de Kennis der Vischfauna van Amboina. Acta Soc. Reg. Scieut. Indo-Neerlandica ; also as Verhand. Konink-Natuurk. Vereen-Nederlandisch-Indie, vol. 8, no. 5, 14 pp. 1860a. Over einige Vischoorten van de Kap de Goede Hoop. Natuurk. Tidjschr. voor Nederlandsch. Indie, vol. 21 (ser. 5, vol. 1), pp. 49-80. Bloch, M. E., and I. G. Schneidek 1801. Systema Ichthyologiae iconibus ex illustratum ... 2 vols., be -f 584 pp., 110 pis. Bocage, J. V. Barboza du, and F. de B. Capello 1864. Sur quelcjues espfeces in^dites de Squalidae. Proc. Zool. Soc. London, 1864, pp. 260-263. 1866. Notes pour servir a I'Ichthyologie du Portugal. Poissons Plagiostomes, part 1. Lisbon. 40 pp., 3 Pis. (French and Portuguese). Bonaparte, Charles Lucien 1832- Iconografia della fauna Italica . . ., vol. 3, Pesci, 78 pis. 1841. (not numbered) and accompanying text (pages not numbered). For dates of publication see Salvadori, Boll. R. univ. Mus. Zool. Anat. Comp., Turin, vol. 3, no. 48, 1888. 1838. Selachorum tabula analytica. Nuov. Ann. Sci. Nat. Bologna, vol. 2, pp. 195-214; also as Mem. Soc. Sci. Neuchatel, vol. 2, no. 4, pp. 1-12, 1839, and sep., Rome. 1846. Catalogo motodico dei pesci Europa. Napoli, 98 pp. Bonde, Cecil von 1924. Shallow water fishes procured by the SS "Pickle". Fisher. Mar. Surv. Union S. Africa, Rept. no. 3 (1922), Spec. Kept, 1, 40 pp., 9 pis. Bonnaterre, p. J. 1788. Ichthyologie, in Tab. Encyc. Method, des trois R^gnes de la nature. Paris, Ivi + 215 pp., pis. A, B + 1-100. Broussonet, P. M. C. 1780. Sur les differentes especes de chiens de mer. Mem. Math. Phys. Acad. Roy. Sci., Paris [1780], pp. 641-680. Burkhardt, Rudolph 1900. Luminous organs of selachian fishes. Ann. Mag. Nat. Hist., ser. 7, vol. 6, pp. 558-568. BIQELOW AND SOHROEDER : SQUALOID SHARKS 139 Cadenat, J. 19ri0. Poissons de Mer du Senegal. Initiations Africaines 3, Inst. Frang. d'Afrique Noire, vol. 3, .S45 pp.. 241 figs. Dakar. Canestrixi, Giovanni 1864. Sopra aleuni pesei poeo noti o nuovi del Mediterraneo. Mem. Acad. Torino, ser. 2, vol. 21, pp. S-iS-SGl, pis. 1, 2. Also separate, Torino, 11 pp., 2 pis. Capello, Felix de Brito 1870. Catalogo dos Peixes . . . Mus. Lisboa [Part 4], Jom. Sci. Math. Phys. Nat. Acad. Lisboa, vol. 2, pp. 131-153, pi. 9. CoLLETT, Robert 1904. Diagnosis of four hitherto undescribed fishes from the depths south of the Faroe Islands. Videns. Selskabs. Forhand. Christi- ania, 1904, no. 9, 7 pp. 1905. Fiske . . . "Michael Sars" Togter i Nordhavet 1900-1902. Kept. Norwegian Fisher, and Marine Invest., vol. 2, part 2, no. 3, 147 pp., 2 pis. Oloquet, Hippolyte 1820. Dictionnaire Sciences Naturelles, suppl. to vol. 1, 138 pp. Paris. rx-viER [G. L. C. F. D.] 1817. La regne animal . . .. Paris, vol. 2, xviii + 532 pp. Daniel, J. 1934. The elasmobranch fishes. Berkeley, xi + 332 pp., 270 figs. Day, Francis 1880- The fishes of Great Britain and Ireland, vol. 1, Ixii + 336 pp., 1884. vol. 2, 388 pp., atlas, 179 pis., London. Deli'in, Frederioo T. 1901. Catalogo de los peces de Chile. Valparaiso, 133 pp. Dickens, D. A. G. 1956. [Observations on Euprotomicrrisl. The Marine Observer, Lon- don, vol. 26, p. 73. DuMERiL, August 1865. Histoire naturelle des poissons . . ., vol. 1, pt. 1, 720 pp., Paris. Eigenmann, Rosa 1891. Description of a new species of Euprotomicrus. Proc. Cali- fornia Acad. Sci., ser. 2, vol. 3, p. 35. 140 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Fang, P. W., and K. F. Wang 1932. The elasmobraneh fishes of Shantung coast. Contrib. Biol. Lab. Sci. See. China, Nanking, zool. ser., no. 8, pp. 213-283, 29 figs. b KLIN, LE Marquis de 1887. Sous les mers. Compagnes d 'explorations du "Travailleur" et du "Talisman". Paris, 337 pp. Fowler, Henry W. 1930. A list of the sliarks and rays of the Pacific Ocean. Proc. 4th Pacific Sci. Congr., Java (1929), vol. 3, pp. 481-508. 1934. Descriptions of new fishes obtained 1907-1910, chiefly in the Philippine Islands and adjacent seas. Proc. Acad. Nat. Sci., Philadelphia, vol. 85, pp. 233-367. 1941. The fishes of the groups Elasmobranchii, Holocephali, Iso- spondyli and Ostariophysi . . . Contributions to the Biology of the Philippine Archipelago and Adjacent Seas. Bull. 100, U. S. Nat. Mus., vol. 13, ix + 879 pp., 30 teit figs. Frade, F. 1929. Une nouvelle espece ou une aberration individuelle de I'Oxyno- tus centrina, (Linn.). Bull. Soc. Portugaise Sci. Nat., vol. 10, no, 22, pp. 263-267. 1932. OxynoUis paradoxus. In: Faune Ichthyol. Cons. Internat. perm. Explor. Mer., PI. (not numbered) and accompanying text. Uarman, Samuel 1899. The fishes. Eepts. Explor. ott' the west coast of Mexico . . ., in charge of Alexander Agassiz. Mem. Mus. Comp. Zool., vol. 24, 430 pp., atlas, pis. 1-85, A-N. 1906. New Plagiostomia. Bull. Mus. Comp. Zool., vol. 46, no. 11. pp. 203-208. 1913. The Plagiostomia . . ., Mem. Mus. Comp, Zool,, vol, 36, xiii + 515 pp,, 75 pis. CJarrick, J. A. F. 1955. Studies on New Zealand Elasmobranchii, Part IV. The sys- tematic position of Centroscymnus waitei (Thompson 1930). Trans. Roy. Soc. New Zealand, vol. 81, pt. 1, pp. 227-239, 2 figs. 1956. Studies on New Zealand Elasmobranchii, Part V. Scymnodala- tias n. g., based on Scymnodon sherwoodi Archey 1921. Trans. Boy. Soc. New Zealand, vol. 83, pt. 3, pp. 555-571, 2 figs. 1957. Studies on New Zealand Elasmobranchii. Part VI. Two new species of Etmopterus from New Zealand. Bull. Mus. Comp. Zool., vol. 116, no. 3, pp. 171-190, 2 pis. BIGELOW AND SCHROEDER : SQUALOID SHARKS 141 Gilbert, Charles H. 1905. The deep sea fishes of the Hawaiian Islands. Bull. U. S. Fish Conim., vol. 23, part 2, pp. xi + 577-713, pis. 66-101. Gilchrist, J. D. F. 1922. Deep sea fishes procured by the SS "Pickle", part 1. Fisher, and Mar. Biol. Surv. Union S. Africa, Eept. 2 (1921), special rept. 3, pp. 41-79, pis. 7-12. Gilchrist, J. D. F., and C. von Bonde 1924. Deep sea fishes procured by the SS "Pickle", part 2, Fisher, and Mar. Biol. Surv. Union S. Africa, Rept. 3 (1922), special rept. 7, 24 pp., 6 pis. Gill, Theodore 1861. Analytical synopsis of the order of Squali, and revision of the nomenclature of the genera. Ann. Lyceum Nat. Hist. New York, vol. 7, pp. 367-413. 1862. On the classification of the families and genera of the Squali of California. Proc. Acad. Nat. Sci. Philadelphia, 1862, pp. 483-501. 1864. Synopsis of the eastern American sharks. Proc. Acad. Nat. Sci. Philadelphia, 1864, pp. 258-265. GiR.\RD, Charles 1854. Characteristics of some cartilaginous fishes of the Pacific coast of North America. Proc. Acad. Nat. Sci. Philadelphia, 1854, pp. 196-197. Gmelin, J. F. 1789. In: Linnaeus, C, Systema Naturae, 13 ed., vol. 1, part 3, pp. 1033-1516. Goode, George Brown, aiul T. 11. Hean 1895. Oceanic Ichthyology . . . Smithsonian Contrib. Knowl. vol. 30, .\x.\v + 553 pp.; vol. 31, Atlas, 123, pis. also 1896, as vol. 22, Mem. Mus. Comp. Zool. Grey, Marion 1956. The distribution of fishes found below 2000 meters. Fieldiana, Chicago Nat. Hist. Mus., vol. 36, no. 2, pp. 75-337. GuNTHER, Albert 1870. Catalogue of the fishes of the British Museum, vol. 8, xxv -j- 549 pp. London. 1877. Preliminary notes on new fishes collected in Japan . . . Ann. Mag. Nat. Hist., ser. 4, vol. 20, pp. 433-446. 142 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1880. Report on the shore fishes . . . "Challenger" Kept. Zool., vol. 1, part 6, 82 pp., 32 pis. 1887. Report on the deep sea fishes. "Challenger" Rept. Zool., vol. 22, Ixv + 1-6 4- 268 pp., pis. 1-66. 1910. Andrew Garrett's Fische der Sudsee . . . Jour. Mue. Godeffroy (Hamburg), vol. 6, pp. 389-515, pis. 141-180. GUNNERUS, JOHAN E. 1766. Vom Haa-Skierding (Carcharias). Drontheimische Geeell. Schrift., vol. 2, pp. 299-307. Helbing, Hermann 1904. Beitrage zur Anatomie und Systematik der Laemargiden. Nova Acta Acad. Leop. Carol., Halle, vol. 82, pp. 335-524, pi. 9. Herrb, a. W. C. T. 1923. Notes on Philippine sharks. I. Philippine Jour. Set., vol. 23, pp. 67-73, pi. 1. 1934. Fishes of the Herre Philippine expedition of 1931. Notes on fishes in the Zoological Museum of Stanford University, Part I. Hongkong, 106 pp. 193.'). Notes on fishes in the Zoological Museum of Stanford Uni- versity. Part 2. Copeia, 1935, pp. 122-127. 1936. Phaenopogon a synonym of Cirrhigaleus. Copeia, 1936, p. 59. 1953. Check list of Philippine fishes. Res. Rept. no. 20, U. S. Fish and Wildlife Serv., 977 pp. llTTBBS, Carl L., and W. I. Follett 1947. Lamna ditropis, new species . . . Copeia, 1947, p. 194. FluBBS, Carl L., and J. L. McHugh 1951. Relationships of the pelagic shark Euprotomicrus bispinatuii . . . Proc. California Acad. Sci., ser. 4, vol. 27, no. 6, pp. 159-176, pis. 4-6. Jaekel, 0. 1890. Ueber die systematische Stellung und iiber fossile Reste der Gattung Pristiophon/s. Zeitschr. deutsche geol. Ges., 1890, pp. 86-120, pis. 2-5. Jenkins, J. Travis 1925. The fishes of the British Isles, vii + 376 pp., 143 pis., London. JOHANN, L. 1899. Ueber eigenthumliche epitheliale Gebilde (Leuchtorgane) bei Spinax niger. Zeitschr. Wiss. Zool., vol. 66, pp. 136-160, pis. 6-8. BIGELOW AND SCHROEDER : SQUALOID SHARKS 14^? Johnson, J. Y. 1867. Description of a new species of Spinacidae . . . Proc. Zool. Soc. London, 1867, pp. 713-715. Jordan, D. S. 1919. The genera of fishes, part 2. Leland Stanford Junior Uni- versity Publ., University Series, pp. i-ix + 163-284 -f i-xiii. 1923. A classification of fishes . . . Leland Stanford Junior University Publ. Biol. Ser,, vol. 3, no. 2, pp. 77-243 + i-x. Jordan, D. S., and B. W. Evebmann 1896. The fishes of North and Middle America. Bull. U. S. Nat. Mus., vol. 47, part 1, Ix -f 1240 pp. 1905. The aquatic resources of the Havcaiian Islands. Part 1. Shore Fishes. Bull. U. S. Fish Conim., vol. 23, part 1, xxviii + 574 pp., 73 pis. 1 chart. Jordan, D. S., and H. W. Fowler 1903. A revievr of the elasmobraneh fishes of Japan. Proc. U. S. Nat. Mus., vol. 26, pp. 593-674, pi. 26. Jordan, D. S., and C. L. Hubbs 1925. Record of fishes obtained by David Starr Jordan in Japan, 1922. Mem. Carnegie Mus., vol. 10, no. 2, pp. 93-346, pi. 12. Jordan, D. S., and J. O. Snyder 1901. A preliminary checklist of the fishes of Japan. Annot. Zool. Japon., vol. 3, pp. 1-159. 1902. Descriptions of two new species of squaloid sharks from Japan. Proc. U. S. Nat. Mus., vol. 25, pp. 79-87. Jordan, D. S., S. Tanaka and J. O. Snyder 1913. A catalogue of the fishes of Japan. Jour. Coll. Sci. Tokyo, vol. 33, art, 1, 497 pp., 396 figs. King, Joseph E., and I. I. Ikehara 1956. Some unusual fishes from the Central Pacific. Pacific Science, vol. 10, pp. 17-24, 2 figs. Krefft, Gerhardt 1955. Ichthyologische Mittheilungen. Zool. Anz., vol. 154, pp. 157-160. Ifig. Lahille, F. 1921. Enumeracion de los peees cartilaginosos. . . . Argentinas Ministr. Agric. Nacion. Buenos Aires, 41 pp., 1 chart. 1928. Nota sobra unos peces elasmobranchios. Ann. Mus. Nac. Buenos Aires, vol. 34, pp. 299-339, 5 pis. 144 BrTI;LETIN : MUSEUM OF COMPARATIVE ZOOLOGY Lesueub, 0. A. 1818. Descriptions of several new species of North American fishes . . . Jour. Acad. Nat. Sci. Philadelphia, new ser., vol. 1, pp. 225-235, 359-368, 3 pis. LiNNAKUS, C. 1758. Sj'stema Naturae, 10th ed., Holmiae, vol. 1, 824 pp. LONNBERO, EiNAR 1907. Fische, in Ergebnisse Hamburger Megalhaenische Sammelreise, Hamburg, vol. 8, no. 6, 16 pp., 1 pi. Lowe, R. T. 1834. [A collection of fishes made in Madeira.] Proc. Zool. Soc, London (1833), part 1, 1834, pp. 142-144. 1839. A supplement to "A synopsis of the fishes of Madeira." Proc. Zool. Soc. London, 1839, part 7, pp. 76-92. 1843. Notices of fishes newly observed or discovered in Madeira. Proc. Zool. Soc. London, 1843, part 11, pp. 81-95. Lucas, A. H. S. 1890. A systematic census of indigenous fish . . . from Victorian waters. Proc. Roy. Soc. Victoria [Melbourne], n. s., vol. 2, pp. 15-47. MacLeay, William 1881. Descriptive catalogue of Australian Fishes, Sydney, 2 vols, -f suppl. 1892. Descriptive catalogue of the fishes of Australia. Proc. Linn. Soc. New South Wales, vol. 6, pp. 202-386. Marshall, N. B. 1956. [Observations on Euprotomicrus.] The Marine Observer, Lon- don, vol. 26, pp. 73-74. Maul, G. E. 1955. Five species of sharks new for Madeira, including two new to science. Notulae Naturae, Acad. Nat. Sci. Philadelphia, no. 279, 13 pp., 3 pis. Matr, Ernst, E. G. Linslet, and R. L. Usinger 1953. Methods and principles of systematic zoology, New York, ix -f 328 pp. McCoy, Frederick 1887. Prodromus of the zoology of Victoria . . . Decade 15, pp. 2, 157- 193, pis. 141-150. Melbourne. BIGELOW AND SCHROEDER : SQUALOID SHARKS 145 McCuLLOCH, Allan R. 1914. Report on some fishes . . . collected by the F. I. S. "Endeavour." Zool. Res. Fishing Exper. ... by F. I. S. "Endeavour," 1909-1914, vol. 2, part 2, pp. 74-165. 1915. Report on some fishes obtained by the F. I. S. "Endeavour" on the coasts of Queensland, New South Wales, Victoria, Tas- mania, south and southwestern Australia. Part. 3. Zool. Res. Fish ing Exper. ... by the F. I. S. "Endeavour," 1909-1914, vol. 3, pp. 97-170, pis. 13-37. 1927. Checklist of the fishes ... of New South Wales. Sydney, ed. 2, xxvi + 104 pp., 43 pis. MiVAET, St. George 1879. Notes on the fins of elasmobranchs. Trans. Zool. Soc. London, vol. 10, pp. 439-484, pis. 74-79. MoLiN'A, Giovanni 1782. Saggio sulla storia naturals del Chile . . . Bologna, 368 pp. MOREAU, Emile 1881. Histoire naturelle des Poissons de la France. Paris, vol. 1, vii + 478 pp. MuLLEB, Johannes, and F. G. J. Henle 1837. Ueber die Gattungen der Plagiostomen . . . Arch. Naturgesch., Jahrg. 3, vol. 1, pp. 394-401, 434. 1837a. Gattungen der Haifische und Rochen . . . Ber. Akad. Wiss. Berlin, pp. 111-118. 1841. Systematische Beschreibung der Plagiostomen. Berlin, xxii + 200 pp., 60 pis. Nichols, J. T., and F. E. Firth 1939. Rare fishes off the Atlantic Coast . . . Proc. Biol. Soc. Washing- ton, vol. 52, pp. 85-88. NOBRE, AUQUSTO 1935. Vertebrados. Fauna marinha de Portugal, Porto, vol. 1, Ixxxiv -f 574 pp., 77 pis. NOEMAN, J. R. 1932. Note on a shark, Oxynotus paradoxus Frade, new to the British Fauna. Proc. Zool. Soc. London, 1932, Part 1, pp. 77-79, 1 pi. 1935. Coast fishes, part 1. "Discovery" Reports, Vol. 12, pp. 1-58. Cambridge. NORONHA, ADOLEO CeSAE DE 1926. A new species of deep water shark (Squaliolus sarmenti) from Madeira. Ann. Carnegie Mus., vol. 16, pp. 385-389, pi. 35. 146 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY OoiLBY, J. Douglas 1888. Catalogue of the fishes . . . Australian Museum. 1. Eecent Palaeichthyian fishes. Australian Mus., Sydney, v + 26 pp. 1889. List of the Australian Palaeichthyes. 2. Proc. Linnaean Soc. New South Wales, set. 2, vol. 4, part 1, pp. 178-186. 1893. Description of a new shark from the Tasmanian Coast. Rec. Canterbury Mus. [New Zealand], vol. 2, pp. 62-63. OSHIMA, H. 1911. Some observations on the luminous organs of fishes. Jour. Coll. Sci., Tokyo, vol. 27, Art. 15, 15 pp., 1 pi. Philippi, Rudolph A. 1887. Historia natural sobre los tiberones y algunoa otros peces de Chile. Ann. Univ. Chile, vol. 71, pp. 1-42, pis. 1-8. Also separate, 42 pp., 8 pis. Santiago de Chile. Phillipps, W. J. 1928. Sharks of New Zealand. 2. New Zealand Jour. Sci. Techn., vol. 10, pp. 221-226, 8 figs. 1931. New species of piked dogfish. New Zealand Jour. Sci. Techn., vol. 12, no. 6, pp. 360-361. PlETSCHMANN, VICTOR 1907. Zwei neue Selachier aus Japan. Anz. Akad. Wiss. Wien, vol. 44, pp. 394-396. 1908. Japanische Plagiostomen. Sitzungsber. Akad. Wiss. Wien, vol. 117, pt. 1, pp. 637-710, pis. 1, 2. 1928. Neue Fiseh-arten aus dem Paeifischen Ozean. Anz. Akad. Wien, vol. 65, pp. 297-298. 1930. Remarks on Pacific fishes. Bull. Bemice P. Bishop Mus. [Honolulu], vol. 73, 24 pp., 4 pis. POEY, Felipe 1861. Memorias sobra la historia natural de la Isla de Cuba, vol. 2, 442 pp., 19 pis. [dates of publication are given on p. 437]. Poll, Max 1951. Poissons, I. Generalites; 2. Selaciens et chimeres. Res. Sci. Exped. Beige Eaux Cotes Afric. Atlant. Sud, vol. 4, Fasc. 1, pp. 1-154, pis. 1-13. QuoY, Jean Rene Constantin, and Paul Gaimard 1824. Description des poissons, in voyage . . . "L'Uranie" et "La Physicienne ", . . ., 1817-20 [vol. 4], Zool., pp. 192-401, pis. 43-65. BIGELOW AND SCHROEDER : SQUALOID SHARKS 147 Rafinesque, Constantine Samuel 1810. Caratteri di alcuni nuovi generi a nuovo specie di animali e piante della sicilia . . ., Palermo, 4 + 105 pp., 20 pis. 1810a. Indice d'ittiologia Siciliana . . . Messina, 70 pp. 2 pis. Regan, C. T. 1906. Descriptions of some new sharks in the British Museum Col- lection. Ann. Mag. Nat. Hist., ser. 7, vol. 18, pp. 435-440. 1908. A synopsis of the sharks of the family Squalidae. Ann. Mag. Nat. Hist., ser. 8, vol. 2, pp. 39-57. 1908a. A collection of fishes from the coasts of Natal, Zululand and Cape Colony. Ann. Natal Mus., vol. 1, pp. 241-255. pis. 37-42. 1912. Philippine sharks. Science, n.s., vol. 36, p. 81. Rbinhardt, J. C. H. 1825. Ichthyologiske Bidrag . . . Oversigt. Dansk. Vid. Selsk. Forh. Kjobenhavn, 1824-25, pp. 2 3. Ret, Luis Lozano 1928. Fauna Iberica. Feces, vol. 1, 691 pp., 20 pis. Inst. Nac. Ciencias, Madrid. Ribeiro, Alipio de M. 1907. Fauna braziliense, Peixes, vol. 2; Ann. Mus. Nac. Rio de Janeiro, vol. 14, pp. 137-217, pis. 2-20. Richardson, L. R. 1956. Probing the edge of the abyss. New Zealand Listener, vol. 35, no. 903, pp. 6-7. Wellington. Richardson, L. R., and J. A. Garrick 1953. A guide to the lesser chordates and cartilaginous fishes. Jour. Biol. Soc. Victoria Univ. Coll., New Zealand, vol. 5, no. 1, pp. 23-37, pis. 1-3. Risso, A. 1826. Histoire uaturelle . . . de 1 'Europe Meridionale . . . Paris, vol. 3, xvi + 480 pp., 16 pis. RiVERO, Luis Howell 1936. A new shark from Tasmania. Occ. Pap. Boston Soc. Nat. Hist., vol. 8, pp. 267-268, pi. 10. 1936a. Some new, rare and little known fishes from Cuba. Proc. Boston Soc. Nat. Hist., vol. 41, pp. 41-76, pis. 9-13. ROCHEBRUNE, A, T, DE 1883- Poissons, in Faune de la Senegambie, Paris, vol. 1 [no. 4], 166 1885. pp., 6 pis. 148 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY RoMER, Alfred S. 1945. Vertebrate Paleontology, 2nd ed., Chicago, viii -|- 687 pp. RouLE, Louis 1919. Poissons provenent des Croisieres du Yacht "Princesse Alice" (1891-1903) et du yacht "Hirondelle" (1914). Result. Camp. Sci. Monaco, vol. 52, 190 pp., 7 pis. Sabmundsson, B. 1922. Zoologiske Meddelelser fra Island, no. 14. Vid. Meddel. Dansk. Naturhist-Foren. Kobenhaven, vol. 74, pp. 159-201, pis. 3-5. Sauvage, H. 1887- Poissons, in Grandidier, Hist. Nat. Polit. Madagascar, vol. 16. 1891. 543 pp., 61 pis. Schmidt, P. J. 1931. Fishes of Japan collected in 1901. Trans. Pacif. Comni. Acad. Sci., Leningrad, U.S.S.R., vol. 2, pp. 1-176. Shaw, George 1804. General Zoology, vol. 5, part. 2, pp. vii -f 251-463, pis. 133-181. Sigalas, R. 1939- Sur Euprotomicrus sarmenti Noronha. Act. Soc. Linn. Bordeaux, 1940. vol. 91, pp. 70-71. Simpson, George G. 1945. The principles of classification . . . Bull. Amer. Mus. Nat. Hist., vol. 85, xvi + 350 pp. Smith, A. 1849 Illustrations of the zoology of South Africa . . London, vol. 7, Pisces, 31 pis. Smith, Hugh M., and Lewis Radcliffe 1912. The squaloid sharks of the Philippine Archipelago . . . Proc. U. S. Nat. Mus., vol. 41, pp. 677-685, pis. 50-54. Smith, J, L. B, 1936. Two interesting new fishes from South Africa. Trans. Boy. Soc. S. Africa, vol. 24, pp. 1-6, pis. 1, 2. 1949. The sea fishes of southern Africa, Central News Agency, S. Africa, 550 pp., 103 pis. Smitt, F. a. 1895. Scandinavian fishes . . . part 2, pp. 567-1240, pis. 28-53. UIGELOW AMD SCHROEDER : SQUALOID SHARKS 149 Snyder, John O. 1904. A catalogue of the shore fishes collected by the "Albatross" about the Hawaiian Islands in 1902. Bull. U. S. Fish Comm., vol. 22, pp. 513-538, 13 pis. Steenstrup, J. J. S. 1861. Om forskiellen imellem beenfiskenes og. bruskfiskenes Skiaeldan nelse. Forhand. Skand. Naturf. Kjobenhavn (1860), 1861, vol. 8, pp. 688-689, also abstracts [French] in Arch. Sci. Phys. Nat. Geneva, 1861, n. s., vol. 11, p. 368, and Ann. Sci. Nat., ser. 4, Zool., vol. 15, p. 368, 1861. Steindachner, Franz 1867. Zur Fisch fauna von Port Jackson. Sitzungsber. Akad. Wien, vol. 66, Abth. 1, pp. 320-335. Tanaka, Shigeho 1911. Figures and descriptions of the fishes of Japan . . . vol. 3, pp. 35-50, pis. 11-15. 1912. Figures and descriptions of the fishes of Japaoi . . ., vol. 5, pp. 71-86, pis. 21-25. 1912a. Figures and descriptions of the fishes of Japan . . ,, vol. 6, pp. 87-108, pis. 26-30. 1912b. Figures and descriptions of the fishes of Japan . . ., vol. 9, pp. 145-164, pis. 41-45. 1917. Figures and descriptions of the fishes of Japan . . ., vol. 26, pp. 455-474, pis. 126-130. 1918. Figures and descriptions of the fishes of Japan, vol. 27, pp. 475-494, pis. 131-135. Thompson, E. F. 1930. New records of the genera CentropJiorus and Eoplichthys in New Zealand. Rec. Canterbury Mus., vol. 3, no. 4, pp. 275-279, pis. 42-44. ToRTONESE, Enrico 1937- Note di ittiologia. Boll. Mus. Zool. Anat. Comp. Torino, vol. 1938. 46, ser. 3, no. 74, pp. 73-102. 1952. Studi sui Plagiostomi . . . Arch. Zool. Ital., vol. 37, pp. 383-398. Tucker, Denys W., and G. Palmer 1949. New British records of two rare deep sea fishes . . . Nature, vol. 164, pp. 930-938. Vaellant, Leon 1888. Expeditions scientifiques du "Travailleur" et du "Talisman" . . . Zool., Poissons, Paris, 406 pp., 28 pis. 150 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1888a. Mission scientifique du Cap. Horn, ] 882-1883, Zool., Poissons, Paris, 35 pp., 4 pis. Waite, Edgae R. 1909. Scientific results of the New Zealand trawling expedition, 1907 (Pisces). Rec. Canterbury Mus., vol. 1, no. 2, pp. 131-156, pis. 12-23. 1910. Notes on New Zealand fishes. Trans. Proc. New Zealand Inst., vol. 42 (1909), pp. 384-391, pis. 37-38. 1914. Notes on New Zealand fishes. No. 4. Trans. Proc. New Zealand Inst., vol. 46, pp. 127-131, pis. 3-6. 1916. Fishes, in Australasian Antarctic Exped., 1911-1914. Sci. Repts., ser. C, Zool. and Bot., vol. 3, part 1, 92 pp., 5 pis., 2 maps. 1921. Catalogue of the fishes of South Australia. Rec. South Aus- tralian Mus., vol. 2, no. 1, pp. 1-199. Weber, Max 1913. Fische. Siboga Exped. uitkom. Zool. Botan. Oceanogr. Geol. Gebied . . ., no. 57, 710 pp., 12 pis. White, Grace 1937. Interrelationships of the elasmobranchs . . . Bull. Amer. Mus. Nat. Hist., vol. 74, art. 2, pp. 25-138, pis. 1-51. Whitley, Gilbert 1931. New names for Australian fishes. Australian Zoologist, vol. 6, pp. 310-334. 1932. Studies in ichthyology, no. 6. Rec. Australian Mus., vol. 18, pp. 321-348, pis. 36-39. 1934. Notes on some Australian sharks. Mem. Queensland Mus., vol. 10, pt. 4, pp. 180-200, pis. 27-29. 1939. Studies in ichthyology, no. 12. Rec. Australian Mus., vol. 20, no. 4, pp. 264-277. 1940. The fishes of Australia, part 1, The sharks . . . Royal Zool. Soc. New South Wales, Australian Zool. Handbook, 280 pp., 303 figs. Woodward, Arthur Smith 1919. On two new elasmobranch fishes . . . Proc. Zool. Soc, London (1918), pp. 231-235, pi. 1. 1932. A Cretaceous pristiophorid shark. Ann. Mag. Nat. Hist., ser. 10, vol. 10, pp. 476-478, pi. 18. PLATES Plate 1. Oxynotus centnna. Above, lower surface of anterior part of head of female, 510 mm. long, coast of Portugal, x about 1.5. Below, roof of mouth of male, 590 mm. long, Portugal, viewed from below, to show arrangement of upper teeth, x about 6, both specimens Mus. Comp, Zool. No. 39576. PLATE 1 Plate 2. Etmopierus bulUsi, new species. Female, 196 mm. long, off northeast coast of Florida, Pelican sta. 42 (see p. 50). Above, side view of mid-sector of trunk, x about 3, to show the linear arrangement of the dermal denticles. Beloiv, ventral view of same, at pectoral fins, to show the random arrangement of the denticles on the lower surface. PLx^TE 2 Plate 3. Above, Somniosus longus, female 1010 mm. Mus. Comp. Zool. No. 39650, upper and lower teeth, left hand side of mouth, x about 3.3. Middle and helow, Somniosus microcephalus, upper and lower teeth, along central part of mouth, from jaws of a Gulf of Maine specimen about 11 feet long, Mus. Comp. Zool. No. 36119, x about 3. PLATE 3 Plate 4. Above, Squaliolus laticaudus, same speeiiiien as in Figure 16A, photograph of mid-portion of trunk, to show papillate appearance of skin, X about 2.8. Below, Euprotomicrus hispinatus, same specimen as in Figure 16F, corre- sponding photograph to show the smoothness of the skin as contrasted with Squaliolus (above), x about 3. PLATE 4 Bulletin of the Museum of Comparative Zoology AT HARVARD COLLEGE Vol. 117, No. 2 CHECK LIST OF THE REPTILES AND AMPHIBIANS OF EAST AFRICA (UGANDA; KENYA; TANGANYIKA; ZANZIBAR). By Arthur Loveridge CAMBRIDGE, MASS., U. S. A. printed for the museum August, 1957 Publications Issued by or in Connection WITH THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE Bulletin (octavo) 18G3 — The current volume is Vol. 117. Breviora (octavo) 1952 — No. 80 is current. Memoirs (quarto) 1864-1938 — Publication was terminated with Vol. 55. Johnsonia (quarto) 1941 — A publication of the Department of Mollusks. Vol. 3, no. 35 is current. Occasional Papers of the Department of Mollusks (octavo) 1945 — Vol. 2, no. 21 is current. Proceedings of the New England Zoological Club (octavo) 1899 1948 — - Published in connection with the Museum. Publication terminated with Vol. 24. The continuing publications are issued at irregular intervals in numbers which may be purchased separately. Prices and lists may be obtained on application to the Director of the Museum of Comparative Zoology, Cambridge 38, Massachusetts. Of the Peters "Check List of Birds of the World," volumes 1-3 are out of print ; volumes 4 and 6 may be obtained from the Harvard University Press; volumes 5 and 7 are sold by the Museum, and future volumes will be published under Museum auspices. Bulletin of the Museiun of Comparative Zoology AT HARVARD COLLEGE Vol. 117, No. 2 CHECK LIST OF THE REPTILES AND AMPHIBIANS OF EAST AFRICA (UGANDA; KENYA; TANGANYIKA; ZANZIBAR). By Arthur Loveridqe CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM August. 1957 No. 2. — Check List of the Reptiles and Amphibians of East Africa (Uganda; Kenya; Tanganyika; Zanzibar) By Arthur Loveridge CONTENTS Page lutroduction 153 Acknowledgements 162 List of the Reptiles (Turtles; Crocodiles; Lizards; Snakes) 163 List of Amphibians (Caeeilians; Toads; Frogs) 305 List of species still unrepresented in the Museum of Comparative Zoology 358 Bibliography 360 Index to all scientific names mentioned i INTRODUCTION "A check list is the lowest class of literature." Though more profound observations have long since faded from memory, the preceding remark — made to me by a former colleague at the National Museum of Wales — has, strangely enough, never been forgotten. Possibly I was startled by G. R. Brook's dictum as the idea of a cheek list being considered as literature had never oc- curred to me. Nevertheless, however classified, a check list may serve a useful purpose in proportion to the time and care devoted to making it as free from error as may be humanly possible. The present list owes its origin to a simple compilation of names extracted from such literature as was available to me in 1914, augmented by records of any specimens that had come my way. It was intended solely as a working basis for my personal use when I should reach East Africa. During the succeeding decade a number of additions resulted from my own field work. Upon my return from East Africa in 1923, I was urged by the then curator of reptiles at the British Museum, Dr. Joan B. Procter, to make it available to other workers. Somewhat reluctantly I did so the following year, assuming I had seen the last of East Africa since I was proceeding to the United States. That check list, of which I never saw proof as it 154 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY was published in Kenya, certainly merited Brook's description. At that time, however, since it was the first list to cover the area, it was probably better than nothing! The area covered w-as precisely the same as in the present list, viz : Kenya Colony and Protectorate . . 224,960 square miles Tanganyika Territory 360,000 Uganda Protectorate 93,981 Zanzibar Protectorate' 1,020 Giving a total area of 679,961 square miles It is, therefore, but little more than a quarter the size of the continental United States," whose herpetofauna has been the sub- ject of intensive study during the past century. East Africa, whose interior was largely a terra incognita sixty years ago, though only about a tenth of the area covered by the North American check list, straddles the equator. Consequently, though solely on an area basis, it enjoys a relatively richer herpetofauna, as shown bv the following figures. Group Total number of forms listed as occurring in East Africa in 11124 3 Kast Africa in 1957 North .Vnierica in 111.").'', I Turtles and Tortoises 13 15 81 Crocodilians o 3 2 Lizards and Chameleons 155 179 148 Snakes 139 161 275 Amphibians 128 169 262 TOTALS 437 527 768 1 Including Pemba Island with an area of approximately 380 square miles. 2 3,022,387 square miles which, added to that of Canada (3,695,189 square miles) gives a total of 6.717,576 for North America — the area covered bv tht 1953 check list. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 155 As so many species are common to more than one territorial area, a regional breakdown of the new figures naturally results in some duplication. So far as Zanzibar Island is concerned, the situation remains obscured by the earlier use of the word Zanzibar (== Negro coast) in a broader sense that covered the mainland littoral of Tanganyika Territory only forty miles away. If only some resident naturalist would conduct a thorough survey of the island's herpetofauna, the matter might be cleared up once and for all. Though the figures for the three offshore islands corre- spond fairly closely to those of Moreau and Pakenham (1941: 109), actually a number of names have been dropped on account of synonymy or for other reasons, while others have been substi- tuted or added. Grouii Uganda I'rotec- torate Kenya Colony Tan- ganyika Territory Peniba Island Zauziliar Island Matia Island Turtles and Tortoises 7 10 12 1 5 1 Croeodili.'Uis o i o 0 1 0 Lizards and Chameleons 41 92 123 10 18 10 Snakes 69 89 113 9 33 12 Caecilians 0 3 7 0 0 0 Toads and Frogs 44 Go 114 6 21 9 TOTALS 1(53 260 371 26 78 32 Apart from the inflation of the figures for Zanzibar's herpeto- fauna for the reasons given above, offset to some extent by the fact that the reptiles of its dependent island Pemba are little known, it is clear that there is a definite relation between speciation and area. 3 The Amphibia are taken from a l!t30 list published in the Proceedings of the Zoological Society of London, pp. 7-32. * As no figures are given in the Check List of North American Amphibians and Reptiles, by K. P. Schmidt (ed. 0. University of Chicago Press), these are my own count in which are included the omitted forms in the addenda (1954, Copeia, pp. 304-30G), but exclusive of recently introduced lizards (8 kinds) and frogs (3 forms). 156 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY At the time of the 1924 check list, when I arrived at the Mu- seum of Comparative Zoology, that institution had only 161 of the 527 species and races now recognized. Today all but 33 are represented. Number of listed forms represented in the Museum of Comparative Zoology Group Present in 1923 Additions since 1924 Still lacking Total Turtles and Tortoises 11 4 0 15 Crocodilians 3 0 0 3 Lizards and Chameleons 49 121 9 179 Snakes 61 95 ; ) 161 Caecilians 0 9 u 9 Toads and Frogs 37 104 19 160 TOTALS 161 333 33 527 Of this total of 527, 1 personally collected 413 in the field, while most of the remaining additions were obtained in exchange for duplicate material secured on the field trips I was privileged to carry out. The majority resulted from the itineraries being care- fully planned to embrace not only type localities of our desider- ata, but also type localities of many species or races that were inadequately described, or known only from one or two indi- viduals. Topotypic series of these creatures resulted in 180 (73 + 107) alleged forms of reptiles and amphibians being relegated to the synonymy. Not always justifiably, though usually so. Nevertheless, anyone who proposes describing additions to the herpetofauna of East Africa, would be well advised to scrutinize with care the synonyms attributed to the species most nearly related to the allegedly new form. It is important to note the distinction between the names listed in the generic synonymies, and those of species or races. Follow- ing the citations for the 135 genera (110 of reptiles; 25 of amphibians) recognized in the present list, are 527 additional LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 157 citations. While the majority are straight synonyms, with a sprinkling of nomina niida, others represent well-defined groups regarded by me as of subgenerie status. This is, of course, a mat- ter of opinion, and to some of them a few of my fellow workers may wish to accord generic rank. Intensive study of any small group is apt to result in observed differences assuming an undue importance, at least temporarily. To adjust one's perspective it is advisable to evaluate the situation on a family basis. In the 1924 check list no synonymies were supplied, so I trust their inclusion in the present one may serve a useful purpose. Is it too much to hope they may act as a deterrent to reckless de- scribers like the late Ernst Ahl, of whose 95 "species" only 28 are considered valid? It is always possible that additional material may demonstrate that an occasional "synonym" represents a recognizable form, but "more material" is just as likely to have the opposite result. That 527 recognized forms should have 830 synonyms does not quite reflect the position, for more than half (274) have no synonyms whatever; on the other hand, several widely distributed species have a dozen or more each. It is but fair to say, however, that of these 1349^ more or less useless names burdening the literature, not one was proposed by an African, but then neither was a single valid one ! This should not be construed as a reflection, for it is little more than a century since any one of the many indigenous tribes of the region had a written language. Not until 1848, 1 am informed, was the first of almost a hundred tongues in this area reduced to writing by a white immigrant. After all, following 300 years of civilizing contacts, it is doubtful whether an indigenous Indian described a single one of the 768 reptiles and amphibians that figure in the latest North American check list. Genera. Chiefly for the benefit of herpetologists in Africa, working without benefit of library facilities, I have listed under each genus all of its synonyms known to me. After this check list was typed, I was given (through the courtesy of Dr. A. S. Romer) the opportunity to scan the extensive synonymies in an advance copy of his "Osteology of the Reptiles" (pp. xxi -\- 772 : Chicago: 1956). This has enabled me to add between 20 and 30 synonyms of world-wide genera to the already lengthy lists. Gen- 5 527 generic -|- 822 species or races. 158 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY erally speaking, I have avoided inclusion of variations due to misprints or misspellings, together with some emendations (cf. footnote to Chamaeleo on p. 36. A few names that appear in the "Osteology" are omitted after due consideration. Unques- tionably, further additions will have to be made in the case of wide-ranging genera with species in Europe or Asia. Many tedious hours have been spent in attempting to ascertain the correct genotypes for each name. Where no type species had been mentioned I have occasionally selected one with the words "by present designation." Such action should be regarded as tentative, for in some instances further research may reveal that a type has already been proposed in some publication I have overlooked. Had I but realized at the outset how much time this uncongenial task would take I might well have shirked it, invok- ing the elder Agassiz's wise counsel to "Study specimens, not l)Ooks." Sound advice which he apparently overlooked when, in compiling his "Nomenclature Zoologicus" (1842-1847), he pro- posed innumerable emendations in the spelling of generic names. However, the chief offenders were Fitzinger and Gray who, ap- parently vying with each other in creating monotypic genera, have each proposed about 90 names — in so far as this list is con- cerned. Of these only 14 and 11 respectively are here regarded as full genera. Their percentages of validity (15% and 12%) compare very unfavorably with the more discriminating work of Giinther (10 out of 31 valid) and of Peters (8 out of 24), each of whom achieved approximately 33 per cent. Fitzinger, taking some work like the "Erpetologie Generale, " gave Dumeril and Bibron as the authors of his type species. As they in turn attributed the authorship to Linnaeus or some other early worker, I have cited the original author of the name with a view to avoiding much unnecessary printing. For example, in the case of Psammorrhoa Fitzinger (1843), instead of saying : "Type by original designation: Agama aculcata Dumeril & Bibron = Agama aculeata Merrem," I have cited Merrem only, omitting Dumeril & Bibron. I ask forgiveness of those taxonomically- minded colleagues who may be shocked by such action. In the course of compiling these generic synonymies, I came across several names that had priority over those currently in use. RuliuR-s setting aside these older names should be obtained from LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 159 the International Commission on Zoological Nomenclature with- out delay. They are : Thermophilus Fitzinger (1843), not Thermophila Hiibner (1816), antedates Ichnotropis Peters (1854) which has been in use as a genus of the LACE RTI DAE for over a century. Ibiba Gray (1825) antedates Boiga Fitzinger (part:1826), both having the same type : Coluber irregularis Merrem. Cerastes Laurenti (1768), long associated with certain vipers, is quite unsuitable as a generic name for the docile opisthoglyphs currently known as Psammophylax Fitzinger (1843), both gen- era having the same type : Coluber rhombeatus Linnaeus. Philodcndros Fitzinger (1843), a most inappropriate name for the terrestrial snakes which have been known as Dromophis Peters (1869) for almost a century. Both genera have the same type : Dendrophis praeornata Schlegel. Chloroechis Bonaparte (1849) antedates Atheris Cope (1862), a name that has been in use for almost a century. Both genera have the same type : Viper a chloroechis Schlegel. At this point I might also mention Cobra lachesis Laurenti (1768). This is an earlier name for Bitis a. arietans Merrem (1820), the common Puff Adder which ranges from the Cape to Morocco. Some years ago Cobra was rejected as generically ap- plicable. It is regrettable that lacJiesis was not dealt Avith at the same time, for reasons I have given elsewhere and have repeated briefly in a footnote to arietans. Eminophis Werner (1924, Sitzb. Akad. Wiss. Wien, 133, Abt. 1, p. 55), whose type (E. lineolata Werner) was said to have come from East Africa, was actually based on an Indian species — TracJiiscJium fuscum (Blyth) — according to Malcolm A. Smith (1928, Ann. Mag. Nat. Hist., (10) 1, p. 496). Species and Races. The 1924 check list was largely binomial, only 7 "varieties" (of Boulengerian nomenclature) and 28 sub- species being cited. Of the 527 reptiles and amphibians figuring in this new list no fewer than 259 are expressed as trinomials — a change reflecting a better understanding of relationships and ranges. Several names appearing in the earlier list, or subse- quently recorded from East Africa, have been omitted after an examination of the material on which they w^re based revealed misidentifications — usually of a subspecific nature. It is to be 160 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY hoped that they will not be reinstated by enthusiastic amateurs without due cause. Various species or races, rather too numerous to list, are now synonymized for the first time. It Avill be observed that citations are given in a carefully standardized format. Precise date of publication is sometimes a problem. For example, it maj^ be noted that the dates I give for many of the 34 species (common to East and South Africa) described by Sir Andrew Smith will be found to differ from those given by Boulenger in his catalogues. My dates were ob- tained from a paper by F. II. Waterhouse (1880, Proc. Zool. Soc. London, pp. 489-491), "On the Dates of Publication of the parts of Sir Andrew Smith's 'Illustrations of the Zoology of South Africa.' " The titles of some journals bear the name of the city of publication, e.g., Bull. Soc. Philom. Paris. Where omitted, and its absence may create difficulties, it has been added in parentheses, e.g., Bull. Mus. llist. Nat. (Paris). Where a figure carries no number or letter, to show that it has not been in- advertently omitted I have expressed it thus : Fig. — . It seems scarcely necessary to point out that capitals for specific or subspecific names are no longer in general use. Like hyphenated names they appear that way in the synonymy only when that was the way they were originally printed ; for example, a specific name like adolfifriderici may have appeared as Adolfi- frider-ici in the first instance. The description of Gecko tubercidosus Daudin (1802) seems to agree with the characters of the Common House-Gecko of East Africa long known as Hemidactylus mabouia (Moreau de Jonnes : 1818). However, the latter name having been in continuous use for 150 years, it is to be hoped a ruling to set tubercidosus aside will be secured from the International Commission on Zoological Nomenclature. The only outstanding change meriting mention here is the use of Boaedon fuliginosiis fidiginosus (Boie:1827) for the Common House-Snake, long known in East and South Africa as Boaedon lineatus lineatus Dumeril & Bibron (1854). As the eastern rep- tile appears indistinguishable from the western, this vexatious change seems unavoidable. Many extraterritorial species, even genotypes that are alien to Africa, are mentioned in footnotes or elsewhere. All are recorded LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 161 in the comprehensive Index which is the key to a mine of infor- mation that might he otherwise overlooked. Ranges. In general, ranges are presented in a standardized format, the included countries — commencing with Spanish Morocco — being taken clockwise around the continent. Circum- stances occasionally make deviations advisable as, for example where forest forms of western Kenya Colony extend due west to Gambia. Recent changes in political status such as Sudan (for the former Anglo-Egyptian Sudan), still differentiated by its spelling from the French Soudan, and Somalia (for Italian Somaliland) have been followed. Eritrea, however, though now merged with Ethiopia is faunistieally so different that it ap- peared advisable to retain it as a separate entity. Other prov- inces that have been similarly treated are the huge units compris- ing the Union of South Africa, and sometimes it has seemed best to be specific with areas like the Gabon, instead of blanketing them in the vast region known as French Equatorial Africa. The territories comprising French West Africa also are at times listed individually. When an animal occurs in any one of the four countries covered by this check list, that country is mentioned specifically. Though Mafia Island, lying south of Zanzibar, forms part of Tanganyika Territory, zoogeographically it is simpler to list it along with Pemba and Zanzibar Islands, from whose sultanate it was once administered. The v\-ord "coastal" is employed to de- note the lowland zone extending from the Indian Ocean to the continental plateau. In width it varies considerably for a dozen degrees south of the equator. Its characteristic herpetofauna reaches inland as far as Lake Nyasa and even beyond. It need hardly be pointed out that the ranges furnished are only as of 195G, and subject to extension as our knowledge in- creases. Caution should be exercised in extending them to include improbable records, for the distribution of many species is con- ditioned by well-defined habitat preferences. The nine principal faunal zones have been discussed elsewhere (Loveridge, 1937, Bull. Mus. Comp. Zool., 79, pp. 481-541), and the majority of species in this check list assigned to one or more of the zones. Our understanding of Hyperolius is so far from complete that to avoid misunderstandings as to the ranges of these sedge- or 162 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY tree-frogs I have departed from the format employed elsewhere by indicating' with an asterisk (*) precisely those countries from which material of the particular species or race is present in the collections of the Museum of Comparative Zoology. As a further safeguard, countries included on the basis of unique records are followed, in parentheses, by the name of the responsible author- ity and the year in which he published the record. This pro- cedure has also been used for members of the RANIDAE. Footnotes. Primarily for the benefit of isolated workers in Africa, explanatory annotations have been given in the form of footnotes. These provide references to, or the reasons for, rela- tively recent taxonomic changes ; others are intended to elucidate ambiguities or to account for seeming omissions, especially where due to erroneous records. A few refer to unsolved problems and indicate profitable lines of investigation that need to be under- taken. ACKNOWLEDGEMENTS The policy of submitting a check list, prior to publication, to the scrutiny of one's colleagues is an admirable one. Unfortu- nately, most of those interested in African herpetology are over- seas, besides which the time factor militates against putting the idea into practice in this instance. However, help has been received from various correspondents and colleagues, and I wel- come this opportunity of expressing my deep appreciation for their assistance. My thanks go to: Mr. J. C. Battersby of the British Museum for his kindness in answering sundry questions. Dr. 0. G. S. Davis of Purdue University for allowing me to see her manuscript synonymy of Python and Eryx. Dr. V. F. FitzSimons of the Transvaal Museum for information regarding certain South African amphibia, and early type localities not to be found on recent maps. Dr. Carl Gans of Harvard Uni- versity for rectifying my concept of the genus Dasypeltis and permitting me to use his amended synonymy. Mons. Jean Guibe of the Museum National d'Histoire Naturelle for examining the Kenya specimen of Biifo latifrons. Henry Horn of Cambridge High and Latin School for his conscientious and painstaking work in checking citations, arranging the vast index, and related matters. C. J. P. lonides, Esq., of Tanganyika Territory, whose LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 163 fine field work has added several new species to East Africa's herpetofauna, and extended the known ranges of others. Miss J. B. MacKenzie and her staff in the library of the Museum of Comparative Zoology, for their ever willing help and skill in tracking down references to obscure or inadequately cited pub- lications. Dr. Ernst Mayr of this museum for advice on various knotty problems of a taxonomic nature. Dr. Robert Mertens of Senckenberg Museum for answering questions respecting early German publications and Boettger types. Dr. H. W. Parker of the British Museum for sundry information, and for pin-pointing obscure localities in Ethiopia and Somalia. Mr. Benjamin Shreve, my colleague for the past twenty-five years, for friendly advice as to the gender and correct suffixes of sundry scientific names. Dr. M. A. Smith of the British Museum for scanning the pages relating to the synonymy of the HYDROPHIIDAE. Dr. E. E. Williams of Harvard University for examining the dentition of Cacostenium. and other favours. It is from our recently published revision of the CRYPTODIRA that the synonymy of the East African members of that suborder have been taken. Miss N. E. Wright of this museum for her pains- taking editorship of this manuscript. Class REPTILIA Subclass ANAPSIDA Older TESTUDINATA'^ Suborder CRYPTODIRA" Family DERMOCHELYIDAE Genus DERMOCHELYS Blainville 1816. Dermocheh/s Blainville, Bull. Soc. Philom. Paris, p. "Ill" (mis- 6 The oldest name for the entire group is TESTUDINES Batsch, 1788 (Anleit. Kennt. Thiere Mineral, 1, p. 437), rejected as being merely a nominative plural of Testudo. f'HELONIA Macartney, 1802 (in Ross, Transl. Cuvier, List. Comp. Anat. 1, pi. iii) is discarded as likely to result in confusion with the .crenus Chelonia and with adjectives derived therefrom. Consequently, as the laws of priority are not applicable to ordinal names. TESTUDINATA Oppel. 1811 (Ordn. Kept., i). o) is preferred. " Other suborders of TESTUDINATA are used by some herpetologists. On account of its peculiar carapacial structure they would isolate Dermochelys as suborder ATHECAE, all other turtles and tortoises being grouped under THECOPHORA. Such a separation has been considered undesirable by some anatomists and other recent workers. 164 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY print = 119). Type by monotypy: Testudo coriacea Linnaeus. 1820. Sphargis Merrem, Vers. Syst. Amphib., p. 19. Type by monotypy: Testudo coriacea Linnaeus. 1822. Coriudo Fleming, Phil. Zool., 2, p. 271. Type by monotypy: Tes- tudo coriacea Linnaeus. 1828. Seytina Wagler, Isis von Oken, 21, col. 861. Substitute name for Sphargis Merrem. 1830. Dermatoclielys Wagler, Nat. Syst. Amphib., p. 133. Emendation for Dcrmochelys. 1832. Chelyra Rafinesque, Atlantic Journ., 1. p. 64. Type by monotypy: Testudo coriacea Linnaeus. Dermochelys coriacea (Linnaeus) Luth or Leathery Turtle 17G(J. Testudo coriacea Linnaeus, Syst. Nat., ed. 12, 1, p. 350: Mediter- ranean Sea. 1771. Testudo arcuata Catesby, Nat. Hist. Carolina, 2. p. 40: Coasts of Carolina and Florida (by inference). 1788. Testudo lyra Lacepede, Hist. Nat. Quad. ovip. Serpens, 1. Synopsis Methodiea (also p. Ill as Le Luth) : Barbary Coast, North Africa. 1792. Testudo tuberculata Schoepff, Hist. Testud., p. 144: No locality. 1814. Chelonia lutaria Eafinesque, Specchio Sci. (Palermo), 2, No. 9, p. 66 : Mediterranean Sea. 1820. Sphargis mercuralis Merrem, Vers. Syst. Amphib., p. 19: Mediter- ranean Sea and Atlantic Ocean. 1829. Dermochelis Atlantica Lcsueur {nomen nudum), in Cuvier, Regne Animal, ed. 2, 2. p. 14, footnote : No locality. 1830. Dermatoclielys porcata Wagler, Nat. Syst. Amphib., pi. i, figs. 1-23 (on p. 133 coriacea is used) : No locality. 1884. Sphargis coriacea var. Sdhlegelii Garman, U. S. Nat. Mus. Bull. No. 25, pp. 295, 303 : Tropical Pacific and Indian Oceans. 1899. Sphargis angusta Philippi, Anales Univ. Chile (Santiago), No. 104, p. 730, pi. i: Tocopilla, Chile. Range. Indian and other tropical oceans, and, as an accidental visitor, the temperate seas. Family CHELONIIDAE Genus CHELONIA Brongniart 1800. Chelonia Brongniart (part), Bull. Soe. Philom. Paris, 2. p. 89. Type designated by Fitzinger:1843: Testudo mydas Linnaeus. 1837. Chelana Buraieister, Handbuch Naturg., 2. Abt. Zool., p. 731. Type by monotypy: Testudo mydas Linnaeus. 1838. Mydas Cocteau (not Fabricius:1799), Rept., in Sagra, Hist. Fis. LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 165 Pol. Xat. Cuba, 4. p. 22. Type by tautonomy: Testudo mydas Linnaeus. 1843. Mydasea Gervais, Diet. Hist. Nat., 3. p. 457. Type by tautonomy: Testudo mydas Linnaeus. 1845. Euchelonia Tschudi, Fauua Peruana, Herp., p. 22. Type by mono- typy: Testudo mydas Linnaeus. 1848. Megemys Gistel, Naturg. Thierr., p. viii. Substitute name for Chelonia Brongniart. 18.58. Euchelys Girard, U. S. Explor. Exped. 1838-1842, Herp-, p. 447. TjT)e by monotypy: E. macropus Girard = Testudo mydas Linnaeus. 1862. Chclone Strauch, Mem. Acad. Imp. Sci. St.-Petersbourg, (7) 5. No. 7, p. 59. Type by original designation: Testudo viridis Schneider = T. mydas Linnaeus. Chelonia mydas (Liunaeus) Green Turtle 1758. Testudo mydas Linnaeus, Syst. Nat., ed. 10, 1, p. 197: Ascension Island. 1782. Testudo (macropus) Walbaum, Chelongr. Schildkroten, p. 112: No locality. 1783. Testudo Viridis Schneider, Natur. Schildkroten, p. 299: No locality. 1787. Testudo japonica Thunberg, Vetensk. Acad. Handl., 8. p. 178, pi. vii, fig. 1 : Japan. 1788. Testudo viarina (seu vulgaris) Lacepede, Hist. Nat. Quad. ovip. Serpens, 1, Sj-nopsis Methodica (also p. 54 as La Tortue Franche) : Torrid Zone. 1788. Testudo viridis-squamosa Lacepede, Hist. Nat. Quad. ovip. Serpens, 1, Synopsis Methodica (also p. 92 as La Tortue ecaille-verte) : Amazon River, Brazil. 1802. Testudo cepedia-na Daudin, Hist. Nat. Rept., 2, p. 50, pi. xvii, fig. 1 : No locality. 1812. Chelonia virgata Schweigger, Konigsberger Arch. Naturwiss. Math., 1. pp. 291, 334, 411: Seas of the Torrid Zone. 1820. Caretta esculenta Merrem, Vers. Syst. Amphib., p. 18: Atlantic Ocean. 1820. Caretta nasicornis Merrem, Vers. Syst. Amphib., p. 18: Ocean near America. 1820. Caretta Thunbergii Merrem, Vers. Syst. Amphib., p. 19: Japan. 1829. Chelonia maculosa Cuvier, R^gne Animal, ed. 2, 2, p. 13: No local- ity. 1829. Chelonia lachrymata Cuvier, RSgne Animal, ed. 2, 2, p. 13: No locality. 1834. Chelonia bicarinata Lesson, in Belanger, Voy. Indies-Orient., Zool., p. 301 : Atlantic Ocean. 166 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1835. Chelonia Marmorata Dumeril & Bibron, Erpet. Gen., 2, p. 546, pi. xxiii, fig. 1 : Ascension Island. 1858. Chelonia formosa Girard, U. S. Explor. Exped. 1838-1842, Herp., p. 456, pi. xxxi, figs. 1-4: Fiji Islands. 1858. Chcloniu tenuis Girard, U. S. Explor. Exped. 1838-1842, Herp., p. 459, pi. xx.xi, fig. 8: Honden Island, Paumotu Group, Tahiti and Eimeo ; Rosa Island. 1868. Chelonia Agassisii Bocourt, Ann. Sci. Nat. Zool. (Paris), (5) 10, p. 122: Nagulate River mouth, Guatemala. 1880. Chelonia depressa Garman (at least in part), Bull. Mus. Comp. Zool., 6, p. 124: East Indies (juvenile). 1887. Chelonia lata Philippi, Zool. Garten, 28. p 84: near Valparaiso, Chile. Range. Indian and other tropical oceans and, as an accidental visitor, the temperate seas. Genus ERETMOCHELYS Fitzinger 1828. Carctta Ritgen (not Rafinesque:1814), Nova Acta Acad. Leop.- Carol., 14. p. 270. Type by monotopy: Testudo imhricata Linnaeus. 1843. Eretmochelys Fitzinger, Syst. Rept., p. 30. Type by original desig- nation: Testudo imbricata Linnaeus. 1868. Herpysmostes Gistel in Blicke, Leben Natur Menschen (Leipzig), p. 145. Type (fide Mertens:1936) : Testudo imbricata Linnaeus. 1873. Onychochelys Gray, Proc. Zool. Soc. London, p. 397, figs. 1-2. Type by monotypy: 0. Icraussi Gray = Testudo imbricata Linnaeus. Eretmochelys imbricata (Linnaeus) Hawksbill Turtle 1766. Testudo imbricata Linnaeus, Syst. Nat., ed. 12, 1. p. 3r)0: American and Asiatic seas. 1834. Chelonia psevdo-mydas Lesson, in Belanger, Voy. Indies-Orient., Zool., p. 299 : Atlantic Ocean. 1834. Chelonia pseudo-caretta Lesson, in Belanger, Voy. Indies-Orient.. Zool., p. 302 : Atlantic Ocean. 1835. Caretta Bissa Riippell, Neue Wirbelth. Fauna Abyssinica, Amphib., p. 4, pi. ii: Red Sea. 1857. Eretmochelys squamata Agassiz, Contr. Nat. Hist. U. S., 1. p. 382: Indian and Pacific Oceans. 1858. Caretta squamosa Girard, U. S. Explor. Exped. 1838-1842, Herp., p. 442, pi. XXX, figs. 1-7: Sulu Seas and Indian Ocean. 1858. Caretta rostrata Girard, U. S. Explor. Exped. 1838-1842, Herp., p. 446, pi. XXX, figs. 8-13: Fiji Islands. 1873. Onychochelys Icraussi Gray, Proc. Zool. Soc. London, p. 398, figs. 1-2: Atlantic Ocean off French Guiana. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 167 Range. Indian and other tropical oceans and, as an accidental visitor, the temperate seas. Genus CARETTA Rafinesque 181-1. Caretta Rafinesque, Speechio Sci. (Palermo), 2, No. 9, p. 66. Type by monotypy: C. nasuta Rafinesque = Testudo caretta Linnaeus. 1835. Thalassochelys Fitzinger, Ann. Wiener Mus., 1, pp, 110, 121, 128. Type by original designation: T. caouana Fitzinger = Testudo caretta Linnaeus. 1838. Caouana Cocteau, Rept., in Sagra, Hist. Fis. Pol. Nat. Cuba, 4. p. 31. Type by monotypy: Thalassochelys cephalo Schneider := Testudo caretta Linnaeus. 1843. UaUchelys Fitzinger, Syst. Rept., p. 30. Type by original designa- tion: H. atra Fitzinger = Testudo caretta Linnaeus. 1873. Cephalochelys Gray, Proc. Zool. Soc. London, p. 408. Type by monotypy: C. oceanica Gray = Testudo caretta Linnaeus. 1873. Eremonia Gray, Proc. Zool. Soc. London, p. 408. Type by mono- typy: E. elongata Gray =^ Testudo caretta Linnaeus. Caretta caretta (Linnaeus) Red-brown Loggerhead 1758. Testudo Caretta Linnaeus, Syst. Nat., ed. 10, 1. p. 197: Islands off America. 1783. Testudo Cephalo Schneider, Natur. Schildkroten, p. 303: No locality. 1788. Testudo camia7ia Lacepede, Hist. Nat. Quad. ovip. Serpens, I, Synopsis Methodica (also p. 95 as La Caouane) : Jamaica, British West Indies (restricted). 1788. Testudo nasicornis Lacepede, Hist. Nat; Quad. ovip. Serpens, 1, Synopsis Methodica (also p. 103 as La Tortue nasicorne) : America. 1814. Caretta nasuta Rafinesque, Speechio Sci. (Palermo), 2. No. 9, p. 66: Sicily (fide Lindholm:1929). 1820. Caretta atra Merrem, Vers. Syst. Amphib., p. 17: Ascension Island. 1833. Chelonia pelasgorum Valenciennes, pi. vi, in Bory, Exped. Sci. Moree, Zool.: Modon, etc., Mediterranean Sea. 1844. Caouana elongata Gray, Cat. Tort. Croc. Amphis. Brit. Mus., p. 53: No locality. 1858. Thalassochelys corticata Girard, U. S. Explor. Exped. 1838-1842, Herp., p. 431, pi. xxix: Madeira Islands. 1873. Cephalochelys oceanica Gray, Proc. Zool. Soc. London, p. 408: Mexico. 1887. Thalassochelys tarapacona (sic) Philippi (identified by subsequent 1 08 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY description of Philippi:1899), Zool. Garten, 28, p. 83: Iquique, Chile. ,1930. Caretia gigas Deraniyagala, Ceylon Journ. Sci., (B), 18, p. 66, figs. 4-6, pi. v: Gulf of Mannar, Ceylon. Range. Indian and other tropical oceans and, as an accidental visitor, the temperate seas. Genus LEPIDOCHELYS Fitzinger 1843. Lepidochelys Fitzinger, Syst. Eept., p. 30. Type by original desig- nation: Chelonia olivacea Eschscholtz. 1880. Colpochelys Garman, Bull. Mus. Coiup. Zool., 8, p. 124. Type by monotypy : C. Tcempi Garman. Lepidochelys oUvacea olivacea (Eschscholtz) Olive Loggerhead 1820. Whelonia multisnitata Kuhl, Beitr, Zool. Anat., 1. p. 78: No locality. 1829. Chelonia olivacea Eschscholtz, Zool. Atlas, p. 3, pi. iii: Manila Bay, Philippine Islands. 1835. Chelonia Bussumierii Dumeril & Bibron, Erpet. Gen., 2, p. 557, pi. xxiv, figs. 1-la: China Sea and Malabar Coast. 1844. Caoiuvna Riippcllii Gray {nomen nudum). Cat. Tort. Croc. Amphis. Brit. Mus., p. 53: India? 1857. Chelonia polyaspis Bleeker, Nat. Tijdschr. Ned. Indies, 14, p. 239: Batavia, Java. 1889. Chelonia duhia Bleeker {nomen nudum), in Boulenger, Cat. Chelon. Ehj'ncho. Croc. Brit. Mus., p. 186: Borneo. 1899. TJialassochelys controversa Philippi, Anales Univ. Chile (Santi- ago), No. 104, p. 732: No locality. 1908. Caretta remivaga Hay, Proc. U. S. Nat. Mus., 34. p. 194, pi. x, figs. 1-3 ; pi. xi, fig. 5 : Ventosa Bay, Gulf of Tehuantepec, Mexico. Range. Indian and other tropical oceans and, as an accidental visitor, the temperate seas. Family TESTUDINIDAE Genus TESTUDO Linnaeus^ 1758. Testudo Linnaeus, Syst. Nat., ed. 10, 1, p. 197. Type by subsequent designation of Fitzinger: Testudo graeea Linnaeus. s In place of the single genus Testudo some authorities would recognize several full genera {Testudo; Geochelonej Psammobates) . Despite possibly ancient sep- aration I personally prefer to regard these groups as of only subgeneric rank. However, this is merely my personal opinion ; for those who think otherwise and consider Geochelone as a full genus, the only synonyms of it would be Centrochelysj Stigmochelys and Megaohersine. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 169 1820. Chersine Merrem, Vers. Syst. Amphib., pp. 12, 29. Type by subse quent designation of Lindholm: Testudo graeca Linnaeus. 1830. Chersus Wagler, Nat. Syst. Amphib., p. 138. TjTje by monotypy: Testudo marginata Schoepff. 1835. Psammobates Fitzinger, Ann. Weiner Mus., 1, pp. 108, 113, 122. Type by original designation: Testudo geometrica Linnaeus. 1835. Geochelone Fitzinger, Ann. Weiner Mus., 1, pp. Ill, 112, 122. Type by original designation: Testudo stellata Schweigger = T. clegans Schoepff. 1869. Peltastes Gray (not Bossi:1807), Proc. Zool. Soc. London, pp. 167, 171. Type by subsequent designation of Lindholm: Testudo graeca Linnaeus. 1870. Chersin.ella Gray, Suppl. Cat. Shield Rept. Brit. Mus., part 1, p. 8. Type by subsequent designation of Lindholm: Testudo graeca Linnaeus. 1870. Testudinella Gray (not Bory:1826), Suppl. Cat, Shield Rept. Brit. Mus., part 1, p. 12. Type by monotypy: Testudo horsfieldi Gray. 1872. Peltonia Gray, App. Cat. Shield Rept. Brit. Mus., p. 4. New name for Peltastes Gray (preoccupied). 1872. Centrochelys Gray, App. Cat. Shield Rept. Brit. Mus., p. 5. Type by monotypy: Testudo sulcata Miller. 1873. Stigmochelys Gray, Hand-List Shield Rept. Brit. Mus., p. 5. Type by monotypy : Testudo pardalis Bell. 1916. Medaestia Wussow, Wochenschr. Aquar.-Terrar. Kunde, 13, p. 170. Type by subsequent designation of Mertens: Testudo graeca Linnaeus. 1933. Megachersine Hewitt, Ann. Natal Mus., 7, p. 257. Type by original designation : Testudo pardalis Bell. 1933. Chersinella Hewitt (not Gray: 1870), Ann. Natal Mus., 7, p. 259. Type by original designation: Testudo geometrica Linnaeus. Testudo pardalis babcocki Loveridge Eastern Leopard Tortoise 1935. Testudo pardalis hahcocki Loveridge, Bull. Mus. Comp. Zool., 79. p. 4: Mount Debasien, Karamojo, Uganda. Range. Sudan and Ethiopia, south through Uganda; Kenya Colony and Tanganyika Territory to Natal, west through Cape Province to Southwest Africa where it meets with the typical form, but is dominant north of 25° S. and in southern Angola. Genus MALACOCHERSUS Lindholm 1929. Malacochersus Lindholm, Zool. Anz., 81, p. 285. Type by original designation : Testudo tornieri Siebenrock. Molaco chersus tornieri (Siebenrock) Pancake Tortoise 1 70 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1903. Testudo tornicri Siebeiirock, Auz. Akad. Wiss. Wien, 40, p. 185 : Busisi, Smith Sound, Lake Victoria, Tanganyika Territory. 1920. Tcstndo Lovcridgii Boulenger, C. R. Acad. Sci. (Paris), 170, p. 263: Dodonia, Ugogo, Tanganyika Territory. 1923. '^Testudo procterae Loveridge, Proc. Zool. Soc. London, p. 298, pis. i-ii: Ikikuyu, south of Gulwe, Tanganyika Territory." Range. Rocky hills of Kenya Colony (Mathews Range south to Njoro. east to Midas Creek), and Tanganyika Territorj^ (fi'oni Busisi, Smith Sound, southeast through Ugogo to Lindi). Genus KINIXYS Bell 1827. Kinixys Bell, Trans. Linn. Soc. Loudon, 15, p. 398. Type by original designation : K. castanea Bell = Testudo erosa Schweig- ger. 1835. CinotJiorax Fitzinger, Ann. Wiener Mus., 1, pp. 108, 111, 121. Type by subsequent designation of Fitzinger : 1843 : Kinixys helliana Gray. 1843. Cinixys Fitzinger, Syst. Kept., p. 29. Emendation for Kinixys Bell. 1873. Klnothorax Gray, Hand-List Shield Eept. Brit. Mus., p. 16. Emendation for CinotJiorax Fitzinger. Kinixys belliana belliana Gray Bell's Eastern Hinged-Tortoise 1831. Kinixys Belliana Gray, Synopsis Reptilium, p. 69: No locality. ("W. Africa ?", later added by Boulenger, appears doubtful.) 1845. Kinixys sehoensis Riippell, Mus. Senckenberg., 3, p. 226, pi. xvi, figs. 1-3 : Shoa, southern Ethiopia. 1863. Kinixys Spcl'ti Gray, Ann. Mag. Nat. Hist., (3) 12, p. 381: "Central Africa" (probably northwest Tanganyika Territory). 1902. Homopus darlingi Boulenger, Proc. Zool. Soc, London, 2, p, 15, pi. iv: Salisbury District, Mashonaland, Southern Rhodesia, 1927. Cinixys lohatsiana Power, Trans. Roy. Soc. S. Africa, 14. p, 410, pis. xix-xx: Lobatsi, Bechuanaland Protectorate. 1931. Kinixys helliana sombensis Hewitt, Ann. Natal Mus., 6, p. 469, fig. Ic, pi. xxxviii, fig, 4: Zomba, Nyasaland, 1931. Kinixys belliana zuluensis Hewitt, Ann, Natal Mus., 6, p. 471, figs. la-lb, pi. xxxviii, figs. 1-3, 5-10: Richard's Bay, Zululand, 1931. Kinixys austraUs Hewitt, Ann. Natal Mus., 6, p. 477, pi, xxxvi, figs. 4-6 : White River, Nelspruit, Lydenburg District, Transvaal. 9 As Procter's Soft-shelled Land-Tortoise is known only from a single specimen, it is tentatively relegated to tlie synonymy on geographical grounds until someone can secure a series from tlie type locality, which should not be confused with the Ikikuyu just outside Dodoma. Not only is M. procterae strikingly different in coloration from every one of the more than a hundred tornicri I have seen, but also its head scalation, especially its large and elongate prefrontals, is unique. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 171 1931. Kinixya jordani Hewitt, Ann. Natal Mus., 6. p. 481, pi. xxxvii, figs. 7-9 {not 1-3 as stated): Isoka, Northern Rhodesia. 1931. Kvnixys youngi Hewitt, Ann. Natal Mus., 6. p. 486, fig. Id, pi. xxxvii. figs. 4-5: Lake Nyasa shore below Livingstonia, Nyasa- land. 1932. Kinixys australis mababiensis FitzSimons, Ann. Transvaal Mus., 15. p. 37: Tsotsoroga Pan, Mababe Flats, Bechuanaland Pro tectorate. 1935. Kinixys natalensis Hewitt, Eee. Albany Mus., 4, p. 353, pi. xxxv, figs. 3-4: Jameson's Drift, Tugela River Valley, Natal. Range. Eritrea southwest to Natal, northwest through Bechu- analand and Angola to French Cameroon where it meets with the western race (which has only four claws on forelimb). Madagascar. Kinixys erosa (Schweigger) Western-Forest Hinged-Tortoise 1812. Tcstudo erosa Schweigger, Konigsberger Arch. Naturwiss. Math.. I. p. 321: "America septentrionali" (error). 1826. Testiido Schopfii Fitzinger (nomen nudum), Neue Class. Rept., p. 44 : No locality. 1827. Kinixys castanca Bell, Trans. Linn. Soc. London, 15. p. 398, pi. xvii, fig. 1 : Africa. Range. Uganda, west through the Belgian Congo to Gambia. Family TRIONYCHIDAE Genus TRIONYX Geoffrey '-^ 1809. Trionyx Geoffroy Saint-Hilaire, Ann. Mus. Hist. Nat. (Paris), 14. p. 1. Type by original designation: T. aegyptiacus Geoffroy = Testudo iriunguis Forskal. 1830. Aspidonectcs Wagler, Nat. Syst. Amphib., p. 134. Type by desig- nation of Stejneger: Trionyx aegyptiacus Geoffroy = Testudo triunguis Forsk°al. 1835. Gymnopus Dumeril & Bibron, Erpet. Gen., 2, p. 472. New name for Aspidonectes Wagler. 1844. Tyrse Gray, Cat. Tort. Croc. Amphis. Brit. Mus., p. 47. Type by tautonomy: Trionyx nilotica Gray = Testudo triunguis Forskal. 1869. Fordia Gray, Proc. Zool. Soc. London, p. 219. Type by monotypy: F. africana Gray = Testudo triimguis Forskal. 10 Only synonyms based on African species are given here as it seems scarcely v.'orth wliile to list the 24 additional ones involving American and Asiatic species. These can be found in the recently published revision of African Cryptodira (see Bibliography). 172 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Trionyx triunguis (Forskal) Nile Soft-shelled Turtle 1775. Testudo triunguis Forskal, Descr. Anim. Avium, Amphib., p. ix: Nile, Egypt. 1809. Trionyx Aegyptiacus Geoffroy Saint-Hilaire, Ann. Mus. Hist. Nat. (Paris), 14, p. 12, pis. i-ii: Nile, Egypt. 1831. Trionyx Niloticus "Shaw" Gray, Synopsis Reptilium, p. 46: Nile, Egyi:)t. 1837. Trionyx lahiatus Bell, Monog. Testudinata, text to pis. — (= xviii-xx) : Sierra Leone. 1844. Trionyx Mortoni Hallow ell, Proc. Acad. Nat. Sci. Philadelphia, p. 120: Africa. 1859. Aspidonectes aspilus Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 295 : Ovenga and Rembo Rivers, Fernan Vaz, French Congo. 1869. Fordia africana Gray, Proc. Zool. Soc. London, p. 219: Nile River at Khartoum, Sudan. 1948. Trionyx triunguis rudolfianus Deraniyagala, Spolia Zeylanica (Colombo), 25, 2, p. 30, fig. 5, pi. xii, fig. c: Ferguson's Gulf, Lake Rudolf, Kenya Colony. Range. Egypt and Eritrea south to Lake Rudolf, Kenya Col- ony; southwest to Lake Albert, Uganda; and Angola; northwest to Senegal. Accidental visitor along the Palestine coast of Asia Minor. Genus CYCLODERMA Peters 1854. Cycloderma Peters, Monatsb. Akad. Wiss. Berlin, p. 216. Type by monotypy : C. frenatum Peters. 1856. Cryptopodiis A. Dum^ril, Revue Mag. Zool. (2), 8, p. 374. Lapsus for Cryptopus Dumeril & Bibron = Cyclanorbis (part). 1859. Heptathyra Cope, Proc. Nat. Sci. Philadelphia, p. 294. Type by monotypy: Cryptopus (sic) aubryi A. Dumeril, 1860. Aspidochelys Gray, Proc. Zool. Soc. London, p. 6. Type by mono typy: A. livingstonii Gray := Cycloderma frenatum Peters. Cycloderma frenatum Peters Zambezi Soft-shelled Turtle 1854. Cycloderma frenatum Peters, Monatsb, Akad. Wiss. Berlin, p. 216: Zambezi River, Mozambique. I860. Aspidochelys livingstonii Gray, Proc. Zool. Soc. London, p. 6, pi. xxii, figs. 1-2: Tributaries of Zambezi, Mozambique. Range. Southern Tanganyika Territory, west and south through Nyasaland (and possibly Northern Rhodesia) to Mozam- bique. LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 17:) Suborder PLEURODIRA Family PELOMEDUSIDAE Genus PELOMEDUSA Wagler 1830. Pelomedusa Wagler, Nat. Syst. Amphib., p. 13G. Type by iiiono- typy: Testudo galeata Schoepff = Testudo suhrufa Lacepede. 1835. Pentonyx Dum^ril & Bibron, Erpet. Gen., 2. p. 389. Type by subsequent designation: P. capensis Dumeril & Bibron =r Tes- tudo stibrufa Lacepede. Pelomedusa subruia olivacea (Seliweigger)" Northeru Marsh-Terrapin 1814. Emys olivacea Schweigger, Prodromi Mon. Chelon., p. 38: "In Fabulosis Nigritae," as Adanson coll. = Senegal. 1835. Pentonyx G.ehafie RiippcU, Neue Wirbelth. Fauna Abyssinica, Amphib., p. 2, pi. 1 : Massaua, Eritrea. 1884. Pelomedusa Gasconi Rochebrune, Faune Senegambie, Eept., p. 25, pi. i, figs. 1-2: Dagana, Senegal (restricted). 1910. Welomedusa galeata var. disjuncta Vaillant & Grandidier, Hist. Phys. Nat. Pol. Madagascar, 17. Eept., p. 56, pi. xx, fig. 3: Du Bourg de Bozas coll., but no locality. Range. Uncertain, but apparently a northern belt extending from Senegal to Eritrea and intergrading with the typical form in northern Uganda and northern Kenya Colony with occasional indi\aduals cropping uj) in the south and Madagascar. Pelomedusa subruia subruia (Lacepede) Southern Marsh-Terrapin 1788. Testudo suhrufa Lacepede, Hist. Nat. Quad. ovip. Serpens, 1, Synopsis Methodica (also p. 173 as La Eoussatre): "de L'Inde, " as collected by Sonnerat restricted to Cape of Good Hope, South Africa. 1792. Testudo galeata Schoepff, Hist. Testud., p. 12, pi. iii, fig. 1: "India orientale, Carolina," restricted to Cape Flats, Cape Province, South Africa. 1798. Testudo Badia Donndorf, Zool. Beytrage Linn. Natur., 3, p. 34 (new name for Lacepede 's La Eoussatre): No locality. 1835. Pentonyx Capensis Dumeril & Bibron, Erpet. Gen., 2, p. 390, pi. 11 Some authorities would relegate this race and its forms to the synonymy of subruja. It is separated here in order to stimulate inquiry as to whether or not 75 per cent of the northern population have their pectoral shields separated on the median line of the plastron — the sole character allegedly separating the two races. P. g. disjuncta, if indeed it actually came from Madagascar, may be a throwbacit for these do occur, though very rarely, within the range of the typical s. subrufa. 174 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY xix, figs. 2-2b : restricted by Mertens to Cape of Good Hope, South Africa. 1849. •Pentonix americana Cornalia, Vert. Synop. Mus. Mediolau, p. 13: "in flumine propre Novaeboracum, " i.e. New York (error). 1855. Pelomedusa Mozamhica "Peters M.S.S. 1848" Gray (Nomen nudutn) , Cat. Shield Rept. Brit. Mus., p. 53: Mozambique. 1856. Felomedusa mossambicensis "Peters" Lichtenstein & v. Martens, {nomen nudum) Nomencl. Eept., p. 2 : Mozambique. 1863. Pelomedusa nigra Gray, Ann. Mag. Nat. Hist., (3) 12, p. 99: Natal, South Africa. 1910. IPelomedusa galeata var. disjuncta VaiUant & Grandidier, Hist. Phys. Nat. Pol. Madagascar, 17, p. 56: Madagascar.^- 1935. Pelomedusa galeata orangensis Hewitt, Rec. Albany Mus., 4, p. 332, pi. xxxi, fig. 3; pi. xxxii, figs. 3-4: Kimberley, Cape Prov- ince, South Africa. 1935. Pelomedusa galeata devilliersi Hewitt, Rec. Albany Mus., 4, p. 337, pi. xxxi, figs. 2 and 4: Besondermeid, Steinkopf, Cape Province, South Africa. 1935. Pelomedusa galeata damarensis Hewitt, Rec. Albany Mus., 4, p. 338, pi. xxxiii, figs. 1-4: Quickborn Farm, near Okahandja, Southwest Africa. 1937. Pelomedusa subrufa wettstcini Mertens, Zool. Anz., 117, p. 141, figs. 1 and 4: Majunga, western Madagascar. Range. Dry savanna of southern Sudan east to British Somali- land, south through Uganda;^" Kenya Colony and Tanganyika Territory to Natal, west through Cape Province to Southwest Africa, north and northeast to Belgian Congo. Genus PELUSIOS Wagler 1830. Pelusios Wagler, Nat. Syst. Amphib., p. 137. Type by monotypy : Einys eastanea Sehweigger = Testudo subnigra Lacepdde. 18G3. Tanoa Gray, Proe. Zool. Soc. London, p. 193. Type by monotypy: Sternothaerus siiiuatus A. Smith. 1863. Notoa Gray, Proc. Zool. Soc. London, p. 195. Type by monotypy: Sternothaerus suhniger Gray = Testudo subnigra Lacepede. Pelusios subniger (Lacepede)^" Black Terrapin 1 7SS. Testudo subnigra Lacepede, Hist. Nat. Quad. ovip. Serpens, I, Synopsis Methodica (also p. 175, pi. xiii as La Noiratre): No locality. 12 But see note under Pelomedusa s. olivacea (Sehweigger). 13 From this s.vnonymy I have removed f^tcrnothcrus niger Dum6ril & Bibron, together with its synonym .Sf. oxiirhuniD Boulenper, as they represent a valid West Afriean siioeies. LOVERJDGE : E. AFRICAN REPTILES AND AMPHIBIANS 17."i 1798. Testudo nigricans Donndorff, Zool. Beytrage Linn. Natur. 3, p. 34 (new name for La Noiratre Lacepede) : No locality. 1814. Emi/s castanea Scliweigger, Prodromi Men. Chelou., p. 45 (new name for siihniffra Daudin in Paris Museum) : No locality. 1825. Sternotltaerus Leachinnus Bell, Zool. Journ., 2, p. 306: No locality. 1844. Sternotherns Dcrbianus Gray, Cat. Tort. Croc. Amphis. Brit. Mus., p. 37: Gambia (restricted). 1851. Sternotlierus nigrescens Bianconi (lapsus for nigricans used in the text). Spec. Zool. Mosamb., Kept., pi. vii. 1886. Pelomedusa Jouberti Eochebrune, Vertebrata nov. Africae Occi- dent., Part 3, p. 10: Noki, Congo Eiver, Angola; Landana, Cabinda ; and Mellacoree Eiver, French Guinea. 1906. Sternothaeriis nigricans seycliellcnsis Siebenrock, in Voeltzkow, Eeise in Ostafrika, 2. p. 38: Gloriosa Island. 1927. Peliisios nigricans rhodesianns Hewitt, Eec. Albany Mus., 4. p. 375, figs, la, Ic ; pi. xxvi. figs. 2-3: Mpika District, Northern Ehodesia. 1931. Peliisios nigricans castanoides Hewitt, Ann. Natal Mus., 6. p. 463, pi. xxxvi, figs. 1-2: Eichard 's Bay, Zululand. 1932. Pelusios bechuanicus FitzSimous, Ann. Transvaal, Mus., 15, p. 37: Thamalakaue Eiver at Maun, Bechuanaland Protectorate. Range. Uganda ; Kenya Colony ; Tanganyika Territory and islands of the Indian Ocean (Pemba; Zanzibar; Seychelles; Madagascar; Mauritius). In east, south to Zululand, Natal, northwest to Angola ; Senegal and Cape A'erde Islands, east to T'ganda. Pelusios sinuatus (Smith) Serrated Terrapin ls38. Sternotherns sinuatus A. Smith, lUus. Zool. S. Africa, Eept., pi. i: In rivers to the north of 25° S., South Africa. 1848. Sternotherns dentatus Peters, Arch. Anat. Phys., p. 494: No locality. 1895. Sternotltaerus bottegi Boulenger, Ann. Mus. Civ. Stor. Nat. Genova, (2) 15. p. 9, pis. i-ii: Bardera, Somalia. 1927. Pelusios sinuatus suluensis Hewitt, Eec. Albany Mus., 4, p. 360. fig. Id, pi. XX, figs. 1-3: near Umsinene Eiver, Zululand. 1933. Pelusios sinuatus leptus Hewitt, Occ. Papers Ehodesian Mus., p. 45, jjI. ix, figs. 1-2: Isoka, Northern Ehodesia. Range. Somalia, south through Kenya Colony and Tanganyika Territory to Natal, northwest to the eastern Belgian Congo." 1+ Unknown from Uganda and Lake Victoria, for early records from tlie Sessc Islands and Hulcoba were based im misidentilied suhttigcr. 176 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Subclass ARCHOSAURIA Order CROCODYLIA^' Family CROCODYLIDAE Genus CROCODYLUS Laurenti''' 1763. Crocodylus Gronovius, Zoophyl. Gronov., fasc. 1, p. 10. Rendered unavailable by the Internat. Comm. on Zool. Nomen.^'^ 1768. Crocodylus Laurenti, Syn. Rept., p. 53. Type by subsequent designation of Stejneger & Barbour : 1917: C. niloticus Laurenti = Lacerta crocodilus Linnaeus (part). 1789. Crocodilus Bonnaterre (part), Tabl. Encycl. Method. Regnes Nat., Erpet., p. 32. Emendation for Crocodylus Laurenti. 1820. Oiampse Merrem, Vers. Syst. Amphib., p. 36. Type by subsequent designation: Crocodilus biscutatus Cuvier = C. acutus Cuvier. 1844. Mecistops Gray (part), Cat. Tort. Croc. Amphis. Brit. Mus. Type by subsequent designation Crocodilus cataphractus Cuvier. 1844. Oopholis Gray, Cat. Tort. Croc. Amphis. Brit. Mus., p. 58. Type by subsequent designation of Deraniyagala: Crocodilus porosus Schneider 1844. Palinia Gray, Cat. Tort. Croc. Amphis. Brit. Mus., p. 60. Type by monotypy: Crocodilus rhomhifer Cuvier. 1844. Motinia'^^ Gray, Cat. Tort. Croc. Amphis. Brit. Mus., p. 60. Type by monotypy: Crocodilus americanus Schneider = C. acutus Cuvier. 1862. Bombifrons Gray, Ann. Mag. Nat. Hist., (3) 10. p. 269. Type by subsequent designation: Crocodilus bombifrons Gray = C. palustris Lesson. 1862. remsacus Gray, Ann. Mag. Nat. Hist., (3) 10, p. 272. Type by monotypy : Crocodilus intermedius Graves. 1874. Philas Gray, Proc. Zool. Soe. London, p. 188. Type by monotypy: Crocodilus johnsoni Krefft := C. johnstoni Krefft. Crocodylus cataphractus Cuvier Long-nosed Crocodile 1801. ^Crocodilus )iiger Latreille, in Buffon, Hist. Nat. Rept, (ed. D6ter- ville), 1, p. 210: Senegal River, Senegal. 15 Generally accepted by herpetologists in preference to LORICATA which lias been applied to pangolins and other groups. 16 The skull from "Afrika ?" named Champse brevirostris (or Crocodilus krevirostrig in the caption to figs.) Werner, 1933, Zool. Anz., 102, p. lOO, figs. 1-2, has been identified as the Asiatic Crocodylus palustris Lesson by VVermuth, 1953, Mitt. Zool. Mus. Berlin, 29, p. 472. 17 See Opinions and Declarations, 1926, in Smithsonian Misc. Coll., 73, pp. 27-33. ISA misprint, corrected to Molinia by Gray, 18(52, Ann. Mag. Nat. Uist., (3) 10, p. 272. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 177 1824. Crocodiliis cataphractm Cuvier, Rech. Ossem. foss. Quad., ed. 2, 5. p. 58, pi. V, figs. 1-2: No locality. 1835. Crocodilus leptorhynchus Bennett, Proc. Zool. Soc. London, p. 129: Fernando Po. 1844. Mecistops Bennettii Gray (new name for C. Icptorhynchns Ben- nett), Cat. Tort. Croc. Aniphis. Brit. Mus., p. 57. Range. Near Ujiji, Lake Tanganyika, Tanganyika Territory, west to Angola, northwest to Senegal. Crocodylus niloticus Laurenti Nile Crocodile 1768. Crocodylus niloticus Laurenti (part), Syn. Kept., p. 53: Egypt (restricted). 1768. Crocodylus africanus Laurenti (part; with part lizard, so unidenti- fiable), Syn. Kept., p. 54: No locality. 1807. Crocodilus vulgaris Cuvier, Ann. Mus. Hist. Nat. (Paris), 10, p. 40, pis. iiii : Egypt. 1807. Crocodilus suchus Geoffroy Saint-Hilairc, Ann. Mus. Hist. Nat. (Paris), 10, p. 84, pi. iii: Nile and Niger Rivers. 1824. Crocodilus Chamses Bory de Saint-Vincent, Diet. Class. Hist. Nat., 5. p. 105: Nile, Egypt. 1826. Crocodilus muULscutatus Riippell, in Cretzschmar, Iris (Frankfurt am Main), 25. p. 99: Soucot, Nile, "Nubia" i.e. Kordofan, Sudan. 1827. Crocodilus marginatus Geoffroy Saiut-Hilaire, Crocodiles, in Descr. Egypte, p. 260: Nile River near Thebes, Egj-pt. 1827. Crocodilus lacunosus Geoffroy Saint-Hilaire, Crocodiles, in Descr. Egypte, p. 261: (mummy) EgjT)t. 1827. Crocodilus complatiatus Geoffroy Saint-Hilaire, Crocodiles, in Descr. Egypte, p. 263: (mummy) Egypt. 1831. Alligator cowieii A. Smith, S. African Quart. Journ., (5) p. 15: "Rivers beyond Currichane, " i.e. Rustenburg, western Trans- vaal. 1831. Crocodilus Octophractus Gray, in Griffith, Animal Kingdom, 9. Synopsis, p. 22: Soucot, Nile River, Kordofan, Sudan. 1857. Crocodilus Binuensis Baikie, Proc. Zool. Soc. London, p. 48: Benue and Niger Rivers, Nigeria. 1872. Crocodilus inadagascariensis Grandidier, Ann. Sci. Nat. Zool. (Paris), 15. art. 20, p. 6: Madagascar. 1872. Crocodilus robustus Grandidier & Vaillant, C. R. Acad. Sci. (Paris), 75. p. 150: (fossil) Amboulintsatre, Madagascar. 1886. Crocodilus hexaphractos Ruppell {nomen nudum), in Schmidt, Ber. Senckenberg. Naturf. Ges., for 1885, p. 131. 1948. Crocodylus niloticus loortliingtoni Deraniyagala, Spolia Zeylanica (Colombo), 25, 2, p. 30: Lake Baringo, Kenya Colony. 178 BULLETIN : MUSEUM OF COMPAEATIVE ZOOLOGY 1948. Crocodylus niloticus pauciscutatus Deraniyagala, Spolia Zeylanica (Colombo), 25. 2, p. 31: Lake Rudolf, Kenya Colony. Range. Uganda ; Kenya Colony ; Tanganyika Territorj^ : also most lakes and rivers of Africa from about 20° N., south to Tugela River, Natal. Zanzibar^'' and other islands (Seychelles; Comoros; Madagascar) of the Indian Ocean; also present in the Zerka River, Jordan. Genus OSTEOLAEMUS Cope 1861. Osteolaemus Cope, Pioc. Acad. Nat. Sci. Philadelphia for I860, p. 549. Type by monotypy: Osteolaemus tetraspis Cope. 1862. Halcrosia Gray, Ann. Mag. Nat. Hist., (3) 10. p. 273. Type by monotypy: Crocodilus frontatus Alurray. 1919. Ostcohlepharon K. P. Schmidt, Bull. Amer. Mus. Nat. Hist., 39, p. 420. Type by original designation : 0. osborni Schmidt. Osteolaemus tetraspis osborni (Schmidt) Eastern Broad-nosed Crocodile 1919. Osteohlepharon osborni K. P. Schmidt, Bull. Amer. Mus. Nat. Hist., 39, p. 421, figs, 2-5, pi. xiii, fig. 1 : Niapu, Belgian Congo. Range. Western Uganda and eastern Belgian Congo.'" Subclass SYNAPTOSAURIA Order SQUAMATA Suborder SAURIA=^^ Family EUBLEPHARIDAE"- Genus HOLODACTYLUS Boettger 1893. Holodactylus Boettger, Zool. Anz., 16. p. 113. Type by monotypy: E. africanus Boettger. Holodactylus africanus Boettger 1893. Holodactylus africanus Boettger, Zool. Anz., 16, p. 114: "Abdal- lah," i.e. Abdulla, north of Webi Shebeli, Ethiopia. ly As an accidental visitor : now mounted in the Peace Memorial Museum, Zanzibar. 20 Fortunately the onl.v Uganda specimen was photographed prior to its escape. The typical form (O. t. tetraspis Cope) ranges from the western Belgian Congo and Angola, northwest to Senegal. 21 LACERTILIA of Owen (1842), Boulenger (1S85) and others. ^2 Geckos with true eyelids. This family, long merged with the GEKKONIDAE, was revived by Underwood (1954, Proc. Zool. Soc. London, 124. p. 476). LOVERIDQE: E. AFRICAN REPTILES AND AMPHIBIANS 179 1915. Eolodactylus aculeatus Calabresi, Monit. Zool. Ital. (Firenze) 26, p. 238, fig. 2: Somalia. Range. Ethiopia and British Somaliland, south through So- malia to Kenya Colony. Family GEKKONIDAE Genus STENODACTYLUS Fitzinger 1826. Stenodactyliis Fitzinger, Xeue Class. Rept, pp. 13, 47. Type by tautonomy : S. elegans Fitzinger, new name for Asmlahotes stenodactyliis (sic) Lichtenstein. 1842. Tolarcnta Gray, Zool. Misc., p. 58. Type by monotypy: T. wilkin- sonii Gray = Ascalabotes sthenodactylus Lichtenstein. 1874. Ceramodactylus Blanford, Ami. Mag. Nat. Hist., (4) 13, p. 454. Type by monotypy: C. doriae Blanford. Stenodactylus sthenodactylus sthenodactylus (Lichtenstein) 1823. Asmlahotes sthenodactylus Lichtenstein, Verz. Doiibl. Mus, Zool. Berlin, p. 102: Egypt and Nubia, i.e. Sudan. 1823. Agame ponctue Geoffroy Saint-Hilaire, Rept., in Descr. Egypte, 1, p. 129, pi. V, fig. 2: Egypt. 1826. Stenodactylus elegans Fitzinger (new name for Ascalabotes steno- dactylus (sic) Lichtenstein), Neue Class. Rept., p. 47. 1827. Trapelus Savignyi Audouin, Hist. Nat. Rept., in Descr. Egypte, 1, p. 167, and Suppl. Rept., pi. i, figs. 31—33; Egj-pt. 1829. Stenodactylus guttatus Cuvier, Regne Animal, ed. 2, 2, p. 58, pi. iv, fig. 2 : Egypt. 1842. Tolarenta Wilkinsonii Gray, Zool. Misc., p. 59: Egypt. 1943. Stenodactylus stenodactylus savattari Seortecci, Miss. Biol. Sagan- Omo, Zool., 1. pp. 268, 294: Elolo, Olo River, Lake Rudolf, EtMopia.2^ Range. Syria ; Arabia ; Egypt, west to Tunisia, south to Lake Rudolf, Kenya Colony. 23 Seortecci (loc. cit.), while stating that his two specimens from Lake Rudolf agree m most respects with S. s. sthenodactylus, claims that they may be dif- ferentiated by two characters. His North African material had 9-10 upper labials, his Ethiopian geckos 13-14. The counts for twelve North African 8. s. stheno- dactylus examined for my 1947 revision of the GEKKONIDAE, ranged from 9-15 (p. 46). A new character used by Seortecci is the number of granules around midbody. In his North African specimens these were 85 and 96 respectively; in his Ethiopian geckos 105 and 106. Owing to the small and irregular size and dispo- sition of these granules it is difficult to obtain the same number in two counts, but our material gives the following : $ (M.C.Z. 31395) Port Sudan, Sudan 91 or 95 9 (M.C.Z. 51659) Ein el Weibeh, Israel 98 or 100 9 (M.C.Z. 21909) Sfax, Tunisia 102 or 105 $ (M.C.Z. 5234) "Dalmatia" 119 (A.L. and B.S.) 180 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Genus CNEMASPIS Straucli 1887. Cnemaspis Strauch, Mem. Acad. Imp. Sci. St.-Petersbourg, (7) 35, No. 2, p. 4. Type by nionotypy: C. hoiilengeri Strauch. 1921. ParagonSmith) Cape Dwarf-Gecko 1849. Ilcmidactylus capensis A. Smith, Illus. Zool. S. Africa, Rept., pi. Ixxv, fig. 3 : Kaffirland and districts north of Cape Province. 18G4. Lygodactylus strigatu-s Gray, Proc. Zool. Soc. London, p. 59: "Southeastern Africa." i.e. Zambezi region. 1932. Lygodactylus bradfieldi Hewitt, February, Ann. Natal Mus., 7, p. 126, pi. vi, fig. 10: Quickborn Farm, near Okahandja, South- west Africa. 1932. Lygodactylus capensis ngarnieyisis FitzSimons, October, Ann. Transvaal Mus., 15, p. 35: Mabeleapudi, Ngamiland, Beehuana- land Protectorate. Range. Kenya Colony and Tanganyika Territory, south to Natal, west through northern Cape Province to Southwest Africa and Angola. Lygodactylus grotei grotei Sternfeld 1911. Lygodactylus Grotei Sternfeld, Sitzb. Ges. Naturf. Freunde Ber- lin, p. 245 : Mikindani, etc., Tanganyika Territory. 1920. Lygodactylus capensis mossamhica Loveridge, Proc. Zool. Soc. London, p. 135: Lumbo, Mozambique. Range. Tanganyika Territory and Mafia island, south to Mo- zambique. Lygodactylus grotei pakenhami Loveridge Pemba Island Dwarf-Gecko 1 88 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1941. Lygodactylus grotei paTcenhami Loveridge, Proe. Biol. Soc. Wash- ington, 54. p. 176: Wete, Pemba Island, Range. Pemba Island. Lygodactylus angularis angularis Giinther 1893. Lygodactylus angularis Giinther, Proe. Zool. Soc. London, 1892, p. 555, pi. xxxiii, figs. 1-3: Shire Highlands, Nyasaland. Range. Tanganyika Territory and Nyasaland, west to North- ern Rhodesia. Lygodactylus picturatus gutturalis (Bocage) 1873. Hemidactylns gutturalis Bocage, Jom. Sci. Lisboa, 4. p. 211: Bissau, Portuguese Guinea. Range. Uganda and Tanganyika Territory (Ujiji only), west to French Congo. Portuguese Guinea. Lygodactylus picturatus keniensis Parker ]93(). Lygodactylus picturatiLS heniensis Parker, Ann. Mag. Nat. Hist., (10) 18, p. 602: Lodwar, Lake Rudolf, Kenya Colony. Range. Kenya Colony (north of the Uaso Nyiro). Lygodactylus picturatus uikerewensis Loveridge 1935. Lygodactylus picturatus uTcerewensis Loveridge, Proe. Biol. Soc. Washington, 48. p. 199: Ukerewe Island, Lake Victoria, Tan- ganyika Territory. Range. Kenya Colony and Tanganyika Territory (in vicinity of Lake Victoria). Lygodactylus picturatus mombasicus Loveridge 1935. Lygodactylus picturatus mombasicus Loveridge, Proe. Biol. Soc. Washington, 48. p. 198: Kilindini, Mombasa Island, Kenya Colony. Range. Coastal Kenya Colony (meets with typical form at both Mombasa and Tanga), south to Tanga, Tanganyika Terri- tory. Lygodactylus picturatus picturatus (Peters) Yellow-headed Dwarf-Gecko 1868. Hemidactylus variegatus Peters (not of Dumeril & Bibron:1836), Monatsb. Akad. Wiss. Berlin, p. 449: "Zanzibar Coast," i.e. Tanganyika Territory. 1870. Hemidactylus picturatus Peters (new name for variegatus Peters: preoccupied), Monatsb. Akad. Wiss. Berlin, p. 115. 1896. Lygodactylus picturatus vars. griseus, septcmlineatus and quin- LOVEBIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 189 quelineatus Tornier (nomina nuda), Die Kriechthiere Deutsch- Ost-Afrikas, p. 15: No tj'pes or type localities designated. 1928. Lygodactylus Duvnni Loveridge, Proc. U. S. Nat. Mus., 72, Art. 24, p. 1, pi. i: Saranda, Ugogo, Tanganyika Territory. Range. (Somalia ? or) Kenya Colony (coastal belt from Mombasa) south to Tanganyika Territory and Mozambique (chiefly coast, rivers and railways) ; Zanzibar and Mafia Islands. Northern Rhodesia and neighboring parts of the Belgian Congo (its present distribution has been complicated by transportation as eggs or adults through human agency). Lygodactylus picturatus williamsi Loveridge Turquoise-Blue Dwarf-Gecko 1952. Lygodactylus picturatus tvilliams-i-^ Loveridge, Jouru. E. Africa Nat. Hist. Soc, vol. "xx," p. "446," = 21. p. 39: Kimboza Forest, 1000 feet. Eastern Province, Tanganyika Territory. Range. Tanganyika Territory (known only from the holo- type). Genus PHYLLODACTYLUS Gray=^ 1828. Phyllodactylus Gray, Spicil. ZooL, p. 3. Type by monotypy: P. pv.lcher Gray. 1843. Eideptes Fitzinger, Syst. Rept., pp. 18, 95. Type by original designation : Phyllodactylus europaeus Gene. 1843. Discodactylus Fitzinger, Syst. Rept., pp. 18, 95. Type by original designation: Pliyllodactyhts tuberculosvs Wiegmaiin. 1845. Gerrliopygus Gray, Cat. Lizards Brit. Mus., p. 150. Type by tan- tonomy: Diplodactylus gerrhopygus Wiegmaun. 1879. Paroedura Giinther, Ann. Mag. Nat. Hist., (5) 3. p. 218. Type by monotypy: P. sanctijohamiis Giinther. Phyllodactylus wolterstorffi (Tornier) ^^ 1900. Diplodactylus wolterstor-jfl Tornier, Zool. Jalirb., Syst., 13, p. 584, fig. A: Tanga, Tanganyika Territory. 26 Altered by the then editor to Williamsi in some printings, to villiamsi iu others ; both volume number and pagination were in error, so corrected in a subse- quent issue by the new editor. 2" Here used in the new sense proposed by Garth Underwood (1954, Proc. Zool. Soc. London, 124, p. 472) involving the transfer to Phyllodactylus of those species with a Gekko-type pupil formerly referred to Diplodactylus, a genus now restricted to Australia. 28 Through the kindness of Dr. Heinz Wermuth, I recently had the opportunity of examining the type and a paratype (now M.C.Z. 54700). They reveal some minor defects in Tornier's description, but though there is little to differentiate them from P. inexpectatus (Stejneger )of the Seychelles, direct comparison with a specimen of the latter (M.C.Z. 49198) shows that the two are distinct. 1!*0 BULLETIN: MUSEUM OP COMPARATIVE ZOOLOGY Range. Tanganyika Territory (known only from tlie four types). Genus EBENAVIA Boettger 1S78. Ebenavia Boettger, Abliaud. Senekenberg. Naturf. Ges., II, p. i]7t>. Type by monotypy: E. inunguis Boettger. Ebenavia sp.'^ Range. Pemba Island. (lenus PHELSUMA Gray 182rj. riiclsiiina Gray, Ann. Philos., (2) 10. p. 199. Type by monotypy: •'P. crepidianus," i.e. GecJco cepedianus Merrem. 1830. Anoplopus Wagler, Nat. Syst. Amphib., p. 142. Type by designa- tion of Fitzinger :1843 : GecJ:o ccpedianiiJi Merrem. Phelsuma dubia dubia (Boettger) 1881. Padiydartylu,s duhius Boettger, Zool. Anz., 4, p. 4G : Madagascar. Range. Coastal Tanganyika Territory; Zanzibar Island ; Com- oro Islands; northwest Madagascar. Phelsuma madagascariensis parkeri Loveridge 1911. Phelsuma madagascariensis parheri Loveridge, Proe. Biol. Sdc Washington, 54. p. 17.5: Kinowe, Pemlia Island. Range. Pemba Island. Genus HOMOPHOLIS Boulenger 188.^). Eoiiiopholis Boulenger, Cat. Lizards Brit. Mus., 1, p. 191. Type l)y monotypy: Gecko walhci-gi (sic) A. Smith. -9 Admission of this genus to the East African list rests solely on some geckos iiewlv hatched from esKs, measuring 5x6 mm., found at Cholvocho and Kinanga- juu, Pemba Island, by R. H. Pakeuham (1947, Ann. Mag. Nat. Hist., (11) 14, V- 135). The assignment to Ehcnaiia is unquestionably correct; it was made by n. W. Parker. At my request he very kindly re-e.xamined them under eery high magnification supijlying me witli a sketch and the substance of the following : Scales of head nuilticarinate, uniform, their size corresponding to the larger liody granules : the latter have some indications of a keel, all the dorsal granules appearing to be sul)iml)ricate with a surface sculpturing of minute papillae ; among tliem are irregular rows of large rounded tubercles ; no enlarged chin sliields ; no claws. Color above, pale straw, with or without a few, very faint, longitudinal lines that converge upon the iiase of tail : the dorsal color sharply demarcated from that on the sides of head, neck and flanks, this is dark sepia passing gradually into the light sepia of the undersurface : tail above dis- plays some paired spots anteriorly, while posteriorly it is banded with darker and ligliter. This is, of course, hatchling coloration and probably differs from that of the adult. Capture of some adults on I'emba is necessary to settle the point as to whether the species is distinct, or whether inunguis Boettger has been introduced fruni Xosy IW-. Rladagascar. and become established on Pemba. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 191 1890. Platjpholis Boulenger, Proe. Zool. Soc. London, p. 80, pi. vii, fig. 2. Type by monotypy: P. fasciata Boulenger. Homopholis fasciata lascicrta (Boulenger) 1890. Platypholis fasciata Boulenger, Proc. Zool. Soc. London, pp. 77. 81, pi. viii, fig. 2: Mombasa, Kenya Colony. Range. Nortlieni Kenya Colony, south to Tanganyika Terri- tory. Genus PACHYDACTYLUS Wiegmann 1834. Pachy dactyl us Wiegmann, Herp. Mexicana, p. 19. Type by mono- typy: P. hcrgii Wiegmann = Laxierta geitje Sparrman. 1843. Colobopus Fitziuger, Syst. Eept., p. 19. Type by original designa- tion: Anoplopus inunguis Wagler := Gecko inimgnis Cuvier, 1817 = Lacerta geitje Sparrman, 1778. 1845. Cantinia Gray, Cat. Lizards Brit. Mus., pp. 167, 168. Type by monotypy: Pachydactylus elegans Gray = Tarentola capensis A. Smith. 1864. Eomodactylus Gray (not Fitzinger :1843), Proc. Zool. Soc. London, p. 59. Type by monotypy: H. turneri Gray. 1894. Elasmodactylus Boulenger, Proe. Zool. Soc. London, p. 724. Tj'pe by monotypy: E. tubereulosiis Boulenger. Pachydactylus bibronii turneri (Gray) 1864. Eomodactylus turneri Gray, Proc. Zool. Soc. Loudon, p. 59, pi. ix, fig. 2 : Tete, Mozambique. 1910. Pachydactylus bibronii var. st.ellatus Werner, Denks. Med.-Nat. Ges. Jena, 16, p. 309: Great Namaqualand, Southwest Africa. Range. Belgian Ruanda-Urundi and Tanganyika Territory, south to Mozambique, west through Nyasaland ; the Rhodesias ; Transvaal ; Bechuanaland and Orange Free State to Little Nama- qualand and adjacent Cape Province, north through Southwest Africa to southern Angola. PachydactYlus tetensis Loveridge 19 '12. Pachydactylus tetensis Loveridge, Bull. Mus. Comp. Zool., 110, p. 17.5, pi. V, fig. 3: MAvanza Eocks, Kasumbadedza, near Tete. Mozambique. Range. Near Liwale, Tanganyika Territory, south to Mozam- bique (on south bank of the Zambezi). Pachydactylus tuberculosus (Boulenger) 1894. Elasmodactylus tuberculosus Boulenger, Proc. Zool. Soc. London. 192 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY pp. 723, 727, pi. xlvii, fig. 2: Lower Congo. 1896. Pachydactylus boulengeri Tornier, Die Kriechthiere Deutsch-Ost- Af rikas, p. 26, pi. ii, fig. 1 : Tabora and Kakoma, Tanganyika Territory. 1913. Elasmodactylus triedrus Boulenger, Eevue Zool. Afr., 3, p. 104, figs. — : Kikondja, Katanga, Belgian Congo. Range. Tanga, Tanganyika Territory, southwest to Nyamkolo, Northern Rhodesia ; west to Lower Congo River ; Belgian Congo. Family AGAMIDAE Genus AGAMA Daudin'° 1802. Agama Daudin (part). Hist. Nat. Eept., 3. pp. 333, 356. Type by subsequent designation: A. colonorum Daudin (part) = Lacerta agama Linnaeus. 1817. Trapelus Cuvier, Eegne Animal, 2, p. 35. Type by monotypy: Agame variahile Geoffrey = Agama m.utabilis Merrem. 1826. Tapaya Fitzinger, Neue Class. Kept., pp. 17, 49. Type by subse- quent designation : Agama orbicularis Cuvier = Lacerta hispida Linnaeus. 1833. Cyclosaurus Wagler, Isis von Oken, col. 894. Type by original designation : Agama orbicularis Daudin = Lacerta hispida Linnaeus. 1843. Phrynopsis Fitzinger, Syst. Kept., pp. 17, 79. Type by original designation: Agama atra Daudin. 1843. Podorrhoa Fitzinger, Syst. Kept., pp. 18, 80. Type by original designation: Agama colonorum Daudin = Lacerta agama Lin- naeus. 1843. Pseudotrapelus Fitzinger, Syst. Kept., pp. 18, 81. Type by original designation: AgamM sinaita Heyden. 1843. Planodes Fitzinger, Syst. Kept., pp. 18, 81. Type by original designation: Agama agilis Olivier. 1843. Trapeloidis Fitzinger, Syst. Kept., pp. 18, 81. Type by original designation: Lacerta sanguinolenta Pallas. 1843. Psammorrhoa Fitzmger, Syst. Kept., pp. 18, 81. Type by original designation: Agama acvleata Merrem. 1843. Eremioplanis Fitzinger, Syst. Kept., pp. 18, 82. Type by original 30 Elimination of Stellio Laurenti (part), 1768, Syn. Kept., p. 56, was achieved by subsequent designation of its type as iS. saxatilis I>aurenti by Stejneser, in M. A. Smith, 1933, Journ. Bombay Nat. Hist. Soc. for 1932, 35, p. 6l9. S. saxatilis, being based on Seba, 1735, Locup. Rerum Nat. Thesaurus, 2, pi. Ixxix, fig. 4, is declared unrecognizal)le. The genus SteUio has been used in other senses by Schneider (1792) ; Latreille (1802) ; Wagler (1830) and others. Intentionally omitted from the synonymy are obvious misspellings such as Trnpetua Oken (1817) ; Tapelus Gray (1825) : Cyoclosaurus Wagler (1833). IX)VERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 193 designation: Trapelus aegypiius Cuvier = Agavia mutabilis Merrem. 1843. Acanihocercus Fitzinger, Syst. Rept., pp. 18, 84. Type by original designation: Stellio cyanogaster Riippell. 184.3. Laudalcia Gray, Cat. Lizards Brit. Mns., p. 254. Type by mono typy: Agama tuberculaia Gray. 1845. Isodactylus Gray, Cat. Lizards Brit. Mus., p. 2J9. Type by niono- typy: Agama sinaita Heyden. 1854. Plocederma Blyth, Journ. Asiatic Soc. Bengal, 23. p. 738. Type by monotypy: Laudalcia {Plocederma) melanura Blyth. 18r)6. Brachysaura Blyth, Journ. Asiatic Soc. Bengal, 25. p. 448. Type by monotypy: B. ornata Blyth = Agama minor Hardwicke & Gray. 1860. Barycephalus Giinther, Proc. Zool. Soc. London, p. 150. Type by monotypy: B. syTcesii Giinther = Agama Uibercidata Gray. 1895. Xenagama Boulenger, Proc. Zool. Soc. London, p. 534. Type by original designation: Uromastix batilliferus Vaillant. Agama ruppelli occidentalis Parker ^^ Western Arboreal-Agama 1932. Agama rueppelli occidentalis Parker, Journ. Linn. Soc. London, Zool., 38. p. 225 : near mouth of Kaliokwell Eiver, Lake Eudolf, Kenya Colony. Range. Southern Ethiopia to northern Kenya Colony. Agama ruppelli septentrionalis Parker Southern Arboreal-Agama 1932. Agama rueppelli septentrion-alis Parker, Journ. Linn. Soc. London. Zool., 38, p. 225: Mount Nyero (as Njiro) ; Madago's village; \'oi and Mbunyi, Kenj-a Colony. Range. Central and southern Kenya Colony. Agama hispida armata Peters^- Peters' Spiny Agama 1854. Agama armata Peters, Monatsb. Akad. Wiss. Berlin, p. 616: Rios de Sena ; Tete, Mozambique. Range. Tanganyika Territory, south through Mozambique; N'yasaland and the Rhodesias to Natal. Agama mossambica mossambica Peters ^Mozambique Agama 1854. Agama mossambica Peters, Monatsb. Akad. Wiss. Berlin, p. 616: Coast of Mozambique. 31 A. vaillanti Boulenger is a synonym of A. r. ruppelli Vaillant of Ethiopia and the Somalilands. Keuya records of vnillaiiti should be referred to one or other of Parker's races. 32 Early East African records of hinpidn or hispida distantl Boulenger slioulil be referred to this race, the other two races being South African. 194 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1874. Agama carmiventris Peters, Monatsb. Akad. Wiss. Berlin, p. 150: Zanzibar Coast; i.e. Tanganyika Territory. Range. Tanganyika Territory, south through northern Mo- zambique and Nyasaland to Northern Rhodesia."^ Agama mossambica montana Barbour & Loveridge^^ Montane lloek-Agama 1928. Agama mossambica montana Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 147: below Bagilo, Uluguru Mountains, Tanganyika Territory. Range. Usambara and Uluguru Mountains, Tanganyika Ter- ritory . Agama agama agama (Linnaeus) Oonnnon Rock-Agama 17.18. Laccrta agama Linnaeus (part), Syst. Nat., ed. 10, 1, p. 207: "America" (error). 1802. Agama colonorum Daudin, Hist. Nat. Kept., 3. p. 3o6: " Amerique meridionale, " etc. (error). 1831. Agama Occipitalis Gray, in Griffith, Animal Kingdom, 9. Synopsis, p. 56 : Africa. 1877. Agama colonorum var. congica Peters, Monatsb. Akad. Wiss. Ber- lin, p. 612 : Chinchoxo, Cabinda. 1877. Agama picticaiula Peters (? part^^) Monatsb. Akad. Wiss. Berlin, p. 612: "Ada Foah, " i.e. ? Adafer, Mauretania, French West Africa; Accra, Gold Coast; also Cameroon. Range. Southern Sudan and Uganda, west to Nigeria, south lo Angola. Agama agama lionotus Boulenger Kenya Rock-Agama 1896. Agama lionotus Boulenger, Proc. Zool. Soc. London, p. 214, pi. viii : southeast of Lake Eudolf , Kenya Colony. Range. Eastern Uganda (Suk), southeast to Voi, Kenya Col- ony. Agama agama usombarae Barbour & Loveridge Usambara Rock-Agama 1928. Agama colonorum lu^ambarae Barbour & Loveridge, Mem. Mus. 33 The two young aganias from Athi Plains, Konya Colony, referred to mossaiit- bica by Moccinard (l'J02, Bull. Mus. Hist. Nat. (Paris), 8, p. 405) are possibly A. a. lionotus, a species that occurs there (M.C.Z. 44260-1). 34 I have long thought that montana is a full species but, pending a thorough revision of the entire genus, consider it is best left as described. A. »i. mossam- hica occurs on the lower slopes of the Uluguru range. 35 While Cameroon specimens are unquestionably A. a. agama, our Gold Coasi material seems nearer to the western race A. a. africana Hallowell. The question cannot be settled until a series from Ada Foah has been studied, being an area where the races meet. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 195 Conip. Zool, 50, p. 150, pi. ii, fig. 1: Soni, Usambara Mountains, Tanganyika Territory. Range. T^sambara Mountains, Tanganyika Territory. Agama agama elgonis Lonnberg Elgon Roek-Agama 1922. Agama elgonis Lonnberg, Arkiv. Zool., 14, No. 12, p. 2: Mount Elgon, Kenya Colony. 1932. Agama agama turuensis Loveridge, Bull. Mus. Conip. Zool., 72, p. 376: Unyanganyi, Turu, Tanganyika Territory. Range. Uganda and Kenya slopes of Mount Elgon, south to Usandawi in central Tanganyika Territory. Agama agama dodomae Loveridge Dodoma Rock- Agama 1923. Agama lionotus var. dodomae Loveridge, Proc. Zool. Soc. London, p. 944: Dodoma, Ugogo, Tanganyika Territory. Range. Central to southwestern Tanganyika Territory. Agama agama ufipae Loveridge IJfipa Roek-Agama 1932. Agama agama ufipae Loveridge, Bull. Mus. Conip. Zool., 72, p. 377: near Kipili, Ufipa, Tanganyika Territory. Range. Western Tanganyika Territory (Lake Tanganyika). Agama planiceps mwanzae Loveridge Mwanza Roek-Agama 1923. Agama lionotus var. mwanzae Loveridge, Proc. Zool. Soc. London, p. 945 : Slianwa, Mwanza, Tanganyika Territory. Range. Northwestern Tanganyika Territory. Agama planiceps caudospina Meek Elmenteita Roek-Agama 1910. Agamn caudospinn Meek, Field Mus. Nat. Hist., Zool. Series, 7, p. 407: Lake Elmenteita, Kenya Colony. 1935. Agama agama Icaimosae Loveridge, Bull. Mus. Comp. Zool., 79. p. 10: near Kaimosi, Kakamega, Kenya Colony. Range. Kenya Colony. Agama annectens Blanford Eritrean Roek-Agama 1870. Agama annectens Blanford, Zool. Abyssinia, p. 446, fig. — : "Soo- roo ' ' i.e. Suru Pass, Eritrea. Range. Eritrea, Ethiopia and the Somalilands, south to north- ern Kenya Colony. Agama cyanogaster (Riippell) Black-necked Arboreal-Agama 1835. Stellio cyanogaster Riippell, Neue Wirbelth. Fauna Abyssinica, Amphib., p. 10, pi. v: Massaua, Eritrea. 1849. Agama atricollis A. Smith, Illus. Zool. S. Africa, Eept., App., p. 14: Xatal, South Africa. 196 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1851. Stellio Capensis A. Dumeril, Cat. Method. Coll. Kept. Mus. Paria, p. 106: "Cape of Good Hope," i.e. South Africa. 1866. Stellio nigricollis Bocage, Jorn. Sci. Lisboa, 1, p. 43: Duque de Braganca, Huila Plateau, Mossamedes, Angola. 1894. Agama gregorii Giinther, Proc. Zool. Soe. London, p. 86: Mkon- umbi, near Lamu, Kenya Colony. Range. Eritrea, south through Ethiopia; Uganda; Kenya Colony and Tanganyiiva Territory (apparently absent from the coastal islands of Pemba, Zanzibar and Mafia ! ) to Natal, north- west through Bechuanaland to Ovamboland, Southwest Africa; Angola and Belgian Congo. Family CHAMAELEONIDAE^'^ Genus CHAMAELEO Laurenti 1763. CJiamaeleon Gronovius,^'^ Zoophyl. Gronov. . . ., p. 12. 1768. Chamaeleo Laurenti, Syn. Rept., p. 45. Type by subsequent desig- nation: C. parisiensiiim Laurenti = Lacerta chamaeleon Lin- naeus. 1839. Diceros Swainson, Nat. Hist. Fishes, Amph. Rept., 2, pp. 347, 369. TjT)e by monotypy: D. bifurcatus Swainson. 1839. Trioceros Swainson, Nat. Hist. Fishes, Amph. Rept., 2, pp. 347, 369. Type by monotypy : T. grayii Swainson. 1843. Trioeras Fitzinger, Syst. Rept., pp. 15, 42. Type by original designation : Chamaeleo oweni Gray. 1843. Furcifer Fitzinger, Syst. Rept., pp. 15, 42. Type by original designation: Chamaeleon Mjidus Brongniart. 1843. Bradypodion Fitzinger, Syst. Rept., pp. 15, 43. Type by original designation: Chamaeleon pumilus Latreille = Lacerta pumUa Gmelin. 1846. Bradypodium Agassiz, Nomen. ZooL, Index Univers., p. 52. Emen- dation for Bradypodion Fitzinger. 186.3. Apola Gray, Proc. Zool. Soc. London for 1864, pp. 467, 473. Type by monotypy : Chamaeleo lateralis Gray. 36 CHAMAELEONTIDAE, derived from CHAMAELEONTES Fitzinger (1843: 41) is rejected. This family was for long referred to a separate suborder (RHIPTOGLOSSA) on account of the peculiar projectile tongue possessed b.v all its members. In recent years, however, this separation has been considered of less significance than the obvious relationship to the AGAMIDAE, from which stock chameleons appear to have been derived. 37 A worlj speciflcall.v rejected from tasonomic purposes in rule 89 of the Internat. Comm. Zool. Nomencl. — cf. 1926, Proc. Biol. Soc. Washington, 39, p. 10;i. Omitted are the numerous variations such as: Chameleo (Bonnaterre ; 1789); Cameleo (Bosc:lS03); Camaeleo (Dumeril :1806) ; Chameleon (Fleming: 1822); Chaelio (A. Smith :1831); Chamaeleon (Fitzinger :1S43) ; Camaehon (DumSrll & Bibron :1863) ; some intentional emendations, others due to care lessuess. LOVERIDQE: E. AFRICAN REPTILES AND AMPHIBIANS 197 186.5. Pterosa^lrus Gray, Proc. Zool. Soc. London for 1864, pp. 467, 473. Type by monotypy: Chamaeleo cristatus Stutchbury. 1865. Mierosaura^'^ Gray, Proc. Zool. Soc. London for 1864, pp. 467, 473. Type by monotypy: M. melanocephala Gray. 1865. Phitvianola Gray, Proc. Zool. Soc. London for 18G4, pp. 467, 474. Type by monotj-py: Chamaeleo namaquensis A. Smith. 1865. Lophosaiira Gray, Proc. Zool. Soc. Loudon for 1864, pp. 468, 474. Type by restriction: Lacerta pumila Gmelin. 1865. Calumma Gray, Proc. Zool. Soc. London for 1864, pp. 468, 476. Type by monotypy: Cliamaeleo cucullatus Gray. 1865. Crassonota Gray, Proc. Zool. Soc. London for 1864, pp. 468, 477. Type by monotypy: Chamaeleo nasntus Dumeril & Bibron. 1865. Ensirostris Gray, Proc. Zool. Soc. London for 1864, pp. 468, 478. Type by monotypy: E. melleri Gray. 1865. Sauroceras Gray, Proc. Zool. Soc. London for 1864, pp. 468, 478. Type by monotypy: Chamaeleo rhinoceratus Gray. 1865. Dicranosanra Gray, Proc, Zool. Soc. London for 1864, pp. 468, 478. Type by restriction : Cliamaeleo hif circus Gray = C. bifidtis Brongniart. 1865. Cyneosaura Gray, Proc. Zool. Soc. London for 1864, pp. 468, 479. Type by monotypy: Chamaeleo pardalis Cuvier. 1865. Calyptrosaura Gray, Proc. Zool. Soc. London for 1864, p. 468. Type by present designation : Chamaeleo calyptratus A. Dumeril. 1865. Erizia Gray, Proc. Zool. Soc. London for 1864, p. 471. Type by present designation: Chamaeleo senegalensis Daudin. 1865. Bilepis Gray, Proc. Zool, Soc, London for 1864, p. 472, Type by tautonomy: Chamaeleo dilepis Leach. 1863. Archaius Gray, Proc. Zool. Soc. London for 1864, p. 475. Type by present designation: Chamaeleo tigris Kuhl. 1956. Bicuspis Loveridge, Breviora (Cambridge, Mass.), No. 59, p. 2. Type by original designation: Ehampholeon marshalli Bou- lenger.^® Chamaeleo senegalensis senegalensis Daudin Senegal Chameleon 1802, Chamaeleo senegalensis Daudin, Hist. Nat. Eept., 4, p. 203 : Eegion watered by the Senegal and Niger Rivers; Gambia and Guinea. 3& Microsanra. Under this name FitzSimons (1943: Transvaal Mus. Mem., 1, p. 158), following Power (1932: Proc. Zool. Soc. London, p. 209) who. however, used Lophosaura. would grant generic status to a group of South African chameleons of which the best known is pumilua. If this course is followed, then Bradypodion {Bradypodium of Gray) Fitzinger takes precedence over Micro- saura, for Fitzinger definitely states : "Typus. Cham, pumilua Latr." Tentatively I prefer to regard the group as subgeneric, but those who differ should removf these three genera from the synonymy of Chamaeleo. 39 A prehensile-tailed species, but with the bicuspid claws of a Rhampholeon; occupying an intermediate position between the two genera, hence Bicuspis.. 198 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1863. Chamaeleo laevigatus Gray, Proc. Zool. Soc. London, p. 95: Five hundred miles south of Khartoum, Sudan. 1864. Chamacleon scnegalensis var. Vciocephahis Gray, Proc. Zool. Soc. London, p. 471 : Type in British Museum. No locality. 1887. Chamacleon sphaeropholis Reiclienow, Zool. Anz., 10, p. 370: Kagera, west of Lake Victoria, Tanganyika Territory. 1951. Chamacleon scnegalensis v. Tihatiensis Monard, Mem. Inst. Franc. Afrique Noire (Sci. Nat.), No. 1, p. 141: Tibati, French Cameroon. Range. Sudan, east to Eritrea, south through Ethiopia to Uganda ; Kenya Colony and western Tanganyika Territory. Re- corded from Angola and the Belgian Congo, northAvest to Sierra Leone (M.C.Z. material) and Senegal. Chamaeleo senegalensis anchietae Boeage*" Angola Chameleon 1872. Chamneleo Anchietae Boeage, Jom. Sci. Lisboa, 4, p. 72, fig. — : Huila, Mossamedes, Angola. 1950. Chamaeleo {Apola) vincTcci Laurent, Eevue Zool. Bot. Afr., 43, p. 349: Swampy grassland sources of the Lofoi Eiver at 1750 metres, near Kundelungu Plateau Research Station, Katanga. Belgian Congo. 1952. IChamaeleo anchietae mertensi Laurent, Revue Zool. Bot. Afr. 46, p. 18: Muneshele, 1400 metres, Fizi Territory, Kivu, Belgian Ruanda-Urundi. 1952. '^.Chamaeleo anehietae marmiguensis Laurent, Eevue Zool. Bot. Afr., 46, p. 19 : Kipiri, 2000 metres, Marungu Plateau, Baudouinville Territory, Tanganyika Province, Belgian Congo. Range. Highlands of southern Tanganyika Territory, west through highlands of eastern and southern Belgian Congo to Angola. Chamaeleo gracilis gracilis Hallowell Graceful Chameleon 1842. Chamaeleo gracilis Hallowell, Journ. Acad. Nat. Sci. Philadelphia, 8, p. 324, pi. xviii: Liberia. 1856. CJiamaeleo granulosus Hallowell, Proc. Acad. Nat. Sci. Philadel- phia, p. 147 : West Africa. 1856. Chamaeleo Burchelli Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 147 : Fernando Po. •10 1 have not seen mertensi and marungensia which were based on scanty material (a ^ , $ and juvenile of the former, two $ $ of the latter), but I have compared a $ paratype of vinckei (M.C.Z. 53262) with a Tanganyika 9 (M.C.Z. 31186) and failed to find differences that would justify its description. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 199 1885. Chamaeleo (Chaviacleo) Sirnoni Bocttger, Ber. Offenbach Ver. Naturk., p. 175, footnote: Mountains of Ashanti, Gold Coast. Range. French Somaliland, south through Uganda and Kenya Colony to northern Tanganyika Territory (Longido and Mern Mountains), west through tlie Belgian Congo (? and Angola) where it meets with the race eUennei Schmidt (characterized by spurless S S ) found around Banana; northwest (though not definitely reported from the French Congo; Nigeria and Da- liomey) to Senegal. Chamaeleo dilepis roperi Boulenger*^ , Spurle.ss Flap-necked Chameleon 1890. Chamaeleon roperi Boulenger, Proc. Zool. Soc. London, p. 85, pi. viii, fig. 4: Kilifi, north of Mombasa, Kenya Colony. Range. Eastern Kenya Colony (Meru to Tana River) and northeast Tanganyika Territory (around Kilimanjaro Mtn. only ) . Chamaeleo dilepis quilensis Bocage*- (doubtfully distinct) 18(3(5. Chamaeleo dilepis var. Quilensis Boeage, Jorn. Sci. Lisboa, 1, p- 59 : Rio Quilo, north of Cabinda, Cabinda. 1887. Chamaeleon parvilohus Boulenger, Cat. Lizards Brit. Mus., 3, p. 449, pi. xxxix, fig. 5: Natal; French Congo; Cameroon. Range, (in part, being based solely on material in the M.C.Z. from ) Kenya Colony ; Tanganyika Territory ; Southern Rho- desia; Transvaal; Zululand; Natal; Southwest Africa; Angola: Belgian Congo ; French Congo ; French Cameroon. Chamaeleo dilepis dilepis Leach*" Spurred F]ai)-necked Chameleon 11 Distinguished only by the ^ ^ being spurless like the $ J ; being identifi- able only on the basis of adult ^ ^ , records require careful screening and the range working out with considerable care. ■12 As this form, whose ^ ^ are usually spurred, is separable from d. dilepis solely on the small size of its occipital flaps, subadult d. dilepis are apt to be reported as d. quilensis with resulting confusion. M.C.Z. material reveals both forms as present in five widely scattered countries. While a good series from a given locality is likely to 1)« readily assignable to one form or the other, solitary specimens are often intermediate in the degree of flap development. That we ai-"e dealing with two sibling species seems improbable. FitzSimons (194.3, Transvaal -Mus. jlem. No. 1, pp. 155-156) resolves that dilemma by according quilensis varietal rank in order to indicate that it is something less than a subspecies. His volume should be consulted for records of both forms south of the Zambezi. 43 Other apparently valid races occurring on the periphery of British East Africa, and approached by individuals within the territories, are : Chamaeleo dilepis petersii Gray, 1864. of Mozambique. Chamaeleo dilepis isabellinus Giinther, 1893 (1892), from the Shire High- lamls of Nyasaland. Chamaeleo dilepis ruspolii Boettger, 1893, from Ogaden, Somalia. 200 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY 1819. Chamaeleo dilepis Leach, in Bowdich, Miss. Ashantee, App., p. 493 : French Congo. 1820. Chamaeleo bilobus Kuhl, Beitr. Zool. Anat., 1, p. 104: French Congo. 1820. Chamaeleo planiceps Merrem, Vers. Syst. Amph., p. 162: Africa. 1864. Clmmaeleon petersii var. MrTcii Gray, Proc. Zool. See. London, p. 470 : East Africa. 1866. Chamaeleo Capellii Bocage, Jorn. Sci. Lisboa, 1, pp. 42, 59 : Ben- guela, Angola. 1903. Chamaeleo anguslicoronatus Barbour, Proc. Biol. Soc. Washington, 16. p. 61 : Zanzibar Island. Range, (in part, being based solely on material in the M.C.Z. from) Kenya Colony; Tanganyika Territory; Pemba and Zanzi- bar Islands ; Nyasaland ; Northern Rhodesia ; Southern Rhodesia ; Cabinda ; Belgian Congo. Chamaeleo bitaeniatus bitaeniatus Fischer" Side-striped Chameleon 1884. Chamaeleo bitaeniatus Fischer, Jahrb. Hamburg, Wiss. Anst., 1. p, 23, pi. ii, figs. 7a-b: Lake Naivasha, Kenya Colony. 1887. Chamaeleo bivittatus (lapsus: nomen nudum) F. Miiller, Yerh. Naturf. Ges. Basel, 8, p. 294: Witu, Kenya Colony. Range. Ethiopia and Somalia, south through Uganda and Kenya Colony to northern and western Tanganyika Territory. Chamaeleo bitaeniatus ellioti Gtinther Montane Side-striped Chameleon 1895. Chamaeleon Ellioti Giinther, Ann. Mag. Nat. Hist., (6) 15, p. 524, pi. xxi, fig. A : Bugoye, east foot of Euwenzori Mountains, Uganda. 1912. Clmmaeleon bitaeniatus graueri Sternfeld, Sitzb. Ges. Naturf. Freunde Berlin, p. 380, pi. xv, figs. 19-21, pi. xvii, fig. 33: Bugoie and Eugege Forests above 2000 metres, Belgian Euanda- Urundi. 1912. Chamaeleon bitaeniatus tornieri Sternfeld, Sitzb. Ges. Naturf. Freunde Berlin, p. 383, pi. xvii, fig. 35 : Lendu Plateau, Ituri District, Belgian Congo. 44 Owing to eesual or individual variation, members of the bitaeniatus group are exceptioually difficult to place. Most large series are likely to contain one or more individuals strikingly different from the rest, and such outstanding speci- mens are apt to receive names. In addition to those listed, or tentatively synonymized. here, two others have recently been described from neighbouring territories. These are C. b. kinetensis K. P. Schmidt (1943) from Mount Kineti at 10,458 feet, Imatong Mountains, southern Sudan, which appears to be inter- mediate between typical bitaeniatus and b. ellioti; also C. b. schoutedeni Laurent (19r>2) from Kabumbe Valley at 2400 metres, Kabobo Mountain, Albertville Ter- ritory, Tanganyika Province, Belgian Congo. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 201 1922. Chamaeleon Bequaerti Witte, Revue Zool. Bot. Afr., 10, p. 69, pi. ii, fig. 1 : Beni, Kivu District, Belgian Congo. Range. Mountains (usually below 9000 feet) of western Uganda ; Belgian Ruanda-Urundi and the adjacent Belgian Congo. Chctmaeleo bitaeniatus rudis Boulenger Ruwenzori Side-striped Chameleon 1906. Chmnacleon rudis Boulenger, Ann. Mag. Nat. Hist., (7) 18, p. 473: Ruwenzori Mountains above 10,000 feet, Uganda. 1983. Chamaeleon Burgeoni Witte, Revue Zool. Bot. Afr., 24, p. 120, fig. — : Mombasa, near Lubero, Kivu District, Belgian Congo. Range. Mountains (usually above 9000 feet) of western Ugan- da ; Belgian Ruanda-Urundi and the adjacent Belgian Congo. Chamaeleo bitaeniatus schubotzi Sternfeld Kenya Side-striped Chameleon 1912. Chamaeleon hitaeniatus schuhotzi Sternfeld, Sitzb. Ges. Naturf. Freunde Berlin, p. 381, pi. xv, fig. 24; pi. xvii, fig. 34: Mount Kenya, 14,000 feet, Kenya Colony. Range. Alpine zones of Mounts Kinangop and Kenya, Kenya Colony ; Mounts Kilimanjaro and Nguru, Tanganyika Territory. Chamaeleo bitaeniatus hohnelii Steindachner Iligli-casqued Chameleon 1891. Chamaeleon Hohnelii Steindachner, Sitzb. Akad. Wiss. Wien., 100, Abt. 1, p. 309, pi. i, figs. 1-la: Laikipia, 6000 feet, Kenya Colony. 1891. Chamaeleon leikipiensis Steindachner, Sitzb. Akad. Wiss. Wien, 100, Abt. 1, p. 311, pi. i, figs. 2-2a: Laikipia, 6000 feet, Kenya Colony. 1912. Chamaeleon bitaeniatus hergeri Sternfeld, Sitzb. Ges. Naturf. Freunde Berlin, pp. 380-381, pi. xiv, fig. 16; pi. xvi, fig. 31: "Sirgoi nordl. von Ravine," i.e. Sergoit, north of Eldama Ravine Station, Kenya Colony (erroneously given as "Sirgoi siidlich von Ravine," in a later paper of 1912). Range. Highlands (usually below 9000 feet) of eastern Uganda and Kenva Colony. Chamaeleo bitaeniatus altaeelgonis Loveridge Alpine High-casqued Chameleon 1935. Chamaeleon hitaeniatus altaeelgonis Loveridge, Bull. Mus. Comp. Zool., 79, p. 15: Kaburomi, 10,500 feet, Mount Elgon, Uganda. 202 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Range. Alpine zoue (above 10,000 feet) of Mount Elgon, Uganda and Kenya Colony. Chamaeleo goetzei goetzei Tornier Tanganyika Goetze Chameleon 1899. Ckamaeleon Goetzei Tornier, Zool. Anz., 22, p. 413, pi. ii, fig. 3: Uhehe, Tanganj-ika Territory. Range. Uzungwe ; Ubena ; Ukinga ; Rungwe and Poroto Moun- tains, Tanganyika Territory. Chamaeleo laterispinis Loveridge Spiny-sided Chameleon 1932. Chnmachon laterispinis Loveridge, Bull. Mus. Comp. Zool., 72, p. 381 : Kigogo, 6000 feet, Uzungwe Mountains, Tanganyika Territory. Range. Uzungwe Mountains, Tanganyika Territory. Chamaeleo incornutus Loveridge Ukinga Hornless Chameleon 1932. Chamaclcon incorniituH Loveridge, Bull. Mus. Conip. Zool., 72, p. 380: Madehani, 7000 feet, Ukinga Mountains, Tanganyika Terri- tory. Range. TTkinga ; Rungwe and Poroto Mountains, Tanganyika Territory. Chamaeleo tempeli Tornier Tubercle-nosed Chameleon 1899. Ckamaeleon Tempeli Tornier, Zool. Anz., 22, p. 411, pi. ii, figs. 2, 4: Uzungwe Mountains, Uhehe, Tanganyika Territory. 1900. Ckamaeleon tempeli var. ivolffi Tornier, Zool. Jahrb. Syst. 13, p. 613, fig. G (captions reversed and an 'f dropped from first wolffi) : "Tardalla, " i.e. Tandala, Ukinga Mountains, Tan- ganyika Territory. Range. Uzungwe ; Ubena and Ukinga Mountains, Tanganyika Territory. Chamaeleo melleri (Gray) Giant One-horned Chameleon 1864. Ensirostris melleri Gray, Proc. Zool. Soc. London, p. 478, pi. xxxii, fig. 1: "Mountains in the interior of East Africa" (as collected by Dr. Meller of Zomba, restricted by Loveridge (1953) to Zomba Mountain, Nyasaland). Range. Savanna Forests of Tanganyika Territory and Nyasa- land. Chamaeleo spinosus Matschie Rosette-nosed Chameleon 1892. Ckamaeleon spinosus Matschie, Sitzb. Ges. Naturf. Freunde Berlin, p. 105: Derema, Usambara Mountains, Tanganyika Territory. LOVERIDOE: E. AFRICAN REPTILES AND AMPHIBIANS 203 Range. Mrgin Forests of Usambara Mountains, Tanganyika Territory. Chamaeleo tenuis Matschie Single Soft-horned Chameleon 1892. Chamaelcon tenuis Matschie, Sitzb. Ges. Naturf. Freiinde Berlin, p. 106; Dererna, Usambara Mountains, Tanganyika Territory. Range. Virgin Forests of Usambara Mountains, Tanganyika Territory. Chamaeleo xenorhinus Boulenger Single Welded-horn Chameleon 1901. Chamaeleon xenorhinus Boulenger, Proc. Zool. Soe. London, 2. p. 135, pi. xii: Ruweuzori Mountains, 6000 feet, Uganda. 1929. Chamaeleo carpentcri Parker, Ann. Mag. Nat. Hist., (10) 3. p. 280, fig.: Ruwenzori Mountains, 6500 feet, Uganda. Range. Virgin Forests (above 6000 feet) of Ruwenzori Moun- tains. Uganda. Chamaeleo fischeri excubitor Barbour Kenya Hornless Chameleon 1911. Chamaeleo tenuis excubitor Barbour, October 31, Proc. Biol. Soc. Washington, 24, p. 219: Meru District, Mount Kenya, Kenya Colony. 1911. Chamaeleon affinis embucnsis Lonnberg, November 16, Svenska Veteusk.-Akad. Handl., 47, No. 6, p. 19, pi. ii, fig. 3 : Forest on eastern slopes of Mount Kenya, one day's march from Embu Boma, Kenya Colony. Range. Virgin forests of Mount Kenya, Kenya Colony. Chamaeleo fischeri uthmolleri Miiller Hanaug Hornless Chameleon'*^ 1938. Chamaeleo uthmdlleri Miiller, Zool. Anz., 122, p. 20, figs. 1-2: Mount Hanang (Gurui or Guruwe) at 2300 feet, central Tan- ganyika Territory. Range. Virgin forest of Mount Hanang, between Mkalama and Kondoa Irangi, Tanganyika Territory. Chamaeleo fischeri tcrvetanus Steindachner Kilimanjaro Two-horned Chameleon 1891. Chamaeleon tavetanus Steindachner, Anz. Akad. Wiss. Wien, 28. p. 142 : Taveta Forest, Kenya Colony. 1891. Chamaeleon tavetensis Steindachner, Sitzb. Akad. Wiss. Wien, 100. Abt. 1, p. 312, pi. i, figs. 3 3a: Taveta Forest, Kenya Colony. ^5 Assuming that the sexlng of the only known specimen is a ^ as stated. 204 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1891. Chamaeleo abbotti Stejneger, Proc. U. S. Nat. Mus., 14, p. 353, fig. — : Mount Kilimanjaro, 4500 feet, Tanganyika Territory. 1893. Chamaeleo taitensis "Steind." (sic) Stejneger (lapsus for tave- tanus), Proc. U. S. Nat. Mus., 16. p. 724. Range. Virgin forests of Teita Mountains, Kenya Colony, west to Kilimanjaro, Tanganyika Territory. Chamaeleo fischeri multituberculatus Nieden Western Usambara Two-horned Chameleon 1913. Cliamaeleon fischeri multituberculatus Nieden, Sitzb. Ges. Naturf. Freunde Berlin, pp. 244, 248, figs. 27-29, pi. xv, fig. 7; pi. xvi, fig. 11: < ' Phillipshof bei Willielmstal, " i.e. Magamba, near Luslioto, western Usambara Mountains, Tanganyika Territory. 1913. Chamaelcon fischeri werncri Nieden (not of Tornier), Sitzb. Ges. Naturf. Freude Berlin, pp. 243, 248, figs. 19-26, pi. xv, fig. 6; pi. xvi, fig. 10: Ambangula and Mlalo, western Usambara Moun- tains, Tanganyika Territory. Range. Virgin forests of western Usambara Mountains, Tan- ganyika Territory. Chamaeleo fischeri fischeri Reicheuow Eastern Usambara Two-horned Chameleon 1887. Chamaelcon Fischeri Eeichenow, Zool. Anz., 10, p. 371: Nguru Mountains, Tanganyika Territory. 1895. CMnmeleon matschiei Werner, Verb. Zool. Bot. Ges. Wien, 45, p. 192 : Usambara Mountains, Tanganyika Territory. 1902. Cliamaeleon tornieri Werner, Zool. Jahrb. Syst., 15, p. 417, pi. xxiv: "Mozambique" (presumably in error). 1913. Cliamaeleon fischeri vosseleri Nieden, Sitzb. Ges. Naturf. Freunde Berlin, pp. 241, 247, figs. 11-18, pi. xv, fig. 5; pi. xvi, fig. 9: Amaui; Bulwa; Nguelo in Usambara Mountains; Magrotto Mountain ; Tanga, Tanganyika Territory.*^ Range. Virgin forests of eastern Usambara Mountains, south to Nguru Mountain, Tanganyika Territory. Chamaeleo fischeri ulugxiruensis Loveridge Uluguru Two-horned Chameleon 1957. Chamaeleo fischeri uluguruensis Loveridge, Tanganyika Notes and Eecords, No. 43, p. 3: Kingokwa, Uluguru Mountains, Tan- ganyika Territory. Range. Virgin forests of the Uluguru Mountains, Tanganyika Territory. *6 Restricted as the author also lists Ukami, which is in the Uluguru Moun- tains ; and TJsaraiuo, i.e. Dar es Salaam region, in error. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 205 Chomaeleo fiillebomi Tornier Poroto Three-horned Chameleon 1900. Chamaelcon filUeborni Tornier, Zool. Jahrb. Syst., 13, p. (314, fig. H: " Nonde-Nike, " i.e. Konde and Nyika (tribes); also "Ngosi- Oder Peroto-Berges um 2200 m." i..e. Ngosi or Poroto Mountains, Tanganyika Territory. Range. Ngosi Volcano, Poroto Mountains, Ukonde, Tangan- yika Territory. Chcanaeleo werneri Tornier Uzungwe Three-horned Chameleon 1899. Chamncleon Werneri Tornier, Zool. Anz., 22. p. 258, fig. 1: ' ' Maschona-Gebiet, ' ' later corrected to Uzungwe Mountains, Uhehe, Tanganyika Territory. 1932. Chamacleon loerneri dahagae Loveridge, Bull. Mus. Comp. Zool., 72, p. 379: Dabaga, Uzungwe Mountains, Tanganyika Territory'. Range. Uluguru and Uzungwe Mountains, Tanganyika Terri- tory. Chomaeleo johnstoni johnstoni Boulenger Ruwenzori Three-horned Chameleon 1901. Chamaelcon johnstoni Boulenger, Proc. Zool. Soe. London, 2, p. 136, pi. xiii: Ruwenzori Mountains, 6000 feet, Uganda. 1911. Chamaeleo'n graueri Steindachner, Anz. Akad. Wiss. Wien, 48. p. 177: Mountains northwest of Lake Tanganyika, Belgian Congo. 1912. Chamaelean johnstoni afflnis Stemfeld (not Riippell), Wiss. Ergebn. Deut. Zentral-Afrika-Exped. 1907-1908, 4. p. 262, pi. vii, fig. 3 : Virgin forest behind boundary mountains on north- west shore of Lake Tanganyika, Belgian Congo. Range. Virgin forests of Uganda; Belgian Ruanda-Urundi and the adjacent Belgian Congo. Chomaeleo jocksonii Boulenger Kikuyu Three-horned Chameleon 1896. Chamaelcon JaclcsonU Boulenger, Ann. Mag. Nat. Hist., (6) 17, p. 376: "Uganda" (error), later amended to Kikuyu, near Nairobi, Kenya Colony. 1903. Chamaelcon jacTcsoni var. vauerescecae Tornier, Zool. Jahrb. Syst., 19, p. 176: Nairobi, 6000 feet, Kikuyu, Kenya Colony. Range. Highlands of Kenya Colony and Tanganyika Territory. Chomaeleo deremensis Matschie Usambara Three-horned Chameleon 1892. Chamaelcon deremensis Matschie, Sitzb. Ges. Naturf. Freunde Ber- lin, p. 103; Derema, Usambara Mountains, Tanganyika Terri- tory. 206 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY Range. Usambara Mountains, Tanganyika Territory. INTRODUCED Chamaeleo oustaleti Mocquard^' Malagasy Giant Chameleon 1894. Chamaeleon Oustaleti Mocquard, C. R. Soc. Philom. Paris, No. 9, p. 3; and 1895 (1894), Bull. Soc. Philom. Paris, (8), 7. p. 113: Betsileo and Diego Suarez, Madagascar. Range. Madagascar; but introduced into Ngong Forest, near Nairobi. Kenya Colony. Genus BROOKESIA Gray 48 1865. Broohesia Gray, Proc. Zool. Soc. London for 1864, pp. 468, 476. Type by monotypy : Chamaeleo supcrciliaris Kuhl. 1874. Itliam'pholeon Giinther, Proc. Zool. Soc. London, p. 443, fig. — . Type by monotypy: Chamaeleo spectrum Buchholz. 1942. Evoluticauda Angel, Mem. Acad. Malgache, No. 36, pp. 154, 178. Type by subsequent selection: Broohesia nasvs Boulenger.*^ Brookesia kerstenii robecchi (Boulenger) Somalia Pigmy-Chameleon 1892. Ehampholeon robecchii Boulenger, Ann. Mus. Civ. Stor. Nat. Genova for 1891, (2) 12. p. 13, pi. i, fig. 3: "Wuorandi," i.e. Warandi, ca. 5° 42' N., 47° 38' E., Somalia. 1897. Uliampholcon maadcra Meek, Field Mus. Nat. Hist., Zool. Series, 1. p. 183: Mandera, British Somaliland. Range. British Somaliland, south through Somalia to north- ern Kenya Colony. Brookesia kerstenii kerstenii (Peters) Kenya Pigmy-Chameleon 1' On April 11, 19G1, Miss V. Hardy-Mason brought to the Coryndon Memorial Museum a iriaut chameleon which she had captured alive at Ngong in an area undergoing deforestation. Tlie reptile, a male, men.sures 18% inches from end of snout to tip of tail, the latter forming 10 inches of the total. The snout does not terminate in a horn, but from tlie crown of the head between the eyes there rises steeply a helmet- like casque with strongly compressed median ridge. From behind this, along the spinal ridge and, to a lesser extent, along the prehensile tail, is a continuou.s serrated ridge formed of spinelike conical scales. A. I.. ■18 Though Rhampholcon is inseparable from Brookesia on the basis of the external characters cited l)y Giinther and Boulenger, it may well rank as a sub- genus on the basis of the anatomical differences described by Parker (1942, Bull. Mus. Comp. Zool., 91, pp. 80-81, hg. 8). In this sense all East African short- tailed chameleons are probably referable to the subgenus Rhampholeon whose range is Tropical Africa and Madagascar. io As Angel designated no type for his genus Evohiticauda, I suggest the iucliulod Hrnokcuiii nasiis be regarded as th(> typt'. LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 207 1868. Chamaeleo Eerstenii Peters, Monatsb. Akad. Wiss. Berlin, p. 449: "Wanga," i.e. Wange, south of Mombasa, Kenya Colony. Range. Northern Kenya Colony, south to Tanganyika Terri- tory. Brookesia nchisiensis Loveridge Pitless Pigmy- Chameleon 1953. Brookesia nchisiensis Loveridge, Bull. Mus. Comp. Zool., 110, p. 190, pi. iii, fig. 1: Nchisi Forest, 5000 feet, Nehisi Mountain, Nyasaland. Range. Virgin forests of southwestern Tanganyika Territory, south to Nchisi Mountain, Nyasaland. Brookesia brevicaudata (Matschie) Bearded Pigmy-Chameleon 1892. Chamaeleon (BrooJcesia) brevicaudatus Matschie (19.viii.92, fide publishers), Sitzb. Ges. Naturf. Freunde Berlin, p. 107: Derema, Usambara Mountains, Tanganyika Territory. 1893. Ehaiiipholeon Boettgeri Pfeffer, Mitt. Naturhist. Mus. Hamburg for 1892, 10. p. 7G, pi. i, figs, (i 7 : Bagamoyo, Tanganyika Terri- tory. Range. Coastal Tanganyika Territory. Brookesia brachyura ionidesi Loveridge Beardless Pigmy-Chameleon 1951. Brookesin ionidesi Loveridge, Bull. Mus. Comp. Zool., 106. p. 179: Kilwa, Southern Province, Tanganyika Territory. Range. Southeastern Tanganyika Territory. Brookesia temporalis (Matschie) Pitted Pigmy- Chameleon 1892. Cliamaeleon (Brookesia) temporalis Matschie, Sitzb. Ges. Naturf. Freunde Berlin, p. 108: Derema, Usambara Mountains, Tan ganyika Territory. Range. Virgin forests of Usambara Mountains, Tanganyika Territory. Brookesia spectrum boulengeri (Steindachner) Spectral Pigmy-Chameleon 1911. Ehampholeon boulengeri Steindachner, Anz. Akad. Wiss. Wien, 48, p. 178: Mountains northwest of Lake Tanganyika, Belgian Congo. Range. \'irgin forests of western Uganda and eastern Belgian Congo. 208 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Family SCINCIDAE Genus MABUYA Fitzinger 1815. Mabuya Eafinesque, Anal. Nat. (Palermo), p. 16. Nomen nudum. 1826. Mabuya Fitzinger (part), Neue Class. Kept., pp. 23, 52. Type by tautonomy: M. dominicensis Fitzinger = Lacertus mubouya Lacepede which Fitzinger lists as a synonym. 1826. Spondylurus Fitzinger, Neue Class. Eept., p. 23 only. Type by monotypy: Seincus sloanii Daudin. 1830. Euprepis Wagler (part), Nat. Syst. Amphib., pp. 132, 161. Type by subsequent designation: Seincus agilis Eaddi = LacerUis mabouya Lacepede. 1839. Herinia Gray, Ann. Nat. Hist., 2, p. 332. Type by monotypy: E. capensis Gray. 1843. Eutropis Fitzinger, Syst. Kept., p. 22. Type by original designa- tion: Euprepes sebae Dumeril & Bibron = Seincus carinatus Schneider (part) + Seineus multifasciatus Kuhl + Euprepes mneularius Blyth. 1843. Trachylepis Fitzinger, Syst. Eept., p. 22. Type by original designation: Euprepes savignyi Dumeril & Bibron = Seincus quinquetaeniatus Lichtenstein. 1843. Oxytropis Fitzinger, Syst. Eept., p. 22. Type by original designa- tion: Euprepes m,erremii Dumeril & Bibron = Tiliqua capensis Gray. 1845. Xystrolepis Tschudi, Fauna Peruana, Herp., p. 44. Type by monotypy: Trachylepis punctata Tschudi (cf. Dunn, 1936, Proc. Acad. Nat. Sci. Philadelphia, for 1935, 87. p. 557). 1845. Copeglossu-m Tschudi, Fauna Peruana, Herp., p. 45. Type by monotypy : C. cinctum Tschudi i= Lacertus mabouya Lac6pede. 1845. Chioninia Gray, Cat. Lizards Brit. Mus., p. 116. Type by present designation: Euprepes delalandii Dumeril & Bibron. 1848. Elabites Gistel, Naturg. Thierr., p. ix. Substitute name for Euprepis Wagler, 1830. 1925. Mabuiopsis Angel, Eept. et Batr. in "Voyage de Ch. Alluaud et E. Jeannel en Afrique Orientale (1911-1912). Eesultats scienti- fiques. Vertebrata." (Paris), 2, p. 21. Type by monotypy Mabuia Jeanneli Angel = M. irregularis Lonnberg. Mabuya quinquetaeniata obsti Werner Tanganyika Five-lined Skink^° 50 Only females and young are five-lined and blue-tailed, the males are strik- ingly different. Ttie number of midbody scale-rows enables this inhabitant of rocky outcrops to be broken up into several races, however the change is so gradual that the limits have to be somewhat arbitrarily delined. North of Uganda one encounters typical M. a. quinquetaeniata (Lichtenstein). South of the Zam- bezi is M. q. margaritifer (Peters), described from Tete, Mozambique (cf. Lover- idge, 1953, Bull. Mus. Comp. Zool., 110. pp. 196-200, 308). LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 20!:) 1913. Mabuia obsti Werner, Jahrb. Hamburg. Wiss. Anst. for 1912, 30, p. 43: Kwa Mtoro, Usandawi, Tanganyika Territory. 1917. Mahuia quinquetaeniata hildebrandti Sternfeld (not Peters: 1874), Wiss. Ergebn. Dent. Zweiter Zentral-Afrika-Exped. 1910-1911. 1. p. 438, pi. x.\iv, fig. 3: "Taita," i.e. Tcita, Kenya Colony. Range. Uganda; Kenya Colony; Tanganyika Territory and Cliapnani Island, Zanzibar (Tornier :1900), south through Nyasa- iand to the Zambezi. Mobuya maculilabris maculilabris (Gray)^^ Speckle-lipped Skink. 1845. Euprepis maculilabris Gray, Cat. Lizards Brit. Mus., p. 114: West Africa. 1866. Euprepes anchietae Bocage, Jorn. Sei. Lisboa, 1, p. 62: "Zaire," i.e. Congo Eiver. 1879. Euprepes noiabilis Peters, Sitzb. Ges. Naturf. Freunde Berlin, p. 36: Chinchoxo, Cabinda; and Pungo Ndongo, Angola. 1912. Ulabum maculilabris major Sternfeld, Wiss. Ergebn. Deut. Zentral- Afrika-Exped. 1907-1908, 4. p. 232: Central Lake Region. 1912. Mabuia vmculilabris var. Tcwidjiviensis Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika-Exped. 1907-1908, 4. p. 233: "Kwidjwi" i.e. Idjwi Island, Lake Kivu, Belgian Congo. 1912. Mabuia macidilabris var. xcauensis Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika-Exped. 1907-1908, 4. p. 233: Wau Island, Lake Kivu, Belgian Congo. 1912. Mabuia, maculilabris var. schubotzi Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika-Exped. 1907-1908, 4, p. 233, pi. vi, fig. 3: Kisenyi, Lake Kivu, Belgian Euanda-Urundi; and Fort Bcni, Belgian Congo. 1912. Mabuia maculilabris var. graueri Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika-Exped. 1907-1908, 4. p. 234, pi. vi, fig. 4: Kisenyi, Lake Kivu, Belgian Ruanda-Urundi ; also the Aruwimi-Ituri region and between the Lualaba River and Lake Tanganyika, Belgian Congo. 1912. Mabuin maculilabris var. rohrbecki Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika-Exped. 1907-1908, A. p. 234: Nguru Mountains: also "Langenburg," i.e. Manda, Lake Nyasa, Tanganyika Territory. 51 This species is mucla in need of critical study on a continental scale. M. in. major, rlesignated as a subspecies by its author to distinguish it from the color varieties he named, appears to be a recognizable race, but when one attempts to define its range diflticulties arise. Seemingly one must accept a great many form?, based on average size and color (subject to fading), or reject them resolutely. In the Usambara Mountains and adjacent coast one meets with a robust ami handsomely coloured form to which, rightly or wrongly, I have been applying the name "comorcntiis Peters." 210 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1912. Mahuia maculilahris var. bergeri Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika-Exped. 1907-1908, 4, p. 235: Dufile, Sudan. Range. Chiefly savanna areas of the southern Sudan and ?? Somalia {fide Calal)resi :1915), south through Uganda; Kenya Colony and Tanganyika Territory (for off-shore islands see races), west through Northern Rhodesia to Angola north and west to French Guinea. Mabuya maculilabris comorensis (Peters) ^' Comoro Speckle-lipped Skink 1854. Eu'prepes comorensis Peters, Monatsber. Akad. Wiss. Berlin, p. 619: "Anjuan, " i.e. Johanna Island, Comoro Islands. 1882. Euprepes angasijanus Peters, Reise nach Mossambique, 3. p. 73 : Angasiya, Grand Comoro Island. 1913. Mahuia comorensis var. infraUnpata Boettger, in Voeltzkow, Reise in Ostafrika, 3, p. 329: Europa Island, Mozambique Channel. Range. Chiefly virgin forest, or recently deforested, areas of coastal Kenya Colony (Mombasa), south through Tanganyika Territory (Usambara and Uluguru Mountains) also Zanzibar, Mafia, Comoro and Europa Islands to Nyasaland (gallery forest along Ruo River, Mlanje Mountain). Possibly Northern Rho- desia (Meruwantipa). Mabuya maculilabris albotaeniata Boettger '^ Pemba Speckle-lipped Skink 1913. MahuUi albotaeniata Boettger, in Voeltzkow, Reise in Ostafrika, 3. p. 350, pi. xxiv, figs. 1-2: Pemba Island. Range. Pemba Island. Mabuya maculilabris boulengeri Sternfeld ^^ Lindi Speckle-lipped Skink 1911. Mahuia hoidengeri Sternfeld, Sitzb. Ges. Naturf. Freunde Berlin, p. 248: Makonde Plateau, Lindi Province, Tanganyika Terri- tory. Range. On tree-trunks in southeastern Tanganyika Territory (Kilwa to Mikindani, inland to Songea), also Nyasaland (Mtim- buka near Fort Johnston) . 52 See preceding footnote. Possibly the synonyms shouhl be recognized. 53 Examination of a cotype (M.C.Z. ;'>.3o2S) and a dozen other specimens fron; seven localities on Pemba, reveal that this is a constantly recognizable color form ; this has led me to reverse my synonyniizing of it in 1928. 5^ Its distribution suggests that boulengeri may be a full species ; the excep- tionally long tail is correlated with a more strictly arboreal life. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 211 Mobuya brevicollis (Wiegmann) 1837. Euprepes brevicollis Wiegmann, Arch. Naturg., 3, Abt. 1. p. 133: * ' Abyssinia, ' ' i.e. Ethiopia. 1837. Euprepes j)yi'>'Jiocephalus Wiegmann, Arch. Naturg., 3. Abt. 1, p. 133 : Ascliik Island, Red Sea, Arabia. 185G. Tiliqita burtoni Blyth, Journ. Asiatic Soc. Bengal, 24. p. 306: ' ' Somali Country. ' ' 1893. Mabnya chanleri Stejneger, Proc. U. S. Nat. Mus., 16. p. 721: Tana River, Kenya Colony. 1895. Mabuya tcssellata Anderson, Proc. Zool. Soc. London, p. 636, pi. xxxvi, fig. 2: near Aden (inferred from title), Arabia. 1905. Mabuin Eothschildi Mocquard, Bull. Mus. Hist. Nat. (Paris), II. p. 286: Endessa, 8° 40' N., 40° E., Ethiopia. 1915. Mabuya pulchra Matschie, Sitzb. Ges. Naturf. Freunde Berlin, p. 29: "Lahadsch, " i.e. Lahej, near Aden, Arabia. Range. Sudan, east through Ethiopia and Eritrea to Asehik Island and Arabia, south through British Somaliland and So- malia to Uganda ; Kenya Colony and Tanganyika Territory. Mabuya plonifrons ( Peters ) ^° 1878. Euprepes (Euprepis) planifrons Peters, Monatsb. Akad. Wiss. Berlin, p. 203, pi. ii, fig. 2 : " Taita, ' ' i.e. Teita, Kenya Colony. 1878. Euprepes (Euprepis) taitanus Peters, Monatsb. Akad. Wiss. Berlin, p. 203, pi. ii, fig. 3: "Taita," i.e. Teita, Kenya Colony. 1911. Mabuia diesneri Sternfeld, Sitzb. Ges. Naturf. Freunde Berlin, p. 248: Kibwezi, Kenya Colony. Range. Ethiopia, east to British Somaliland and Somalia, south through Kenya Colony to Tanganyika Territory. ^^ Mabuya megalura (Peters) Grass-top Skink 1878. Euprepes (Mabuia) megalura Peters, Monatsb. Akad. Wiss. Berlin, p. 204, pi. ii, fig. 4: "Taita," i.e. Teita, Kenya Colony. 1884. Eumeces massaianus Fischer, Jahrb. Hamburg, Wiss. Anst., 1. p. 18, pi. ii, fig. 5: Lake Naivasha Highlands, Kenya Colony. Range. Ethiopia and Somalia, south through Uganda; Kenya Colony and Tanganyika Territory to Belgian Ruanda-Urundi and the adjacent Belgian Congo. Mabuya bayonii keniensis Loveridge Eastern Bayon's Skink ^^ 35 Specimens from northeast Tanganyika Territory are typical ; in the re- mainder of the Territory southwest to Nyamkolo, Lake Tanganyika, Northern Rhodesia, they are intermediate towards perrotetli (Dum^ril & Bibron) as I have discussed elsewhere (1937, Tanganyika Notes & Records, No. 43, p. 4). 56 East African skinks formerly referred to bayonii (Bocage), differ from that Angola species in possessing tricarlnate scales. Possibly the eastern form extends to the Belgian Congo (Laurent :1952) and northeastern Angola. 212 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1956. Mabuya bayonii Iccniensis Loveridge, Breviora (Cambridge, Mass.), No. 59, p. 2: Northern Uaso Nyiro, Sotik, Kenya Colony. Range. Kenj^a Colony and Tanganyika Territory.^® Mabuya varia varia (Peters)^' Savanna Variable Skiuk 1867. Euprepes (Euprepis) varius Peters, Monatsb. Akad. Wiss. Berlin, p. 20: Tete, Mozambique. 1867. Euprepes Olivieri var. albo-punctatus Bocage, Jorn. Sci. Lisboa, 1, p. 223 : Benguela and Catumbela, Angola. 1869. Euprepes damaranus Peters, Ofvers. Kongl. Vet. -Akad. Forh., p. 660: Damaraland, Southwest Africa. 1869. Euprepes {Mabuya) laevigatus Peters, Monatsb. Akad. Wiss. Berlin, p. 434: Gerlachshoop, north of Middleburg, Transvaal. 1871. Euprepes {Exiprepis) Isselii Peters, Monatsb. Akad. Wiss. Berlin, p. 567: "Keren," i.e. Cheren, Bogos, Eritrea. 1872. Euprepes angolensis Bocage, Jorn. Sci. Lisboa, 4, p. 78: Bibala, Mossamedes; also Dondo, Angola. 1874. Euprepes {Euprepis) hildebrandtii Peters, Monatsb. Akad. Wiss. Berlin, p. 372, pi. — , fig. 4: "Barawa, " i.e. Brava, Somalia. Range. Sudan, east to British Somaliland, south through Uganda; Kenya Colony and Tanganyika Territory to Natal, west to Southwest Africa, north to Boma, Belgian Congo. Mabuya varia braiini Tornier'^* Ukinga Montane Skink 1902. Mabuia brauni Tornier, Zool. Jahrb. Syst., 15, p. 585: Ukinga Mountains, Tanganyika Territory, Range. Tanganyika Territory (known onW from the holo- type). Mabuya irregularis Lonnberg Alpine-meadow Skinli 1922. Mabuia {striata ? var.) irregularis Lonnberg, Arkiv. Zool. 14, No. 12, p. 4: Soy, Kenya Colony. 1923. Mabuia Jeanneli Angel, Bull. Mus. Hist. Nat. (Paris), 29, p. 490: Mount Kinangop, Aberdare Mountains, Kenya Colony. Range. Alpine meadows of Uganda (Mount Elgon) and Kenya Colony (Mounts Kinangop and Kenya). 5' The Museum of Comparative Zoology has a cotype of isselii with 3 (instead of 1) frontoparietals. Parker's comments (1942, Bull. Mus. Comp. Zool., 91, p. 84) notwithstanding, I think the status of hildebrandtii is more probably that of a synonym than a full species. Its distribution scarcely permits of its being a recognizable race. All the available material should be restudied. 58 I am now inclined to think this skink may be a race of bocagii related to M. b. mlanjenaia Loveridge of Nyasaland. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 213 Mobuya striata striata (Peters) ^^ Common Two-striped Skiiik 1844. Tropidolepis77ia striatum Peters, Ber. Akad. Wisa. Berlin, p. 36: Mozambique. 1849. Euprepcs punctatissimus A. Smith, Illus. Zool. S. Africa, Rept., pi. xxxi, fig. 1: northeastern districts of Cape Province, South Africa. 1849. Euprepes sunderallii A. Smith, Illus. Zool. S. Africa, Kept., App., p. 11: Interior of South Africa. 1864. Euprepes grantii Gray, Proc. Zool. Soc. London, p. 62: "South- Eastern Africa, ' ' i.e. Zambezi region. 1869. Euprepes vari.egatus Peters, Of vers. Kongl. Vet.-Akad. Forh., p. 660 : Damaraland, Southwest Africa. 1869. Euprepes wahlhergii Peters, Ofvers, Kongl. Vet. Akad. Forh., p. 661 : Damaraland, Southwest Africa. 1869. Euprepes (Euprepis) Griitzneri Peters, Monatsb. Akad. Wiss. Berlin, p. 433 : Gerlaehshoop, north of Middleburg, Transvaal. Range. Ethiopia and British Somaliland, sonth through Uganda; Kenya Colony and Tanganyika (also Peniba ; Zanzibar and Mafia Islands) to Natal, west to Southwest Africa, possibly north to the Belgian Congo. Genus RIOPA Gray"" 1839. Eiopa Gray, Ann. Nat. Hist., 2. p. 332. Type by subsequent desig- nation: Laeerta punctata Linnaeus. 1839. Chiamela Gray, Ann. Nat. Hist., 2. p. 332. Type by subsequent designation : C. lineata Gray. 1839. Hagria Gray, Ann. Nat. Hist., 2, p. 333. Type by monotypy: E. vosmaerii Gray.®-"- 1839. Campsodactylus Dumeril & Bibron, Erpet. Gen., 5, p. 761. Type by monotypy : C. lamarrei Dumeril & Bibron = Hagria vosmaerii Gray. 1843. Liosoma Fitzinger (not of Brandt: 1834), Syst. Rept., p. 22. 59 This synonymy requires restudying and the range correlated with those of -If. 8. ellenbergeri Chabanaud of Zambezia, and the various forms described from Angola and Southwest Africa. 60 Treated by Boulenger (1887, Cat. Lizards Brit. Mus., 3, p. 220) as a "Section" of the genus Lygosoma. Subsequently, by common consent, raised to full generic status, an action confirmed by Malcolm A. Smith (1937, Rec. Indian Mus., 39, p. 227). It is in this sense it is used here. More recently Mittleman (1952, Smithsonian Misc. Coll., 117, No. 17, pp. 8-10, 18-21), in a survey of Lygosomine genera, has suggested further subdivision to a quite unacceptable degree. He divides Riopa into six full genera. Of the species listed here he would retain only mabuiiformis and tanae. To Mochltis he refers fernandi, sundevallii, modesta and pembana, reviving Eumecla for the reception of anchietae. 61 Scincus vosmaerii Cocteau. which Gray cites as a synonym, appears to be only a manuscript name. 214 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Type by original designation : Eumeces viicrolepis Dumeril & Bibron. 1843. Sphenosoma Fitzinger, S3'st. Rept., p. 23. Type by original desig- nation : Lacerta jmnciata Linnaeus. 1843. Evgongylus Fitzinger, Syst. Eept., p. 23. Type by original designa- tion: Euvieces oiypellii Dumeril & Bibron =: Lacerta rufescens Shaw (part). 1864. Mochlm Giinther, Proc. Zool. Soe. London, p. 308. Type by mono- typy: M. punctnlatus Giinther = Eumeces sundevallii A. Smith. 1870. Eumecia Bocage, Jorn. Sci. Lisboa, 3, p. 67. Type by monotypy: E. anchietae Bocage. 1880. Sci)acontias Giinther, Ann. Mag. Nat. Hist., (5) 6. p. 235. Type by monotypy : S. modestus Giinther. 1892. Lcpidoihi/ris Cope, Journ. Morph., 7. p. 233. Tj^pe by subsequent designation: Tiliqita fernmidi Burton. 1952. Squamicilia Mittleman, Smithsonian Mise. Coll., 117, No. 17, p. 9. Type hj original designation: Eumeces isodactylus Giinther. 1952. Tachiigia Mittleman, Smithsonian Misc. Coll., 117. No. 17, p. 10. New name for Liosoma Fitzinger : 1843 (preoccupied). Riopa mabuiiformis Loveridge Mabuya-like Skink 193.'>. Riopa ntabuiifonnis Loveridge, Bull. Mus. Comp. Zool., 79. p. 12: Ngatana, Tana River, Kenya Colony. Range. Kenya Colony (known only from the type series). Riopa tanae Loveridge Tana-Delta Skink 1935. Eiopa tanae Loveridge, Bull. Mus. Comp. Zool., 79, p. 11: Kan. near mouth of the Tana River, Kenya Colony. Range. Coastal Kenya Colony. Riopa fernandi (Burton) Red-and-blaek Skink 1836. Tiliqua fernandi Burton, Proc. Zool. Soc. London, p. 62: Fernando Po, Gulf of Guinea. 1845. Plestiodon Harlani Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 170: Liberia. 1854. Euprepis striata Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 98: Liberia. 1883. Eupr.epes (Tiliqua) clegans J. G. Fischer (not of Peters :1854), Oester-Progr. Akad. Gymn. Hamburg, p. 3, pi. — , figs. 12-15: Sierra Leone. 1884. Euprepcs leoninus Fischer (new name for elegans Fischer, preoc- cupied), Abhand. Nat. Ver. Hamburg, 8, p. 7. 1885. Tiliqua nigripes F. Miiller, Verh. Naturf. Ges. Basel, 7. p. 704: Gold Coast. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 215 Range. Uganda and Belgian Ruanda-Uruudi, west through Belgian Congo to French Guinea. Riopa sundevallii sundevallii ( Smith )*"'- Suudevall 's Writhiug-Skink 1849. Eximices (Eiopa) siinderallii A. Smith, Illus. Zool. S. Africa, E«pt., App., p. 11: Country eastward of Cape Colony, i.e. Natal, South Africa. 1854. Eumeces afer Peters, Monatsb. Akad. Wiss. Berlin, p. 619: Mozambique Island; Boror; Inhambane and Mossimboa, Mozam- bique. 1862. Eumeces reticulatvs Peters, Monatsb. Akad. Wiss. Berlin, p. 23: New Barmen, Hereroland, Southwest Africa. 1864. Mochlus pimctulatus Giinther, Proc. Zool. Soc. London, p. 308: Zambezi Expedition. 1868. Eumeces perdicicolor Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 317: Zanzibar Island. 1870. Euprepes (Senira) Dumerili Steindaclmer, Sitzb. Akad. Wiss. Wien, 62, Abt. 1, p. 341, pi. iii, fig. 5: Zanzibar Island. 1898. Lt/gosonia ferrandii Boulenger, Ann. Mus. Civ. Stor. Nat. Genova (2), 18, p. 718, pi. ix, fig. 2: Lugh, Somalia. Range. Sudan, east to French Somaliland (and Arabia), south through Uganda ; Kenya Colony ; Tanganyika Territory ; Zanzibar and Mafia Islands to Natal, northwest through the Transvaal to Southwest Africa, north to French Equatorial Africa where one encounters intermediates with a western race."'^ Riopa modesta modesta (Giinther) Brown Writhing-Skink 1880. Sepacontias modestus Giinther, Ann. Mag. Nat. Hist., (5) 6. p. 235 : Mpwapwa, Ugogo, Tanganyika Territory. Range. Ethiopia (1895) and Uganda (1902), east and south through Kenya Colony and Tanganyika Territory direct to Nyasaland; also recorded from Belgian Ruanda-Urundi (1912). the Belgian Congo (1952) and Angola (19'S7).'^ 82 Omissions from the synonymy refer to the western subspecies. 63 /J. s. guinrensis (Peters :1879), erroneously recorded from East Africa hy Tornier (1S96), is a barely recognizable form on the basis of fewer subdlgital lamellae beneath the fourth toe. It (including chaperi Chabanaud :1SS4_) ranges from French Efiuatorial Africa (intermediates) west to Portuguese Guinea. In this connection the status of houyi (Sternfeld :1916) and mocquardi (Chabanaud : 1917) require investigating. fi-i The resemblance to sundevallii of occasional examples of modesta make it uecessar.v to view with suspicion the five solitary records of its occurrence outside its custoniarv ran^re. British Somaliland records are referable to R. m. somalica Parker (1942). 216 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Riopa pembana (Boettger) Pemba Island Writhing-Skink 1913. Lygosoma (Eiopa) pembanum Boettger, in Voeltzkow, Eeise in Ostaf rika, 3, p. 350, pi. xxiv, figs. 4-5 : Pemba Island. Range. Coastal Kenya Colony (probably introduced) and Pemba Island. Riopa anchietae (Bocage) Western Serpentiform Skink 1870. Eumecia anchietae Boeage, Joru. Sci. Lisboa, 3, p. 67, pi. 1: Huilla Plateau, Mossamedes, Angola. Range. Western Kenya Colony and Tanganyika Territory, southwest to Northern Rhodesia and Angola, north to Belgian Congo. Genus LYGOSOMA Hardwicke & Gray"" 1827. Lygosoma Hardwicke & Gray, Zool. Journ., 3, p. 228. Type by monotypy: Lacerta serpens Bloch = Anguis quadrupes Linnaeus. Subgenus LEPTOSIAPHOS Schmidt 1943. Leptosiaphos K. P. Schmidt, Field Mus. Nat. Hist. Zool. Ser., 24. p. 332. Type by original designation: Lygosoma meleagris Boulenger. Lygosoma kilimense Stejneger^ Kilimanjaro Five-toed Skink 1891. Lygosoma Icilimensis Stejneger, Proc. U. S. Nat. Mus., 14, p. 405: Mount Kilimanjaro, Tanganyika Territory. 1900. Lygosoma dathrotis Boulenger, Ann. Mag. Nat. Hist., (7) 6, p. 194: foot of Mount Kenya, Kenya Colony. 1903. Lygosoma thomasi Tornier, Zool. Jahrb. Syst., 19, p. 175: Nairobi, Kenya Colony. 1907. Lygosoma aloysii-sahaudiae Peracca, Boll. Mus. Zool. Anat. Comp. Univ. Torino, 22, No. 553, p. 1: Mitiana and Toro, Uganda. 65 Departing from the custom employed elsewhere, I have omitted the lengthy and quostionable synonymy of Lygosoma. Its East African representatives all belong to the subgenus Leptosiaphos, separated by Schmidt from 8aiph08 (correct spelling) of Gray :1831 whose range is restricted to southeast Asia and Australia. Mittleman (1952, Smithsonian Misc. Coll., 117, No. 17, pp. 6, 20) regards Leptosiaphos as a full genus, separating it from Saiphos on characters I have had tested on 4 African species (177 specimens) and 6 Australian (22 speci- mens), and I found the following to hold good. Ear-opening present though often minute ; anal not or but slightly enlarged ; range : East and Central Africa Leptosiaphos Ear-opening absent ; anal noticeably enlarged ; range : Malaya ; Indonesia ; Australia Saiphos 66 De Witte, who subsequently (1953) secured 182 of these skinks, concurs with my action in synonymizlng dewittei. L. gromieri, referred by its author to Riopa, had a longitudinally divided frontonasal and so simulated a pair of supranasals ; its synonymizlng with kilimensis can be regarded as provisional. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 217 1925. 'iLygosoma gromieri Angel, Bull. Mus. Hist. Nat. (Paris), 31. p. 419: (On termitarium near station) Tsavo, Kenya Colony. 1933. Lygcysomu (Siaphos) coviprcssicauda Witte (not Werner: 1897), Revue Zool. Bot. Afr., 23, p. 175, figs. 1-4: Sandoa, Belgian Congo. 1934. Siaphos dewittei Loveridge (new name for compressicauda Witte, preoccupied), Copeia, p. 184. 1943. Lygosoma (Lcptosiaphos) weberi K. P. Schmidt, Field Mus. Nat. Hist. Zool. Scr., 24. p. 335, fig. 27: Lotti Forest at about 3300 feet on west side of Imatong Mountains, Sudan. Range. Virgin forest, usually on mountains, from southern Sudan through Uganda; Kenya Colony and Tanganyika Terri- tory, west through the Belgian Congo to Congulu, Angola. Lygosoma kutuense Lonnberg Kenya Five-toed Skink r.ni. Lygosoma ]:utuent>is Lonnberg, Kungl. Svenska. Vetensk.-Akad. HandL, 47, No. 6, p. 17 fig. 3: Near Kutu Village, between Embu and Fort Hall, Kenya Colony. Range. Kenya Colony (known only from the holotype). Lygosoma graueri graueri Sternfeld Ruanda Five-toed Skink 19lL\ Lygosoma graueri Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika- Exped. 1907-1908, 4, p. 240, fig. 3: Locality given only for races, so suggest restricting it to bamboo forest at foot of Mount Karisimbi, Belgian Ruanda-Urundi. hH'2. Lygosoma graueri quvnquedigiiata Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika-Exped. 1907-1908, 4. p. 241, fig. 3b, pi. vi, fig. 5: Bamboo forest at foot of Mount Karisimbi, Belgian Euanda- Urimdi. Range. Virgin forests of Uganda ; Belgian Ruanda-Urundi and the adjacent Belgian Congo. Lygosoma meleagris Boulenger"' Ruwenzori Four-toed Skink 1907. Lygosoma meleagris Boulenger, Ann. Mag. Nat. Hist., (7) 19, p. 488: Mubuku Valley, 7000 feet, Ruwenzori Mountains, Uganda. 1932. Siaphos meleagris helleri Loveridge, Proc. Biol. Soc. Washington, 45, p. 113: Bugongo Ridge, 9500 feet, Ruwenzori Mountains, Belgian Congo (not Uganda side). 1933. Lygosoma (Siaphos) Burgeoni de Witte, R^vue Zool. Bot. Afr., 24, p. 116, figs. 1-2: Kalonge, 6725 feet, Ruwenzori Mountains, Belgian Congo. 1941. Lygosoma meleagris haclcarsi de Witte, Inst. Pares Nat. Congo Beige, Expl. Pare Nat. Albert, Miss. G. F. de Witte (1933- 67 Dr. de Witte (1953) concurs with the synonymy as given here. 218 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1935), 33. p. 138, figs. 39-40, pi. ii, figs. 4-6: Kamatembe, 6825 feet, Pare National Albert, Belgian Congo. Range. Virgin forests of Uganda ; Belgian Ruauda-Urundi and the adjacent Belgian Congo. Genus ABLEPHARUS Fitzinger 6S 1823. Ahlepharus Fitzinger, in Lichtenstcin, Verz. Doul)l. Zool. Mus. Berlin, p. 103. Type by monotypy: A. pannonicus Fitzinger. 1833. Lerista Bell, Proc. Zool. Soc. London, p. 99. Type by monotypy: L. lineaia Bell. 1834. Cryptohlepharnm (sic) Wiegmann, Herp. Mexicana, p. 12. Type by subsequent designation: Ahlepharris poecilopleuro (sic) Wiegmann. 1839. Petia Gray, Ann. Nat. Hist., 2, p. 335. Substitute name for " Cryptohlcpharus Wiegmann." 1843. Microblepharis Fitzinger, Syst. Kept., p. 23. Type by original designation: Ahlepliarus menestriesii Dumeril & Bibron = Scincus hivii tains Menestries. 1843. Ophiopsis Fitzinger, Syst. Eept., p. 23. Tj-pe by original designa- tion: Lerista lincata Bell. 1845. Morethia Gray, Cat. Lizards Brit. Mus., p. 65. Type by monotypy: M. anomahis Gray. 1845. Menetia Gray, Cat. Lizards Brit. Mus., pp. 65, 66. Type by monotypy : M. greyii Gray. 1845. Miculia Gray, Cat. Lizards Brit. Mus., p. 66. Type by monotypy: M. elegams Gray. 1872. Blepharosteres Stoliczka, Proc. Asiatic Soc. Bengal, p. 74. Type by monotypy : B. grayanus Stoliczka. 1881. Phaneropis Fischer, Arch. Naturg., 47. Abt. 1, p. 236. Type by monotypy: P. muelleri Fischer. Ablephorus boutonii airicanus Sternfeld Coral-rag Snake-eyed Skink 1918. Ablepharus boutoni africanus Sternfeld, Abhand. Senckenberg. Nat. Ges., 36, p. 423 : Manda Island and Malindi, Kenya Colony ; Pemba Island. Range. Rocky coastal stretches of Somalia; Kenya Colony (including Manda, Lamu and Mombasa Islands) ; and Tan- 68 Recently Mittlenian (19.52, Smithsonian Misc. CoU., 117. No. 17, pp. 13. 19), following Stejneger, bas revived Crytoblepharua as a full genus on the basis of its frontoparietal being fused with the interparietal, together with the size of its ear-upHiiing. To uie this seems a wholly unnatural division (of. footnote to A. wahlbergii). It is customary to cite "Herp. Mesicana" for Cryptohlephwrus, but Wiegmann gives Cryptoblepharum as a synonym of Ablepharus without inten- tion of proposing a new genus. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 219 ganyika Territory; also Pemba, Zanzibar and Mafia Islands. Ablepharus wahlbergii (Smith)"" Savanna Snake-eyed Skink 1849. Cnjptohlcpharus icalilhergii A. Smith, Illus. Zool. S. Africa, R*pt., App., p. 10: Natal, South Africa. 1894. Ablephurus carsonii Bouleuger, Proc. Zool. Soc. Loudon, p. 735, pi. xlLx, figs. 4-4a: "Fwambo," i.e. Fwamba, Northern Rhodesia. 1924. Ablepharus massaiensis Augel, Bull. Mus. Hist. Nat. Paris, 30. p. 52: Masai Plains near Nairobi, Kenya Colony. Range. Ethiopia and British Sornaliland, south through So- malia, Kenya Colony and Tanganyika Territory (reported from Grave Island, Zanzibar, but possibly in error for boutonii afri- canus; also Mafia Island) to Natal, west to Southwest Africa, northeast through Angola to Belgian Congo. Unknown from I^ganda. Ablepharus megalurus Nieden Blue-tailed Snake-eyed Skink 1913. Ablepharus megalurus Nieden, Mitt. Zool. Mus. Berlin, 7, p. 89: "Kinjanganja in Turu, 4° 50' S." i.e. Unyanganyi, Turn, Tanganyika Territory. Range. Central Tanganyika Territory (from Unyanganyi south to Saranda on the Central Railway). Genus SCELOTES Fitzinger'° 1826. Scelotes Fitzinger, Neue Class, Rept., pp. 23, 53. Type by niono- iypj: Bipes anguineus Merrem == Anguis bipes Linnaeus. 1830. Zygnls Wagler (not Oken:1816), Nat. Syst. Amphib., p. 160. Type by monotypj': Anguis bipes Linnaeus. 1839. merinia Gray, Ann. Mag. Nat. Hist., 2. p. 332. Type by monotypy: 69 A. carsonii Is only tentatively referred to the synonymy. It was based on a single skink whose interparietal was fused with the frontoparietal, probably an individual aberration as wahlhergii is common at Abercorn, from which Pwamba is not very distant. The status of A. tancrcdi Boulenger (1909) from Dabarif, Ethiopia, and A. u-ilsani Werner (1914) from Talodi, Sudan, also merit investi- gation as both are still known only from their respective holotypes. 70 Overlooking Gongylomorphus of Fitzinger, Witte & Laurent (1943, Mem. Mus. Royal Hist. Nat. Belgique, (2) Ease. 26, p. 3) revived its synonym Thyrus as a monotypic genus. They described Proscelotes for the reception of two species (eggeli and ulu- be regarded as a recognizable subgenus. So far as this check list is concerned, Witte & Laurent would refer the two forms of tetraduvtijlun to Scpsina which, together with Herpetosaura they raise to full generic rank, while synonymizing Jlelanosepa with Scelotes. A multiplicity of small genera emphasizes diflferences at the expense of relationships, though in this instance the authors claim that similarities are the result of convergences. 220 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY H. capensis Gray, which therefore antedates Gongylus capensis sp. n. of A. Smith: 1849 Ji 1839. AmpUglossus Dumeril & Bibron, Erpet. Gen., 5, pp. 606, 608. Type by monotypy: A. astrolahi Dumeril & Bibron. 1843. Gongylomorphus Fitzinger, Syst. Bept., p. 22. Type by monotypy: Scincus hojerii Desjardin. 1845. Thyrtis Gray, Cat. Lizards Brit. Mus., p. 124. Type by monotypy: Scincus iojerii Desjardin. 1849. Lithophilus A. Smith (not of Fi-61ich:1799), Illus. Zool. S. Africa, Rept., App., p. 12. Type by subsequent designation: L. inornatiis A. Smith. 1854. Hfirpetosaura Peters, Monatsb. Akad. Wiss. Berlin, p. 619. Type by monotypy: B.. arenicola Peters. 1861. Scpomorphus Peters, Monatsb. Akad. Wiss. Berlin, p. 422, Type by monotypy: S. caffer Peters. 1866. Sepsina Boeage, Jorn. Sci. Lisboa, 1, p. 62. Type by monotypy: S. angolensis Boeage. 1866. Dumerilia Boeage, Jorn. Sci. Lisboa, 1, p. 63. Type by monotypy: D. bayonii Boeage. 1874. llhinoscincus Peters, Monatsb. Akad. Wiss. Berlin, p. 374. Type by monotypy: Sepsina (Bhinoscincus) tetradactyla Peters. 1875. Scincodipus Peters, Monatsb. Akad. Wiss. Berlin, p. 551. Type by monotypy: S. congicus Peters = Dumerilia bayonii Boeage. 1887. Herpetoseps Boulenger, Cat. Lizards Brit. Mus., 3. p. 416. Tj^je by monotypy: E. anguineris Boulenger. 1943. Proscelotes Witte & Laurent, Mem. Mus. Eoyal Hist. Nat. Bel- gique, (2) Fasc. 26, pp. 9, 13. Type by original designation: Scelotes eggeli Toruier. Scelotes eggeli Toruier Usambara Five-toed Scelotes 1902. Scelotes eggeli Tornier, Zool. Anz., 25, p. 700: Kwai, western Usambara Mountains, Tanganyika Territory. Range. Usambara Mountains Tanganyika Territory. Scelotes uluguruensis Barbour & Loveridge Uluguru Five-toed Scelotes 1928. Scelotes uluguruensis Barbour & Loveridge, Mem. Mus. Comp. vi Fi-om the brief description. Dr. V. FitzSimous considers thiat H. capensis may well be a Scelotes. Mr. J. C. Battersby writes me that the type of H. capensis Gray, 1838, is uot to be found in the British Museum, and apparently was not there in 1887 when Boulenger wrote the Cat. Lizards Brit. Mus., 3, p. 209 and listed -"fHerinia" in the synonymy of Lygosoma. In 1935 Malcolm Smith (Fauna Brit. India. Rept. Amph., 2, p. 216) transferred it to the synonymy of Muhuya, possibly on the assumption that it was identical with Tiliqua capensis Gray, 1830, which was also described from the Cape of Good Hope. The latter species, as Mahuya capensis (Gray), 1830, is in general use for Scincus trivittatus Cuvier, 1829, preoccupied by trivittatus Hardwicke & Gray. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 221 Zool., 50. p. 167, pi. iv, fig. 6: Bagilo, 5000 feet, Uluguru Moun- tains, Tanganyika Territory. Range. Uluguru Mountains, Tanganyika Territory. Scelotes tetradactylus tetradactylus (Peters) Eastern Four-toed Scelotes 1874. Sepsina (Ehinoscincus) tetradactyla Peters, Monatsb. Akad. Wiss. Berlin, p. 374: "Zanzibar Coast," i.e. Tanganyika Territory. Range. Eastern Tanganyika Territorj^ south to Nyasaland. Scelotes tetradactylus hemptinnei (Witte) Western Four-toed Scelotes 1933. Sepsitia Hemptinnei de Witte, Eevue Zool. Bot. Afr., 23. p. 188: Lukafu, Kundelungu, Katanga, Belgian Congo. Range. Near Ujiji, Tanganyika Territory, west to Belgian Congo. Genus MELANOSEPS Boulenger" 1887. Melanoscps Boulenger, Cat. Lizards Brit. Mus., 3, p. 422. Type by monotj'py : Herpetosaura atra Giinther. Melanoseps ater longicouda Tornier Long-tailed Limbless-Skink 1900. Melanoseps ater var. longicaiula Tornier, Zool. Jahrb. Syst., 13, p. 602: Masailand, and " Karagwe am Pangani, " i.e. Korogwe, Pangaui River, Tanganyika Territory. Range. Southern Kenya Colony and northern Tanganyika Territory (whether the skink from Uluguru Mountains should be included is uncertain). Melanoseps ater rondoensis Loveridge Rondo-Plateau Limbless-Skink 1942. Melanoseps ater rondoensis Loveridge, Bull. Mus. Comp. Zool., 91. p. 360: Nchingidi, 2700 feet, Rondo Plateau, Lrndi District, Tanganyika Territory. Range. Rondo Plateau, Tanganyika Territory. "2 To the synopsis of the races furnished by me (1942, Bull. Mus. Comp. Zool., 91, p. 359) should be added the Nyasaland forms described (1943, Bull. Mus. Comp. Zool., 110. pp. 220-223). Witte & Laurent (1943, Mem. Mus. Royal Hist. Nat. Belgique (2), Fasc. 26, p. 32) may well he right in merging Melanoseps with Scelotes but the decision merits careful revisionary study of the entire complex. These limbless skinks are associated with primary forest, either montane or gallery. 222 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Melanoseps crter matengoensis Loveridge Mateugo-Highlands Limbless-Skink 1942. Melanoseps atcr matengoensis Loveridge, Bull. Mus. Conip. Zool., 91, p. 361 : Ugano, Matengo Highlands west of Songea, Tan- ganyika Territory. Range. Matengo Highlands, Tanganyika Territory. Melanoseps ater uzxxngwensis Loveridge Uzungwe-Mountains Limbless-Skink 1942. Melanoseps ater usungtcen.sis Loveridge, Bull. Mus. Comp. Zool., 91, p. 361: Kigogo, Uzungwe Mountains, Tanganyika Territory. Range. Uzungwe Mountains, Tanganyika Territory. Genus SCOLECOSEPS Loveridge 1920. Scolecoseps Loveridge, Proc. Zool. Soc. Loudon, p. l.')9. Type S. boulengeri Loveridge. Scolecoseps acontias (Werner) Sandy-soil Limbless-Skink 1913. .]felanoseps aeontias Werner, 1912, Jalirb. Hamburg. Wiss. Anst. for 1912, 30, p. 19: Dar es Salaam, Tanganyika Territory. Range. Coastal (Dar es Salaam to Kilwa) Tanganyika Terri- tory. Genus ACONTIAS Cuvier 1817. Acontius Cuvier, Regne Animal, 2, p. 60. Type by monotypy: Angtiis mcleagri^ Linnaeus. Acontias percivali Loveridge Teita Mountains Limbless-Skink 19.15. Acontias percivali Loveridge, Bull. Mus. Comp. Zool., 79. p. 13: Foot of Mount Mbololo, Teita Mountains, Kenya Colony. Range. \'ieinity of Voi, Kenya Colony. Genus FEYLINIA Gray 1845. Feylinia Gray, Cat. Lizards Brit. Mus., p. 129. Type by monotj-py: F. currori Gray. 18.")6. Anehjirops A. Dumeril, Revue Mag. Zool., (2) 8, p. 420. Type by monotypy: Acontias elega/ns Hallowell. 1857. Sphcnorhina Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 52. Typo by monotypy: Acontias elegans Hallowell. 1943. Chahanamlia Witte & Laurent, Mem. Mus. Royal Hist. Nat. Belgique, (2) Fasc. 26, p. 37. Type by original designation: Feylinia boiiU tifferi Cliabanaud. 'n LOVEBIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 223 Feylinia currori elegotis (Ilallowell)'^ Western-Forest Limbless-Skink 1852. Acontias clegans Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 65 : " Liberia, ' ' probably in error for Gabon. Range. U. 56: "in Zeylania ad littora maris. ' ' 1783. Lacerta capensis Sparrman, Eesa till Goda Hoppsudden, p. 749: " Hinterbruyntjes Heights," = Bruintjeshoogte, a village be- tween Pearston and Somerset East, eastern Cape Province.^^ 1788. " L. tupinamhis" LacepMe (part),*^ Hist. Nat. Quad. ovip. Serpens, 1, Synopsis Methodica, p. 251, pi. xvii: restricted to Cape of Good Hope, i.e. South Africa. 1802. Tupinavihis elegans Daudin (part). Hist. Nat. Rept., 3, p. 36: "Amerique meridionale" (error). 1802. Tupinamhis stellatus Daudin, Hist. Nat. Eept., 3, p. 59, pi. xxxi: Senegal (restricted). 1819. Monitor pultiher Leach, in Bowdich, Miss. Ashantee, App., p. 493: "Fantee, " i.e. Fanti, Ashanti, Gold Coast. 1844. Monitor elegans senegalensis Schlegel, Abbild. Amphib., p. x: Senegal. Range. Africa (exclusive of arid areas in the north and south- west, also the rain-forest region of the west^'), including Uganda; Kenya Colony ; Tanganyika Territory ; Zanzibar and Mafia Islands. Voranus exanthematicus microstictus Boettger''* Eastern Savanna-Monitor 1845. Mo>nitoT microstictus Riippell {nomen nudum), Mus. Senckenberg., 3. p. 301: " Abyssinien, " i.e. Ethiopia. 1893. Varanus microstictus Boettger, Kat. E^pt. Saminl. Mus. Sencken- berg. Naturf. Ges., Part 1, pp. viii, 72: Ethiopia. Range. Ethiopia and British Somaliland, south through Uganda ; Kenya Colony ; Tanganyika Territory and Zanzibar Island to Mozambique (where it meets with the southern race, V. e. alhigularis (Daudin)). »£- Dr. V. F. FitzSimons, to whom I am indebted for the above information, informs me that this locality appears on Sparrman's map (1876) as "Agter- bruntjes Hoogte." 86 Only tlie vernacular name is given, not "Lacertus tupinamhis" as cited by some authors. In fact it is listed with the crocodiliaus in the Synopsis Methodica. 87 The only race, V. n. ornatus (Daudin), has a discontinuous distribution in the surviving patches of rain forest from the western Belgian Congo and Cameroons to Liberia. The species is referable to the subgenus Polydaedalus. 88 This is the monitor which for a century has been miscalled Varanus ocellatus (Heyden in Kuppell), a name that is actually synonymous with V. e. exanthe- maticus (Bosc) which ranges from Eritrea to Senegal. LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 237 Family AMPHISBAENIDAE Genus ANCYCLOCRANIUM Parker 1942. Ancyclocranium Parker, Bull. Mus. Comp. Zool., 91, p. 37: Type by original designation: Anops somalicus Scortecci. Ancyclocronimn borkeri Loveridge Lindi Sharp-snouted Worm-Lizard 1942. Ancyclocranium iarlceri Loveridge, Proc. Biol. Soc. Washington, 59. p. 73, pi. xiii: Mbemkuru Eiver, Lindi district, Tanganyika Territory. Range. Tanganyika Territory (known only from holotype). Ancyclocranium ionidesi Loveridge Kilwa Sharp-snouted Worm-Lizard 1955. Ancylocranium ionidesi Loveridge, Journ. E. Africa Nat. Hist. Soc, 22, p. 177, pi. — : Kilwa, Southern Province, Tanganyika Territory. Range. Tanganyika Territory (known only from the type series). Genus AMPHISBAENA Linnaeus *=' 1758. * AmpMsljaena Linnaeus, Syst. Nat. ed. 10, 1, p. 229. Type by sub- sequent designation: A. fidiginosa Linnaeus. 1844. Sarca Gray, Cat. Tort. Croc. Amphis. Brit. Mus., p. 71. Type by monotypy: Amphisbaena caeca Cuvier. 1844. *Cynisca Gray, Cat. Tort. Croc. Amphis. Brit. Mus., p. 71. Type by monotypy : Amphisbaena leucura Dumeril & Bibron. 1861. *Diphaluti Cope, Proc. Acad. Nat. Sei. Philadelphia, p. 75. Type by original designation : D. fenestratua Cope. 1865. *Bronia Gray, Proc. Zool. Soc. London, p. 448. Type by monotypy: B. brasillana Gray. 1878. Ophloproctes Boulenger, Bull. Soc. Zool. France, 3, p. 300. Type by monotypy : 0. liberiensis Boulenger. 1885. Aporarclius Cope, Proc. American Philos. Soc, 22, p. 187. Type by monotypy: Amphisbaena prunicolor Cope. 1885. Zygnis Cope (not Oken:1816), Proc. American Philos. Soc, 22, p. 188. Type by monotypy: Amphi-sbaen-a quadrifrons Peters. 1907. *Cluriiulia Boulenger, Ann. Mag. Nat. Hist., (7) 20, p. 48. Type by monotypy: C. swynnertoni Boulenger. 89 Seven of these names, indicated by asterisks, here regarded as subgenera, are considered full genera by Vauzolini (1951, Herpetologica. 7, pp. 113-123). Five of the six East African species would be assigned to as many different subgenera, with his Hhrevea replacing Zygnis Cope (preoccupied). 238 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 1911. Amphisbaeniila Sternfcld, Sitzb. Ges. Naturf. Fieunde Berlin, p. 246. Type by monotypy: A. orientalis Sternfeld. 1951. *Loveridgea Vanzolini, Herpetologica, 7. p. 114. Type by original designation: Aviphishaena phylofiniens Tornier. 1951. *Shrevea Vanzolini, Herpetologica, 7. p. 115. Type by original designation: Amphisbaena quadrifrons Peters. Amphisbaena ionidesii Battersby Liwale Worm-Lizard 1950. Ampliishaena ionidesii Battersby, Ann. Mag. Nat. Hist., (12) 3, p. 413, fig. — : Liwale, Southern Province, Tanganyika Terri- tory. Range. Tanganyika Territory. Amphisbaena phylofiniens Tornier Ujiji Worm-Lizard 1899. Amphishaoui phylofiniens Tornier, Zool. Anz., 22. p. 260: Ujiji, Tanganyika Territory. Range. Tanganyika Territory.''" Amphisbaena mpwapwaensis Loveridge Mpwapwa Worm-Lizard 1932. Amphisbaena mpwapwaensis Loveridge, BuU. Mus. Comp. Zool., 72. p. 378: Mpwapwa, Ugogo, Tanganyika Territory. Range. Tanganyika Territory.^" Amphisbaena ewerbecki (AVeruer) Mbanja Worm-Lizard 1910. Chirindia ewerbecM Werner, 1909, Jalirb. Hamburg, Wiss. Anst., 27, p. 37: Mbanja near Lindi, Tanganyika Territory. Range. Tanganyika Territory.^" Amphisbaena rondoensis Loveridge Nchingidi Worm-Lizard 1941. Amphishaena rondoensis Loveridge, Bull. Mus. Comp. Zool., 87, p. 394, fig. 23: Nchingidi, Rondo Plateau, Lindi district, Tan- ganyika Territory. Range. Tanganyika Territory.®** Amphisbaena orientalis (Sternfeld) Mikindani Worm-Lizard 1911. Amphisbaenula orientalis Sternfeld, Sitzb. Ges. Naturf. Freunde Berlin, p. 246: Mikindani, Southern Province, Tanganyika Ter- ritory. Range. Tanganyika Territory .°" Genus GEOCALAMUS Giinther 1880. Geocalamus Giinther, Ann. Mag. Nat. Hist., (5) 6. p. 234. Type by monotypy: G. modestiis Giinther. 90 Only from ^icinity of the t.vpe localit.v. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 239 Geocolomus acutus Sternfeld \"oi Wedg'e-siKJiited AVorm-Lizard 1912. CTeocalamii.s acutus Sternfeld, Wiss. Ergebn. Deut. Zentral-Afrika- Exped. 1907-1908, 4. p. 209: Voi. Kenya Colony. 1913. deocalamus noltci Boettger, in Voeltzkow, Eeise in Ostafrika, 3, p. 366, pi. xxvi, tig. 6: Moslii, Tanganyika Territory. Range. Kenya Colony and Tanganyika Terirtory. Geocalamus modestus Giinther ]\I])^vap^va Wedge-.snouted Worm-Lizard 1880. Geocalamus modestvs Giinther, Ann. Mag. Nat. Hist., (5) 6. p. 23-1: Mpwapwa, Ugogo, Tanganyika Territory. Range. Tanganyika Territory. Suborder SERPENTES '' Family TYPHLOPIDAE Genus TYPHLOPS Oppel" 1811. Tijphlops Oppol, Ordn. Fam. Gattung Eept., p. 54. Type by sub- sequent designation: Anguis lumbricalis Linnaeus. 1830. Typhlin-a Wagler (not TypJtline Wiegmann), Xat. Syst. Amphib., p. 196. Type by subsequent designation: " Acontias lineatiis Eeinw. " = Typhlops Uneatus Boie.®" 1843. AspUlorln/nclius Fitzinger, Syst. Rept., p. 24. Type by original designation: Typhlops eschrichtii Schelegel ^ Acontias pioictatus Leach. 1843. Gerrhopilus Fitzinger, Syst. Eept., p. 24. Type by original designa- tion: Typhlops ater Schlegel. 1843. Pseudotyphlops Fitzinger, Syst. Eept., p. 24. Type by original designation : Typhlops polygrammicus Schlegel. 1843. Rhinotyphlops Fitzinger, Syst. Eept., p. 24. Type by original designation: Typhlops lahindli Schlcgc-l. 1843. Eamphotyphlops Fitzinger, Syst. Eept., p. 24. Type by original designation: Typhlops multilineatits Schlegel. 910PHIDIA of Macartney (1802), Boulengcr (1893) and others. 93 In case Fitzinger's (1843) designation or later ones be considered inadequate, the type of TyphUna is hereby fixed as above to avoid any possibility of Typhlops Keptemstrintus Schneider being selected, for this would affect the use of Lepto- tuphlops and LEPTOTYPHLOPIDAE. 240 BULLETIN: MUSEUM OP COMPARATIVE ZOOLOGY 1844. Pilidion Diimeril & Bibron, Erpet. Gen., 6, p. 257. New name for Typhliiia lineatus Wagler ^ TypMops lineatus Boie. 1844. Ophthalmidium Dumeril & Bibron, Erp6t. G6n., 6. p. 262. Type by subsequent designation: 0. longissimum Dumeril & Bibron. 1844. Cathetarhinus Dumeril & Bibron, Erp6t. Gen., 6. p. 268. Type by monotypy: C. melanoeephalus Dumeril & Bibron. 1844. Onychocephalus Dumeril & Bibron, Erpet. Gen., 6. p. 272. Type by subsequent designation: TypMops lalandii Sclilegel. 1845. Onychophis Gray, Cat. Lizards Brit. Mus., p. 132. Type by monotypy: 0. franMinii Gray = TypMops lalandii Schlegel. 1845. TypMinalis Gray, Cat. Lizards Brit. Mus., p. 134. New name for TypMina Wagler. 1845. Anilios Gray, Cat. Lizards Brit. Mus., p. 135. Type by present designation: A. Icachii Gray = Anguis lumbricalis Linnaeus. 1845. Argyrophis Gray, Cat. Lizards Brit. Mus., p. 136. Type by present designation: A. bicolor Gray = TypMops nigroalbiis Dumeril & Bibron. 1845. Meditoria Gray, Cat. Lizards Brit. Mus., p. 139. Type by mono- typy: M. nasuta Gray = Anguis lumbricalis Linnaeus. 1861. DiaphorotypMops Jan, Arch. Zool. Anat. Fisiol., 1, p. 185. Type by subsequent designation: D. disparilis Jan. 1868. Letheobia Cope, Proe. Acad. Nat. Sci. Philadelphia, p. 322. Type by designation of Dunn & Dunn (1940: Copeia, p. 75): L. pallida Cope. 1881. G-rypotypMops Peters, Sitzb. Ges. Naturf. Freunde Berlin, p. 70. Type by original designation: Onydhocephalus acutus Dumeril & Bibron. Typhlops blanfordii bloniordii Boulenger Ethiopian Blind-Snake 1889. Typhlops Blanfordii Boulenger, Ann. Mag. Nat. Hist., (6) 4. p. 363: Senafe, Ethiopia. Range. Eritrea and Ethiopia, south to northern Kenya Col- ony (at Moyale; fide Scortecei:1940). Typhlops blanfordii lestradei Witte^ Ruanda Blind-Snake 1933. Typhlops Lestradei de Witte, Revue Zool. Bot. Afr., 23. p. 206, figs. 1-3: "Rubengera" i.e. Ruhengeri, Belgian Ruanda-Urundi. Range. Uganda, west through Belgian Ruanda-Urundi to Belgian Congo. Typhlops schlegelii brevis Scortecci Northern Schlegel's Blind-Snake 9* For detailed discussion of relationsliips with adolfi Sternfeld and dtihius Cliabanaud, sec Loveridge, 1942, Bull. Mus. Comp. Zool., 91, pp. 254-255, pi. 11, fig. 2. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 241 1929. Typhlops brevis Scortecci, Atti Soc. Ital. Sci. Nat. (Milano), 68, p. 287: Chisiniaio (Kismayu), Somalia. Range. Southern Sudan and Uganda (Lado Enclave) ; British Somaliland south through Somalia (probably to northern Kenya Colony). Typhlops schlegelii excentricus Procter Dark-bellied Blind-Snake 1922. Typhlops excentricus Procter, Ann. Mag. Nat. Hist., (9) 9, p. 685: Kilosa, Tanganyika Territory. Range. Tanganyika Territory (an east-central area from Mpwapwa to Morogoro). Typhlops schlegelii mucruso (Peters)®'^ Eastern Schlegel's Blind-Snake 1854. Onychoccphalus mucruso Peters (part), Monatsb. Akad. Wiss. Berlin, p. 621 : Maeanga, Mozambique. 1873. Onychocephahis Pctersii Bocage, Jorn. Sci. Lisboa, 4. p. 249: Bibala, Mossamedes, Angola. 1886. Typhlops {Onychoccphalus) hnmbo Bocage, Jem. Sci. Lisboa, 11- p. 171 : Quisange, Benguela, Angola. 1893. Typhlops hottentotus Bocage, Jorn. Sci. Lisboa, (2) 3. p. 117: Quindumbo, Benguela, Angola. 1893. Typhlops mandensis Stejneger, Proc. U. S. Nat. Mus., 16, p. 725: Wange, mainland opposite Manda Island, Kenya Colony. 1910. Typhlops latirostris Stemfeld, Mitt. Zool. Mus. Berlin, 5. p. 70: Tabora, Tanganyika Territory. 1912. Typhlops viridiflavus Peracca, Ann. Mus. Zool. Univ. Napoli, (2) 3, No. 25, p. 3 : Lake Bangweulu, Northern Rhodesia. Range. Kenya Colony; Tanganyika Territory and Zanzibar Island, south to Mozambique north of the Zambezi River,^'^ west through Nyasaland and Northern Rhodesia to Angola. Typhlops punctalus punctatus (Leach)®' Spotted Blind-Snake 1S19. Acontias punctatus Leach, in Bowdich, Mis.s. Ashantec, App., p. 493: "Fantee," i.e. Fanti, Ashanti, Gold Coast. 1844. Typhlops cschrichiii Schlegel, Abbild. Amphib., p. 37, pi. xxxii, figs. 13, 16: Gold Coast. ^■' Owing to young individuals having rounded snouts resembling those of })unctatus, they have often been reported as that species. Direct comparison oC the types of the numerous synonyms of both species is very mucli uceded. "•5 An arbitrary boundary, for the two races merge in this vicinity. T. s. schlegelii (Peters) ranges over the whole of Africa south of the river. 9" Several other forms, which should porhaps be added to this synonymy, have been described in the last thirty-live years. 242 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 1844. Onydhocephalus congestus Dum^ril & Bibron, Erpet. Gen., 6. p. 265 : No locality. 1845. OnychopJiis Barroivii Gray, Cat. Lizards Brit. Mus., p. 133: "India?" 1848. Onycltoccphalus Liheriensis Hallowell, Proc. Acad. Nat. Sci. Phila- delphia, p. 59: Liberia. 1848. Onycliocephalus niciro-lineatus Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 60 : Liberia. 1864. Typhlops liheriensis var. intermedia Jan, Icon. Gen. Ophid., p. 24, and 1, livr. 5, pis. v-vi, fig. 2: Liberia. 1864. Typhlops Kraussi Jan, Icon. Gen. Opliid., p. 26, and 1, livr. 3, pi. vi, tig. 2 : Gold Coast. 1864. TypMops lineolatus Jan. Icon. Gen. Ophid., p. 24, and 1. livr. 9, pi. i, fig. 4 : Sierra Leone. 1866. Onycliocephalus angolensis Bocage, Jorn. Sci. Lisboa, 1, p. 46: Duque de Braganca, Angola. 1887. Typhlops (Onychocephalus) congicus Boettger, Zool. Anz., 10, p. 650: Povo Netonna, near Banana, Belgian Congo. 1893. Typhlops houlengcri Bocage, Jorn. Sci. Lisboa, (2) 3. p. 117: Interior of Benguela, Angola. 1904. Typhlops iocagei, Perreira, Jorn. Sci. Lisboa, (2) 7, p. 114: Cabicula, Bom Jesus, Angola. 1910. Typhlops Tornieri Sternfeld, Mitt. Zool. Mus. Berlin, 5. p. 69: Mount Kilimanjaro, Tanganyika Territory. 1910. Typhlops Adolphi Sternfeld, Mitt. Zool. Mus. Berlin, 5. p. 70: "Fort Blus," later corrected to Port Beni, Belgian Congo. 1917. Typhlops dtibiiis Chabanaud, Bull. Mus. Hist. Nat. (Paris), 1916, 22. p. 364: "Kivori, " i.e. Kivu Volcanoes, 1500 metres, Belgian Congo. 1920. Typhlops Milleti Chabanaud, Bull. Mus. Hist. Nat. (Paris), 26. p. 463: Togo. Range. Forested areas of Uganda; montane forests of Kenya Colony and Tanganyika Territory, southwest to Angola and northwest to Senegal (frequently recorded in error for other species). Typhlops punctatus gierrcd Mocquard Usambara Spotted Blind-Snake 1897. Typhlops Gierrai Mocquard, Bull. Mus. Hist. Nat. (Paris), 3. p. 122: Tanga, Tanganyika Territory. Range. Virgin forests of the Usambara and Uluguru Moun- tains (possibly the gallery forests of adjacent lowlands also), Tanganyika Territory. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 248 Typhlops kaimosae Loveridge Kakamega Spotted Blind-Snake 1935. Typhlops kaimosae Loveridge, Bull. Mus. Comp. Zool., 79, p. 5: Kaimosi, Kakamega, Nyanza Province, Kenya Colony. Range. Kenya Colony (known only from the holotype; pos- sibly an aberrant jninctatiis in which the preocular is separated from the upper labials by nasal and ocular). Typhlops tettensis tettensis (Peters) Tete Blind-Snake 18(30. Onycliocephahis tettensis Peters, Monatsb. Akad. Wiss. Berlin, p. 80 : Tete, Mozambique. Range. Kilwa and Liwale, Tanganyika Territory, southwest to Tete, Mozambique. Typhlops tettensis rondoensis Loveridge Rondo Blind-Snake 1942. Typhlops tettensis rondoensis Loveridge, Bull. Mus. Comp. Zool., 91, p. 25(5: Xchingidi, Rondo Plateau, Lindi district, Tanganyika Territory. Range. Rondo Plateau, Tanganyika Territory. Typhlops unitaeniatus unitaeniatus Peters Stripe-backed Blind-Snake 1878. Typhlops (Letheoiia) unitaeniatus Peters, Monatsb. Akad. Wiss. Berlin, p. 205, pi. ii, fig. 5: "Taita," i.e. Teita, Kenya Colony. Range. Somalia (where the race ataeniatus Boulenger also occurs), south through coastal Kenya Colony to Buhuri, north- east Tanganyika Territory. Typhlops platyrhynchus Sternfeld Tanga Blind-Snake^^ 1910. Typhlops platyrhynchus Sternfeld, Mitt. Zool. Mus. Berlin, 5, p. 69 : Tanga, Tanganyika Territory. Range. Tanganyika Territory (known only from the type locality). Typhlops graueri Sternfeld Lake Tanganyika Blind-Snake 1912. Typhlops graueri Sternfeld, Wiss. Ergebn. Deutsch. Zentral- Afrika-Exped. 1907-1908, 4. p. 264: Virgin forest behind bound- ary mountains on northwest of Lake Tanganyika, Belgian Congo. Range. Ujiji, Tanganyika Territory, northwest to Medje, Belgian Congo. Typhlops gracilis Sternfeld Urungu Blind-Snake 1910. Typhlops gracilis Sternfeld, Mitt. Zool. Mus. Berlin, 5. p. 70: Kit- ungulu, Urungu, Tanganjoka Territory, 9» Typhlops opisthopachya Werner (1917), allegedly from Tanga, has been shown by me to be a synonym of the Australian pinguia Waite. 244 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Range. Southwestern Tanganyika Territory and Northern Rhodesia. Typhlops pallidus (Cope) Zanzibar Blind-Snake 1868. Letheohm pallida Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 322: Zanzibar Island. Range. Southern Sudan; coastal belt of Kenya Colony and Tanganyika Territory ; Pemba and Zanzibar Islands. Typhlops uluguruensis Barbour & Loveridge TTluguru Blind-Snake 1928. Typhlops uluguruensis Barbour & Loveridge, Mem. Mas. Conip. Zool., 50, p. 104: Nyange, Uluguru Mountains, Tanganyika Territory. Range. Tanganyika Territory (known only from the type locality). Typhlops braminus (Daudin) Brahniiny or Flower-Pot Blind-Snake 1803. Eryx hrami'nus Daudin, Hist. Nat. Kept., 7, p. 279: Bengal, India. 1820. Tartrix Russellii Merrem, Vers. Syst. Amphib., p. 85: India orientali. 1838. Onychocephalus Capensis A. Smith, Illus. Zool. S. Africa, Eept., pi. ii, fig. 3; pi. liv, figs. 9-16: South Africa ("Interior of" may be considered to be erroneous). 1845. Argyrophis truncatus Gray, Cat. Lizards Brit. Mus., p. 138: Philipx)ine Islands. 1860. Ophthalmidium tenue Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 497 (type believed lost) : Hong Kong. 1863. Typhlops inconspieuo'us Jan, Elenco Sist. Degli Ofidi, p. 11: Madagascar. 1864. Typhlops pammeces Giinther (new name for tenue Hallowell: pre- occupied), Kept. British India, p. 176, pi. xvi, fig. C. 1882. Typhlops (Typhlops) euproctus Boettger, Zool. Anz., 5, p. 479: Loucoube, Nossi Be, Madagascar. 1889. Typhlops comorensis Boulenger, Ann. Mag. Nat. Hist., (6) 4, p. 361 : Comoro Islands. 1906. Typhlops braminus var. arenicola Aimandale, Mem. Asiatic Soc. Bengal, 1. p. 192: Bamnad, South India. 1906. Typhlops limbricM Annandale, Mem. Asiatic Soc. Bengal, 1. p. 193 : Ramnad, South India. 1909. Typhlops braminus var. pallidus Wall (not Cope:1868), Journ. Bombay Nat. Hist. Soc, 19, p. 609: Dibrugarh, Upper Assam. 1909. Typhlops microcephalus Werner, Jahresb. Ver. Nat. Wiirttemberg LOVERIDQE: E. AFRICAN REPTILES AND AMPHIBIANS 245 (Stuttgart), 65. p. 60: Madagascar. 1910. Glauconia braueri Stenifeld, Mitt. Zool. Mus. Berlin, 5, p. 69: Baganioyo, Tanganyika Territory. 1918. Typhlops capenMs Rendahl (sp. n.; not of A. Smith), Arkiv. Zool., 11. No. 17, p. 1, figs. 1-3: "Capeland," i.e. South Africa. 1919. Typhlops fletcheri Wall, Journ. Bombay Nat. Hist. Soc, 26. p. 556: Nilgiris, India. Range. Coastal zone of Kenya Colony ; Tanganyika Territory ; Mozambique and South Africa. On islands of Socotra ; Zanzibar ; Comoro; Madagascar and Mauritius.^^ Southern Europe and Asia (Arabia to China) ; islands of the Indian Ocean; China Sea; Pacific Ocean (Mariauna; Caroline; Marshall; Solomon; Hawaiian); West Indies; South America (Mexico to Argentina). Typhlops lumbriciformis (Peters) Wormlike Blind-Snake 1874. OnycJioceplialus (Lethcobin) lumbriciforviis Peters, Monatsb. Akad. Wiss. Berlin, p. 377: "Zanzibar Coast," i.e. Tanganyika Territory. 1904. Typhlops Meebergi Werner, Zool. Anz., 27. p. 464: Usambara, Tanganyika Territory. Range. Coastal Kenya Colony and Tanganyika Territory; Zanzibar Island.'"" Family LEPTOTYPHLOPIDAE Genus LEPTOTYPHLOPS Fitzinger 1824. Stenostoma Wagler (not Latreille:1810), in Spix, Serp. Brasil., p. 68. Type by monotypy: S. albifrons Wagler. 1843. Leptotyphlops Fitzinger, Syst. Rept., p. 24. Type by original designation: Typhlops nigricans Schlegel. 1843. Bucephalus Fitzinger, Syst. Rept., p. 24. Type by original designa- tion: Typhlops bilineatus Schlegel. 99 The world-wide distribution of this tiny snake is attributable to transport in soil surrounding the roots of introduced plants, at other times in ballast. Its first appearance in a country is apt to lead to redescription. The above list of synonyms is not necessarily complete. Possibly some of the five Typhlops with 20 midbody scale-rows, described from Ceylon by Taylor (1947, Univ. Kansas Sci. Bull., 31, pp. 287-298) may be based on individual variants. The Malagasy species rewndii Hallowell. Thelotornis kirtlandii Idrtlondii (Hallowell) Northern Vine-Snake ^^^ 1844. Leptophis Kirtlandii Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 62 : Liberia. 1854. Oxyhelis Lecomtei Dumeril & Bibron, Erpet. Gen., 7. p. 821: Gabon, French Congo. 1854. Tragops nifulns Dumeril & Bibron, Erpet. Gen., 7. p. 827: Senegal. 185(5. Oxyhelis violacea Fischer, Abhand. Nat. Ver. Hamburg, 3. p. 91, pi. ii, fig. 7: Edina, Grand Bassa County, Liberia. Range. Somalia, south through Uganda and Kenya Colony to northern Tanganyika Territory (where it meets and merges with the southern race) southAvest to central Angola, north and west to Portuguese Guinea. Thelotornis kirtlandii capensis Smith Southern Vine-Snake 1849. Thelotornis capensis A. Smith, Illus. Zool. S. Africa, Rept., App., p. 19: Kaffirland and the country towards Port Natal, i.e. Durban, Natal. 1895. Thelotornis Kirtlandii var. mossamhicana Boeage, Herp. Angoht Congo, p. 119: Manica, Mozambique (restricted). Range. Southern Tanganyika Territory and Mafia Island, 133 A name slightly preferable to Blrrl-Snake, for, though its principal prey may consist of birds and nestlings at certain seasons or in some localities, else- where chameleons and other reptiles figure largely in its menu. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 275 south (except in area occupied by T. k. oatesii Giinther) to Natal northwest to northern Southwest Africa and southern Angola (where it meets and merges with the northern race). Genus HEMIRHAGERRHIS Boettger 136 1893. Hemirhagerrhis Boettger, Zool. Anz., 16, p. 129. Type by mono- typy : H. kelleri Boettger. Hemirhagerrhis kelleri Boettger Stripe-bellied Snake 1893. Hemirhagerrhis kelleri Boettger, Zool. Anz., 16, p. 129: "Ah- dallah," i.e. Abdulla, north of Webi Shebeli, Ethiopia. 1908. Amplarhinus taeniatus Sternfeld, Mitt. Zool. Mus. Berlin, 4. p. 244, fig. 3: Lamu Island, Kenya Colony. Range. Ethiopia and British Somaliland, south through So- malia to Mombasa and Voi, Kenya Colony. Hemirhagerrhis nototaenia nototaenia ((liinther) Eastern Spot-striped Snake 1864. Coronella nototaenia Giinther, Proe. Zool. Soc. London, p. 309, pi. xxvi, fig. 1 : Eios de Sena, Zambezi, Mozambique. 1878. Ablabes Eildebrandtii Peters, Monatsb. Akad. Wiss. Berlin, p. 205, pi. ii, fig. 6: Kitui, Ukamba, Kenya Colony. 1901). Amplorhinus Giintheri Mocquard, Bull. Mus. Hist. Nat. (Paris), 12, p. 251: Lake Ngami, Bechuanaland Protectorate. Range. Sudan and British Somaliland south through Kenya Colony and Tanganyika Territory to Mozambique and Bechuana- land. north through the Rhodesias to Belgian Cono-o. Genus PSAMMOPHYLAX Fitzinger 137 1768. Cerastes Laurenti, Syn. Kept., p. 81. Type by subsequent designa tion of rejervary:1923: Coluber rhombeatus Linnaeus. Type by subsequent designation of Stejneger :1936: Cerastes candidus Laurenti =: Coluber haje Linnaeus, i.e. Naja haje (Linnaeus). 136 Bogert (1940. Bull. Amer. Mus. Xat. Hist., 77, p. 75) considers Amplorhintig A. Smith (1847. 111. Zool. S. Africa. Kept., text to pi. Ivii) must for the present be regarded as a monotypic genus, with A. multimaciilatus Smith the only species. 137 Technically Cerastes is correct, but in view of its application to the horned vipers of North Africa for so long (I have over 100 citations in this sense between 1880 and 1950), to use it for the more or less innocuous opisthoglyphous grass-snakes of East and South Africa seems to imply a pedantic devotion to legality. In view of the medical implications involved in anti-cerastes serum and the frequency with which the name Cerastes has appeared in medical literature, it should be re.iected, or Stejneger's solution of the difficulty accepted. 276 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1843. Psammophylax Fitzinger, Syst. Kept., p. 26. Type by original designation: Coluber rhombeatus Linnaeus. 1847. T rimer orhinus A. Smith, Illus. Zool. S. Africa, Eept., text to pi. Ivi. Type by monotypy: Coluber rhombeatus Linnaeus. Psammophylax tritaeniatus multisqucanis ( Loveridge ) ^' " Many-scaled Grass-Snake 1932. T rimer orhinus tritaeniatus multisquamis Loveridge, Proc. Biol. Soc. Washington, 45, p. 84: Nairobi, Kenya Colony. Range. Highlands of Ethiopia and Kenya Colony, south (in highlands) to the Central Railway, Tanganyika Territory. Psammophylax tritaeniatus variabilis Giinther ^■^'' Gray-l)e]lied Grass-Snake 1893. Psammophylax variabilis Giinther, Proc. Zool. Soc. London, 1892, p. 5ri7, col. pi. XXXV : Shire Highlands, Nrasaland. Range. Highlands of Tanganyika Territory (south of the Cen- tral Railway), south (in highlands) to Mozambique and Nyasa- land. Psammophylax tritaeniatus tritaeniatus (^Giinther) White-bellied Grass-Snake 1868. Rhageri'his tritaeniatus Giinther, Ann. Mag. Nat. Hist., (4) 1. p. 423, pi. xix, fig. H: Southeastern Africa. 1881. Coronella tritaenia Giinther, in Frank Gates, Matabeleland and the Victoria Falls (London), App. 3, p. 329, pi. C: "South- eastern Africa," i.e. Matabeleland, Southern Rhodesia. Range. Dry savanna of Tanganyika Territory-, south to Natal, west to Southwest Africa, north to Angola, east through south- ern Belgian Congo to Lake Tanganyika. Genus RHAMPHIOPHIS Peters 18.54. Rhamphiophis Peters, Monatsb, Akad. Wiss. Berlin, p. 624. Type by monotypy: B. rostratus Peters. 1863. Dipsina Jan, Arch. Zool. Anat. Fisiol., 2, pp. 215, 313. Type by monotypy: Coronella multimaculata A. Smith. 138 Early Kenya and Tanganyika records appeared as Rhagerrhis tritaeniata , my own first captures were erroneously recorded as Psammophis hrevirostrin. Boulenger's (1896:649) record of Trimerorhinus tritaeniatus from "Uganda," like otliers from the same source, should read Kenya Colony. 139 The "Trimerorhinus rhombeatus" record of Sternfeld (1910) from Ukinga. southern Tanganyika Territory, was almost certainly based on a misidentitiPii P. t. variabilis, the form which is abundant throughout the southern highlands. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 277 Rhconphiophis acutus acutus (Giinther) Southern Striped Beaked-Snake 1888. Psammophis acutus Giinther, Ann. Mag. Nat. Hist., (6) 1, p. 327, pi. xix, fig. D: Pungo Ndongo, Angola. Range. Southern Tanganyika Territory west through North- ern Rhodesia to Angola, northeast through the Belgian Congo to Belgian Kuanda-Urundi. Rhconphiophis rubropunctatus (Fischer) Red-spotted Beaked-Snake 1884. Dipsina ruhropunctaia Fischer, Jahrb. Hamburg. Wiss. Anst., 1, p. 7, pi. i, fig. 3: Near Arusha, Tanganyika Territory. Range. Sudan east to British Somaliland, south through So- malia and eastern Kem'a Colony to northeastern Tanganyika Territory, and {jide Moreau and Pakenham) Zanzibar Island. Rhamphiophis oxyrhynchus rostratus Peters Eastern Brown Beaked-Snake 1854. Pihumphiophis rostratus Peters, Monatsb. Akad. Wiss. Berlin, p. 624: Tete; Mesuril and Quitangonha, Mozambique. 1868. Fihagerrhis unguiculata Giinther, Ann. Mag. Nat. Hist., (4) \. p. 422, pi. xix, fig. G: Zanzibar. 1870. Coelopeltis porrectus Jan, Icon. Gen. Ophid., livr. 34, pi. ii, fig. 1: Africa? Range. Sudan east to Somalia, south through Kenya Colony ; Tanganyika Territory and Zanzibar Island to Mozambique, west through Nyasaland and Northern Rhodesia to southeastern Belgian Congo. Rhamphiophis oxyrhynchus oxyrhnchus (Reinhardt) Western Brown Beaked-Snake 1843. Psammophis oxyrhynchus Reinhardt, Dansk. Vidensk, Selsk. Skrift., 10. p. 244, pi. i, figs. 10-12: Guinea. 1929. Ehamphiophis connali Parker, Ann. Mag. Nat. Hist., (10) 4, p. 450: Accra, Gold Coast. Range. Uganda, west through northern Belgian Congo to Nigeria; Togo; Gold Coast; French Soudan. Genus DROMOPHIS Peters "° 1843. Philodendros Fitzinger, Syst. Rept., p. 26. Type by original 140 Unfortunately this name is antedated by Philodendros, a most misleading name for a terrestrial snake that bears a strong resemblance to Paammophia. The sooner a ruling to reject Philodendros can be obtained, the better. 278 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY designation: Dendrophis praeornata Schlegel. 1846. Philodendrus Agassiz, Nomen. Zool., Index Univers., p. 285. Emendation for Philodendros Fitzinger. 1869. Dromophis Peters, Monatsb. Akad. Wiss. Berlin, p. 447. Type by monotypy: Dendrophis praeornata Schlegel. Dromophis lineatus (Dumeril & Bibron) Buff-striped Grass-Snake 1854. Dryophylax lineatus Dumeril & Bibron, Erpet. Gen., 7. p. 1124: White Nile, Sudan. 1901. Psammopliis sibilans tumbensis Schenkel, Verb. Naturf. Ges. Basel, 13, p. 172: Tumbo Island, French Guinea. 1902. Psammopliis hrevirostris temporalis Werner, Verb. Zool. Bot. Ges. Wien, 52. p. 335: Coja, Togo. Range. Sudan and Uganda, south along Lake Tanganyika, Tanganyika Territory to noilhern Nyasaland and Northern Rhodesia, northwest through the Belgian Congo to Portuguese Guinea (though as yet not recorded from the Gold Coast and Sierra Leone ) . Genus PSAMMOPHIS Boie"^ 1819. Macrosoma Leach (not Hubner:1818), in Bowdich, Miss. Ashantec, App., p. 493. Tj'pe by monotypy: Coluber elegans Shaw. 1826. Psammophis Boie, Isis von Oken, 19, col. 982. Type by monotypy: Coluber sibilans Linnaeus. 1826. Psammophis Fitzinger, Neue Class. Rept., pp. 29, 30. Type accord- ing to Boie: Coluber sibilams Linnaeus. 1827. Psa7nmophis Boie (part), Isis von Oken, 20, col. 521. Type by specific designation: Cohiber sibilans Linnaeus. 1838. Taphrometopon Brandt, Bull. Acad. Sci. St. Petersburg, 3, p. 243. Type by monotypy: Coluber (T.) Uneolatus Brandt. 1854. Chorisodon Dumeril & Bibron, Erpet. Gen., 7. p. 901. Type by monotypy: C. sibiricum Dumeril & Biljron = Coluber Uneolatus Brandt. 1854. Monodiastema G. Bibron, in Dumeril & Bibron, Erpet. Gen., 7, p. 901. Type by monotypy: Chorisodon sibericum Dumeril & Bibron = Coluber (Taphrometopon) Uneolatus Brandt. 1868. Phayrea Theobald, Cat. Rept. Asiatic Soc. Mus., p. 51. Type by monotypy: P. isabellina Theobald = Coluber condanarits Mer- rem. 141 Owing to some confusion as to authorship I have cited all three of tlic earliest records ; Fitzinger attributes the name to Boie. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 279 1872. Amphiophis Boeage, Jorn. Sci. Lisboa, 4. p. 81. Type by mono- typy: A. angolensis Boeage. 1892. Ehamphophis Boulenger, Zool. Eec. for 1891, p. 12. New name for Amphiophis Boeage, preoecupied. 1924. Mike Werner, Sitzb. Akad. Wiss. Wien, 133, Abt. 1, p. 51. Type by monotypy: M. elcgantissimus Werner = Coluber condanarus Merrem. Psommophis punctulcrtus trivirgotus Peters Southern Speckled Sand-Snake 1878. Psavimophis punctulatus var. trivirgatus Peters, Monatsb, Akad. Wiss. Berlin, p. 206 "Taita," i.e. Teita, Kenya Colony. Range. British Somaliland and Somalia, south through Uganda ( ?) and Kenya Colony to Arusha and Same, Tanganyika Territory. Psommophis sibilons sibilons (Linnaeus) Hissing Sand-Snake 1758. Coluber sibilans Linnaeus (part), Syst. Nat., ed. 10, 1, p. 222: "Asia," (error). 1802. Coluber Gemmatus Shaw, Gen. Zool., 3, p. 535: No locality. 1803. Coluber moniliger Daudin, Hist. Nat. Kept., 7, p. 69: No locality. 1827. Coluber auritus Geoffroy Saint-Hilaire, Kept., in Descr. Egypte, pp. 147, 151, pi. viii, fig. 4: Egypt. 1856. Psammophi^i irregularis Fischer, Abhand. Nat. Ver. Hamburg, 3. p. 92: Peki, Gold Coast. 1881. Psammophis brevirostris Peters, Sitzb. Ges. Naturf. Freunde Berlin, p. 89: Xa-Matlale, Mozambique. 1882. Psammophis sibilaiis var. mossambica Peters, Reise nach Mossam- bique, 3, p. 122: Mozambique Island, Mozambique. 1882. Psammophis sibilant var. tettensis Peters, Reise nach Mossambique, 3, p. 122: Tete and Mozambique Island, Mozambique. 1884. Psammophis sibilans var. intermedius Fischer, Jahrb. Hamburg. Wiss. Anst., 1, p. 14: Arusha, Tanganyika Territory. 1887. Psammophis sibilans var. leopardinus Boeage, Jorn. Sci. Lisboa, 11. p. 206: Catumbela and interior Mossamedes, Angola. 1908. Psammophis thomasi Gough, Ann. Transvaal Mus., 1, p. 30, fig. — : Salisbury, Southern Rhodesia. Range. In suitable savanna country from Egypt south through Uganda; Kenya Colony; Tanganyika Territory and Zanzibar Island to Natal ; northwest to Southwest Africa and on to Mauretania (outside of forested areas occupied by P. s. phil- lipsii) ; east to the Sudan. 280 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Psominophis subtaeniatus sudanensis Werner Northern Stripe-bellied Sand-Snake 1919. Psainmophis suhtaeniattis var. sudanensis Werner, Denks. Akad. Wiss. Wien, 96. p. 504: Kadugli, Sudan. Range. Drier regions of the southern Sudan and southern Ethiopia, south through Uganda; Kenj^a Colony; Tanganyika Territory ; Zanzibar and Mafia Islands, to northern Mozambique and Nyasaland."^ Psammophis biseriatus tangcmicus Loveridge Western Link-marked Sand-Snake 1940. Psammophis biseriatus tanganicus Loveridge, Bull. Mus. Comp. Zool., 87, p. 57: Mangasini, Usandawi, Tanganyika Territory. Range. Southern Libya, southeast through the Sudan to Eritrea, British Somaliland and Ethiopia (where it meets with P. h. hiseriatus along the Ethiopian-Somali border north of the Nogal River) ; south through Uganda to western Tanganyika Territory. Psammophis biseriatus biseriatus Peters Eastern Link-marked Sand-Snake 1881. Psammophis biseriatus Peters, Sitzb. Ges. Naturf. Freunde Berlin, p. 88: "Taita," i.e. Teita, Kenya Colony. 1913. Psammophis bitaeniatus (lapsus for biseriatus) Boettger, in Voeltzkow, Eeise in Ost-Af rika, 3, p. 355 : Patta Island, Kenya Colony. Range. Somalia, south through eastern Kenya Colony to northeastern Tanganyika Territory (where it meets with P. h. tanganicus in the vicinity of Lake Manka). Psammophis angolensis (Boeage) Pigmy Sand-Snake 1872. Amphiophis angolensis Socage, Jom. Sci. Lisboa, 4. p. 82: Dondo, Cuanza River, Angola. 1877. Ablabes Homeyeri Peters, Monatsb. Akad. Wiss. Berlin, p. 620: Pungo Ndongo, Angola. Range. Tanganyika Territory and Zanzibar Island, south to Mozambique, west through Nyasaland and Northern Rhodesia to Angola and the Belgian Congo. 142 The typical form P. s. subtaeniatus Peters (with its synonyms P. moniliger var. bilineaius Peters :1S67 ; P. locagii Boulenger :1895 ; and P. tratisvaalenais Gough:1908), characterized by three labials entering the orbit, occurs in central Mozambique and extends right across southern Africa south of the Zambezi. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 281 Genus CALAMELAPS Gunther"' 1849. Choristodon A. Smith (not Jonas :1844), Illus. Zool. S. Africa, Kept., App., p. 18: Type by monotypy: C. concolor Smith. 1866. Calamelaps Giinther, Ann. Mag. Nat. Hist., (3) 18, p. 26. Type by original designation: Calamaria unicolor Eeinhardt. 1947. Choristocalamus Witte & Laurent, Mem. Mus. Hist. Nat. Belgique, 29, p. 16. New name for Choristodon Smith (preoccupied). Calamelaps unicolor unicolor (Reinhardt) "* Seventeen-scaled Purple-Glossed Snake 1843. Calamaria, unicolor Eeinhardt, Dansk. Vidensk. Selsk. Skrift., 10, p. 236, pi. i, figs. 1-3: Guinea. 1877. Atractaspis Eildebrandtii Peters, Monatsb. Akad. Wiss. Berlin, p. 616, pi. — , fig. 3: Zanzibar Coast, i.e. Tanganyika Territory. 1923. Calamelaps niangarae Schmidt, Bull. Amer. Mus. Nat. Hist., 49, p. 117, fig. 12: Niangara, Belgian Congo. Range. Uganda ; Kenya Colony and Tanganyika Territory, west to Sierra Leone. Calamelaps unicolor warreni Boulenger Nineteen-scaled Purple-glossed Snake 1908. Calamelaps warreni Boulenger, Ann. Natal Mus., 1. pp. 230, 234, fig. 3 : Kosi Bay, Zululand. 1915. Calamelaps mellandi Boulenger, Proc. Zool. Soc. London, p. 214: Chirini Island, Lake Bangweulu, Northern Rhodesia. Range. Kenya Colony and Tanganyika Territory, south through Mozambique and the Rhodesias to the Transvaal and Zululand, Natal. Calamelaps unicolor polylepis Bocage Twenty-one scaled Purple-glossed Snake 1873. Calamelaps polylepis Bocage, Jorn. Sci. Lisboa, 4, p. 216: Dondo, Cuanza River, Angola. i-*3 Under the name of Choristocalamus, Witte and Laurent (loc. cit.) would revive Smith's monot.vpic genus Choristodon on the grounds that it has retained an anterior temporal, a seventh upper labial, and posterior sublinguals — charac- ters that seem inadequate to me. On the other hand their action in synonymizing Rhinocalamus with Cala- melaps has much to commend it, but the genotype - — which is the only species known to me — occupies so intermediate a position in relation to certain other genera that I have preferred to retain it until further work has been done on the group. 144 Only 17-scaied snakes occur in West Africa north of the Equator ; in the east they appear to be associated with the surviving montane or gallerv forests, but occasional individuals (about one in twenty) occur at Llwale in southeastern Tanganyika Territory where 19-scaled specimens predominate. For these the nam*- C. u. warreni is retained — at least until its distribution is worked out in rela- tion to that of C. u. polylepis which appears to have only 21 scales in Angola. 282 BU1.LETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1888. Calnmelaps miolepis Giinther, Anu. Mag. Nat. Hist., (6) 1, p. 323: Cape Maclear, Lake Nyasa, Nyasaland. Rcmge. Tanganyika Territory {fide a single old record from Tukuyu), south to Nyasaland; Southern Rhodesia; Angola. Genus RHINOCALAMUS Giinther 1888. Ehinocalatmis Giinther, Ann. Mag. Nat. Hist., (6) 1, p. 322. Type by monotypy : ff. dimidiatus Giinther. Rhinocolaxnus dimidiatus Giinther 1888. Bhino calamus dimidiatus Giinther, Ann. Mag. Nat. Hist., (6) 1, p. 322, pi. xix, fig. C: Mpwapwa, Ugogo, Tanganyika Territory. Range. Tanganyika Territory (known only from the type locality). Genus MICRELAPS Boettger 1880. Micrelaps Boettger, Ber. Senckenberg. Naturf. Ges., p. 136. Type by monotypy : M. mixlleri Boettger. 1888. Elaposchema Mocquard, Mem. Cent. Soc. Philom. Paris, p. 122. Type by monotypy: E. vaillanti Mocquard. Micrelaps bicolorotus Sternfeld 1908. Micrelaps bicoloratus Sternfeld, Sitzb. Ges. Naturf. Freunde Ber- lin, p. 93 : Kibwezi, Kenya Colony. 1908. BhinocMamus melcagris Sternfeld, Mitt. Zool. Mus. Berlin, 4, p. 244, fig. 4: Lamu Island, Kenya Colony. Range. Kenya Colony and Tanganyika Territory (N.E. of Lake Manyara). Genus AMBLYODIPSAS Peters 1856. Amblyodipsas Peters, Monatsb. Akad. Wiss. Berlin, p. 592. Type by monotypy: Calamaria microphthalma Bianconi. Amblyodipsas katangensis ionidesi Loveridge 1951. Amblyodipsas hatangensis ionidesi Loveridge, Bull. Mus. Comp. Zool., 106, p. 193: Tunduru, Southern Province, Tanganyika I'erritory. Range. Tanganyika Territory. LOVERIDQE: E. AFRICAN REPTILES AND AMPHIBIANS 283 Genus MIODON Dumeril"' 1858. Microsoma Jan (not Macquart:1855), Revue Mag. Zool,, (2) 10. p. 519. Type by monotypy: M. neuwiedii Jan. 1859. Miodon A. Dum6ril, Arch. Mus. Hist. Nat. (Paris), 10. p. 206. Type by monotypy: Elapomorphus gahonensis A. Dum^ril. 1860. Urobelus Reinhardt, Vidensk. Meddel, Kjebenhavn, p. 229. Type by monotypy : U. acanthias Reinhardt. 1902. Cynodontophis Werner, Verb. Zool. Bot. Ges. Wien, 52. p. 346. Type by monotypy: C. aemulans Werner = Microsoma notatum Peters. 1941. Melanocalamu^ Witte, Explor. Pare. Nat. Albert, Miss. G. F. de Witte, Fasc. 33, p. 216. Type by monotypy: M. leopoldi Witte. Miodon colloris collaris (Peters) Pale-collared Snake-eater 1881. Microsoma collare Peters, Sitzb. Ges. Naturf. Freunde Berlin, p. 148: Macange, "Cuango" i.e. Kwango, French Congo. 1887. Microsoma fulvicollis Mocquard, Bull. Soc. Philom. Paris, (7) 11. p. 65 : Franceville, French Congo. 1888. Elapomorphus caeutiens Giinther, Ann. Mag. Nat. Hist., (6) 1, p. 323, pi. xix, fig. B: Cameroon Mountain, British Cameroon. 1910. Cynodontophis werneri L. Miiller, Abhand. Bayer Akad. Wiss., 2 Kl., 24. p. 612: Cameroon. Range. Southwestern Uganda, west to Cameroon and possibly Nigeria. NCodon collaris chxistyi Boulenger Eastern Snake-eater 1903. Miodon Christyi Boulenger, Ann. Mag. Nat. Hist., (7) 12. p. 354: Uganda. 1923. Miodon unicolor Schmidt, Bull. Amer. Mus. Nat. Hist., 49, p. 119, fig. 13 : Poko, Belgian Congo. Range. Central Uganda, south to western Tanganyika Terri- tory, west to eastern Belgian Congo. 1*5 It is impossible at this time to investigate the numerous changes proposed by Witte and Laurent (1947, Mem. Mus. Kov. Hist. Nat., 29, pp. 58-89) but two ^ 5 gahonensis (M.C.Z. 49727, 53737) from the Gold Coast (which ther would perhaps assign to neuwiedi Jan) have 218 and 233 ventrals respectively, while two $ $ (M.C.Z. 53738, etc.) have 247 and 259. This would appear to effectively dispose of the alleged race M. g. schmidti Witte and Laurent from the Congo. On zoogeographical grounds, together with size and color pattern, I accept their separation of gahonensis from the collaris group, but not on the basis of adult eye diameter — being at least twice its distance from the mouth in gahonensis^ as opposed to less than twice its distance in collaris, fulvicollis, etc. Our material fails to support this allegation for the lirst two collaris — taken at random — had eye diameters that were included 1 Ms (M.C.Z. 42956 :Congo) to 2 or rather more than 2 times (M.C.Z. 9254 :Cameroon) their distance from the border of the mouth — a very poor character indeed. 284 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Miodon collaris groueri Sternfeld Central-Lake-Region Snake-eater 1908. Miodon Graueri Sternfeld, Sitzb. Ges. Naturf. Freuude Berlin, p. 94: "Entetbe," i.e. Entebbe, Uganda."^ Rcoige. Central Uganda, southwest to eastern Belgian Congo (L. Kivu). Genus CHILORHINOPHIS Werner"' 1907. ChilorhinopMs Werner, Akad. Anz. Wien, 44, p. 479 (brief notice) and 1908 (for 1907), Sitzb. Akad. Wiss. Wien, 116. Abt. 1, p. 1881. Type by monotypy: C. butleri Werner. 1927. Parkerophis Barbour & Amaral, Bull. Antivenin Inst. America, 1. p. 25. Type by original designation: Apostolepis gerardi Bou- lenger. Chilorhinophis gerardi tanganyikae Loveridge Tanganyika Two-headed Snake 1951. Chilorhinophis gerardi tanganyikae Loveridge, Bull. Mus. Conip. Zool., 108, p. 195: Nyarakolo, Lake Tanganyika, Northern Rho- desia. Range. Western Tangaujaka Territory (Ujiji), south to Northern Rhodesia (Nyamkolo), west to southern Belgian Congo (Lukonzolwa). Chilorhinophis carpenteri liwalensis Loveridge Liwale Two-headed Snake 1951. Chilorhinophis carpenteri liwalensis Loveridge, Bull. Mus. Comp. Zool., 106, y. 196: Liwale, Southern Province, Tanganyika Terri- tory. Range. Southeastern Tanganyika Territory. Genus APARALLACTUS Smith 1849. Aparallactus A. Smith, lUus. Zool. S. Africa, Kept., App., p. 15. Type by monotypy: A. capensis Smith. 1849. Elapomorphus A. Smith (not of Wiegmann : in Fitzinger :1843), ^46 Entebbe may be viewed with suspicion for Capt. C. R. S. Pitman, wlio re- sided at Entebbe for many years, failed to find it there and doubts its occurrence. Allegedly Grauer, like some other collectors, failed to label his material immedi- ately. The nearest point to Entebbe from which the Museum of Comparative Zoology has c. graueri, is Kingani, 5000 feet, near Fort Portal (M.C.Z. 54737 : coll. J. H. Blower, Esq. xi.l955). 147 The Amani, Usambara Mountains, record of the Sudanese butleri is believed to have lieen based on a misidentification, possibly of an Aparallactus werneri. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 285 Illus. Zool. S. Africa, Kept., App. p. 16, footnote. Type by monotypy: E. capensis Smith. 1854. Uriechis Peters, Monatsb. Akad. Wiss. Berlin, p. 623. Type by subsequent designation : U. lunulatus Peters. 1859. Elapops Giinther, Ann. Mag. Nat, Hist., (3) 4. p. 161. Type by monotypy : E. modestus Giinther. 1860. Periaspis Cope, Proe. Acad. Nat. Sci. Philadelphia, pp. 241, 266. Type by monotypy: P. plumheatra Cope = Elapops modestus Giinther. 1863. Cercocalamus Giinther, Ann. Mag. Nat. Hist., (3) 11. p. 21. Type by monotypy: C. collaris Giinther = Aparallactus capensis Smith. 1870. Metopophis Peters, Monatsb. Akad. Wiss. Berlin, p. 643, pi. i, fig. 3. Type by monotypy: Uriechis lineatus Peters. 1917. Houleophvi Chabanaud, Bull. Mus. Hist. Nat. (Paris), 1916, 22. p. 379. Type by monotypy: R. ch.evalieri Chabanaud z= Uriechis lineatus Peters. 1923. GuyoTtutrchia Angel, Bull. Mus. Hist. Nat. (Paris), 29, p. 348. Type by monotypy: G. unicolor Angel = Elapops modestus Giinther. Aparallactus modestus (Giinther)"^ Western-Forest Centipede-eater 1859. Elapops modestus Giinther, Ann. Mag. Nat. Hist., (3) 4. p. 161, pi. iv, fig. C: West Africa. 1860. Periaspis plumheatra Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 242 : Liberia. 1896. Aparallactus houlengeri Werner, Verh. Zool. Bot. Ges. Wien, 46. p. 363, pi. vi, figs. 6-6b : Cameroon. 1897. Aparallactus peraffinis Werner, Verh. Zool. Bot. Ges. Wien, 47. p. 404, pi. ii, fig. 3 : Interior of Cameroon. 1897. Aparallactus ubangensis Boulenger, Ann. Mag. Nat. Hist., (6) 19. p. 279, fig. — : Zongo, Ubangi Rapids, Belgian Congo. 1901. Aparallactus flavitorques Boulenger, Ann. Mus. Congo, (1) 2, p. 11, pi. iv, fig. 3: Lubue, Kasai, Belgian Congo. 148 Material of modestus in the Museum of Comparative Zoology lends no sup- port whatever to the action of Witte and Laurent (1947, pp. 99, 103) in reviving ubangensis as a race of modestus, to whose synonymy I had relegated it in 1944 (p. 187). The lower labials do not conform to their key and, taken from East to West we find the parietal is in contact with the 5th upper labial only (1 side) or .5th and 6th (5 sides) or 6th upper labial (2 sides) in Uganda snakes. 5th only (2 sides) on a ^ (M.C.Z. 53994) from Mayala, B. C. parietal in contact with 6th upper labial (M.C.Z. 42944) at Makaia, B. C. 5th and 6th (4 ex.) or 6th upper labial (2 ex.) at Bitye, F. C. oth and 6th (1 ex.) at Accra, G. C. 5th and 6th (1 ex.), or 6th upper labial (1 ex.) in i.iberia. In other words, both m. modestus and m. ubangensis occur in Liberia ; Gold Coast (or possibly not ni. modestus); French Cameroon; Belgian Congo and Uganda. 286 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1902. Aparallactus dolloi Werner, Verh. Zool. Bot. Ges. Wien, 52, p. 346: Banzyville, Ubangi River, Belgian Congo. 1902. Aparallactus congicus Werner, Verh. Zool. Bot. Ges, Wien, 52. p. 346 : Lingunda, Belgian Congo. 1907. Aparallactus Bat.esii Boulenger, Ann. Mag. Nat. Hist., (7) 19, p. 325: 5 miles inland from Kribi, French Cameroon. 1910. Aparallactus Cliristyi Boulenger, Ann. Mag. Nat. Hist., (8) 5, p. 512: Mabira Forest, "Chagwe, " i.e. Kyagwe, Uganda. 1917. Aparallactus nigrocollaris Chabanaud, Bull. Mus. Hist. Nat. (Paris), 1916, 22. p. 377, figs. 18-19: French Congo. 1917. Aparallactus nigrocollaris Roucheti Chabanaud, Bull. Mus. Hist. Nat. (Paris), 1916, 22. p. 378, figs. 20-21: French Congo. 1923. Guyomarchia unicolor Angel, Bull. Mus. Hist. Nat. (Paris), 29. p. 348, figs. 1-4: (probably near Sangha) French Congo. 1924. Aparallactus Graueri Werner, Sitzb. Akad. Wiss. Wien, 133. Abt. 1, p. 42: Beni, Belgian Congo. Range. Uganda, west to Sierra Leone. Aparallactus lunulotus lunulotus (Peters) Southeastern Plumbeous Centipede-eater 1854. Uriechis lunulotus Peters, Monatsb. Akad. Wiss. Berlin, p. 623: Tete, Mozambique. Range. Sudan and Belgian Congo, south to Uganda (Mt. Rom) and Tanganyika Territory ; Mozambique ; Nyasaland ; the Rho- desias and Transvaal. Aparallactus liuiulatus concolor (Fischer) Northeastern Plumbeous Centipede-eater 1884. Uriechis concolor Fischer, Jahrb. Hamburg. Wiss. Anst., I. p. 4, pi. i, fig. 1 : Arusha, Tanganyika Territory. 1931. Aparallactus concolor boulengeri Scortecci (not Werner :1896), Atti. Soc. Sci. Nat. Ital. (Milano), 70. p. 212: Villaggio Duca degli Abruzzi and inland from Mogadiscio, Somalia. Range. Sudan, east to Eritrea, south through eastern Kenya Colony to northern Tanganyika Territory. Aparallactus jacksonii jacksonii (Giinther) Kilimanjaro Centipede-eater 1888. Uriechis Jacksonii Giinther, Ann. Mag. Nat. Hist., (6) 1, p. 325, pi. xix, fig. E: Foot of Mount Kilimanjaro, Tanganyika Terri- tory. Range. Southern Ethiopia, south through Kenya Colony to Tanganyika Territory."** 149 A race (A. j. oweni Loveridge) occurs at Torit in the southern Sudan. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 287 Aparollactus wemeri Boiilenger Usambara Centipede-eater 1895. Aparallactus Wemeri Boulenger, Ann. Mag. Nat. Hist., (6) 16< p. 172: Usambara Mountains, Tanganyika Territory. Range. Eastern Tanganyika Territory. Aparallactus turner! Loveridge Malindi Centipede-eater 1935. Aparallactus turneri Loveridge, Bull. Mus. Comp. Zool., 79, p. 9: Sokoki Forest, near Malindi, Kenya Colony. Range. Coastal Kenya Colony (from Lamu to Mombasa). Aparallactus guentheri Boulenger ^^° Zanzibar Centipede-eater 1895. Aparallactus guentheri Boulenger, Ann. Mag. Nat. Hist., (6) 16. p. 172: "East and Central Africa" (omit Angola). Subse- quently types were designated as from Zanzibar; Shire High- lands and Lake Nyasa, Nyasaland. 1928. Aparallactus uluguruensis Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 132: Nyange, Uluguru Mountains, Tanganyika Territory. Range. Coastal Kenya Colony; Tanganyika Territory and "Zanzibar," south to Nyasaland. Aparallactus capensis capensis Smith Cape Centipede-eater 1849. Aparallactus capensis A. Smith, Illus. Zool. S. Africa, Eept., App., p. 16: "Kaffirland eastward to Cape Colony," i.e. Natal, South Africa. 1849. Elapomorphus capensis A. Smith, Illus. Zool. S. Africa, Rept. App., p. 16: "Kaffirland eastward of Cape Colony," i.e. Natal, South Africa. 1863. Cercocalamus collaris Giinther, Ann. Mag. Nat. Hist., (3) 11. p. 21, pi. iii, fig. A: "Central America" (error). 1895. Aparallactus punctatolineatus Boulenger, Ann. Mag. Nat. Hist., (6) 16. p. 173: Angola. Range. Tanganyika Territory; Zanzibar and Mafia Islands: south through Mozambique ; Nyasaland and the Rhodesias to Cape Province (East London), northwest through Bechuanaland to Angola. Subfamily DASYPELTINAE Genus DASYPELTIS Wagler'^' 150 A., guentheri was based on the ring-necked young, uluguruensis on the uni- formly plumbeous adult. 151 statements to the contrary notwithstanding, the gender of Dasypeltis is feminine. At the time this check list goes to press the genus is undergoing re- visionary study by Carl Gans. 288 , BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 1829. Anodon A. Smith (not of Oken:1815),i" Zool. Journ., 4, p. 443. Type by monotypy: A. typiis Smith = Coluber scaber Linnaeus. 1830. Dasypeltis Wagler, Nat. Syst. Amphib., p. 178. Type by mono- typy: Coluber scaber lavamsievis. 1830. Analcis Wagler, Nat. Syst. Amphib., p. 191, footnote 3. New name for Anodon A. Smith (preoccupied). 1833. Rachiodon Jourdan, Le Temps (Paris:13.vi.l833), p. ?. New name for Anodon A. Smith (preoccupied). 1834. Eaphiodon Jourdan (misprint) L'Institut (de la France: Paris) 2, p. 214, corrected to Rachiodon on p. 223. 1845. Deirodon Owen, Odontography; or, A Treatise on the Comparative Anatomy of the Teeth . . . (London), p. 220. New name for Anodon A. Smith (preoccupied). 1846. Dirodon Agassiz, Nomen. Zool., Index Univers., p. 127. Emenda- tion for Deirodon Owen. Dasypeltis scabra (Linnaeus) ^'"^ Common Egg-eater 1758. Coluber scaher Linnaeus, Syst. Nat., ed. 10, 1, p. 223: "in Indis, " i.e. South Africa.i^* 1829. Anodon typus A. Smith, Zool. Journ., 4, p. 443: Near Capetown, Cape Province, South Africa. 1853. Rachiodon abyssinus A. M. C. Dumeril {nom&n nudum), Mem. Acad. Sci. (Paris), 23, p. 71: A casual reference to Cartin (misspelled Quartin) Dillon's specimen from Ethiopia. 1854. Rachiodon Abyssinus Dumeril & Bibron, Erpet. Gen., 7, p. 496: * ' Abyssinia, ' ' i.e. Ethiopia. 1864. Dasypeltis scaber var. capensis Peters, Monatsb. Akad. Wiss. Ber- lin, p. 644, footnote: Cape of Good Hope, i.e. South Africa. 1864. Dasypeltis scaber var. mossambicus Peters, Monatsb. Akad. Wiss. Berlin, p. 644, footnote: Boror and Tete, Mozambique. 1864. Dasypeltis scaber var. breviceps Peters, Monatsb. Akad. Wiss. Ber- lin, p. 645, footnote: "Kaf&rland," i.e. South Africa. 1878. Dasypeltis lineolatris Peters, Monatsb. Akad. Wiss. Berlin, p. 206: Kitui, Ukamba, Kenya Colony. 1912. Dasypeltis scabra var. atra Sternfeld, Wiss. Ergebn, Deutsch. Zentral-Afrika-Exped. 1907-1908, 4, p. 272 : Virgin forest behind boundary mountains on northwest shore of Lake Tanganyika, Belgian Congo. 1954. Dasypeltis scabra loveridgei Mertens, Zool. Anz., 152. p. 213: 152 Suppressed in favor of Dasypeltis in 1956 (cf. Opiu. Declar. Inter. Comm. Zool. Nomencl., 12, p. 241. Opinion 387). 153 Dr. Gans points out that the species scabra, whicli may possibly be subject to some racial subdivision, includes the specimens that in 1942 I referred to "s. scaber" and "s. pabnariim." It is on Dr. Gans' advice that medici and fasciata are now treated as full species. 15* Cf. Flower, 1933, Proc. Zool. Soc. London, p. 818. LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 289 Finkenstein Farm near Windhoek, Southwest Africa. Range. Southern Arabia and Somalia, south to Natal, west to the Cape, northwest to the Gold Coast, east through the southern Sudan to Ethiopia. Dasypeltis medici lamuensis Gans 1957. Dasypeltis medici lamuensis Gans, Breviora (Cambridge, Mass.), No. 79, p. 1 : Lamu Island, Kenya Colony. Range. Reddish laterite or gray alluvial soils of coastal Som- alia, south through Kenya Colony to the vicinity of Kilimanjaro on the frontier of Tanganyika Territory. Dasypeltis medici medici (Bianconi) 1859. Dipsas Medici Bianconi, Mem. Accad. Sci. Bologua, 10, p. 501, pi. xxvi (reprinted as p. 277 of Spec. Mossamb.) : Mozambique. 1868. Dasypeltis scaber var. fasciolata Peters, Monatsb. Akad. Wiss. Berlin, p. 451 ; Zanzibar Island. 1888. Dasypeltis elongata Mocquard, Mem. Cent. Soc. Philom. Paris, p. 131, pi. xii, fig. 2: Zanzibar Island. 1893. Dasypeltis scabra var. bianconii "Med." Boettger (lapsus for var. medici Bianconi), Zool. Auz., 16. p. 133. Range. Reddish laterite soils of the Kenya Colony — Tan- ganyika Territory frontier (as far inland as Kilimanjaro) ; Zan- zibar and Mafia Islands ; south to Mozambique and Nyasaland. Dasypeltis fasciata (Smith) Western-Forest Egg-Eater 1844. Dipsas carinatus Hallowell (not of Sehlegel:1837), Proc. Acad. Xat. Sci. Philadelphia, p. 119: "Africa," later (1857, loc. cit.. p. 69) amended to Liberia. 1849. Dasypeltis fasciaius A. Smith, Illus. Zool. S. Afi-ica, Eept., foot note to pi. Ixxiii : Sierra Leone. 1859. Bachiodon siibfasciatus A. Dumeril (nomen nudum), Arch. Mus. Hist. Nat. (Paris), 10, p. 198: casual mention of name. 18G3. Eachiodon. scaber Tar. sntbfasciatus Jan (nomen nudum), Elenco Sist. Degli Ofidi, p. 100: Gold Coast. 1863. BacModon scaber var. %micolor Jan (part: ■nomen nudum), Elenco Sist. Degli Ofidi, p. 106: Gold Coast. 1907. Dasypeltis macrops Boulenger, Ann. Mag. Nat. Hist., (7) 19, p. 324: Efulen, French Cameroon. Range. Western Uganda, west through the Belgian Congo and Cameroons to Gambia. 290 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Family ELAPIDAE Geniis ELAPSOroEA Bocage 1866. Elapsoidea Boeage, Jorn. Sci. Lisboa, 1. pp. 50, 70. Type by monotypy : E. giintherii Boeage. 1896. Elapechis Boulenger, Cat. Snakes Brit. Mus., 3, p. 358, footnote. Substitute name for Elapsoidea Boeage. EQopsoidea sundevallii giintherii Boeage Western Garter-Snake 1866, Elapsoidea Giintherii Boeage, Jorn. Sei. Lisboa, 1. pp. 50 ,70, pi. i, figs. 3-3b: Bissau, Portuguese Guinea and Cabinda, Portuguese Congo. 1897. Elapechis moebiusi Werner, Verb. Zool. Bot. Ges. Wien, 47. p. 400: Kete, Gold Coast. Range. Northern and Western Tanganyika Territory, through western Kenya Colony and Uganda, west to Senegal. Elapsoidea sundevallii loveridgei Parker Northeastern Garter-Snake 1949. Elapsoidea sundevallii loveridgei Parker, Zool. Verhand. Rijksmus, Nat. Hist. Leiden, No. 6, p. 95 : Machakos, Kenya Colony. Range. Central and eastern Kenya Colony. Elapsoidea sundevallii nigra Giinther Eastern Garter-Snake 1888. Elapsoidea nigra Giinther, Ann. Mag. Nat. Hist., (6) 1, p. 322: ' ' Ushambola, Zanzibar, ' ' i.e. Usambara Mountains, Tanganyika Territory. Range. Virgin forests of the Usambara; Magrotto and Ulu- guru Montains, eastern Tanganyika Territory. Elapsoidea simdevallii decosteri Boulenger Southeastern Garter-Snake 1888. Elapsoidea Decosteri Boulenger, Ann. Mag. Nat. Hist., (6) 2. p. 141 : Delagoa Bay, Mozambique. 1895. Elapsoidea Boulengeri Boettger, Zool. Anz., 18, p. 62: Boroma, Zambezi, Mozambique. Range. Southern Province of Tanganyika Territory, south through Mozambique, Nyasaland and the Rhodesias to the Trans- vaal. Genus BOULENGERINA Dollo'^' 1886. Boulengerina Dollo, Bull. Mus. Roy. Hist. Nat. Belgique, 4. p. 155 For Bogert's reasons for svnonymlzing Limnonaja, see Bogert, 1943, BuU. Amer. Mus. Nat. Hist., 81. pp. 296, 300. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 291 159. Type by monotypy: B. storTusi Dollo. 1923. Limnonaja Schmidt, Bull. Amer. Mus. Nat. Hist., 49, p. 124. Type by monotypy: Boulengerina christyi Boulenger. Bonlengerina annulata stormsi Dollo Tanganyika Water-Cobra 1886. Boulengerina stormsi Dollo, Bull. Mus. Eoy. Hist. Nat. Belgique, 4. p. 160, fig. — : Lake Tanganyika, Belgian Congo. Range. Lake Tanganyika and its shores off Tanganyika Terri- tory and Belgian Congo. ^^^ Genus NAJA Laurenti 1768. Naja Laurenti, Syn. Kept., p. 90. Type by tautonomy: Coluber naja Linnaeus. ^'^ 1820. Naia Merrem, Vers. Syst. Amphib., pp. 15 (bis), 147, 190. Emendation for Naja Laurenti (Naja on p. 15 (primus)). 1830. Uraeus Wagler, Nat. Syst. Amphib., p. 173. Type by monotypy: Coluber haje Linnaeus. 1830. Aspis Wagler (not Laurenti: 1768), Nat. Syst. Amphib., p. 173. Type by monotypy: Coluber naja Linnaeus. 1831. Tomyris Eichwald, Zool. Spec. Eossiae Poloninae, Part 3, p. 171. Type by monotypy : T. oxiana Eichwald ^ Coluber naja Lin- naeus. Naja haje haje (Linnaeus) Egyptian Cobra 1758. Coluber haje Linnaeus, Syst. Nat., ed. 10, 1. p. 225: Lower Egypt. 1768. Cerastes candidus Laurenti, Syn. Kept., p. 83: Libya. 1789. Coluber Candidissimus Lacep^de, Hist. Nat. Quad. Ovip. Serpens, 2. Table Methodique, p. 86, text p. 118: Libya. 1854. Naja Int.ermixta Dumeril & Bibron, Erpet. Gen., 7, p. 1299: Based on pi. xix of A. Smith, 1849, so South Africa. 1854. Naja haje var. annulifera Peters, Monatsb. Akad. Wiss. Berlin, p. 624: Tete, Mozambique. 1873. Naja haje var. viridis Peters, Monatsb. Akad. Wiss. Berlin, p. 411, pi. i, fig. 1 : West Africa. 156 The locality data for the two British Museum specimens allegedly from Nyasaland had best be regarded with suspicion pending confirmation. Mr. J. C. Battersby informs me (16.i.57) that they were collected by Surgeon H. M. Hanschell, R.N., for the London School of Tropical Medicine which, in 1926. turned them over to the British Museum. 157 Stejneger (1936, Copeia, p. 140) states that Laurenti's six included species are all synonyms of Coluber naja Linnaeus. Prior action by Boie (1826. Isis von Oken, col. 981) in designating Vipera haemachatea Latreille as type of Naja is inadmissable. Included in the synonymy by some authors is Lepidon Swainson (1839, 2, p. 345). However, no type was designated, the name simply appearing in a key between Naia and Elaps with the scant description : Head broad, neck not expansive. // East African this might apply to Pseudohaje or Elapsoidea. 292 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Range. Spanish Morocco east to Egypt, south through Uganda, Kenya Colony, Tanganyika Territory and Zanzibar Island. In fact almost all Africa except for the area of N. h. harotseensis Angel (with 15 midbody scale-rows) known only from Lealui, Northern Rhodesia ; and tlie area occupied by N. h. anchietae Bocage (with 17 midbody scale-rows) which ranges through the Rhodesias, Bechuanaland, Southwest Africa and Angola. An- other race occurs in Arabia. Naja nigricollis pallida Boulenger Pink-or-red Spitting-Cobra 1896. Naia nigricollis var. pallida Boulenger, Cat. Snakes Brit. Mus., 3. p. 379: Inland from Berbera, British Somaliland (restricted by Bogert:1942), also Lake Eudolf. Range. On red murram soils from northern Ghana and Sudan east to British Somaliland, south through Kenj^a Colony (Voi to Kibwezi) to Tanganyika Territory (vicinity of Longido Moun- tain) . Naja nigricollis nigricollis Reinhardt Common Spitting-Cobra 1843. Naja nigricollis Reinhardt, Danks. Vidensk. Selsk. Skrift., 10, p. 369, pi. iii, figs. 5 and 7: Guinea. 1854. Naja mossamhica Peters, Monatsb. Akad. Wiss. Berlin, p. 625: Sena and Tete, Mossambique. 1893. Naja nigricollis var. crawshayi Giinther, Proc. Zool. Soc. London, p. 620: Lake Mweru, Northern Rhodesia. 189.J. N'aja nigricollis var. occidontalis Bocage, Herp. Angola Congo, p. 135; Dondo; Huilla; Humbe; Quisange; and Quilengues, Angola. 1895. Naja nigricollis var. melanolcuca Bocage (not of Hallowell:1857), Herp. Angola Congo, p. 136: Caeonda and Catumbella, Angola. 1955. Naja nigricollis atriceps Laurent, Revue Zool. Bot. Afr., 51, p. 135 : Mugera, 1450 metres, Ruyigi Territory, Urundi, Belgian Ruanda-Urundi.^^^ Range. Savanna areas of almost all Africa south of 25° N. (except where occupied by recognizable races), thus including Uganda ; Kenya Colony ; Tanganyika Territory ; Pemba and Zanzibar Islands. 158 Specimens from Mangasini and other localities in Tanganyika Territory are listed among the paratypes. Like most cobras nigricollis is very variable as regards color, and sevei'al striking variants may occur in a single locality. Many races not listed heie have been described. An overall study of the genus is much to be desired. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 293 Naja melanoleuca Ilallowell Blaek-and-White Cobra ^'® 1857. Xaia haie var. melanoleuca Hallowell, Proc. Acad. Nat. Sci. Phila- delphia, p. 61 : Gabon, French Congo. 1885. Naja haje var. leucosticta Fischer, Jahrb. Hamburg. "Wiss. Anst., 2, p. 115, pi. iv, fig. 11 : Cameroon and Ogooue Kiver, French Congo. 1955. ^Naja melanoleuca sutfulva Laurent, Eevue Zool. Bot. Afr., 51, p. 132: Lwiro, 1850 metres, Kabare Territory, Kivu, Belgian Congo. Range. Forested areas of much of Africa south of 15° N. in the "West, to the Equator in the East, including Uganda, Kenya Colony, Tanganyika Territory and Zanzibar Island. Genus PSEUDOHAJE Gunther 1858. Pseudohaje Giinther, Cat. Snakes Brit. Mus., p. 222. Type by monotypy : P. nigra Giinther. Pseudohaje goldii (Boulenger) Black Forest-Cobra 1S95. Xaia goldii Boulenger, Ann. Mag. Nat. Hist., (6) 16, p. 34: near Asaba, Nigeria. 1896. NaAa yahoma Moequard, C.-E. Ill Congres Internat. Zool., Leyde, 1895, p. 233: Yakoma, Abiras District, northern Belgian Congo. Range. Forested areas of Uganda, southwest to Southwest Africa, northwest to Nigeria. Genus DENDROASPIS Schlegel 1848.^^° Dendroaspis Schlegel, Versl. Zool. Gen. Amsterdam, p. 5. Type by monotypy: Elaps jamesoni Traill. 1852. Dinophis Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 203. Type by monotypy: D. hammondii Hallowell = Leptophis viridis Hallowell. 1855. Dendroechis Fischer, Programm Realschule, Johann. Hamburg, pp. 20, 68. Type by monotypy: D. reticulata Fischer = Leptophis viridis Hallowell. 1856. Dendraspis A. Dum^ril (not of Fitzinger:1843), Revue Mag. Zool., (2) 8, p. 558. Emendation for Dendroaspis Schlegel. 159 The few specimens I have taken in the southeast (Mikindani, T. T. ; Nyasaland) were a variegated brown, not black as at Kaimosi, K. C. Whether the brown cobras really occupy a well-defined range so that Laurent's name might be applied to them, is not clear. Material in the Museum of Comparative Zoology suggests there Is no such division In scale-counts between southeast and west as is shown by Laurent's data. 160 Date uncertain, but communicated by Schlegel to a meeting held on March 13, 1848 (fide Brongersma, 1936, Zool. Meded., 19, p. 136). 294 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Dendroospis iamesoni koiinosae Loveridge Eastern Jameson's Mamba 1936. Dendraspis jainesoni kaimosae Loveridge, Proc. Biol. Soc. Wash- ington, 49. p. 64: Kaimosi, Kakamega, Kenya Colony. Range. Western Kenya Colony, west through Uganda to Belgian Ruanda-Urundi and the eastern Belgian Congo. Dendroaspis angusticeps (Smith) Common Green Mamba 1849. Naia angusticeps A. Smith, lUus. Zool. S. Africa, Rept., pi. Ixx: Natal, South Africa. 1865. Dendraspis intermedins Giinther, Ann. Mag. Nat. Hist., (3) 15, p. 97 : Zambezi Eiver, Mozambique. 1907. Dendraspis sjosiedti Lonnberg, ia Sjostedt, Wiss. Ergeb. Schwed. Zool. Exped. KUimandjaro, Meru umgeb Massaisteppen, No. 4, p. 17, pi. i, fig. 2: "Kibonoto, " i.e. Kibongoto, Mount Kiliman- jaro, Tanganyika Territory. Range. Kenya Colony; Tanganyika Territory and Zanzibar Island, south through Mozambique, Nyasaland and the Rhodesias to Natal.'" Dendroaspis polylepis antinorii Peters'"" Northern Brown-Mamba 1873. Dendraspis antinorii Peters, Monatsb. Akad. Wiss. Berlin, p. 411, pi. i, fig. 2: "Ansaba," i.e. Anseba, Eritrea. Range. Eritrea, south through Ethiopia and British Somali- land to Mount Moroto, northeastern Uganda, and Murri, northern Kenya Colony. Dendroaspis polylepis polylepis Giinther Southern Brown-Mamba 1864. Dendraspis polylepis Giinther, Proc. Zool. Soc. London, p. 310: Zambezi River, Mozambique. 1907. Dendraspis mamba Gough, Zool. Anz., 32, p. 454: White River Settlement, near Nelspruit, Lydenburg District, Transvaal. Range. Southern Kenya Colony and Tanganyika Territory, south through Mozambique ; Nyasaland and the Rhodesias to Natal, west to Southwest Africa, northeast through Angola to the eastern Belgian Congo. 161 This is the correct distribution, confused for a century by the inclusion of records pertaining to D. p. polylepis, etc. 162 For discussion on the validity of this race, see Parker (1949, Zool. Verb. Rijksmus. Nat. Hist. Leiden, No. 6, p. 98). This, and typical polylepis are the snakes commonly called "Black Mambas," though they are not black ; the young are grey green, somewhat darker than the vivid fresh green of young and adult angusticeps. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 295 Family HYDROPHIIDAE * Genus PELAMIS Daudin 1803. Pelajnis Daudin (part), Hist. Nat. Rept., 7, p. 361. Type by desig- nation of Fitzinger:1843 : P. bicolor Daudin =: Anguis platvra Linnaeus. ^^^ 1816. Pelamys Oken, Lehrb. Naturg., 3. Part 2, pp. vi, xv, 279. New name for Pelamis Daudin (part) with type by monotypy: Anguis platuro (sic) Linnaeus. 1817. Ophincctes Rafinesque, Amer. Monthly Mag., 1, p, 432. Type by present designation: 0. luteus Rafinesque.'^®* 1848. Elaphrodytes Gistel, Naturg. Thierr., p. ix. New name for Hydnis Schneider {fide Stejneger:1907. Not seen.). 1910. Pelamydrus Stejneger, Proc. U. S. Nat. Mus., 38, p. 111. Type by monotypy: Anguis platura Linnaeus. Pelomis ploturus (Linnaeus) Parti-colored Sea-Snake 1766. Anguis platura Linnaeus, Syst. Nat., ed. 12, 1, p. 351: No locality. 1799. Hydrus Bicolor Schneider, Hist. Amphib., 1, p. 242 (based on Scba, 1735, Thesaurus 2, pi. Ixxvii, fig. 2): Mexico? 1817. Pelamis sclineideri Rafinesque (new name for Hydrus bicolor Schneider, var.), Amer. Monthly Mag., 1, p. 432. 1817. Ophinectes luteus Rafinesque, Amer. Monthly Mag., 1. p. 433: No locality.^®^ 1837. Hydrophis pelamis Sehlegel (new name for Anguis platura Lin- naeus), Essai Phys. Serp., 1, p. 508. 1842. Pelamis ornata Gray, Zool. Misc., p. 60: Borneo. 1854. Pelamis bicolor var. variegata Dumeril & Bibron, Erpet. Gen., 7. p. 1337: Celebes. 1854. Pelamis bicolor var. sinuata Dumeril & Bibron, Erpet. Gen., 7, p. 1338 : No locality. 1856. Hydrophis (Pelamis) bicolor var. alternans Fischer (new name for variegata Dumeril & Bibron), Abhand. Nat. Ver. Hamburg, 3, p. 63. 1872. Hydrophis bicolor var. macidata Jan, Icon. Gen. Ophid., livr. 40, pi. iii: Indian Ocean. 163 For lengthy discussion on rejection cf Hydrus Schneider by Malcolm Smith (1926, Monogr. Sea-Snakes, p. 116), see H. M. Smith and B. H. Taylor (1945. U. S. Nat. Mus. Bull. 187, p. 176, footnote). 164 Rafinesque states that his new genus differs from Pelamis in its "carinated or angular abdomen," and the species luteus is characterized as "entirely yellow." It, and doubtless some or all of the other "species" of Ophinectes he describes, is apparently but a variant of the highly variable Anguis platura Linnaeus. 165 As pointed out in the preceding footnote to the genus Ophinectes, possibly several other "species" attributed to this genus are also synonyms that should be added to the above list. 296 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1921. Hydrus platurus var. subohscurus Wall, Snakes of Ceylon, p. 422 (name for var. C of Boulenger:1896) : Ceylon. 1921. Hydrus platurus var. pallidus Wall, Snakes of Ceylon, p. 422 (name for var. D of Boulenger :1896) : Travancore, India. 1955. Pelamis platura vara, hrunnea ; fasciata; leucostriata ; neurileu- cura; neuricatenata ; tricolor Deraniyagala, Colored Atlas of some Vertebrates of Ceylon (Colombo), 3, p. 80; no precise localities mentioned, merely "Ceylon waters." Range. Indian and other tropical oceans as far as the coast of Mexico in the Pacific Ocean. Occasionally temperate seas such as those of South Africa and around New Zealand in the southern hemisphere ; as far north as the Russian Coast Province in the northern hemisphere.^*'*' Family VIPERIDAE Genus ATRACTASPIS Smith 1849. Atractaspis A. Smith, Illus. Zool. S. Africa, Rept., footnote to text of pi. Ixxi. Type by monotypy: A. bibronii Smith. 1854. Brachycranion Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 99. Type by monotypy: B. corpulentum Hallowell. 1862. Eurystephus Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 337. Type by monotypy: Elaps irregularis Reinhardt. 1895. Clothelaps Cope, Trans. Amer. Philos. Soc, 18. p. 211. Type by present designation; Atractaspis congica Peters. 1906. Melanelaps Wall, Journ. Bombay Nat. Hist. Soc, 17. p. 27. Type by monotypy: M. mcphersoni Wall = Atractaspis andersonii Boulenger. Atractaspis microlepidota microlepidota Glinther^®^ Small-scaled Burrowing-Adder 166 1 am unaware of any actual specimens of platurus from Kenya Colony, Tanganyika Territory or Zanzibar Island : reported "sea-snakes" invariably proved "to be highly colored muraenid eels. However, P. platurus has been taken along the Red Sea coast of Somalia ; in the Seychelles and Madagascar, while •"It is known from all parts of the coast of East Africa, where it is not uncom- mon south of the equator," according to Malcolm Smith (1943, Fauna British India, 3, Serpentes, p. 477). 167 Laurent, in his revision of the genus Atractaspia (1950:10) regards fallow as an eastern race of microlepidota, whose typical form he restricts to the south- ern Sudan,, but I am far from happy regarding the numerous races of microlepi- dota postulated by Laurent. In the first place the type locality of fallax is not in Tanganyika Territory as assumed by Laurent, but "drei Tagereisen von Kiriama" and that is clearly shown on the map of Von der Decken's itinerary (1S69, Reisen in Ost-Africa, 1, part 1, map iii ; also cf. part 2, 1871, pp. 64, 435) as about halfway between Bura and Mombasa. In later years (1897) Tornier corrupted the spelling to "Kircamo." When adequate series are available it is interesting to note the extensive range of variation of typical microlepidota. For example 14 ^ ^ from Torit, Sudan, have a ventral range of 216-239; subcaudals 25-29; 9 99, also from Torit, have 233-242 ventrals ; 21-24 subcaudals. The sexes and extremes have been carefully checked in these snakes (M.C.Z. 53535-53557). LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 297 1866. Atractaspis microlepidota Giinther, Ann. Mag. Nat. Hist., (3) 18. p. 29, pi. vii, fig. 3: "Probably Weat Africa" (error). 1866. Atractaspis fallax Peters, Monatsb. Akad. Wiss. Berlin, p. 890: "Kiriame" corrected to "Kiriama" = Giriama, Kenya Colony. 1913. Atractaspis phillipsi Barbour, Proc. Biol. Soc. Washington, 26, p. 148: Singa, Sennar Province, Sudan. Range. Southern Sudan, east through Ethiopia to Somalia, south to Kenya Colony.^^* Atractaspis irregularis bipostocularis Boulenger^®' Eastern Variable Burrowing-Adder 1905. Atractaspis bipostocularis Boulenger, Ann. Mag. Nat. Hist., (7) 15, p. 190: Fort Hall near Mount Kenya, Kenya Colony. 1908. Atractaspis conradsi Sternfeld, Sitzb. Ges. Naturf. Freunde Ber- lin, p. 9-1: Ukerewe Island, Lake Victoria, Tanganyika Territory. 1930. Atractaspis schoutedeni "Witte, Eevue Zool. Bot. Afr., 19, p. 224, figs. 1-3: "N'Goma," i.e. Goma, Lake Kivu, Belgian Congo. 1934. Atractaspis bahaulti Angel, Bull. Soc. Zool. France, 59, p. 169: Kadjudju, Kivu region, Belgian Congo. 1945. Atractaspis irregularis loveridgei Laurent, Eevue Zool. Bot. Afr., 38, p. 322: Bunia, Ituri region, Belgian Congo. Range. Kenya Colony and northwest Tanganyika Territory, west through Uganda and Belgian Ruanda-Urundi to the eastern Belgian Congo. ^o"- Atractaspis oterrima Giinther Western-Forest Burrowing-x\dder 1863. Atractaspis aterrima Giinther, Ann. Mag. Nat. Hist., (3) 12, p. 363 : West Africa. it>s Unknown from Uganda, for Lorlwar, though west of Lake Riulolf is in Kenya Colony (not Uganda as stated by Parlter :1949 :109) ; the unique $ with 37 midbody scale-rows and 245 ventrals, allegedly from "Lake Tanganyilca" accord- ing to Sir Jolin Kirk who sent it to the British Museum, is scarcely assignable to this form. 169 Proceeding from east to west across Africa there is a gradual, but definite, increase in scale-rows from 21 (Fort Hall) to 25-27 (i. irregularis at Ganta, Liberia). A. bipostocularis was founded on a juvenile with two postoculars (no second specimen has been taken in fifty years). I regard this condition as aberrant, for In our series of 2o m. microlepidota from Torit, Sudan, two $ J (M.C.Z. 53554-5) have two postoculars on the right side of the head but the normal single scale on the left. If this view is accepted, then the name bipotitocnlaris should take precedence over conradsi (which was also founded on an aberrant individual as I demonstrated by collecting normal "irregularis" on Ukerewe Island. Apart from Laurent's Eritrean i. angeli, and his i. uelensis from the Congo, which are both apparently recognizable by high ventral counts, the j^uestion arises as to how many forms one should recognize — 21-23, 23-25. 25-2 (, or just two, viz. 21-23 with some overlap of 25 In Uganda and eastern Congo, and 25-27 for the western i. irregularis (inc. parkeri Laurent, if the subcaudal counts are not found to overlap with more material). 298 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Range. Virgin forest of Uluguru Mountains, Tanganyika Ter- ritory, northwest to Uganda, west to Portuguese Guinea. Atractaspis bibronii rostrata Giinther^'" Zanzibar Bibron's Burrowing-Adder 1868. Atractaspis rostrata Giinther, Ann. Mag. Nat. Hist., (4) 1, p. 429, pi. xix, fig. 1 : Zanzibar. 1901. Atractaspis Icatangae Boulenger, Ann. Mus. Congo, (1) 2, p. 13, pi. V, fig. 2: Lofoi, Katanga, Belgian Congo. 1901. Atractaspis coarti Boulenger, Ann. Mus. Congo, (1) 2, p. 14, pi. v, fig. 3 : Albertville, Belgian Congo. Range. Kenya Colony and Tanganyika Territory (Zanzibar coast; possibly Zanzibar Island), south to Mozambique (north of the Zambezi), west through Nyasaland and Northern Rhodesia to southern Belgian Congo. Genus CAUSUS AVagler 1830. Causus Wagler, Nat. Syst. Amphib., p. 172. Type by monotypy (as Naja V-nigrum Boie is a synonym): Sepedon rhombeata Lichtenstein. 1842. Distidhuriis Hallowell, Journ. Acad. Nat. Sei. Philadelphia, 8, p. 337. Type by monotypy: D. viaculatiis Hallowell = Sepedon rhombeata Lichtenstein. 1862. Eeterophis Peters, Monatsb. Akad. Wiss. Berlin, p. 276. Type by monotypy: H. resimus Peters. 1882. Dinodipsas Peters, Sitzb. Akad. Wiss. Berlin, p. 893. Type by monotypy : D. angidifera Peters = Aspidelaps lichtensteinii Jan. Cousus rhonxbeatus (Lichtenstein)^" Rhombic Night- Adder 1823. Sepedon rhombeata Lichtenstein, Verz. Doubl. Mus. Zool. Berlin, p. 106 : No locality. 1827. Naja Col(uber) v nigrum Cuv. {sic) Boie, Isis von Oken, 20, col. 557: Africa. 1842. Bistichurus Maculatus Hallowell, Journ. Acad. Nat. Sci. Phila- delphia, 8, p. 337, pi. xix: Liberia. 170 To this form all East African records of hibrotiii (to which can be added Tornier's misidentiflcations of '■irregularis" from Tanga, etc.) are referable. Laurent has shown that rostrata is distinguishable by normally having 23 mid- body scale-rows (rarely 21 or 22), whereas h. bibronii A. Smith — with a range south of the Zambezi to Angola — has 21 scale-rows (rarely 23). 171 I am not prepared to say that this common reptile, which (except for Eritrea, Southwest Africa, and the offshore islands) has been reported from every political unit in Africa south of 20° N., does not break up into some races. I would point out, however, that small populations of uniformly colored, an'l presumably striped individuals also, are scattered over much of the range and occur together with normally marlied specimens. They do not seem to be any more entitled to racial recognition than Vipera dorsalis Gray does in Britain. LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 299 1905. Cajisus rhomheatus var. hilirifiatus Boulenger, Ann. Mag. Nat. Hist., (7) 16, p. 114: Between Benguela and Bihe, Angola. 1912. Causus rhomheatus var. taeniata Sternfeld, Wiss. Ergebn. Deutscb. Zentral-Afrika-Exped. 1907-1908, 4. p. 276: No. locality (but Sehubotz coll., .so probaljiy between Lakes Victoria and Kivu). 1955. Caucus lineatus Laurent, Eevue Zool. Bot. Afr., 51, p. 136: Serani Research Station, Kundelungu Plateau, Kasenga Territory, Upper Katanga, Belgian Congo. Range. Sudan east to British Soinaliland, south through Uganda, Kenya Colony and Tanganyika Territory to Natal, northwest to Angola and French West Africa (Mauretania and French Soudan). Causus resimus (Peters) Velvety-green Night- Adder 1862. Heterophis resimus Peters, Monatsb. Akad. Wiss. Berlin, p. 277. pi. — , fig. 4: Jebel Ghule, Sennar, Sudan. 1888. Causus Jacksonii Giinther, Ann. Mag. Nat. Hist., (6) 1, p. 331: Lamu, Kenya Colony; Laie Tanganyika, Tanganyika Terri- tory. 1893. Causus 7iasalis Stejneger, Proc. U. S. Nat. Mus., 16, p. 735: Tana River, Kenya Colony; Cunga, Angola; also West Africa. 1895. Causus resimus var. angolensis Bocage, Herp. Angola Congo, p 148: Rio Dande and eight other localities in Angola. Range. Sudan, east to British Somaliland, south through Uganda; Kenya Colony and Tanganj'ika Territory to Mozam- bique, west tlirough Northern Rhodesia to eastern Belgian Congo and Belgian Ruanda-Urundi. Causus defilippii (Jan) Snouted Night- Adder 1862. H {.eterodon) Be Filippii Jan, Arch. Zool. Anat. Fisiol., 2. p. 225: Africa. 1863. Causus {Eeteropliis) rostratus Giinther, Ann. Mag. Nat. Hist., (3) 12, p. 363: Ugogo, Tanganyika Territory. Range. Coastal Colony ; Tanganyika Territory and Zanzibar Island (at Mtende), south through Mozambique; Nyasaland and the Rhodesias to Natal. Cavisus lichtensteinii (Jan) 1859. Aspidclaps Lichtensteinii Jan, Revue Mag. Zool., p. 511: Gold Coast. 1882. Dinodipsas angulifera Peters, Sitzb. Akad. Wiss. Berlin, p. 893, pi. XV : "Laguna, near Puerto Cabello, Venezuela" (error). Range. Western Kenya Colony and Uganda, west to Spanish (luinea ; French and British Cameroons ; Nigeria ; Liberia. 300 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Genus VIPERA Laurenti 1768. Fipera Laurenti (part), Syn. Kept., p. 99. Type by designation of Fitzinger:1843 : Vipera redii Gmelin r= Coluber francisci redi Laurenti = Cohiier aspis Linnaeus. 1816. Bems Okeu, Lehrb. Naturg., 3, Part 2, p. 234. Type by tautonomy: Coluber berus Linnaeus. 1820. Pelias Merrem, Vers. Syst. Amphib., p. 148. Type by monotypy: Coluber berus Linnaeus. 1822. Chersea Fleming, Phil. Zool., 2. p. 295. Type by monotypy: ' ' Cerastes vulgaris " ? rr Vip.era vulgaris Latreille = Coluber aspis Linnaeus. 1834. ETtinaspis Bonaparte, Icon. Fauna. Ital., Anfibi, part. 42. Type by monotypy : Coluber ammodytes Linnaeus. 1842. Daboia Gray, Zool. Misc., p. 69. Type by subsequent designation: Vipera elegans Daudin = Coluber russellii Shaw. 1843. Ehinechis Fitzinger, Syst. Eept., p. 28. Type by original designa- tion: Coluber ammodytes Linnaeus. 1843. Chersophis Fitzinger, Syst. Kept., p. 28. Type by original designa- tion: Vipera elegans Daudin = Coluber russellii Shaw. 1844. Echidnoides Mauduyt, Herpetol Vienne, p. 29. Type by original designation: E. trilamina Millet =: Coluber berus Linnaeus. 1!)27. Mesocoronis Eeuss, Zool. Anz., 71. p. 216. Type by original desig- nation : .V. c.oronis Eeuss = Vipera berus var. bosniensis Boett- ger. 1927. Teleovipera Eeuss, Zool. Anz., 73. p. 125. Type by original desig nation : Coluber ammodytes Linnaeus. 1927. Acridophaga Eeuss, Zool. Anz., 73. p. 126. Type by original designation: Pelias ursinii Bonaparte. 1927. Mesovipcra Eeuss, Zool. Anz., 73. p. 126. Type by original desig- nation: Coluber aspis Linnaeus. 1927. Macrovipera Eeuss, Zool. Anz., 73. p. 126. Type by original desig- nation: Coluber lebetinus Linnaeus. 1927. Mesohorinis Eeuss, Naturforseher, 4. p. 129. Misprint for Meso- coronis Eeuss. 1929. Latastea Eeuss, Wochenschr. Terrar. Aquar.-Kunde, 26. p. 64. Type by original designation: Vipera latastei Bosca. ]!i29. Tzarevscya Eeuss, Wochenschr. Terrar. Aquar.-Kunde, 26. p. 64. Type by original designation: Mesocoronis {Tzarevscya) tigrina Eeuss =^ Vipera haznalcovi Nikolski. 19?.;". Latasteopara Eeuss, Nachr.-Blatt. Aquar.-Terrar.-Ver., p. 216. Type by original designation : Vipera hugyi Schinz. Vipera hindii Boulenger Montane Viper 1910. Vipera hindii Boulenger, Ann. Mag. Nat. Hist., (8) 5. p. 513: LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 301 Fort Hall, 4000 feet, Kenya Colony."^ Range. Kinangop and Aberdare Mountains, Kenya Colony. Vipera supercilioris Peters Lowland Viper 1854. Vipera superciliaris Peters, Monatsb. Akad. Wiss. Berlin, p. 625: Mainland opposite Querimba Island, Cape Delgado, Mozam- bique. Range. Mwaya, Lake Nyasa, Tanganyika Territory and Mo- zambique. Genus BITIS Gray 1768. Colra Laurenti,"^ Syn. Kept., p. 103. Type by designation of ritzinger:1843: Colvher atropos Linnaeus. 1820. Echidna Merrem (part; not Forster :1788), Vers. Syst. Amphib., p. 150. Type by designation of Fitzinger :1843: E. arietans Mer- rem. 1842. Clotho Gray (not Faujas de Saint-Fond: 1808), Zool. Misc., p. 69. Type by tautonomy: Cohra clotho Laurenti. 1842. Bitis Gray, Zool. Misc., p. 69. Type by tautonomy: Coluber bitis (sic: error for bitin) Bonnaterre = Cobra lachesis Laurenti. 1845. Calechidna Tschudi, Fauna Peruana, Herp., p. 60. Type by mono- typy: Echidna ocellata Tschudi = Coluber atropos Linnaeus. Bitis worthingtoni Parker Kenya Horned Viper 1932. Bitis worthingtoni Parker, Journ. Linn. Soc. London, Zool., 38, p. 221 : Lake Naivasha, Kenya Colony. Range. Uplands of Kenya Colony. Bitis arietans arietans (Merrem) ^'^ Puff Adder 1768. Cobra lachesis Laurenti, Syn. Eept., p. 104: No locality. 1768. Cobra clotho Laurenti, Syn. Kept., p. 104: "Ceylon and Cuba" (error). 1789. Cioluber) Bitin Bonnaterre, Tabl. Encycl. Method. Ungues Nat., Ophiol., p. 22: "Ceylon" (error). 172 One wonders whether the type actually came from Fort Hall, or was merely sent by Dr. S. L. Hinde when stationed there. 173 Suppressed in favor of Bitis in 1945 (cf. Opin. Declar. Inter. Comm. Zool. Nomencl., 3, pp. 77-89 : Opinion 188). 174 Actually the correct name is lachesis, which has been used less than a dozen times since it was revived by Stejneger. As it has long been associated generically with the Tropical American pit-vipers, the sooner it is officially suppressed the better. Between 1849 and 1949 Bitis arietans has been con- sistently used over 250 times in the literature. The English name, which has come to stay, is also somewhat unfortunate for, while the burrowing-adders and night-adders are all oviparous, the "puff adder" is a truly viviparous "viper." 302 BULLETIN : MUSEUM OF COMPAEATIVE ZOOLOGY 1798. Coluber intumescens Donndorf, Zool. Beytrage Linn. Natur., 3. p. 209: (not seen). 1802. Vipera severa Sonnini & Latreille (part), Hist. Nat. Kept., 3. p. 335, pi. — , fig. 1: "Japan" (error). 1820. Vipera {Echidna) arietans Merrem, Vers. Syst. Aniphib., p. 152: Cape of Good Hope, i.e. South Africa. 1822. Vipera inflata Burchell, Travels S. Africa, 1. p. 469: "Cape Colony," i.e. Cape Province, South Africa. 1829. Vipera brachyura Cuvier, Kegne Animal, ed. 2, 2. p. 90: No locality. 1842. Clotho lateristriga Gray, Zool. Misc., p. 69: Gambia. Range. Africa (outside forested areas) from southern Morocco east to Arabia; south (except for an undefined area centering about British Somaliland where B. a. somalica Parker occurs) through Uganda; Kenya Colony; Tanganyika Territory (and possibly Zanzibar on the basis of a single specimen collected by Kirk, though not necessarily on the island) to Natal, west through Cape Province, northwest to Morocco. Bitis arietans somalica Parker 1949. Bitia lackcsis somalicu Parker, Zool. Verhand. Eijksmus. Nat. Hist. Leiden, No. 6, p. 101: Bohodle, 2100 feet, British Somaliland. Range. British Somaliland, south to northern Kenya Colony. Bitis gobonica gabonica (Dumeril & Bibron) Central African Gaboon Viper ^" 1854. Eohidna Gabonica Dumeril & Bibron, Erpet. Gen., 7. p. 1428, pi. Ixxx b: Gabon, French Congo. Range. Southern Sudan and Uganda ; eastern Tanganyika Territory ^"'^ south to Mozambique ; west through Northern Rho- desia to Angola; north and West to Togoland where (or in the Gold Coast) it meets with the race B. g. rhinoceros Schlegel which ranges westward to French Guinea. *t>^ Bitis nasicornis (Shaw) Nose-horned Viper 1802. Coluber Nasicornis Shaw, Nat. Miscell., 3. pi. xciv: Interior of Africa (from tlie master of a Guinea vessel). 175 Also known as the River-Jack : the name Rhinoceros Viper, though properly pertaining to the West African race of Gaboon Viper, is best aliandoned as it has been so often applied to the Nose-horned Vi])er. 1T6 "Zanzibar" was included by me in 1924 on the basis of Sternfeld's 1910 listing. Doubtless his record, like that of Boulenger in 1896, referred to the opposite coast (now Tanganyika) as it is unknown on the island today. Stern- feld also lists gabonica from Damaraland. i.e. Southwest Africa, and Stevenson- Ilamiltou includes it in the herpetofauna of Kruger I'ark, Transvaal. Both are omitted from the above range pending contirmation. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 303 1854. Vipera Eexacera Dum^ril & Bibron, Erpet. Gen., 7. p. 1416, pi. Ixxviii h, fig. 2 : No locality. Range. Southern Sudan; western Kenya Colony; Uganda; southwest through Belgian Ruanda-Urundi and the Belgian Congo to Angola; north and west to French Guinea (but un- recorded from Togo and Sierra Leone). Genus ECHIS Merrem 1820, Echis Merrem (part). Vers. Syst. Amphib., p. 149. Type by designation of Boie (1826) or Fitzinger (1843) : Pseudoboa carinata Schneider. 1849. Toxicoa Gray. Cat. Snakes Brit. Mus., p. 29. Type by subsequent selection: Echis arenicola Boie := Scythale pyramidum Geoffrey. Echis ccorinotus pyramiduin (Geoflfroy) Egyptian Saw-scaled Viper 1827. Scythulc pyramidum Geoffrey Saint-Hilaire, Kept., in Descr. Egypte, p. 152, pi. vii, fig. 1: Egypt. 1827. Echis arenicola Boie, Isis von Oken, 20. col. 558: North Africa. 1834. Echis pavo Eeuss, Mus. Senckenberg., 1. p. 157: Egypt. 1834. Echis varia Eeuss, Mus. Senckenberg., 1, p. 160: Ethiopia. 1837. Fipcra echis Schlegel, Essai Phys. Serp., 2. p. 583. pi. xxi, figs. 10-11: Africa; Arabia; Bengal, India. Range. Northern India, west through Persia and Arabia to North Africa, south in arid regions to northern Kenya Colony ; Uganda and Gold Coast. ^'' n' Genus ATHERIS Cope 1849. Chloroechis Bonaparte, Proc. Zool. Soe. London, p. 145, footnote. Type by tautonomy: Vipera chloroechis Schlegel.^'* 1862. Aiheris Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 337. Type by monotypy: Vipera chloroechis Sclilegel. 1863. Poecilostolus Giinther, Ann. Mag. Nat. Hist., (3) 11. p. 25. Type liy monotypy: P. hurtonii Giinther = Echis sqiiatxic/er Hal lowell. 177 Possibly the Gold Coast form is separable. E. c. carinatus Schneider occurs in a limited area around Madras, India. 178 Chloroechis was rejected by Boulenger on the grounds that it was inade- (|uately described. However, Bonaparte refers to the green color of this true, but arboreal, viper from Ashanti, and his evident intention was to erect a genus whose type by tautonomy was Vipera chlnroechis Schlegel. 304 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Atheris squamiger (Hallowell) Green Bush-Viper^'® 1854. Echis squainigera Hallowell, Proc. Acad. Nat. Sci. Philadelphia, p. 193 : Near Gabon Eiver, French Congo. 1863. Poecilostolus Biirtonii Giinther, Ann. Mag. Nat. Hist. (3), 11, p. 25: Cameroon. 1864. Atheris polylepis Peters, Monatsb. Akad. Wiss. Berlin, p. 642: Liberia. 1885. Atheris Lucani Rochebrune, Bull. Soc. Philom. Paris (7), 9. p. 89: Landana, Cabinda. 1885. Atheris proximus Rochebrune, Bull, Soc. Philom. Paris (7), 9, p. 90: Bissarie, Casamance River district, Senegal. 1887. Atheris anisolepis Moequard, Bull. Soc. Philom. Paris (7) 11. p. 89 : " Alim Leketi, ' ' presumably Alima Leketi River, French Congo. 1887. Atheris laeviccps Boettger, Zool. Anz., 10, p. 651: Povo Netonna, near Banana, Belgian Congo. 1888. Atheris suhocularis Fischer, Jahrb. Hamburg. Wiss. Anst., 5, p. 5, pis. i, fig. 1, and ii, fig. 11 : Cameroon. 1955. Atheris hispida Laurent, Revue Zool. Bot. Afr., 51, p. 138: Lutungura, Kivu District, Belgian Congo. Range. Western Kenya Colony and Uganda; west through Belgian and French Congo to Cameroons and Fernando Po; south to Angola. Atheris nitschei nitschei Tornier Great Lakes Bush-Viper 1902. Atheris nitschei Tornier, Zool. Jahrb. Syst., 15, p. 589, fig. — : Mpororo Swamp, Tanganyika Territory. 1906. Atheris woosnami Boulenger, Ann. Mag. Nat. Hist., (7) 18, p. 37: Mubuku Valley, Euwenzori Mountains, Uganda. Range. Uganda ; western Tanganyika Territory ; Belgian Ruanda-Urundi ; Belgian Congo. Atheris nitschei rungweensis Bogert Rungwe Mountain Bush-Viper 1940. Atktris nitschei rungweensis Bogert, Bull. Amer. Mus. Nat. Hist., 77, p. 104, fig. 18: Rungwe Mountain, Tanganyika Territory. Range. Southwest Tanganyika Territory ; Misuku IMountains, northwest Nyasaland and Northern Rhodesia. 1'9 Though commonly referred to as "tree-vipers," so far as my experience goes these snakes are usually found in bushes or shrubbery at the forest-edge. Pos- sibly trionomials should be applied to this form if it can be demonstrated that chloroechis (Schlegel) is but a western race ranging from the Gold Coast to Liberia. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 305 Athens cerotophorus Werner Usambara Mountain Bush-Viper 1895. Atheris ccratophorus Werner, Verb. Zool. Bot. Ges. Wien, 45, p. 194, pi. V, fig. 1: Usambara Mountains, Tanganyika Terri tory. Range. Usambara Mountains, Tanganyika Territory .^^" Atheris barbouri Loveridge Uzungwe Mountain Bush-Viper 1930. Atheris barbouri Loveridge, Proe. New England Zool. Club, 11, p. 107: Dabaga, Uzungwe Mountains, Tanganyika Territory. Range. T'zungwe and Ukinga Mountains, Tanganyika Terri- tory. Class AMPHIBIA Order GYMNOPHIONA^'^ Family CAECILIIDAE^'- Genus SCHISTOMETOPUM Parker 1941. Schistometopum Parker, Ann. Mag. Nat. Hist.. (11) 7, p. 17, jfig. 4. Type by original designation: Dermophis gregorii Bou- lenger. Schistometopum gregorii (Boulenger)^*'' Mud-dwelling Caecilian 1894. Dermophis gregorii Boulenger, Proe. Zool. iSoc. London, p. G4tt, pi. .\1, fig. 4: Ngatana, Tana River, Kenya Colony. 1912. Boulengerula denhardti Xieden, Sitzb. Ges. Naturf. Freunde Ber- lin, p. 199: Tana Eegion, Kenya Colony. 1913. Bdellophis unicolor Boettger, in Voeltzkow, Reise in Ostafrika, 3, p. 353, pi. xxiii, fig. 18: Lake Peecatoni, near Witu, Kenya Colony. Range. Coastal Kenya Colony, south to Wami River near Bagamoyo. and Rufigi River near Kilindi, Tanganyika Territory. Genus BOULENGERULA Tornier 1896. Boxdengcrula Tornier, Die Kriechthiere Deutsch-Ost-Afrikas, p. 164. Type by monotypy: B. boulengeri Tornier. ISO The Togo and Gold Coast records are rejected pending definite capture. 181 APODA of r.nulenger, 1882, Cat. Batr. Gradientia s. Caudata and Batrachia Apoda ill the Coll. Brit. Mus., p. 88. '>^S2 Hypogrophis guentheri Boulenger, 1SS2, described as from "Zanzibar," has been showu by Parker. 1941, to be a synon.vm of H. rontrata (Cuvier) of the Sp.v- chelles. It is consequently omitted from this list. 1S3 The reasons for amalgamating these three "species," following the collectiou of topotypes of all three, will be found in Lo%eridge, 1936, Bull. Mus. Comp. Zool., 79, pp. 373-:>78. 306 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY Boulengerula taitanus Loveridge Taita Mountains Caecilian 1935. Boulengerula taitanus Loveridge, Bull. Mus. Comp. Zool., 79. p. 16: Mount Mbololo at 4800 feet, Taita Mountains, Kenya Colony. Range. Taita Mountains, Kenya Colony. Boulengerula changamwensis Loveridge Cliangamwe Lowland Caecilian 1932. Boulengerula cliangamwensis Loveridge, Bull. Mus. Comp. Zool., 72. p. 381 : Changamwe at 192 feet, near Mombasa, Kenya Colony Range. Kenya Colony (known only from the type locality). Boulengerula boulengeri Tornier Usambara Bluish-Gray Caecilian 1896. Boulengerula boulengeri Tornier, Die Kriechthiere Deutsch-Ost- Afrikas, p. 16-i: Usambara Mountains, Tanganyika Territory. Range. Usambara and Magrotto Mountains, Tanganyika Ter- ritory. Boulengerula uluguruensis Barbour & Loveridge Uluguru Pink Caecilian 1928. Boulengerula uluguruensis Barbour & Loveridge, Mem. Mus. Comp. Zool., 50. p. 183: Vituri, 2000 feet, Uluguru Mountains, Tan- ganyika Territory. Range. T^higuru Mountains, Tanganyika Territory. Genus SCOLECOMORPHUS Boulenger 1883. Scolecomorplius Boulenger, Ann. Mag. Nat. Hist., (5) 11, p. 48. Type by nionotypy : S. kirkii Boulenger. 1895. Bdellophis Boulenger, Proc. Zool. Soe. London, p. 412. Type by monotypy : B. vittatus Boulenger. Scolecomorphus vittatus (Boulenger) 1895. Bdellophis vittatus Boulenger, Proc. Zool. Soc. London, p. 412, pi. xxiv, fig. 4: Usambara Mountains, Tanganyika Territory. Range. Usambara; Magrotto and Uluguru Mountains, Tan- ganyika Territory. Scolecomorphus attenuatus Barbour & Loveridge 1928. Scolccomorplnis attenuatus Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 181: Nyingwa, 7500 feet, Uluguru Mountains, Tanganyika Territory. Range. THuguru Mountains, Tanganyika Territory. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 307 Scolecomorphus kirkii uluguruensis Barbour & Loveridge 1928. Scolecomorphus uIugurKensis Barbour & Loveridge, Mem. Mus. Comp Zool, 50, p. 180: Nyingwa, 7500 feet, Uluguru Mountains. Tanganyika Territory. Range. Uluguru IMountaius, Tanganyika Territory. Scolecomorphus kirkii kirkii Boulenger 1883. Scolecomorphus Kirl-ii Boulenger, Ann. Mag. Nat. Hist., (5) 11, p. 48: "Lake Tanganyika?" Range. Highlands of .southwestern Tanganyika Territory;"^ Nyasaland and Northern Rhodesia ( Nyamkolo :/ic?e Pitman). Order SALIENTIA Suborder OPISTHOCOELA Family PIPIDAE^^' Genus XENOPUS AVagler 1827. Xenopus Wagler, Isis von Oken, 20, col. 726. Type by monotypy: X. boiei Wagler = Bufo laevi-s Daudin. 1829. Dactylethra Cuvier, Regne Animal, ed. 2, 2, p. 107. Type by monotypy: "crapaud lisse de Daudin" = Bufo laevis Daudin. 1864. Silurana Gray, Ann. Mag. Nat. Hist., (3) 14. p. 315. Type by monotypy: S. tropicalis Gray. Xenopus clivii Peracea Eritrean Smooth Clawed-Prog 1898. Xenopus Clivii Peracea, Boll. Mus. Zool. Anat. Comp. Univ. Torino, 13. No. 321, p. 3: Adi Caie and Saganeiti, Eritrea. Range. Eritrea, south through Ethiopia to Lake Rudolf, Kenya Colony. Xenopus laevis borealis Parker Kenya Smooth Clawed-Frog 1936. Xenopus laevis borealis Parker, Ann. Mag. Nat. Hist., (10) 18, p. 596: Mount Marsabit, Kenya Colony. Range. Kenj^a Colony. Xenopus laevis victorianus Ahl Tanganyika Smooth Clawed-Frog 1.S4 I regard Nieden's record of Bdellophis vittatus from Ubena as a misidenti- lied kirkii for I personally collected this specie.s on the Mutindi-Njombe road, in addition to a large series from Nyasaland. 1H5 Formerly referred to the suborder AGLOSSA, the African tongueless frogs are separated by some authors as XENOPIDAE. Besides PIPIDAp; the OPIS- THOCOELA includes only the DISCOGLOSSIDAE. 308 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1924. Xenopus victorianus Ahl, Zool. Anz., 60, p. 370: Busisi, Smith Sound, Lake Victoria, Tanganyika Territory. Range. Uganda ; Kenya Colony and western Tanganyika Ter- ritory, west to Belgian Congo. Xenopus laevis bunyoniensis Loveridge Ruanda Smooth Clawed-Frog 1932. Xenopus laevis hunyoniensis Loveridge, Proc. Biol. Soc. Washing- ton, 45, p. 114: Bufundi, Lake Bunyoni, Uganda, Range. Highland lakes of southwest Uganda ; Belgian Ruanda Urundi and Belgian Congo. Xenopus laevis petersii Bocage Angola Smooth Clawed-Frog 1895. Xenopus Petersii Bocage, Herp. Angola Congo, p. 187: Anibaca ; Bcnguela; Caconda; Catumbela; Chiniboto; Dondo; Huila; Quibula; Quindumbo and Sao Salvador, Angola. 1927. Xenopus poiceri Jlavntt, Ree. Albany Mus., 3, p. 413, pi. xxiv, fig. 3 : Victoria Falls, Zambezi, Northern Rhodesia. Range. Southwestern Tanganyika Territory, southwest through Northern Rhodesia to Angola. Xenopus muelleri (Peters) Miiller's Smooth Clawed-Frog 1844. Dactylethra Muelleri Peters, Ber. Akad. Wiss. Berlin, p. 37: Mozambique (restricted to Tete, Zambezi, by Loveridge). Range. Southern Sudan (at Gondokoro) ; Uganda; Kenj^a Colon}- ; Tanganyika Territory ; Zanzibar and Mafia Islands : Mozambique soutli to Transvaal {fide Mertens) ; w^est through Nyasaland and Northern Rhodesia to Angola ; north and west in arid areas to Dahomey {fide Chabanaud). Suborder PROCOELA Family BUFONIDAE Genus BUFO Laurenti^'" 1758. Bufo A. J. Roesel von Rosenhof, Hist. Nat. Ranarum, p. 69, pi. xxiv. Type by present designation : B. aquaticus von Rosenhof 186 Bufo of von Rosenhof appears to have been ignored as pre-Linnaean and not consistently binomial. Hylaplcsla Schlegel, 1826 (which antedates Boie, 1827) has been relegated to the synonymy of Dendrobates by Davis (1935, Zool. Ser. Field Mus. Nat. Hist., 20, p. 88), whose genus Cacophryne (Type: Hi/l(i borbonica Tschudi) is a synonym of Leptophryne Fitzinger, Syst. Rept. p. 32. Type by original designation : Bm/o cruentatiis Tschudi. Phryniscus Wiegmann. 1834, whose type by monotypy is P. nigricans Wiegmann = Atelopua stelzneri Weyenberg. is also one of the ATELOPODIDAB. f^cutiger Theobald, 1868, was removed from the synonymy by Boulenger. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 309 equals and antedates B. calamita Laurenti. 1768. Bufo Laurenti (part), Syn. Eept., p. 25. Type by tautonomy: B. vulgaris Laurenti = Bona bufo Linnaeus. 1824. Oxyrhynchus Spix, Anim. Spec. Testud. Eana. Brasiliam, p. 49. Type by monotypy: 0. naricus Spix = Eana typhonm Lin- naeus. 1828. Chaunu^ Wagler, Isis von Oken, 21, col. 744. Type by monotypy: C. marmoratus Wagler (inc. C. glohulosiis Wagler) both = Bufo granulosus Spix. 1838. Osilophu^ "Cuv." Tschudi, Class. Batr. Kept., p. 89. Type by monotypy: " Osilophus typJionius Cuv. " = Eana typhonw Linnaeus. 1843. Phrynoidis Fitziuger, Syst. Eept., p. 32. Type by original designa- tion: Bufo asper "Kuhl, " i.e. Gravenhorst. 1843. Fhrynomorphus Fitzinger, Syst. Eept., p. 32. Type by original designation: Bufo leschenaultii Dumeril & Bibron =: Bufo guttaius Schneider. 1843. Bocidophryne Fitzinger, Syst. Eept., p. 32. Type by original designation : Bufo agiia Daudin = Eana marina Linnaeus. 1843. PeJtophryne Fitzinger, Syst. Eept., p. 32. Type by original desig nation: Biifo peltocephalus Dumeril & Bibron, i.e. Tschudi. 1843. Otolophus "Cuv." Fitzinger, Syst. Eept., p. 32. Type by original designation: Bufo margaritifer Daudin =: Eana typhonic Lin- naeus. 1843. Eurhina Fitzinger, Syst. Eept., p. 32. Type by original designa tion: Bufo prohoscideus Wagler := Eana typhonia Linnaeus. 1843. Chilophryne Fitzinger, Syst. Eept., p. 32. Type by original desig- nation: Bufo d'oriignyi Dumeril & Bibron. 1843. Phryne Fitzinger, Syst. Eept., p. 32. Type by original designation : Bufo vulgaris Laurenti = Eana bufo Linnaeus. 1845. Anaxyrus Tschudi, Fauna Peruana, Herp., p. 78. Type by mono- typy: A. melancholicus Tschudi = Bufo compactilis Wiegmann. 1849. Schismaderma A. Smith, Illus. Zool. S. Africa, Eept., App., p, 28. Type by original designation : S. lateralis Smith = Bufo carens Smith. 1860. Adenomus Cope, Proe. Acad. Nat. Sci. Philadelphia, p. 371. Type by monotypy: A. badioflavus Cope = Bufo Tcelaarti Giinther. 1862. Elmebo Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 357. Type by monotypy: Bufo haematiticus Cope. 1865. Epidalea Cope, Nat. Hist. Eeview, 5, p. 102. Type by monotypy: Bufo calamita Laurenti. 1868. Otaspis Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 312. Type by monotypy : Peltaphryne empusa Cope. 1868. Pegaeus Gistel, in Blicke, Leben Natur. Menschen (Leipzig), p. 310 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 161. Type by designation of Mertens:1936: Bona bufo Linnaeus. 1870. Nannophryne Giinther, Proc. Zool. Soc. London, p. 401. Type by monotypy: N. variegata Giinther. 1870. Ansonia Stoliczka, Journ. Asiatic See. Bengal, 39, p. 152. Type by monotypy: A. penangensis Stoliczka. 1876. Cranopsis Cope, Journ. Acad. Nat. Sci. Philadelphia, (N.S.) 8, p. 96. Type by monotypy: C. fastidiosus Cope. 1876. Crfipidius Cope, Journ. Acad. Nat. Sci. Philadelphia, (N.S.) 8, p. 97. Type by monotypy: C. epioticus Cope. 1876. OUotis Cope, Journ. Acad. Nat. Sci. Philadelphia, (N.S.) 8. p. 98. Type by monotypy: 0. caerulescens Cope. 1879. Dromoplectus Camerano, Atti. R. Acad. Sci. Torino, 14, p. 882. Type by monotypy: Bufo anomalus Giinther = B. compactilis Wiegmann. Bufo cccrens Smith 1848. Bufo car.ens A. Smith, lUus. Zool. S. Africa, Rept., pi. Ixviii, fig, 1 : Interior of South Africa. 1849. Schismaderma lateralis A. Smith (new name for Bufo car ens Smith), Illus. Zool. S. Africa, Rept., App., p. 28. Range. Kenya Colony and Tanganyika Territory, south to Natal, northwest through Bechuanaland to the Belgian Congo. Bufo brauni Nieden Dead-leaf Toad 1910. Bafo braimi Nieden, Sitzb. Ges. Naturf. Freunde Berlin, p. 450: Amani, Usambara Mountains, Tanganyika Territory. Range. Mrgin forests of the Usambara ; Magrotto and Ulu- guru Mountains, Tanganyika Territory. Bufo regularis reguloris Reuss Egyptian Square-marked Toad 1834. Bufo regularis Reuss, Mus. Senckenberg., 1, p. 80: Egypt. 1841. Bufo pantherinus Dumeril & Bibron (part)^®^ Erpet. Gen. 8. j). 687: Egypt (A. Lefebvre coll.) by present restriction. 1855. Bufo mosaica Seetzen, Reise Syria, Palast., Aegypt. . . ., 3, p. 492: Egypt. 1858. Bufo guineensis Giinther (part). Cat. Batr. Sal. Brit. Mus., p. 59: Coast of Guinea (restricted). 1868. Bufo spinosus Bocage (not of Daudin:1803), Proc. Zool. Soc. London, 1867, p. 845: Benguela, Angola. 1897. Bufo garmani Meek, Field Mus. Nat. Hist. Zool. Series, 1, p. 176: * ' Haili, ' ' i.e. Halleh, British Somaliland. 187 Sometimes attributed to Boie, but his pantherinus, as well (is its use by Tschudi (1838), constitute nomina ituda. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 311 1932. Bufo regularis kisoloensis Loveridge/®^ Occ. Papers Boston Soc. Nat. Hist., 8, p. 52: Kisolo, Kigezi district, Uganda. Range. Except in the northeast (Morocco and Tunisia), pres- ent in all African countries from Algeria (Hoggar and A'ir) and Libya (Fezzan) east to Egypt, south through Uganda, Kenya Colony, Tanganyika Territory, Pemba and Zanzibar Islands, to Natal and the Cape, though in some areas as more or less recognizable races, of which many have been described. Bufo regularis marakwetensis Roux^*'' 1935. Bufo regularis marakivetensis Eoux, Mission Sci. de I'Omo, 3, Zool., p. 180: Mount Marakwet at 2500 metres, Kenya Colony. Range. Kenya Colony (known only from the two type S S ). Bufo funereus Bocage 1866. Bufo funereus Bocage, Jorn. Sci. Lisboa, 1, p. 77: Duque de Braganca, Angola. 1882. Bufo henguelensis Boulenger, Cat. Batr. Sal. Brit. Mus., pp. 299, 475, pi. xix, fig. 3: Benguela, Angola (restricted). Range. Uganda, southwest to Angola ; northwest to Cameroon and Fernando Po (also Dahomey according to Chabanaud :1916). • Bufo camenuiensis camerunensis Parker 1936. Bufo camerunensis camerunensis Parker, Proe. Zool. Soc. London, p. 153 : Oban, Calabar, Nigeria Range. Uganda, west to Liberia (but as yet unrecorded from Dahomey and Togo). Bufo latifrons Boulenger Spiny-flanked Toad 1900. Bufo latifrons Boulenger, Proc. Zool. Soc. London, p. 435, pi. xxvii, fig. 1 : Benito River, Gabon, French Equatorial Africa. Range. Kenya Colony;"" French Congo; French and British Cameroons. 188 Revived by Laurent (1952, Herpetologica. 8, p. 53) as a full species sym- patric with regularis. After careful reappraisal of the paratypes, together with the large series that led me to synonyinize kisoloensis, I find Laurent's allegedly diflerential characters offer no grounds for recognizing kisoloensis. 189 Without topotypical material it is unwise to suggest that this is a recogniz- able race, a full species, or perhaps a synonym of regularis or some allied species 190 The inclusion of this species rests on a toad taken on Mount Kinangop between 2600 and 2700 metres, according to Angel (1925, Rept. et Batr., p. 56, in "Voyage de Ch. Alluaud et R. Jeannel en Afrique Orientale (1911-1912). Resultats Scientifiques. Vertebrata" (Paris). At my request Mons. Guib6 has kii.dly re-examined this So (not 55) mm. specimen, and states (19.1.1957) that in all respects it agrees with the deflnition of latifrons set forth in Parker's (1936, Proc. Zool. Soc. London, p. 153) key, which should be consulted to avoid confusion with B. c. camerunensis Parker. 312 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Bufo steindachnerii Pfeffer 1893. Bufo Steindachnerii Pfeffer, Jahrb. Hamburg. Wiss. Anst., 1892, 10, p. 103, pi. ii, fig. 8: Kihengo, Tanganyika Territory. Range. Somalia, south through coastal Kenya Colony to Tan- ganyika Territory. Bufo vittatus Boulenger 1906. Bufo vittatus Boulenger, Proc. Zool. Soc. London, p. 573, fig. 98: Entebbe, Uganda. Range. Lower Egypt ; Uganda and western Tanganyika Ter- ritory. Bufo parkeri Loveridge 1932. Bufo parkeri Loveridge, Bull. Mus. Comp. Zool., 72. p. 382: Mangasini, Usandawi, Tanganyika Territory. Range. Tanganyika Territory (known only from the type series). Bufo urvmguensis Loveridge 1932. Bufo urunguensis Loveridge, Bull. Mus. Comp. Zool., 72, p. 383: Kitungulu, Urungu, Tanganyika Territory. Range. Tanganyika Territory (known only from the type series). Bufo ushoranus Loveridge 1932. Bufo ushoranus Loveridge, Occ. Papers Boston Soc. Nat. Hist., 8, p. 45: Ulugu, Ushora, Tanganyika Territory. Range. Tanganyika Territory (known only from the type series). Bufo lindneri Mertens 1955. Bufo lindneri Mertens, Jahrb. Ver. Vaterl. Naturk. Wurttemberg, 110, p. 48: Dar es Salaam, Tanganyika Territory. Range. Tanganyika Territory (known only from the type S and 9 ). Bufo lonnbergi lonnbergi Andersson 1911. Bufo lonnhergi Andersson, Svenska Vetensk.-Akad. Handl., 48, No. 8, p. 35, pi. ii, figs. 4a-b and 6a-b: Mount Kenya, Kenya Colony. Range. Kenya Colony. Bufo lonnbergi nairobiensis Loveridge ^^^ 191 On geographical grounds this race Is very doubtfully distinct, since it has now been taken on Mount Kinangop and the Mau. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 313 1932. Bufo lonnbergi nairobiensis Loveridge, Occ. Papers Boston Soc. Nat. Hist., 8, p. 48: Nairobi, Kenya Colony. Range. Kenya Colony. Bufo mocquardi Angel 1924. Bufo mocquardi Angel, Bull, Mus. Hist. Nat. Paris, 30, p. 269: Mount Kenya and Mount Kinangop, Kenya Colony. Range. Kenya Colony.^^" Bufo taitonus taitanus Peters ^''^ 1878. Bufo taitanus Peters, Monatsb. Akad. Wiss. Berlin, p. 208, pi. ii, fig. 8: "Taita," i.e. Teita, Kenya Colony, Range. Kenya Colony; Tanganyika Territory (Iramba Pla- teau) ; Mozambique (Chifumbazi). Bufo taitanus uzunguensis Loveridge 1932. Bufo taitanus uzunguensis Loveridge, Occ. Papers Boston Soc. Nat. Hist., 8. p. 44: Kigogo, Uzungwe Mountains, Tanganyika Terri- tory. Range. Uzungwe, Ubena and Poroto Mountains, southern Tanganyika Territory. Bufo micranotis micranotis Loveridge 1925. Bufo micranotis Loveridge, Proe. Zool. Soc. London, p. 770, pi. i, fig. 1 : Kilosa, Tanganyika Territory. Range. (Possibly Coastal Kenya Colony on basis of tadpoles from Diani Beach) ; Tanganyika Territory. Bufo micranotis rondoensis Loveridge 1942. Bufo micranotis rondoensis Loveridge, Bull. Mus. Comp. Zool., 91. p. 387: Nchingidi, Rondo Plateau, Tanganyika Territory. Range. (Coastal Kenya Colony if Diani Beach tadpoles are referable to this race) ; Hondo Plateau, Tanganyika Territory; •langombe Swamp, Zanzibar (in M.C.Z.). Bufo anotis Boulenger li'07. Bufo anotis Boulenger, Ann. Mag. Nat. Hist., (7) 20. p. 48, pi. iii: Chirinda Forest, southeast Mashonaland, Southern Rhodesia. 192 The Angola record of Menard (1937) is certainly erroneous. This little toad is quite distinct from nairobiensis. 193 Many species or races of earless toads have been recorded under this name from neighbouring territories, including Songosongo Island, Tanganyika Terri- tory. All such require careful reappraisal before being included in the range. Other races occur in Nyasaland and Mozambique. 314 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Range. Kilwa, Tanganyika Territory and Southern Rho- desia.^^* Genus NECTOPHRYNOIDES Noble 1926. Nectophrynoides Noble, Amer. Mus. Novit., No. 237, p., 15. Type by original designation: Nectophryne tornieri Koux. Nectophrynoides tornieri (Ronx) Usambara Viviparous-Toad 1906. Nectophryne tornieri Koux, Proc. Zool. Soc. London, 1, p. 63, pi. ii, fig. 4: Ukami, i.e. Uluguru Mountains, Tanganyika Terri- tory. 1910. Nectophryne wcrthi Nieden, Sitzb. Ges. Naturf. Freunde Berlin, p. 439 : Dar es Salaam, Tanganyika Territory.^^^ Range. Usanil)ara ; Magrotto and Uluguru Mountains, Tan- ganyika Territory. Nectophrynoides vivipara (Tornier) Rungwe Viviparous-Toad 1905. Pseudophrync vivipara Tornier, Sitzb. Akad. Wiss. Berlin, 2, p. 855 : Rungwe and Kinga Mountains, Tanganyika Territory ( restricted ).-^^^ Range. Uluguru ; Uzungwe ; Ukinga ; Rungwe and Poroto Mountains. Tanganyika Territory. Suborder DIPLASIOCOELA Family RHACOPHORIDAE'* Genus CHIROMANTIS Peters 1855. Chiromantis Peters, Arch. Naturg., 21, Abt. 1, p. 56. Type by monotypy: C. xerampelina Peters. 194 My early (1925) record of the occurrence of this species in western Tan- ganyika Territory was based on misidentifled vittatus (Biikoba) or the subse- quently described iishoraHus (Ushora). 195 Dar es Salaam is rejected pending confirmation. It may be observed that botli wcrthi and some of the cotypes of viripara were allegedly collected in Dar es Salaam by Werth. Werth was a chemist who had a shop in Dar es Salaam. Possibly, like other residents at the coast, he may have spent a vacation at one or other of the mission stations in the nearby Uluguru Mountains where both species occur — and failed to label his specimens. Alternatively some of these arboreal toads may liave been brought to Dar es Salaam in c;umps of bamboos for planting in the Botanical Gardens. That they would survive is unlikely. 196 Includes POLYPEDATIDAE. Though included families are predominantly diplasiocoelous. occasional memliers of the RHACOPIIORIDAE and MICROHY- LIDAE have precocious vertebrae. Formerly the four African families of this check list were grouped together as FIRMISTERNIA, but even that arrangement had its arciferal exceptions. Laurent (1951, Revue Zool. F>ot. Afr., 45, p. 120) suggests a new grouping In which he would remove Hyperolius to a new family, the HYPEROLIIDAE, and with it place such ranid genera as Astylosternus, Arthroleptis, Hemisus, etc. LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 315 Chiromontis rufescens (Giiuther) Western Foam-nest Tree-Frog 1868. Polypedates rufeseerm Giiuther, Proc. Zool. Soc. London, p. 486: West Africa. 1875. Chiromantis guineensis Buchholz & Peters, Monatsb. Akad. Wiss. Berlin, p. 203, pi. i, fig. 1 : Victoria, British Cameroon. Range. Uganda,^^' west through the Belgian Congo to Nigeria. Chiromantis xerampelina Peters Southeastern Foam-nest Tree-Frog 1854. Chiromantis xerampelina Peters, Monatsb. Akad. Wiss. Berlin, p. 627 : Sena and Tete, Mozambique. 1920. Chiromantis umhelluzianus Ferreira, Joru. Sci. Lisboi (2), 8. p. 205, pis. i-ii: Umbelluzi Bridge, Lourengo Marques district, Mozambique. Range. Coastal Kenya Colony ; Tanganyika Territory and Zanzibar Island, south through Mozambique, Nyasaland and the Rhodesias to Zululand, Natal. Chiromantis petersii kelleri Boettger Northern Foam-nest Tree-Frog 1893. Chiromantis Tcelleri Boettger, Zool. Anz., 16, p. 131: Lake steppe near Laku, and "Abdallah, " i.e. Abdulla country north of Webi Valley, Ethiopia. 1900. Chiromantis hachowshii Nikolsky, Ann. Mus. Zool. Acad. Sci. St. Petersbourg, 5, p. 246: Ferad, Ethiopia. 1916. Hylamhatcs cnantiodactylics Calabresi, Monit. Zool. Ital. (Fir- enze), 27. p. 36, pi. ii, fig. 2: Bardera, Somalia. 1929. Chiromantis maerops Ahl, Zool. Anz., 80. p. 29: Ganda Ali, "Annia" i.e. Ennia Galla, Ethiopia. Range. Ethiopia and British Somaliland south through So- malia to uorthern Kenj'a Colony. Chiromantis petersii petersii Boulenger Central Foam-nest Tree-Frog 1882. Chiromantis petersii Boulenger, Cat. Batr. Sal. Brit. Mus., p. 93, pi. X, figs. 1-la: "Interior of East Africa," i.e. Mpwapwa, Ugogo, Tanganyika Territory. 19' The only Uganda record is that of Loveridge (1942) for a nest taken in Budongo Forest, northwestern Uganda. In 1902 Boulenger listed acerampcUna from Uganda (apparently without supporting material), but it is far removed from the range of that species. 316 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY 1929. Chiromantis albescens Ahl, Zool. Anz., 80, p. 30: Pokomoni, (oppo- site Lamu Island), Kenya Colony. 1929. Chiromantis fasciatus Ahl, Zool. Anz., 80, p. 31 : Teita, Kenya Colony; also Ikoma, Tanganyika Territory. 1930. Chiromantis pygm/ieus Ahl, Zool. Anz., 88, p. 219: Kibwezi, Kenya Colony. 1931. Cliiromantis pictus Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 213: Kilimatinde, Tanganyika Territory. 1931. Chiromantis rugosus Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 215: " Langenburg, " i.e. Manda, Lake Nyasa, Tanganyika Territory. Range. Kenya Colony (south of the equator), southwest to central Tanganyika Territory .^^* Genus LEPTOPELIS Giinther 1858. Leptopelis Giinther, Cat. Batr. Sal. Brit. Mus., p. 89. Type by monotypy: Hyla aubryi A. Dumeril. 1924. Pseudocassina Ahl, Mitt. Zool. Mus. Berlin, 11, p. 8. Type by monotypy: P. ocellata Ahl (not Hylambates ocellatus Mocquard) as it includes P. rugosa Ahl {jide Ahl). 1941. Elaphromantis Laurent, Revue Zool. Bot. Afr., 35, p. 98. Type by original designations: Hylambates notatus Peters. 1941. Ueteropclis Laurent, Revue Zool. Bot. Afr., 35, p. 99. Type by original designation: Leptopelis parTceri Barbour & Loveridge. 1941. Taphriomnntis Laurent, Revue Zool. Bot. Afr., 35, p. 101. Type by original designation: Cystignathus bocagii Giinther. 1950. Cryptothylax Laurent & Combaz, Revue Zool. Bot. Afr., 43, p. 276. Type by monotypy: Hylambates greshoffl Schilthuis. Leptopelis grcnnineus (Boulenger)^"'' 1898. Megalixalus gramineus Boulenger, Ann. Mus. Civ. Stor. Nat. Genova (2) 18, p. 721, pi. x, fig. 2: Between Badditu and Dime, Somalia. 19!* It might be pointed out tliat Kenya is included solely on the basis of the types of albescens, fasciatus and pygmaetis. Much of Ahl's material was sub- adult and it was on the descriptions that I synonyuiized them with petcrsU in 19oo (Bull. Mus. Comp. Zool., 74, p. 390). When the opportunity occurs to exam- ine the Pokomoni type it should be compared with C. p. killeri; also with xeram- piiina which I obtained in series at nearby Witu. The Ikoma fasciatus is certainly p. petersii, but the Teita specimen was apparently referred to xeram- pclina by an earlier worker. Teita and nearby Kibwezi are faunistically similar, so pygmaens is likely to be specifically identical with the Teita frog. 199 Admitted to this list on the somewhat slender evidence of four tadpoles from Domorso waterhole near Moyale, recorded by Scortecci in 1940. In the same paper he records four juveniles from Javello Plain near Agheremariani. The only other Ethiopian record is that of Parker (1930) for Wourambouechi. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 317 Range. Ethiopia and Somalia, south to northern Kenya Col- ony. Leptopelis bocagii (Giinther) 1864. Cystignathus bocagii Giinther, Proe. Zool. Soc. London, p. 481, pi. xxxiii, fig. 2: Duque de Braganca, Angola. 1893. Hylamhates angolensis Bocage, Jorn. Sci. Lisboa, (2) 3. p. 119: Caeonda, Angola. 1893. Hylamhates cynnamomeus Bocage, Jorn. Sci. Lisboa, (2) 3. p. 120: Quilengues, Angola. 1904. Hylamiates hocagei var. leiicopunctata Ferreira, Jorn. Sci. Lisboa, (2) 7. p. 113: Gumba, Angola. 1937. Leptopelis hocagei haasi Mertens, Abhand. Senckenberg. Naturf. Ges., No. 435, p. 21, fig. 2: Nsombo, Lake Bangweulu, Northern Rhodesia. Range. Kenya Colony (from Northern Uaso Nyiro), south through Tanganyika Territory to Nyasaland, west through Northern Ehodesia to Southwest Africa, north through Angola to the Belgian Congo. Leptopelis anchietae (Bocage) 1873. Eylambatcs anchietae Bocage, Jorn. Sci. Lisboa, 4, p. 226: Mos- samedes, Angola. 1895. Hylamhates marginatus Bocage, Herp. Angola Congo, p. 178: Quisange, Benguela, Angola. 1930. Hylamhates hrevipalmatus Ahl, Zool. Anz., 87, p. 228: Unyika, Tanganyika Territory. Range. Southwestern Tanganyika Territory ,^°° west to Angola, north through Belgian Congo to the Bamenda District of French Cameroon. Leptopelis argenteus (Pfeffer)-"^ 1893. Hylamhates argenteus Pfeffer, Jahrb. Hamburg. Wiss. Anst., 1892, 10, p. 100, pi. ii, fig. 3: Marsh south of Bagamoyo, Tanganyika Territory. Range. Coastal (Bagamoyo to Kuponda) Tanganyika Terri- tory. 200 Known onlj- from a single record from the southwest (Ahl's type of brevi- paltnatus) for the frogs assigned to anchietae by Tornier (1896) were made paratypes of signijer by Ahl =; lermiculatus (Boiilenger). 201 Erroneously referred to the synonymy of Hylamhates maculatus by me (1930, Proe. Zool. Soc. London, p. 24). 318 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Leptopelis concolor Alil"°" 1929. Leptopelis concolor Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 129: "Wito," i.e. Witu, Kenya Colony. Range. Coastal Kenya Colony, south to Mikindani, Tan- ganyika Territory. Leptopelis flavomaculatus (Glintlier)^**'^ 1864. Hyperolius flavomaculatus Giinther, Proc. Zool. Soc. London, p. 310, pi. xxvii, fig. 1: Eovuma Bay, Tanganyika Territory. 1897. Hylamhates johnstoni Boulenger, Proc. Zool. Soc. London, pp. 801, 803, pi. xlvi, fig. 4: Kondowe to Karonga, Nyasaland (re stricted). Range. Kenya Colony (Ngatana), south through Tanganyika Territory to Mozambi(iue, west through Nyasaland to Southern Rhodesia; also eastern Belgian Congo. Leptopelis aubryi (Dumeril) 18.14. Hyla punctata Hallowell (not of Schneider : 1799), Proc. Acad. Nat. Sci. Philadelphia, p. 193 : near Gabon River, French Congo. 1856. Hyla aubryi A. Dumeril, Revue Mag. Zool., (2) 8, p. 561: Gabon, French Congo. 1929. Leptopelis harhouvi Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 119: Mount Lutindi, Usanibara Mountains, Tanganyika Terri tory.2o* Range. Usanibara Mountains, Tanganyika Territory ; Belgian Ruanda-Urundi, west to Angola, north to Cameroon, west (with gaps) to French Guinea. Leptopelis notolus christyi (Boulenger) 1906. Uylamhates cuhito-alhiis Boulenger (part), Ann. Mag. Nat. Hist., (7) 17. p. 323: only the juvenile eotypes from Lake Albert, Bunyoro, Uganda.-"'' 1912. Hylamhates christyi Boulenger, Ann. Mag. Nat. Hist., (7) 12, p. 556 : Mabira Forest, ' ' Chagwe, ' ' i.e. Kyagwe, Uganda. 1929. Leptopelis budducnsis Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 119: Northwest Buddu Forest, "Deutsch-Ost-Afrika," i.e. Uganda. 202 The Mombasa record of nattilensis (Smith) by Peters (1869) is probably referable to concolor, the two species being superficially very alike. 203 Based on a juvenile, while the synonym was described from an adult ; it serves to emphasize the strilving age variation presented by members of this genus. 204 The types (M.C.Z. 13561-71), which neither I nor others who have examined them can distinguisli from Cameroon aubryi except in size, were never seen by Ahl who had no representatives of the "species." 205 Tile name cuhito-alhus should be restricted to the eotypes from Zima. Cameroon LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 319 Range. Virgin forests of Uganda. Leptopelis vermiculotus (Boiilenger)""" 1909. nylambatcs vcrmiculatus Boulenger, Ann. Mag. Nat. Hist., (8) 4, p. 497: Amani, Usambara Mountains, Tanganyika Territory. 1929. Leptopelis sl(jnifer Ahl, Sitzb. Ges. Naturf. rreunde Berlin, p. 216: Devema, Usambara Mountains, Tanganyika Territory. Range. Virgin forests of the Usambara, Magrotto and Rimgwe Monntains, Tanganyika Territory. Leptopelis karissimbensis Ahl 1929. Leptopelis Tcarissimhfinsis Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 195: Mtulia Gama, Mt. Karisimbi, Belgian Euanda. 1929. Leptopelis Icivuensis Ahl, Sitzb. Ges. Naturf. Freunde Berlin, i>. 206 : Kisenyi, Lake Kivu, Belgian Ruanda. 1929. Leptopelis graueri Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 213: 70 km. west of south end of Lake Edward, Belgian Congo. 1929. Leptopelis rugegensis Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 218: Rugege Forest, Belgian Ruanda. Range. Virgin forests of southwest Uganda; Belgian Ruanda- Trundi and the Belgian Congo. Leptopelis parkeri Barbour & Loveridge ""' 1928. Leptopelis parJceri Barbour & Loveridge, Mem. Mus. Comp. Zool.. 50, p. 236, pi. iv, figs. 9-10: Vituri, Uluguru Mountains, Tan- ganyika Territory. 1929. Leptopelis martiensseni Ahl, Sitzb. Ges. Naturf. Freunde Berlin. p. 219 : ? Amani or Tanga, Tanganyika Territory. Range. Virgin forests of ITsambara and Uluguru Mountains, Tanganyika Territory. Leptopelis ulugxuxiensis Barbour & Loveridge 192S. Leptopelis ulugnrKeitsis Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 235, pi. iii, fig. 3 : Nyange, Uluguru Mountains, Tanganyika Territory. 1929. Leptopelis amaniensis Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 197 : Amani, Usambara Mountains, Tanganyika Territory. 1929. Leptopelis usamharae Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 205: Usambara Mountains, Tanganyika Territory. 1929. Leptopelis tanganus Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 221: Tanga, Tanganyika Territory. 206 The numerous specimens from U.sambara localities listed as rujus Reiche- iiow by Nieden (191.5) are referable to vermiculatiis. 207 Some early and one recent record of cM/f/s Reichenow are referable to this species. 320 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1929. Leptopelis grandiceps Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 207 : Amani, Usambara Mountains, Tanganyika Territory. Range. Virgin forests of Usambara and Uluguru Mountains, Tanganyika Territory. Genus HYLAMBATES Dumeril 1858. Eylambates A. Dumeril, Ann. Sei. Nat., (3) 19. p. 162, Type by monotypy: H. maculatus A. Dumeril. 1950. Phlyctimantis Laurent & Combaz, Revue Zool, Bot. Afr., 43. p, 274: Type by original designation: Eylambates leonardi Bou- lenger. Hylambotes maculatus Dumeril Red-blotched Black Frog 1853. Hylamhates maculatus A. Dumeril, Ann. Sci. Nat., (3) 19. p. 165, pi. vii, figs, lib and 4: Zanzibar Island. 1944. Kassina poweri Hoffman, Soolog. Navors. Nas. Mus. Bloemfontein, 1, p. 169, figs. 1-3: St. Lucia, Estuary, Zululand. Range. Kenya Colony ; Tanganyika Territory ; Zanzibar Island ; Mozambique and Nyasaland ; Zululand, Natal. Hylambates verrucosus Boulenger 1912. Ilylambates verrucosus Boulenger, Ann. Mag. Nat. Hist., (8) 10. p. 141: Mabira Forest, "Chagwe, " i.e. Kyagwe, Uganda. Range. Uganda and eastern Belgian Congo (fide No])le; also Witte). Genus KASSINA Girard 208 1853. Kassina Girard, Proc. Acad. Nat. Sci. Philadelphia, p. 421. Type by monotypy: Cystignathus senegalensis Dumeril & Bibron. 1864. Cassina Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 182. Emenda- tion for Kassina Girard. 1907. Paracassina Perraca, Boll. Mus. Zool. Anat. Comp. Univ. Torino, 22. No. 553, p. 3. Type by misidentification : Cystignathus sene- galensis Dumeril & Bibron.^o^ 1924. ITornierella Ahl, Mitt. Zool. Mus. Berlin, 11. p. 10. Type by subse- quent designation of Ahl: 1926: T. pulclira Ahl = ? 208 This genus, so frequently redescribed on the basis of juveniles or adults in which the vomerine teeth are absent, is much in need of a comprehensive revision. 209 The description of Paracassina was based on two frogs (senegalensis) from Toro, Uganda, which Peracca had misidentifled as Cassitia obsciira Boulen- ger. Consequently, as pointed out by Pariser (1930, Proc. Zool. Sec. London, p. 31), Paracassina cannot be referred to the synonymy of Mocquardia Ahl (new name for Rothschildia Mocquard, preoccupied by Rothschildia Grote:lS96). LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 321 1937. Cassiniopsis Monard, Bull. Soc. Neuchatel. Sci. Nat., 57. pp. 13, 41. Type by original designation: C. kuvangensis Monard. 1939. Scmnodactyhts Hoffman, Soolog. Xavors. Nas. Mus. Bloemfontein, 1, p. 89. Type by monotypy: S. thahanchuensis Hoffman. 1940. Kassinula Laurent, Revue Zool. Bot. Afr., 33, p. 313. Type by original designation: K. ivittei Laurent = Cystignathus sene- galensis Dumeril & Bibron (juv.). Kassina senegalensis (Dunieril & Bibron )-'° 1841. Cystignathus ScJicgalensis Diuneril & Bibron, Erpet. Gen., 8, p. 418: Galam Lakes, Senegal. 1854. Cystignathus argyreivittis Peters, Monatsb. Akad. Wiss. Berlin, p. 626 : Cabaceira, Mozambique. 1917. Megali.ralus macnlosu-s Sternfeld, Wiss. Ergebn. Zweiter Deutseh. Zentral-Afrika-Exped. 1910-1911, 1, p. 501: Duma, Ubangi, French Equatorial Africa.-'^ 1930. Kassina descrticola Alil, Zool. Anz., 83. p. 280: Windhuk, South- west Africa. 1930. Kassina modesta Ahl, Zool. Anz., 88. p. 281; Marianhill, Natal. 1933. Cassina angeli Witte, Revue Zool. Bot. Afr., 23. p. 172: Lukafu, Kundelungu, Belgian Congo. '^- 1940. Kassinula ivittci Laurent, Revue Zool. Bot. Afr., 33. p. 314, figs. lb and 2 (based on juveniles of from 12.5 to 14 mm.) : Kan- senia, Belgian Congo. 1942. Kassina senegalensis ovambocnsis Hoffman, Soolog. Navors. Nas. Mus. Bloemfontein, 1. pp. 150, 155, figs. 26-27: Ovamboland, Southwest Africa. Range. Sudan, east to Ethiopia (and possibly British Somali- land), south through Uganda, Kenya Colony and Tanganyika Territory to Natal, west to Cape Province, north to the Cain- eroons, west (with some gaps) to Senegal. Genus AFRIXALUS Laurent -^^ -10 Dr. V. FitzSimons informs me (22..X.56) that he regards Cassina s. inter media Werner (1898) as a synon.vm of wcalii Boulenger : while A', poiceri Iloff man (1944) was based ou a specimen of Ili/lamhatiS maculatus. 211 Dr. R. Mertens has found the type of maciilosus to be a juvenile senegalen- sis. 212 Dr. K. Laurent recentl.v (1954) revived angeli (which I synonymized in 193(5) as a subspecies, but on grounds that appear imlikely to stand the test of a thorough, continent-wide study of this group. Such a revisionary examination may well result in additions to the foregoing synonymy. 213 Originally proposed as a subgenus without designation of type. Whether there is suthcient justitication for the separation by Laurent of Acanthixalus (for spinosus Buchholz & Peters) or Hctaixalas (for Malagasy members for- merly assigned to Mcgali.valus) remains to be seen. Megalixalus is certainly monotypic with seijchillensis as the only species. 322 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1944. Afrixalus Laurent, Eevue Zool. Bot. Afr., 38. p. 113. Type by present designation: M.egalixalus fornasinii congicus Laurent = EyperoUus dorsalis Peters. Afrixalus fornasini fornasini (Bianconi) Mozambique Bauana-Frog 1849.-^'* Euchnemis Fornasini Bianconi, Nuovi Ann. Sci. Nat., (2) 10. 1848, p. 107, pi. V, fig. 1; also 18o0, Spec. Zool. Mozamb., Eept., p. 23, pi. V, fig. 1: Mozambique. 1854. Hyperolius hivittatus Peters, Monatsb. Akad. Wiss. Berlin, p. 627: Boror, Mozambique. 1862. Hyperolius spinifrons Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 342 : Umvoti, Natal. 1913. Megalixalus fornasinii var. unicolor Boettger, in Voeltzkow, Reise in Ostaf rika, 3. p. 349 : Pemba Island, Zanzibar. 1920. Megalixalus loveridgii Procter, Proc. Zool. Soc. London, p. 418, fig. 4 : Morogoro, Tanganyika Territory. 1930. Megalixalus dorsimaeulatus A hi, Sitzb. Ges. Naturf. Freunde Berlin, p. 92: Magrotto, near Tanga, Tanganyika Territory. Range. Coastal Kenya Colony ; Tanganyika Territory ; Pemba, Zanzibar and Mafia Islands ; Mozambique ; Natal and extreme eastern Cape Province (unless spinifrons proves to be a recogniz- able race). Afrixalus fornasini dorsalis ( Peters j Cameroon Banana-Frog 187.5. Hyperolius dorsalis (Schlegel) Peters, Monatsb. Akad. Wiss. Ber- lin, p. 206, pi. i, fig. 2: Boutry, Ashanti, Gold Coast, and Victoria, British Cameroon. 1889. Megalixalus schneideri Boettger, Ber, Senckenberg. Naturf. Ges., p. 276: Bonamandune (King Bell Dorf), Cameroon. 1941. Megalixalus fornasinii congicus Laurent, Revue Zool. Bot. Afr., 35. p. 120, fig. 1: Buta, Uele, Belgian Congo. 1951. Afrixalus dorsalis regularis Laurent, Ann. Soc. Roy. Zool. Belgique, 82, p. 25, fig. 1 : Lambarene, Ogooue River, French Congo. Range. Uganda, west through Belgian Congo to French Guinea. Afrixalus fulvovittatus leptosomus (Peters)^" 214 April, 1S40, according to Sherborn, but Bianconi (1S59, Mem. Accad. Sci. Bologna, 10. p. 498) claims the name (with a single i ending) was first pub- lished on 4.ii.l847 ' 215 In earlier times there was much confusion regarding these frogs ; both names were freely applied to specimens from the eastern seaboard which were, in reality, A. b. brachi/cncmis Boulenger. The only places in East Africa where I have encountered leptosomus is at Kaimosi. K. C., and Ujiji, T. T. The latter were recorded by me as brachi/vnvmis, and I am indebted to Dr. Laurent for examining them and pointing out that they are actually juvenile leptosomus. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 323 1877. Hyperolius Icptosomus Peters, Monatsb. Akad. Wiss. Berlin, p. 619, pi. — , fig. 5: Chinchoxo, Cabinda. 1931. Hr/pcroUus brevipalviatus Ahl, Mitt. Zool. Mus. Berlin, 17, p. 25: Saiignielima, French Cameroon. 1941. MegaJixaUts leptosomus vpembae Laurent, Eevue Zool. Bot. Afr., 35. p. 125, fig. 3: Nyonga, Lake Upemba, Belgian Congo. Range. Sudan; Uganda; western Kenya Colony; western Tanganyika Territory, west to Angola, north to the Cameroons, and possibly Dahomey (fide Laurent). ^^^ Alrixalus brachycnemis brachycnemis (Boulenger)^^^ Short-limbed Banana-Frog 1896. Megalixalus brachycnemis Boulengcr, Ann. Mag. Nat. Hist., (6) 17, p. 403, pi. xvii, fig. 2: Chiradzulu, Nyasaland. 1931. Hyperolius pygmaeus Ahl, Mitt. Zool. Mus. Berlin, 17. p. 22: Tanga, Tanganyika Territory. 1931. Eyperolius viultifasciatus Ahl, Mitt. Zool. Mus. Berlin, 17, p. 24: Rungwe, Tanganyika Territory. 1931. Hyperolius ipianae Ahl, Mitt. Zool. Mus. Berlin, 17, p. 43: Ipiana, Tanganyika Territory. 1931. Hyperolius unicolor Ahl, Mitt. Zool. Mus. Berlin, 17, p. 122: Ipiana, Tanganyika Territory. Range. Eastern Kenya Colony ; eastern Tanganyika Terri- tory; Zanzibar and Mafia Islands; Mozambique; Nyasaland. Afrixalus uluguruensis (Barbour & Loveridge) Montane Banana-Frog 1928. Megalixalus uluguruensis Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 231, pi. iii, fig. 2: Vituri, Uluguru Mountains, Tanganyika Territory. Range. Virgin forests of Usambara; Magrotto and Uluguru Mountains, Tanganyika Territory. 216 A. /. fulvovittatue: (Cope), inc. rittiger Peters, ranges from northern Nigeria west to Senegal. « 217 Manv earlv East African records of '■fuliovittatus" and "leptosomus" refer to this species. " Trinomials are used on account of the western A. b. wiedholzi (Mertens :1938), of which Mcgalixaliis schoiitedenl Laurent is a synonym. As the range of typical brachycnemis is largely conhned to the coastal region inland to Lake Nvasa, I (195.3* was probably wrong in synonymizing the slightly larger A. b. orophilus Laurent from the Central .\frican highlands. 324 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Genus HYPEROLIUS Rapp'^' 1842. Hyperolius Kapp, Arch. Naturg., 8. Abt. 1, p. 289. Type by mono- typy : H. marmoratiis Rapp. 1864. Eappia Giinther, Zool. Kec, 1, p. 130. New name for Hyperolius Rapp, thought to be preoccupied by Uperolia Gray. A. Species represented in the Museum of Comparative Zoology by material (often types, paratypes or topotypes) from those countries indicated liy an asterisk (*) in Range. Hyperolius viridiflavus viridiflavus (Dumeril & Bibron) 1841. Euon.emis viridiflavus Dumeril & Bibron, Erpet. Gen., 8, p. 528: Ethiopia. 218 This, to me the most perplexing of all African genera, is still the least understood and most unsatisfactory. In 1931 Hyperolius was reviewed by Ernst Ahl, who described 98 additional species, many of which were composites of distinct forms having somewhat similar patterns. .\hl failed to realize that the juvenile livery of many "species" or races is indistinguishable, though dif- fering niarivedly from that of their respective subadults, which again may be totally unlike the mature frogs. The males of the latter sometimes have entirely different markings from those of their own females. In ]9;U Ahl published two voluminous papers, that in Das Tierreich (No. 5.'), Amphibia Anura, III, xvi + 477 pp.) bears the date "March 19.31," and, having figures of the numerous species, is the more useful. The other (Mitt. Zool. Mus. Berlin, 17, pp. 1-132), presumably intended to appear first for it carries the abbreviation "spec, nov." after each of the new names, is dated "1 April 1931." From it one can obtain type localities and get an idea of what paratypic material (in many instances mistakenly called "cotypen") was at the author's disposal. The seemingly meticulous, and extremely detailed descriptions actually do not cover the variations of Ahl's available material ; often they are flagrantly careless and misleading. It is but fair to warn workers that the keys in Das Tierreich are worse than useless, being based on characters of a trivial or variable nature common to a host of forms that differ solely in their striking color patterns. When Ahl's series of "cotypen" was adequate, a selection of them was obtained in 1932 by exchange for the Museum of Comparative Zoology. The avail- ability of this material made it possible for me to synonymize almost 50 of Ahl's alleged species ; Laurent has similarly dealt with some others. We have not always agreed : in some instances, no doubt, because of the composite nature of Ahl's "species," its faded condition, or other cogent reasons. Laurent, who has done more work on this genus than anyone, has himself proposed at least GO new Hyperolius. Following a decade of bewildering nomen- clatorial changes in which species were synonymized or revived as racial entities of this or that form, which in turn was liable to be demoted to subspecific rank of something else, Laurent (1951) endeavored to bring order out of chaos by assigning many of these names to subspecific status of much earlier forms like marmordtus, viridiflarus, etc. In preparing this check list I hoped to follow Laurent in this laudable effort, but encountered so many geographical difficulties that — in the interests of stability — it seemed advisable to adhere to binomials except where I have previously used trinomials. Departing from the format used elsewhere in this check list, I have divided these sedge- or tree-frogs into two groups. The first (A) contains those species or subspecies represented by material in the Museum of Comparative Zoology (mdicated by an asterisk*) that, as of today, do not seem capable of further synonymization. The ranges given have usually been restricted to the country of origin of the type series and its synonyms, together with those countries from which I have, or have seen, material whose identifications I have been able to verify. From what I have said, I hope it will be clear to enthusiastic local amateurs that ranges should not be extended by the inclusion of early extra- territorial records of the same "species," even though identified by a competent herpetologist of the period. Such book work merely perpetuates confusion. To the second group (B, see p. 334) are assigned those "species," usually known only from a single holotyi>e, that I have not seen. In some instances it is possible that the species in question may not even be referable to Hyperolius, whose horizontal pupil is the only external character differentiating it from Afrixalua — whose pupil is vertical. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 325 Range. Ethiopia, south to Bannissa, Northern Frontier Prov- ince, *Kenya Colony. Hyperolius pachydenna "Werner ^^^ 1908. Bappia pachyderma Werner, Sitzb. Akad. Wiss. Wien, 1907, 116, Abt. 1, p. 1403: Goudokoro, Sudan. 1921. Rappia burgeoni Witte, R^vue Zool. Afr., 9, p. 19, pi. v, figs. 2-2b: ' ' Madyu, Uele ' ' = Gangula Madi on Uele River, midway between Bambili and Niangara, Belgian Congo. 1925. Rappia Bossii Calabresi, Atti. Soc. Ital. Sci. Nat. (Milano), 64, p. 121, fig. — : Upper Uele region, Belgian Congo. 1931. Hi/peroUus phrynoderma Ahl, Das Tierreich, no. 55, p. 344, fig. 218: No locality, but collected by Deutsche Zentral-Afrika Expedition. 1931. Hyperolius ocidatus Ahl, Das Tierreich, no. 55, p. 377, fig. 252: Balaibo, banks of Duki River, Belgian Congo. Range. *Sudan; *Uganda; *Kenya Colony; *Belgian Cono;o. Hyperolius picturatus Peters"" 1875. Hyperolius picturatus "Schlegel" Peters, Monatsb. Akad. Wiss. Berlin, p. 206, pi. ii, fig. 2: Boutry, Ashanti, Gold Coast. 1875. Hyperolitis nitidulus Peters, Monatsb. Akad. Wiss. Berlin, p. 209, pi. iii, fig. 4: Yoruba, Lagos, Nigeria. Range. *Western Kenya Colony ; *Uganda ; *Belgian Congo ; *French Congo; Sao Thome; *French Cameroon; Togoland; Gold Coast ; *Ivory Coast ! *Liberia ; *Sierra Leone. Hyperolius marginatus Peters 1854. Hyperolius marginatus Peters, Monatsb. Akad. Wiss. Berlin, p. 627: Maqanga, Mozambique. 1931. Hyperolius pictus Ahl, Das Tierreich, no. 55, p. 301, fig. 176: Crater Lake, Ngosi Volcano, Tanganyika Territory. 1931. Hyperolius ngoriensis Ahl, Das Tierreich, no. 55, p. 324, fig. 198: Crater Lake, Ngosi Volcano, Tanganyika Territory, Range. Montane marshes of southern * Tanganyika Terri- tory ; Mozambique and Nyika Plateau, northern *Nyasaland. Hyperolius goetzei Ahl 1931. Hyperolius goetaei Ahl, Das Tierreich, no. 55, p. 413, fig. 286: Uhehe, Tanganyika Territory. 219 With respect to the first two names (Rappia pachyderma was based on a juvenile of generalized appearance), I have followed Laurent (1951) who regards pachyderma as a race of viridiflavus. Hitherto I have referred to this frog as rosgii. 220 The alleged example from Pemba Island, reported by Boettger (1913), on being re-examined by Mertens (1940) was found to be a juvenile Afrixalus f. fornasini. 326 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Range. Uplands of *Kenya Colony and *Tanganyika Terri- tory. Hyperolius discodactylus Ahl 1931. Hyperolius discodactylus Ahl, Das Tierreich, no. 55, p. 363, fig. 239 : Rugege Forest, Belgian Euanda-Urundi, and west of Lake Edward, Belgian Congo. 1931. Hyperolius alticola Ahl, Das Tierreich, no. 55, p. 379, fig. 255: at 1800 metres, Euwenzori Mountains, Belgian Congo. Rcmge. *Ruwenzori Mountains, Uganda; *Belgian Congo. Hyperolius montanus (Angel) 1924. Rappia montana Angel, Bull. Mus. Hist. Nat. (Paris), 30. p. 269: Mount Kinangop, Aberdare Mountains, Kenya Colony. Range. Montane marshes of Mts. Kinangop and Kenya, *Kenya Colony. Hyperolius pantherinus (Steindachner) 1891. Megalixalus pantherinus Steindachner, Anz. Akad. Wiss. Wien, 28. p. Ii2: Laikipia, Kenya Colony. Range. Kenya Colony (known only from the holotype : photo inM.C.Z.V Hyperolius bayoni (Boulenger)"^ 1911. Uappla bayoni Boulcnger, Ann. Mus. Civ. Stor. Nat. Genova, (3) 5. p. 1G8: Bulolo; Busu ; Enteblje; Jinja; Kabulamuliro ; and Mbale, Uganda. 1931. Hyperolius stuhmarini Ahl, Das Tierreich, no. 55, p. 395, fig. 271: Vitchumbi, south end of Lake Edward, Belgian Congo. 1940. Hyperolius ornatus Laurent, Eevue Zool. Bot. Afr., 34, p. 4: Eutshuru, Belgian Congo. Range. *Uganda, southwest to Belgian Ruanda-Urundi (also reported from Mozambique by Parker in 1931). Hyperolius kivuensis Ahl""' 1931. Hyperolius kivuensis Ahl, Das Tierreich, no. 55, p. 280, fig. 151 (34 mm. $, Kandt coll.): Lake Kivu, Belgian Congo. 1931. Hyperolius kwidjwiensis Ahl, Das Tierreich, no. 55, p. 296, fig. 172 (32 mm. S, Schubotz coll.): "Kwidjwi," i.e. Idjwi Island, 221 A 9 frog (M.C.Z. 10287) from the Uganda shore of Lake Eflward differs slightly from a topotype stuhlmanni (M.C.Z. 2S126). 222 In 1951 Laurent regarded kicidjiciensis (with which he would include kandt i) as distinct. He would transfer koehi to the synonymy of schubotzi Ahl (in whose synonymy I place the 37 mm. macrodaeti/lus) . The series of frogs from Nyamliolo, Lake Tanganyika, N. Rhodesia which in 19:50 I referred to rhodoscelis in error, appear to be indistinguishable from Kivu frogs. LOVERIDGE : E. AFRICAN REPTILES AND AMPHIBIANS 327 Lake Kivii, Belgian Congo. 1931. Hyperolius kandti Ahl, Das Tierreich, no. 55, p. 327 (33 mm. S, Kandt coll.) : Lake Kivu, Belgian Congo. 1931. Hyperolius l-oehli Ahl, Das Tierreich, no. 55, p. 405 (33 mm. S , Koehl coll.) : Kisenyi, Lake Kivu, Belgian Euanda-Urundi. Range. *Soiitlnvest Uganda ; *western Tanganyika Territory ; ^Northern Rhodesia ; *eastern Belgian Congo ; *Belgian Ruanda- Urundi. Hyperolius variabilis Ahl""^ 1931. Hyperolius variabilis Ahl, Das Tierreich, no. 55, p. 297, fig. 173: restricted to Bukoba, Tanganyika Territory. 1931. Hyperolixis wettsteini Ahl, Das Tierreich, no. 55, p. 343, fig. 217: Bukoba, Tanganyika Territory. 1931. Hyperolius mohasicus Ahl, Das Tierreich, no. 55, p. 360, fig. 236: Lake Mohasi, Belgian Euanda-Urundi. 1931. Hyperolius monticola Ahl, Das Tierreich, no. 55, p. 376, fig. 251 : "Niansa," i.e. Nyanza, Belgian Euanda-Urundi. 1931. Hyperolius irregularis Ahl, Das Tierreich, no. 55, p. 396, fig. 272: Lake Mohasi, Belgian Euanda-Urundi. Range. Montane meadows of southwest *Uganda; Belgian Ruanda-Urundi and adjacent *Belgian Congo. Hyperolius latifrons Ahl"-* 1931. Hyperolius latifrons Ahl, Das Tierreich, no. 55, p. 335, fig. 208: Bamboo forest at 2400 metres, near Mhi-abu Clahama village. Mount Karisimbi, Belgian Euanda-Urundi. 1931. Hyperolius Icarissimhiensis Ahl, Das Tierreich, no. 55, p. 348, fig. 223 : Bamboo forest at 2400 metres, near Mhcabu Gahama village, Mount Karisimbi, Belgian Euanda-Urundi. Range. *Western Tanganyika Territory ; Belgian Ruanda- Urnndi and *Belgian Congo. Hyperolius multicolor Ahl"-' 1931. Hyperolius multicolor Ahl, Das Tierreich, no. 55, p. 367, fig. 243: Bamboo forest at 2400 metres, near Mhcabu Gahama village, 223 Considered a race of viridiflavus by Laurent (1951) who would add punc- tntis-simus Ahl and flaroguttatus (part) to the above synonymy. While Ahl's material was often composite I scarcely think it warrants such action ; however, icettsteini and monticola are added fide Laurent. 224 Both latifrons and karissimbiensia were based on juveniles of 21 and 23 mm. respectively. Laurent (1951) regards karissimbiensis as a distinct race of viridi- flavus to which he would add multicolor Ahl. 225 Our cotype and series of multicolor from Lake Bunyoni (with whose identi- tication Laurent concurs) do not support its synonymization with "karissimbien- sis." 828 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Mount Karisimbi, Belgian Ruanda-Urundi. Range. Montane meadows of southwest *Uganda; *Belgian Ruanda-Urundi and adjacent Belgian Congo. Hyperolius bitaeniatus Ahl"^'' 1931. Eyperolhts hitaeniatvs Ahl, Das Tierreich, no. 55, p. 322, fig, 196: Ukonde-Unyika, Tanganyika Territory. Rcmge. Southwestern Tanganyika Territorj^ and nearby *Northern Rhodesia. Hyperolius fuelleborni Ahl"^ 1931. Hyperolius fuellchorni Ahl, Das Tierreich, no. 55, p. 349, fig. 224: ' ' Langenburg, ' ' i.e. Manda, Lake Nyasa, Tanganyika Territory. Range. Region around northern end of Lake Nyasa, southwest *Tanganyika Territory. Hyperolius mariae Barbour & Loveridge^^^ 1928. Hyperolius mariae Barbour & Loveridge Mem. Mus. Comp. Zool., 50, p. 217, pi. iii, fig. 1: Derenia, Usambara Mountains, Tan- ganyika Territory. 1931. Hyperolius melanophthalmus Ahl, Das Tierreich, no. 55, p. 341, fig. 215 : Zanzibar. 1931. Hyperolius renschi Ahl, Das Tierreich, no. 55, p. 397: Zanzibar. Range. Coastal *Tanganyika Territory and *Zanzibar Island."^ Hyperolius flavoguttatus Ahl'^° 1931. Hyperolius flavoguttatus Ahl, Das Tierreich, no. 55, p. 369, fig. 245 : Bukoba, Tanganyika Territory. Range. *Kenya Colony and *Tanganyika Territory. 226 At one time (1943) Laurent thought bitaeniatus might be a synonym of rhodoscelis Bonlenger (with which he correctly synonymlzes iinduJatus Boulen- gcr). A comparison of our andidatus cotype with our cotype of hitaeniatiis con- vinces me thpy are distinct. Wo have also a good series of quite typical bitaeniatus from Abercorn. More recently (19.51) Laurent has, mistakenly I think, referred bitaeniatus to the synonymy of H. mariae Barbour & Loveridge. 227 Laurent (1951) refers fuelleborni to the synonymy of nyassae Ahl. We have cotypes of both and I cannot agree ; to my thinking "nyassae" is a synonym of rhodoscelis (Boulenger). 228 Laurent (1951) adds noblel Ahl to the synonymy; I adhere to my (1936) alignment of it with puncticulatus (Pfeffer). 229 My earlier record from Lake Nyasa was based on a mlsidentified fuelleborni. 230 AH that the Museum of Comparative Zoology has is a bleached cotype from Bukoba ; a correctly colored topotype from Mount Kenya, and a frog from Mount Mbololo somewhat doubtfully attributable to flavoguttatus. These do not suggest that Laurent is correct in synonymizing flavoguttatus with variabilis Ahl. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 329 Hyperolius udjidjiensis AhP^^ 1931. Hyperolms udjidjiensis Ahl (part), Das Tierreich, no. 55, p. 370, fig. 246: restricted to Ujiji, Tanganyika Territory. Range. * Tanganyika Territory. Hyperolius striolatus Peters ^^^ 1882. Hyperolius striolatus Peters, Sitzb. Ges. Naturf. Freunde Berlin, p. 9: "Taita," i.e. Teita, Kenya Colony. 1902. Eappia ferniquci Mocquard, Bull. Mus. Hist. Nat. (Paris), 8, p. 407: Athi River, Kenya Colony. 1931. Hyperolius caeruleopunctatus Ahl, Das Tierreich, no. 55, p. 350, fig. 225: Nairobi and Kibwezi, Kenya Colony. 1931. Hyperolius pulchromarmoratus Ahl, Das Tierreich, no. 55, p. 367, fig. 242: "British East Africa" (Hiibner of Kibwezi). 1931. Hyperolius udjidjiensis Ahl (part). Das Tierreich, no. 55, p. 370, fig. 246 (only the paratypes from Kibwezi, Kenya Colony). 1931. Hyperolius scheffleri Ahl, Das Tierreich, no. 55, p. 387: Kibwezi. Kenya Colony. Range. *Kenya Colony. Hyperolius rhodoscelis (Boulenger) 1901. Eappia rhodoscelis Boulenger, Ann. Mus. Congo, (1) 2, fasc. 1, p. 3, pi. ii, figs. 1-la: Pweto, Lake Mweru, Belgian Congo. 1901. llappia undulata Boulenger, Ann. Mus. Congo, (1)2, fasc. 1, p. 4, pi. ii, fig. 2: Pweto and Lofoi, Lake Mweru, Belgian Congo. 1902. Eappia symetrica Mocquard, Bull. Mus. Hist. Nat. (Paris) 8. p. 408: Athi Eiver, Kenya Colony. 1920. Eappia platyrhinus Procter, Proc. Zool. Soc. London, p. 416, fig. 3 : Nairobi, Kenya Colony. 1931. Hyperolius asper Ahl, Das Tierreich, no. 55, p. 310: Nairobi, Kenya Colony. 1931. ^Hyperolius hreviceps Ahl, Das Tierreich, no. 55, p. 316, fig. 190 (at least the paratypes from Eldama Ravine, Kenya Colony, are rhodoscelis): "Tschimbo" = Chimbo, Mozambique. 1931. Hyperolius nyassae Ahl, Das Tierreich, no. 55, p. 339, fig. 213: "Langenburg, " i.e. Mauda, Lake Nyasa, Tanganyika Terri- tory. Range. ? *Southeru Sudan; *Kenya Colony; *Tanganyika Territory; ? Mozambique; ? *Angola; southern *Belgian Congo. 231 From Ujiji we have a topotype series consisting of a ^ and five 9 9 that I collected In 1939. I regard them as distinct from striolatus with which species Laurent (1951) would synonymize udjidjiensis (part). It seems ques- tionable whether Ahl's cotype from Kibwezi is conspecific. 232 To the synonymy of striolatus, Laurent would add symetrica, platyrhinus, asper and the Baringo specimen of breviceps Ahl. Geographically this seems plausible but is incorrect unless the species is dichromatic. 330 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Hyperolius punctatissimus Ahl"^'^ ]P31. Hyperolius punctatiftsiniun Alil, Das Tierreich, no. 55, p. 299, fig. 174: restricted to Bukoba, Tanganyika Territory. Range. * Tanganyika Territory. Hyperolius vermicularis Ahl""* 1893. Fappin vermiculata Pfeffer (not of Peters: 1882), Jahrb. Ham- burg. Wiss. Anst., 1892, 10. p. 98, pi. i, fig. 12: Zanzibar. 1931. Hyperolius vermiailaris Ahl (new name for vermiculata Pfeffer, preoccupied by an Angolan species), Das Tierreich, no. 55, p. 275, fig. 145. Range. *Kenya Colony; Zanzibar Island. Hyperolius marmoratus onunatostictus Laurent "'^^ 1951. Hyperolius marmoratus ommatostictus Laurent, Ann. Soc. Roy. Zool. Bclgique, 82. p. 36: "Kibouoto," i.e. Kibongoto, Mount Kilimanjaro, Tanganyika Territory. Range. ^Tanganjnka Territory. Hyperolius orgentovittis Ahl"^'' 1931. Hyperolius argcntovittis Ahl, Da.s Tierreich, no. 55, p. 345, fig. 220: Ujiji, Tanganyika Territory. 1931. Hyperolius callichromus Ahl (part), Das Tierreich, no. 55, p. 372, fig. 248: Western banks of "Russisi, " i.e. Ruzizi River and northwest shores of Lake Tanganyika, Belgian Congo. 233 Laurent (1951) would synonymize this also with i-ariabUis Ahl, but our solitary specimen from Kahare. r.ukol)a District, seems to show relationship with strioldtiis rather than with variahUis. -31 Laurent (1931) would add sci-iptux Ahl lo the synonymy. However, I failed to meet with vermicularis in the vicinity of Tanga (from whence came the holo- type of scriptus), obtaining our abundant material from seven localities north of Mombasa. 235 The Museum of Comparatve Zoology topotypes (M.C.Z. 21009-11) were taken at the same time as the type series. Kibougoto is not very high, viz. 1300 metres, roughly 4223 feet. -36 In their tibial pattern our long series of argcntorittis topotypes form a linU with the speckled tibial pattern exhibited by our callicluomus cotypes (M.C.Z. 17630-1). In the opposite direction — on the southeast shore of I>ake Tanganyika — the even larger series I captured at Nyamkolo are remarkably uniform, dis- playing a single red-brown streak on the tibia that is only occasionally broken up into two or three elongate blotches. Some of our cotypes of brieni Laurent (1943, Ann. Mus. Congo Beige, (1) 4, fasc. 2, p. 119, fig. 30 only), viz ^ and 5 (M.C.Z. 26264, 26266) have a dorsal pattern, and differ but slightly in tibial pattern, from argentovittis. Yet our other cotypes of brieni ^ and J (M.C.Z. 26263, 26205), also from Nyonga, Upper Luapula, have the vermiculated dorsal patterns so admirably portrayed in Laurent's figure 37. Thej' are unquestionably referable to the same form as the frogs from Kabengere. Belgian Congo, that I have referred to graucri Ahl. Either brieni is a cuiiiposite. being part argcnto- vittis and part graucri, or the species is dichromatic, in which case Laurent is correct in adding graueri to the synonymy of ai gcntoritlis and brieni should gc too. Our cotype (M.C.Z 17633) of decipiens Ahl from the Ruzizi lends no su]' port to Laurent's synonymizing of that species with nrgcnturil tis. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 331 Range. *Tanganyika Territory; *Nortlieru Rhodesia; *Bel- gian Congo. Hyperolius puncticulatus substriatus Ahl Northern Broad-striped Sedge-Frog 1931. Hypcroliios substriatus Ahl, Das Tierreich, no. 55, p. 358, fig. 234: Magrotto Mountain, Tanganyika Territory. Range. Coastal *Kenya Colony (Malindi), south through the Usambara Mountains to Morogoro, *Tanganyika Territory ; *Zan- zibar Island.'^' Hyperolius puncticulatus puncticulatus (Pfeffer)^^® Southern Broad-striped Sedge-Frog 1893. L'ap2)ia puricticulata Pfeffer, Jahrb. Hamburg. Wiss. Anst., 1892, 10. p. 31, pi. ii, fig. 2: Zanzibar. 1931. Hyperolius noblei Ahl, Das Tierreich, no. 55, p. 400, fig. 275: Kihva, Tanganyika Territory. Range. Zanzibar Island ; Uluguru and Rungwe Mountains of *Tanganyika Territory ; Misuku ; Nyika and Mlanje Mountains of *Nyasaland. Hyperolius argus ahli Loveridge"^^ Northern Argus-spotted Sedge Frog 1936. Hyperolius ahli Loveridge, Bull. Mus. Comp. Zool., 79, p. 402 : Lake Pcccatoni, northeast of Witu, Kenya Colony. Range. Coastal *Kenya Colony, south to Lindi. *Tanganyika Territory. Hyperolius argus argus Peters"*" Southern Argus-spotted Sedge-Frog 1854. Hyperolius arfjus Peters, Monatsb. Akad. Wiss. Berlin, p. 628 (l)ased on 9 > : Boror, Mozambique. 23T M.C.Z. 17162 from Mwera, Zanzibar Island, is undoubtedly referable to this form. It remains to be seen which race is dominant on the island. -38 I disagree with Laurent's action in removing noblei from the synonyiny of puncticulatus to that of mariae Barbour & Loveridge. Uganda and other records outside the range given here, are considered questionable. 239 For a coloured plate showing the striking sexual dichromatism of this raci'. whose ^ is git't'u tlt'i'Ucd with tilacK. see l.oviTidg.- 1 1942. Hull. Mus. t'oinii Zool., 91, pi. iii). 240 Admission to this list depends on the identification of the juveniles from Mafia Island reported by Parker (1937), and from the J from Zanzibar recorded by Laurent (1943) : both are somewhat questionable. My Tanganyika material identified as nrqus by Procter (1920), and consequentlv some of my own, were actually referable to p. substriatus Ahl (1931). See Parker (1931, Proc. Zool. Soc. London, pi. i) for astouishiug age and sexual dichromatism in a near topo- typic series. 332 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 1854. Hyperolius fiavoviridis Peters, Monatsb. Akad. Wiss. Berlin, p. 628 (based on S ) : Boror, Mozambique. 1854. Hyperolius Tettensis Peters, Monatsb. Akad. Wiss. Berlin, p. 628 (based on $): Tete, Mozambique. 1893. Bappia platycephala Pfeffer, Jahrb. Hamburg. Wiss. Anst., 1892, 10, p. 96, pi. ii, fig. 2 (based on $) : Quilimane, Mozambique. Range. *Mozambique ; *Nyasalancl. Hyperolius concolor sonsibaricus (Pfeffer) Northeastern Straw-or-Greeu Sedge-Frog 1893. Eappia sansibarica Pfeffer, Jahrb. Hamburg. Wiss. Anst., 1892. 10. p. 97, pi. ii, fig. 4: Zanzibar. 1931. Hyperolius ruhripes Ahl, Das Tierreich, no. 55, p. 361: Kililana, mainland opposite Lamu Island, Kenya Colony. Range. *Keiiya Colony ; ^Tanganyika Territory ; Zanzibar Island.-" Hyperolius concolor tuberilingvds Smith "*^ Southeastern Straw-or-Green Sedge-Frog ] 849. II yperolius tub.erilinguis A. Smith, Illus. Zool. S. Africa, Rept., App., p. 26: country eastward of Cape Colony, i.e. Natal. 1862. nyperolius coccotis Cope, Proe. Acad. Nat. Sci. Philadelphia, p. 342: Umvoti, Natal. 1864. Hyperolius citrinus Giinther (part), Proe. Zool. Soc. London, p. 311, pi. xxvii, fig. 2: restricted to "Zambezi Expedition." 1S67. ''.Hyperolius granulosus Peters, Monatsb. Akad. Wiss. Berlin, p. 891, footnote: Mozambique. 1947. Hyperolius Jcivucnsis smaragdinxis Laurent, Ann. Mag. Nat. Hist., (11) 14. p. 292: Charre, Mozambique. 1947. Hyperolius sansiharicus lovcridgei Laurent, Ann. Mag. Nat. Hist.. (11) 14. p. 294: Kitaya, Rovuma River, Tanganyika Territory. Range. Southern *Tanganyika Territory; Mozambique; *Ny- asaland; *Natal. Hyperolius porkeri porkeri Loveridge Northeastern Brown-or-Green Sedge-Frog 1933. Hyperolius parTceri Loveridge, Bull. Mus. Comp. Zool., 74. p. 410: Bagamoyo, Tanganyika Territory. -11 The Uganda record of Roux (1910) must be considered doubtful. 242 The Charre series on which Laurent based his smaragdinus, were originallj referred by Parker (1931) to concolor (Hallowell) of Liberia. With this deter- mination I can And no fault as Liberian and Natal frogs appear Indistinguishable though it appears advisable to treat them as distinct until we linow more of the intermediate races — of which kivuensis may well be one. The 133 frogs from Kitaya, on some of which Laurent based his loveridgei, were originally assigned by me to c. oitrimis Giinther. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 333 Range. Coastal *Kenya Colony, south to Dar es Salaam, *Tanganyika Territory. Hyperolius porkeri rovumae Loveridge"^^ Southeastern Bro\vn-or-Green Sedge-Frog 1942. Hyperolius parTceri rovumae Loveridge, Bull. Mus. Comp. Zool., 91. p. 410, pi. iii, figs. 5-6: Kitaya, Rovuma River, Tanganyika Territory. Range. Coastal (Kilwa to Kitaya) *Tanganyika Territory. Hyperolius nasutus Glinther Sharp-and-blunt -snouted Sedge-Frog 1864. Hyperolius nasutus Giinther, Proc. Zool. Soc. London, p. 482, pi. xxxiii, fig. 3 : Duque de Braganca, Angola. 1895. Eappia pwnctulata Bocage, Herp. Angola Congo, p. 168: Banks of the ' ' Quanza ' ' i.e. Cuanza River, Angola. 1901. Eappia grannlata Boulenger, Ann. Mus. Congo, (1) 2, fasc. 1, p. 4, pi. ii, fig. 3 : Pweto, Lake Mweru, Belgian Congo. 1901. Eappia oxyrhynchus Boulenger, Ann. Mus. Congo, (1) 2, fasc. 1, p. 0, pi. ii, fig. 4: Pweto and Lofoi, Lake Mwenu, Belgian Congo. 1931. ^Hyperolius petersi Ahl, Das Tierreich, no. 55, p. 274, fig. 144: Mombasa, Kenya Colony. 1931. Hyperolius acuticeps Ahl, Das Tierreich, no. 55, p. 282, fig. 153: TJkonde-LTnyika, Tanganyika Territory. Range. Ethiopia {fide Parker) ; Uganda {fide Loveridge) ; *Keuya Colony; ^Tanganyika Territory; *Nyasaland; *North- ern Rhodesia; ^Southern Rhodesia; *Angola; *Belgian Congo; French Cameroon {fide Mertens :1940) ; Liberia (as nasutus subsp. Laurent :1951) ; French Guinea (as oxyrJujnchus subsp. Laurent :1951). Hyperolius pusillus (Cope)"*"* Transparent Pigmy Sedge-Frog 1862. Crumenifcra pusilla Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 343 : Umvoti, Natal. 1864. Hyperolius microps Giinther, Proc. Zool. Soc. London, p. 311, pi. xxvii, fig. 3 : Rovuma Bay, Tanganyika Territory. 243 The brown ^ and green J with white side-stripe are shown in color on plate iii, together with the two sexes of the typical form which lacks a light side stripe. 244 The heavily spotted milnei from northeast Kenya may appear to differ from pusillus of Natal, but so many intermediate conditions occur along the 2000 miles that separate them it seems advisable to ignore subspecies. Many West African frogs have been referred to pusillus in error. 334 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1932. Hyperolius iisaramoae Loveridge, Proc. Biol. Soc. Washington, 45, p. 63 (based on 5 ) : Mogogoni Swamp, south of Dar es Salaam, Tanganyika Territory. 1935. Hyperolius inilnei Loveridge, Bull. Mus. Comp. Zool., 79. p. 18 (based on $ ) : Witu, Coast Province, Kenya Colony. 1935. Hyperolius translucens Power, Proc. Zool. Soc. London, p. 339, figs. 6-7, pi. i, figs. 3-4 (based on $ and $ ) : Port St. Johns, Pondoland, South Africa. Range. Coastal *Keuya Colony ; *Tangaiiyika Territory ; *Nyasaland ; *Natal and adjacent *Cape Province. B. Species usually known only from the original descrip- tion, fre(iuently based on a single individual. This list is alphabetically arranged. Hyperolius albifrons Ahl 1931. Hyperolius albifrons Ahl, Das Tierreich, no. 55, p. 354, fig. 230: ' ' Africa. ' ' Range. Mafia Island (//'rfe Parker :1937). Hyperolius alfaofrenatus Ahl 1931. Hyperolius albofrenatus Ahl, Das Tierreich, no. 55, p. 315, fig. 189 (coll. Ule or Ulil) : "German East Africa." Range. Tanganyika Territory. Hyperolius albolabris Ahl 1931. Hyperolius albolabris Ahl, Das Tierreich, no. 55, p. 287, fig. 101 (coll. Neumann: 26.ii.1894) : Kwa Buosch or Bnorch, Tan ganyika Territory. Range. Tanganyika Territory. Hyperolius bergeri Ahl 1931. Hyperolius bergeri Ahl, Das Tierreich, no. 55, p. 346, fig. 221 : "Guaso" i.e. Uaso Narok, Keuj-a Colony. Range. Kenya Colony. Hyperolius glandicolor Peters 1878. Hyperolius glandieolor Peters, Monatsb. Akad. Wiss. Berlin, p. 208, pi. ii, fig. 9: "Taita," i.e. Teita, Kenya Colony. Range. Kenya Colony. Hyperolius guttolineatus Ahl 1931. Hyperolius guttolineatus Ahl, Das Tierreich, no. 55, p. 321, fig. 195 (coll. Uhl): "German East Africa." 19)1. HyperoUu.'^ vuinuorafus guttatoUneatus Laurent {lapsus for gtit- LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 335 tolineatus Ahl, though page and figure numbers are wrong and should be as eited above), Ann. Sei. Roy. Zool. Belgique, 82. p. 389. Range. Tanganyika Territory. Hyperolius mctmioratus campylogrammus Laurent '^^ 19.'51. Hyperolius marmoratus campylogrammus Laurent, Ann. Soc. Roy. Zool. Belgique, 82. p. 36: Bura, 1050 metres, Teita, Kenya Colony. Range. Kenya Colony. Hyperolius quadrotomaculotus Ahl 1931. Hyperolius quadratomaculatus Ahl, Das Tierreich, no. 55, p. 412. lig. il85 : Mohoro, Tanganyika Territory. Range. Tanganyika Territor}^ Hyperolius scriptus Ahl 1931. Hyperolius scriptus Ahl, Das Tierreich, no. 55, p. 287, fig. 160: Tanga, Tanganyika Territory. Range. Tanganyika Territory. Hyperolius tomieri Ahl 1931. Hyperolius tomieri Ahl, Das Tierreich, no. 55, p. 304, fig. 179: I'kami, i.e. Uluguru Mountains, Tanganyika Territory. Range. Tanganyika Territory. Hyperolius viridiflavus pitmoni Laurent 1951. Hyperolius virieliflavus pitmani Laurent, Ann. Soc. Roy. Zool. Belgique, 82, p. 394: Lake Bunyoni, southwest Uganda. Range. Uganda. Family RANIDAE^^"" Genu.s ARTHROLEPTIDES Nieden 1910. Arthroleptides Nieden, Sitzb. Ges. Naturf. Freunde Berlin, p. 445. Type by monotypy : A. martiensseni Nieden. Arthroleptides martiensseni Nieden I^sambara Montane-torrent-Frog 245 To which Laurent wonhl assign the Mount Mbololo frog that I refer with misg^ivings to flavoyuttatua .\hl. 246 This is eniploj-pd in its generall.v accepted sense. Laurent (1951, Revue Zool. Bot. Afr.. 45, p. 120) has proposed radical changes in which he would transfer Arthrolepth and Hrmi!3 Though qtivckrtti was svuonymi/.ed liy Boulenger himself with /. fuscigula, with which it appears to agree in limb length. Dr. V. F. PitzSimons informs me ili2.x.5«» that it does not in the least resemble typical fuscigula and should, in bis opinion, be assigned to the synonymy of /. angolensis. He regards /. fuscigula as being mure or less confined to the Cape ; unfortunately short-linibed individu- als are present throughout much of the range of /. nngolensis. 340 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Mozambique ; *Nyasaland ; Northern and Southern Rhodesia ; Bechuanaland ; *Transvaal; *Natal; *Orange River; *Cape Province ; *Angola ; *Belgian Congo ; *Belgian Ruanda-Urundi. Rana fuscigula fuscigula Dumeril & Bibrou Short-limbed Dusky-throated-Frog 1841. Eajia fuscigula Dumeril & Bibron, Erpet. Gen., 8. p. 386: Cape of Good Hope, i.e. South Africa. Range. Owing to long confusion with E. f. chapini and B. f. angoJensis little reliance can be placed on records in the litera- ture. I have examined the material on which those from Kenya Colony and Tanganyika Territory were based and found them to be R. f. chapini. I have collected frogs conforming to /. fusci- gula at MiLshongero, Lake Mutanda, southwest *Uganda (which may be an area of intermediates as /. angolensis was also pres- ent there); *Nyasaland; *Bechuanaland Protectorate; *Orange River; *Cape Province. Rana christyi Boulenger 1919. Bana christyi Boulenger, Revue Zool. Afr., 7. p. 5: "Madie, " i.e. Medje, Ituri, Belgian Congo. Range. *Uganda and *Belgian Congo. Rana oxyrhynchus gribingviiensis Angel Montane-forest Sharp-nosed-Prog 1922. Bana (Ptychadena) Grihinguiensis Angel, Bull. Mus. Hist. Nat. (Paris), 28. p. 399, fig. — : Fort Crampel, Lake Chad, French Equatorial Africa. Range. Streams in montane forest, or recently deforested areas, of *Uganda; *Kenya Colony; *Tanganyika Territory; *Nyasaland ; *Belgian Congo ; French Congo ; French Equatorial Africa ; *French Cameroon ; Nigeria ; Ivory Coast ; *Liberia. Rana oxyrhynchus oxyrhynchus A. Smith Savanna Sharp-nosed-Frog 1849. Bana oxyrhynchus "Sundevall" A. Smith, Illus. Zool. S. Africa, Kept., pi. Ixxvii, figs. 2 and 2a-2c : Kaffirland and the region of Port Natal, i.e. Durban, Natal. 1868. Bana anchietae Bocage, Proc. Zool. See. London, 1867, p. 843, fig. 1 (48 mm. $ ) : Benguela, Angola. 1897. Phrynobatrachns hailensis Meek, Field Mus. Nat. Hist. Zool. Series, 1, p. 175: "Haili," i.e. Halleh, British Somaliland. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 341 1906. Sana Theileri Mocquard, Bull. Mus. Hist. Nat. (Paris), 12. p. 252: Nelspruit, Transvaal. 1923. Ptychadena aberae Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 97: Abera, near "Jandjan" (I Jiran), southwest Ethiopia. 1933. Bana oxyrliynchus migiurtina Scortecei, Atti Soc. Ital. Sci. Nat. (Milano), 72. pp. 12, 51, pi. i, fig. 3: Uadi Arro, Somalia. Range. Chiefly in stream^s of lowland or plateau savanna in the *Sudan, Eritrea, *Ethiopia, French and *British Somaliland, Somalia, *Uganda Protectorate, *Kenya Colony, ^Tanganyika Territory, *Pemba, Zanzibar and Mafia Islands, *Mozambique, *Nyasaland, Northern and *Southern Rhodesia, Bechuanaland Protectorate, *Transvaal, *Natal, *Cape Province, Southwest Africa, Cabinda, Angola, Belgian Congo and *Belgian Ruanda- Urundi. Reported from almost every country in Africa south of 15° N., but many western records may have resulted from con- fusion with the montane forest race listed above or with R. o. superciliaris Giinther of Sierra Leone. Rana floweri Boulenger-^* 1917. Bana -floweri Boulenger, Ann. Mag. Nat. Hist., (8) 20. p. 417: Rosaires, Blue Nile, Sudan. 1923. Bana erlangeri Ahl, Mitt. Zool. Mus. Berlin, 11, p. 5: Lake Abaya, northeast of Lake Stephanie, Ethiopia. 1925. Bana harhouri Loveridge, Proc. Zool. Soc. London, p. 776, pi. i. fig. 2; Nyambita, Mwanza, Tanganyika Territory. 1931. Ahrana cotti Parker, Proc. Zool. Soc. London, 1930, p. 898, fig. 1: Charre, Mozambique. Range. *Egypt (Giza: M.C.Z. 3138-40), Sudan, *Kenya Col- ony, *Tanganyika Territory, *Mozambique. (I regard the French Cameroon record of Monard, 1951, confirmed by Laurent, as somewhat questionable.) Rana mascoreniensis mascareniensis "^^ Dumeril & Bibron Common Mascarene-Frog 254 For long confused with o. oxyrhynchus, from which floweri may be dis- tinguished by the fine black vermiculatlons on its buttocks and shorter hindllmb, whose tibio-tarsal articulation, when directed forwards, reaches only to the eye or nostril (in o. oxyrhynchus to end of snout, usually far beyond). 255 Unless a $ (M.C.Z. 28622) from Liwale, southeastern Tanganyika Terri- tory is correctly assigned, so far as I am aware the perfectly valid species or subspecies R. mascareniensis mossambica Peters (1854), together with its synonyms trinotlis var. Bottgeri Pfefifer (1893) and vernayi FitzSimons (1932) does not occur in the area covered by this list. However, early records from British East Africa of the Senegalese species trinodis Boettger (1881), of which schubotzi Sternfeld (1917) is a synonym, may have been based on mossambica'! 342 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 1841. Eana Mascarcniensis Dumeril & Bibron, Erpet. Gen., 8. p. 350: Madagascar; Mauritius; Seychelles. 1855. Eana nilotica Seetzen, Eeise durch Syrian, Palast., AegjT)t, etc., 3, p. 490 : Egypt. 1857. Eana savignyi Jan, Cenni. Mus. Civ. Milano, p. 52 (probably Egypt, but this work is not available to me). 1864. ■^^Eama idae Steindachner, Verb. Zool.-Bot. Ges. Wien, 14. p. 266, pi. xii, fig. 1 : Madagascar. 1864. lEana nigrescens Steindachner, Verb. Zool.-Bot. Ges. Wien, 14. p. 268, pi. xii, figs. 2 and 2a-2c : Madagascar. 1865. Eana spinidactyla Cope, Proc. Acad. Nat. Sci. Philadelphia, p. 197: Natal. 1881. Eana ahyssinica Peters, Sitzb. Ges. Naturf. Freunde Berlin, p. 163: Ailet near Massawa; and Keren, Bogos, Eritrea. 1886. Eana Neivtanii Bocage, Jorn. Sci. Lisboa, 11, pp. 70, 73: Sao Tome, Gulf of Guinea. Range. *Egypt ; *Sudan; *Etliiopia; * Uganda; *Kenya Col- on}^ ; *Tanganyika Territory ; *Zanzil)ar and Mafia Islands ; *Madagascar : *Mozambique ; *Nyasaland ; *Northern and ^Southern Rhodesia; *Transvaal; *Angola ; *French Congo; *Nigeria; *Gold Coast; *Ivory Coast; *Liberia ; *Sierra Leone. Besides which it has been reported from almost every country in Africa, but due to confusion with closely related forms I have listed only those countries represented in the collection of the Museum of Comparative Zoology. Rana mascareniensis venusta Werner Primary-forest Mascarene-Frog 1908. Eana vcnusia Werner, Sitzb. Akad. Wiss. Wien, 1907, 116, Abt. 1. pp. 1889, 1892, pi. iv, fig. 11: Mongala, Sudan; Entebbe. Uganda ; Lagos, Nigeria. Range. Primary or gallery forested areas of the southern Sudan; *Uganda; *Kenya Colonj^; *Tanganyika Territory, west through *Belgian Ruanda-Urundi and *P)elgian Congo; *French and British Cameroon to Lagos, Nigeria. Rana mascareniensis uzungwensis LoAcridge Montane Mascarene-Frog 1932. Eana mascareniensis u:;ungivensis Loveridge, Bull. Mus. Comp. Zool., 72, p. 384: Dabaga, Uzungwe Mountains, Tanganyika Territory. 1952. Ptycliadena macrncepliala Laurent, Revue Zool. Bot. Afr., 46, ii. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 343 22: Upper Lubitshako River, 1900-2000 metres, south Kivu, Fizi Territory, Belgian Congo. Range. Upland marshes of southern *Tanganyika Territory ; *N3'asaland ; *Southeru Rhodesia ; *Angola ; Belgian Congo and Belgian Ruanda-Urundi. Rana loveridgei ( Laurent )"^^ 1954. Piychadena loveridgei Laurent, Ann. Mus. Roy. Congo Beige, 34, p. 14, pi. i, fig. 4; pi. ii, fig. 1; pi. iii, figs. 3-4: Tare, Busanza, ca. 1800 metres, Astrida region, Belgian Ruanda-Urundi. Range. South-west Uganda; *Tangauyika Territory; Belgian Ruanda-Urundi ; Belgian Congo ; Angola. Rana ansor(^i Boulenger 1905. Eana ansorgii Boulenger, Ann. Mag. Nat. Hist., (7) 16. p. 107, pi. iv, fig. 1 : Between Benguela and Bihe, Angola. Range. Southwest *Tanganyika Territory ; Mozambique ; *Ny- asaland; *Northern Rhodesia; * Angola north to Lower Congo River and Katanga, Belgian Congo. Rana fasciata merumontana Lonnberg Northern Tanganyika Striped Grass-Frog 1907. Rana merinnontana Lonnberg, in Sjostedt, Wiss. Ergeb. Schwed. Zool. Exped. Kilimandjaro, Meru umgeb, Massaisteppen, No. 4, p. 21, pi. i, figs. 4a-b: Mount Meru, Tanganyika Territory. Range. Montane meadows of northern *Tanganyika Terri- tory. Rana fasciata fiillebomi Nieden Southern Tanganyika Striped Grass-Frog 1910. Rana fiillehorni Nieden, Sitzb. Ges. Naturf. Freunde Berlin, p. 436: Crater Lake, Ngosi Volcano, Poroto Mountains, Tan- ganyika Territory. Range. Montane meadows of southern *Tanganyika Terri- tory and *Nyasaland. Rana stenocephala Boulenger Uganda Striped Grass-Frog 1901. Rana stenocepliala Boulenger, Ann. Mag. Nat. Hist., (7) 8, p. 515: Entebbe, Uganda. Range. *L'ganda (known only from the type locality). 256 Whether this is a valid "species" remains to be seen ; it is based in part on small ^ paratypes of R. m. uzungwcnsiH Loveridge (see above) from Dabaga and Kigogo, T. T. However, Laurent had abundant material and the countries given in the range are taken from his paratypes. 344 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Rcrna occipitalis Giinther Groove-crowned Bullfrog 1858. Bana occipitalis Giinther, Cat. Batr. Sal. Brit. Mus., p. 130, pi. xi: Gambia ; West Africa ; Africa. 1863. Ban^ hydraletis "Boie" Peters (nomeJi nudum), Monatsb. Akad. Wiss. Berlin, p. 78 : Type in Leiden Museum examined by Peters. 1864. Bona bragantina Bocage, Revue Mag. Zool. (2) 16, p. 253: Duque de Braganca, Angola. 1924. Banosoma sehereri AM, Arch. Naturg., 90, Abt. A, p. 250 : Senegal. 1925. Bana miuanzae Loveridge, Proc. Zool. Soc. London, p. 772, pi. ii: Simiyu River at Sagayo, Mwanza Province, Tanganyika Terri- tory. 1954. Bana accidentalis Lamotte & Zuber-Vogeli (lapsus for occipitalis Giinther), Bull. Inst. Franc. Afr. Noire, 16, Ser. A, p. 940. Range. Reported from French Morocco ; Algeria ; Libya ; *Sudan; Eritrea; *Uganda; Kenya Colony; *Tanganyika Terri- tory ; ^Northern Rhodesia ; Angola ; Cabinda ; *Belgian and French Congo ; French Equatorial Africa ; Sao Thome ; Fer- nando Po ; *French and British Cameroon ; Dahomey ; Togo ; *(jlold Coast; *Liberia; *Sierra Leone; French and Portuguese Guinea ; *Gambia ; *Senegal. Rana adspersa edvdis (Peters) Eastern Burrowing-Bullfrog 1854. Pyxicephnlus edulis Peters, Monatsb. Akad. Wiss. Berlin, p. 626: Boror ; Tete ; and Mozambique Island, Mozambique. 1893. Phrynopsis Boiilengerii Pfefifer, Jahrb. Hamburg. Wiss. Anst., 1892, 10. p. 101, pi. ii, figs. 5-7: Mozambique. 1916. Pyxicephalus flavigida Calabresi, Monit. Zool. Ital. (Firenze), 27, p. 34, pi. ii, fig. 1 : Oroflillo, Somalia. 1924. Phrynopsis usambarae Ahl, Zool. Anz., 60, p. 271: Usambara, Tanganyika Territory. 1927. ? Pyxicephalus obbianus Calabresi, Atti Soc. Ital. Sci. Nat. (Milano), 66, pp. 15, 35, pi. i, fig. 1: Dolobseio, near Obbia, Somalia. Range. Somalia, south through *Kenya Colony ; *Tanganyika Territory ; *Mozambique ; Nyasaland and Northern Rhodesia to the Zambezi River. ^''' Rana delalandii delalandii (Dumeril & Bibron) Delaland 's Burrowing-Frog 1841. Pyxicephalus Delalandii Dumeril & Bibron, Erpet. Gen., 8. p. 445, pi. Ixxxvii, figs. 1 and lab : South Africa. 257 For detailed locality records, see Loveridge, 1950, Journ. E. Afr. Nat. Hist. Soc. for 1947-1948, 19, pp. 253-255. The Museum of Comparative Zoology has typical It. a. adspersa (Tschudi) from Southern Rhodesia and the Transvaal, south to Cape Province. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 345 1854. Pyxicephalus marmoratus Peters, Monatsb. Akad. Wiss. Berlin, p. 626: Boror, Mozambique. 1884. Phrynobatrachus monticola Fischer, Jahrb. Hamburg. Wiss. Anst., 1, p. 26: Southern Uaso Nyiro, Masailand, Kenya Colony. 1927. Eana (Tomopterna) hieroglyphica Ahl, Sitzb. Ges. Naturf. Freuude Berlin, p. 42: So. Omadu, Somalia. 1936. Arthroleptis agadezi Angel, Bull, Soc. Zool. France, 61. p. 275: Agadez, Air, Sahara, French West Africa. 1938. Arthroleptis ahli Deckert, Sitzb. Ges. Naturf. Freunde Berlin, pp. 127, 168, fig. 46: Lichtenstein (approx. 32°50' S., 26°E., between Bedford and Cheviot Falls, eastern Cape Province), South Africa. 1944. Arthroleptis rosei Hoffman, Soolog. Navors. Nas. Mus., Bloem- fontein, 1. p. 174, figs. 1-4: "Chitiala," i.e. Chitala River, Nyasaland. Range. Arid savanna regions in French West Africa (at Agadez: fide Guibe), west through the Sudan to Eritrea; south through *Ethiopia; *British Somaliland; *Somalia; Uganda; *Keuya Colony ; *Tanganyika Territory ; Mozambique ; *South- ern Rhodesia ; Bechuanaland ; *Transvaal ; *Orange River ; *Cape Province to Southwest Africa where it meets with the race R. (]. cryptotis Boulenger. Rona tuberculosa (Giinther)-^** Angolan Burro wing-Frog 1864. Pyxicephalus tuberculosa Giinther, Proc. Zool. Soc. London, p. 479, pi. xxxiii, fig. 1 : Pungo Ndongo, Angola. 1896. Rana pulchra Boulenger, Ann. Mag. Nat. Hist., (6) 18. p. 468: Lake Tanganyika (without indication as to which coast). Range. ''Lake Tanganyika;" Angola. Rana ornata ornata (Peters) Northern Ornate Burrowing-Frog 1878. Pyxicephalus ornatus Peters, Monatsb. Akad. Wiss. Berlin, p. 207, pi. ii, fig. 7: "Taita," i.e. Teita, Kenya Colony. 1912. Eana macrotym'paniim Boulenger, Ann. Mag. Nat. Hist., (8) 10, p. 140: West of Juba River, Gallaland, Ethiopia. 1919. Eana (Hildebrandtia) togoensis Boulenger,-^" Trans. Roy. Soc. S. Africa, 8, p. 34: Mangu, Togo. Range. Somalia; Kenya Colony; *Tanganyika Territory; -5S Miss A. G. C. Grandison informs me (29.xi.56) that Boulenger himself synonymized pulchra with tuberculosa — of which we have only Angolan ma- terial. In details our frogs do not agree too well with the description of pulchra, and the relationship with delalandii requires investigation. 259 Kew name for H. omatisslma Nieden (not of Bocag^), proposed by Boulen- ger, sight unseen. 346 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY *BeIgian Congo; French Equatorial Africa (det. A.L.) ; Togo; *CT0ld Coast. Genus PHRYNOBATRACHUS Glinther 260 1849. Stenorhynchus A. Smith (not of Hemprich: 1820), Illus. Zool. S. Africa, Eept., App., p. 23. Type by monotypy: S. natalensis Smith. 1862. Phrynobatrachus Giinther, Proc. Zool. Soc. London, p. 190. Type by monotypy: P. natalensis Giinther (correct). 1863. Leptoparius Peters, Monatsb. Akad. Wiss. Berlin, p. 452. New name for StenorhyncJnis A. Smith (preoccupied). 1925. Hylarthroleptis Ahl, Sitzb. Ges. Naturf. Freunde Berlin, 1923, p. 98. Type by present designation: H. accraensis Ahl. 1926. MicrobatracJius Hewitt (not of Koux:1910), Ann. S. African Mus., 20, p. 420. Type by original designation: Phrynobatrachus capensis Boulenger. 1926. Microbatrachella Hewitt, Ann. S. African Mus., 20, p. 420, cor- rigenda slip. Xew name for Microbatrachus Hewitt (preoccu- pied). Phrynobatrachus perpahnatus Boulenger 1898. Phrynobatrachus perpalmatu.'i Boulenger, Proc. Zool. Soc. London, p. 479, pi. xxxviii, fig. 1 : Lake Mweru, Northern Ehodesia. 1924. Phrynobatrachus perpahnatus werneri Ahl, Zool. Anz., 60. p. 273: El Grassi, etc., Sudan. Range. Sudan ; *Tanganyika Territory ; ^Mozambique ; *Ny- asaland ; *Northern Rhodesia ; *Belgian Congo. Phrynobatrachus acridoides (Cope) 1867. Staurois acridoides Cope, Journ. Acad. Nat. Sci. Philadelphia, 6, p. 198 : Zanzibar Island. 1919. PhrynoboArachus Boulengeri Witte, Revue Zool. Afr., 6, p. 225: Beira and Coguna, Mozambique. 1923. Hylarthroleptis medilineatus Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 100: " Tscharra, Dauafluss," i.e. Charra, Tana River. Kenya Colony. 1923. Hylarthroleptis janenschi Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 100: Tendaguru, near Lindi, Tanganyika Territory. Range. Sudan ; Eritrea ; Somalia ; *Kenya Colony ; *Tan- 260 Despite Hewitt's defense (1926, Ann. S. African Mus., 20, p. 425) of the generic status of Natalohatrachua (Hewitt & Methueu, 1913, Trans. Roy. Soc. S. Africa, 3, p. 107. Type by monotypy N. boncberyi Hewitt & Metliuen) it should probably be regarded as a subgenus of Phrynohutrackun of which Noble regarded it as a synonym. LOVERIDQE: E. AFRICAN REPTILES AND AMPHIBIANS 347 ganyika Territorj' ;~^^ Pemba, Zanzibar and Mafia Islands; Mozambique ; Northern Rhodesia ; French Equatorial Africa (fide Sternfeld:1917) ; *French and British Cameroon; also re- ported from Gold Coast (Witte :1919) ; French Guinea (Klap- tocz:191;3; Chabanaud:1921) ; Gambia (Witte :1919) ; Senegal (Mertens:1938). Phrynobcrtrachus pakenhami Loveridge 1941. Phnjnohatrachus i^akenlMmi Loveridge, Proc. Biol. Soc. Washing- ton, 54. p. 178: Machengwe Swamp, near Wete, Pemba Island. Range. Pemba Island. Phrynobcrtrachus kinangopensis Angel 1924. Fhrynobairachns Kinangopensis Angel, Bull. Mus. Hist. Nat. (Paris), 30, p. 130: Mount Kinangop, Aberdare Mountains, Kenya Colony. Range. From 5000 (at Nairobi) to 11,000 feet (on Mount Kenya), in highlands of Kenya Colony. Phrynobatrachus plicatus (Giinther) 1858. Byperoliu.s plicatus Giinther, Cat. Batr. Sal. Brit. Mus., p. 88, pi. vii, fig. C: Coast of Guinea. 1900. Phryjiobat radius auritus Boulenger, Proc. Zool. Soc. London, 2, p. 440, pi. xxviii, fig. 2: Benito Eiver, French Congo. 1919. Phrynobatrachus discodactylus Boulenger, Revue Zool. Afr., 7, p. 7: "Madie," i.e. Medje, Ituri, Belgian Congo. Range. *Ilganda (Budongo Forest) and Belgian Ruanda- Urundi, west through *Belgian Congo to Cabinda, north and west through French Congo ; Rio Muni ; Fernando Po ; *French and British Cameroon; Togo; Gold Coast; *Liberia to *Sierra Leone. Phrynobatrachus krefftii Boulenger 1909. Phrynobatrachus krefftii Boulenger, Ann. Mag. Nat. Hist., (8) 4, p. 49t> : Amani, Usambara Mountains, Tanganyika Territory. Range. Usambara and Magrotto Mountains, Tanganyika Ter- ritory.-®" 261 Frogs from the Uluguru Mountains of Tanganyika, referred with a query to ogoensis Boulenger by P.arbour and Loveridge (1928), are referable to acri- doidea. The two species are so similar that one is inclined to doubt all West .\frican records of "acridoidcs." 2G2 Progs from Kuwenzori and Kuanda-Urundi referred to krefftii by Nieden (1915) were subsequently renamed versicolor by Ahl (1924). 348 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Phxynobotrachus dendrobates (Boulenger) 1919. Arthrolepiis dendrohates Boulenger, Revue Zool. Afr., 7, p. 8: "Madie, " i.e. Medje, Ituri Forest, Belgian Congo. 1924. Phrynohatrachus pctropedetoides Ahl, Zool. Anz., 61, p. 102: Ruwenzori Mountains and west of Lake Edward, Belgian Congo. Range. *Uganda ; Belgian Riianda-Urundi and the *Belgian Congo.-"' Phrynobcrtrachus versicolor Ahl 1924. Phrynobatrachiis versicolor Ahl, Zool. Anz., 61. p. 100: Ivugege Forest, Belgian Ruanda-Urundi. Range. *Uganda ; *Belgian Ruanda-Urundi and the *Belgian Congo. Phrynobatrachus graueri (Nieden) 1910. Arthrolcptis graueri Nieden, Sitzb. Ges. Naturf. Freunde Berlin, p. 441 : Rugege Forest, Belgian Ruanda-Urundi. Range. Western *Kenya Colony ^*^* and *Uganda, west through *Belgian Ruanda-Urundi to the *Belgian Congo. Phrynobatrachiis natalensis (Smith) 1849. Stcrnorhynclius natalensis A. Smith, Illus. Zool. S. Africa, Rept., App., p. 24: Natal, South Africa. 1862. Phrynobatrachus ■natalensis Giinther, Proc. Zool. Soc. London, p. 190 : Natal, South Africa. ]8()2. Bicroglossus angustirostris Cope, Proc. Acad. Nat. Sci. Philadel- phia, p. 341: Umvoti, Natal (by inference only). 1875. Arthrolcptis natalensis var. irrorata Peters, Monatsb. Akad. Wiss. Berlin, p. 210: Accra, Gold Coast. 1894. Phrynobatrachus ranoides Boulenger, Proc. Zool. Soc. London, pp. 641, 644, pi. xxxix, fig. 2 : Pietermaritzburg, Natal. 19U4. Phrynobatrachus natalensis fonna gracilis Andersson, Results Swed. Exped. Zool. to Egypt and the White Nile 1901, \. No. 4, p. 10, figs. — : Ghrab el Aish, south of Kaka on the White Nile, Sudan. Range. *Sudan; Ethiopia; *Uganda; *Kenya Colony; *Tan- ganyika Territory ; Zanzibar Island ; Mozambique ; *Nyasaland ; *Northern and *Southern Rhodesia; Bechuanaland ; *Transvaal ; *Xatal; Orange River; *Cape Province; Southwest Africa; *Angola; *Belgian Congo; Belgian Ruanda-Urundi; French 263 The record from Accra, Gold Coast, of Deckert (1938), certainly requires verilicatiou. -64 That is, Kaimosi and the Yala River region of Kakamega ; the Mount Kinangop record of Angel (1925) is unacceptable. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 349 Cameroon; Dahomey; Togo; *Liberia; *Sierra Leone; French Guinea; Portuguese Guinea; *Gambia; French West Africa. Phrynobcrtrachns keniensis Barbour & Loveridge 1928. Phrynobatrachus Tceniensis Barbour & Loveridge, Proc. New Eng- land Zool. Club, 10, p. 89: Mount Kenya, Kenya Colony. Range. Upland meadows (Kikuyu; Molo; Mt. Kinangop ; Mt. Kenya and Uplands) of Kenya Colony. [Phrynobatrachus bottegi (Boulenger)] 1894. Arthroleptis hottegi Boulenger, Ann. Mus. Civ. Stor. Nat. Geneva, (2) 15. p. 16, pi. iv, fig. 3: Auata Eiver, Somalia Range. Ethiopia and Somalia.^*^^ Phrynobatrachus minutus (Boulenger) 1895. Arthroleptis minutus Boulenger, Proc. Zool. Soc. London, p. 539, pi. XXX, fig. 4: "Durro," i.e. Duro, Ethiopia. 1910. Arthroleptis scheffleri Nieden, Sitzb. Ges. Naturf. Freunde Berlin, p. 438: Nairobi and Kibwezi, Kenya Colony; Mpwapwa, Tan- ganyika Territory; Zanzibar. 1924. Arthroleptis aliifer Ahl, Arch. Naturg., 90. Abt. A, p. 251: Usaramo, Tanganyika Territory. Range. Sudan; Somalia; *Uganda; *Kenya Colony; *Tan- ganyika Territory ; Zanzibar Island ; *Northern Rhodesia ; Trans- vaal (Mertens:1937) ; Angola (Monard:1937) ; *Belgian Congo. Also recorded from French Equatorial Africa (Sternfeld :1917) ; French Guiuea (Chabanaud :1921) and Portuguese Guinea (Boulenger :1906), but all three require investigation. Phrynobatrachus rxingwensis (Loveridge) 1932. Arthroleptis rwngwensis Loveridge, Bull. Mus. Comp. Zool., 72. p. 386: Ilolo, Eungwe Mountain, Tanganyika Territory. Range. Southwestern *Tanganyika Territory and adjacent *Belgian Congo.^^^ 265 Records from Uganda (Peracca :1909 ; Boulenger :1911) and Belgian Congo (Noble :1924) require verifying. I examined the Kibongoto, T. T. specimens of Lonnberg (1907) and found they were based on minutus together with some juvenile natalensis. The record from Singida, T. T. (FitzSimons :1930) was also a juvenile natalensis. 266 A series of frogs from Lukafu received from the Congo Museum as gut- terosus Chabanaud, a species described from French Guinea. 350 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Phrynobatrachus ukingensis mababiensis FitzSimons"®' 1932. Phrynobatrachiis mnbabicnsis FitzSimons, March, 1932, Ami. Transvaal Mus., 15. p. 40, and 1935, ibid., 16, p. 390, fig. 26: Tsotsoroga Pan, Mababe Flats, Bechuanaland Protectorate. 1940. Phryyiobatrachus vanrooyeni Hoffman, Soolog. Navors. Nas. Mus. Bloemfontein, 1. p. 99, fig. 2B: Broedershoek, Greytown, Natal, South Africa. 1944. Phrynobatrachus chiikilaensis Hoffman, Soolog. Navors. Nas. Mus. Bloemfontein, 1. p. 177, figs. 5-6: "Chitiala," i.e. Chitala Elver, Nyasaland. Range. Marshes, chiefly upland, of southern *Tangan3'ika Territory ; Mozambique ; *Nyasaland ; *Northern Rhodesia ; *Beehuanaland ; *Natal ; *Belgian Congo. Phrynobcrirachus uldngensis uldngenesis (Loveridge) 1932. Arthroleptis iikiitgaisis I.ovoridge, Bull. AIus. Comp. Zool., 72, p. 385: Madehani, Ukinga Mountains, Tanganyika Territory. Range. Virgin forests of Ukinga and Ruugwe Mountains, southern *Tanganyika Territory and Misuku Mountains, north- ern Nyasaland. Phxynobatrachus rouxi (Nieden) 1912. Arlliroleptis roiuri Nieden, Wiss. Ergebn. Deutsch. Zentral-Afrika- Exped. 1907-1908, 4. p. 178, pi. v, figs. 5a-b: Northwest Buddu Forest, Uganda. Range. Uganda and adjacent *Belgian Congo.^^^ Genus ARTHROLEPTIS Smith ^"^ 1849. Arthroleptis A. Smith, Illus. Zool. S. Africa, Kept., App., p. 24. Type by monotypj': A. ica:hlbcrgii Smith. 1863. Hemimantis Peters, Monatsb. Akad. Wiss. Berlin, p. 451. Type by monotypy: E. calcaratus Peters. 267 Frogs from N.vainkolo, southeast end of Lake Tanganyika, Northern Rho- desia, topotypic of moorii Boulenger — to which I (1933) referred them in error — are referable to mababiensis. The extensive series from the Uzungwe, Ubena and Ukinga Mountains assigned by me (1933) to parrulus Boulenger, are also mababiensis, the description of which appeared onl.v after my manuscript had gone to press. Subsequently I had the opportunity of examining the types of moorii and parrnlus in the British Museum. 268 The Mount Kenya record of Angel (1920) is considered doubtful. 269 Laurent (1954, Ann, Mus. Congo, Zool., 1. Misc. Zool. H. Schouteden, p. 3.'j) presents a chart setting forth the osteological characters hy which he sep- arates t€3 (M.C.Z. 21718- 21) of sylvatica from the original series collected by de Witte (in February, 1925) and see no justification for regarding sylvatica as distinct. 352 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Poroto Mountains, southern * Tanganyika Territory and Misuku Mountains, northern *Nyasaland. Arthroleptis poecilonohis Peters 1863. Arthroleptis poecilonotus "Schlegel" Peters, Monatsb. Akad. Wiss. Berlin, p. 446: Boutry, Ashanti, Gold Coast. 1882. Arthroleptis macrodactylus Boulenger, Cat. Batr. Sal. Brit. Mus., p. 117, pi. xi, fig. 5: Gabon, French Congo. 1885. Arthroleptis Mvittatus F. Miiller, Verb. Naturf. Ges. Basel, 7. p. 671, pi. ix, figs, k-1: Tumbo Island, French Guinea. 1900. Arthroleptis inguinalis Boulenger, Proc. Zool. Soc. London, p. 442, pi. xxvii, fig. 2: Benito Eiver, Gabon, French Congo. Range. Virgin forests of southern *Sudan ; *Uganda"'^? ; west through *Belgian and *French Congo ; Fernando Po ; *French and British Cameroon ; *Nigeria ; Dahomey ; Togo ; Gold Coast ; *Ivory Coast; *Liberia; and *Sierra Leone to Portuguese Guinea. Artholeptis adoliifriderici adolfifriderici Nieden^'^ 1910. Arthroleptis adolfi-friderici Nieden, Sitzb. Ges. Naturf. Freiinde Berlin, p. 440: Eugege Forest, Belgian Euanda-Urundi. 1939. Arthroleptis affinis Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 303. fig. 1: (Type ? 39 mm.) Amani, Usambara Mountains, Tan ganyika Territory. 1939. Arthroleptis schocneheclci Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 305, fig. 2: (Type $ 22 mm.) Amani, Usambara Mountains, Tanganyika Territory. Range. Virgin forests of (Mbololo Mountain) *Kenya Colony and (Usambara; Magrotto; Uluguru and Poroto Mountains) *Tanganyika Territory ; west to Belgian Ruanda-Urundi and the *Belgian Congo. Arthroleptis stenodactylus whytii Boulenger-" 271 Included on the basis of a not-too-well-preserved frog (M.C.Z. 25402) that I took at Mubango, Mabira Forest. 272 Trinomials are employed because of a southeastern race {A. a. jrancei Loveritlge :1953) occurring in Ruo River forest, Mlanje Mountain, Nyasaland. 273 This montane-forest, burrow-breeding Arthroleptis differs from typical s. stenodactylus in having a metatarsal tubercle that is usually shorter than, though occasionally as long as, the inner toe ; also the tibio-tarsal articulation of its forward-pressed hindlimb reaches the tympauiim or, rarely, the eye. On the other hand, the lowland-savanna, burrow-breeding, s. stenodactylus has a metatarsal tubercle that is usually longer than the inner toe, though in juveniles it is occasionally only as long as the toe ; the tibio-tarsal articulation of its adpressed hindlimb reaches only to the elbow or shoulder, very rarely as far as the tympanum. Owing to the difficulty of separating juveniles, and the general similarity of the two forms, much confusion exists in the literature. A. whytii itself was a composite of the two races ; ulugiiruensis was described by me owing to reliance on Ahl's advice when he said that my Uluguru frogs were not subspccitically identical with lonnhergi — of which I had no cotypes at that time. IS'ow the Museum of Compai-ative Zoology has many topotypes of both whytii and all its synonyms. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 353 1897. Arthroleptis whyiii Boiilcnger, Proc. Zool. Soe. London, pp. 801, 802, pi. xlvi, fig. 3: '"Masuka," i.e. Misuku Mountains, Nyasa- land (restricted). 1915. Arthroleptis Jonnbergi Nieden, Mitt. Zool. Mus. Berlin, 7. p. 3G1 : Mombo, foot of Usambara Mountains, Tanganyika Territory. 1932. Arthroleptis stenodactylus uluguruensis Loveridge, Proc. Biol. See. Washington, 45, p. 61 : Nyingwa, Uluguru Mountains, Tan- ganyika Territory. 1939. Arthroleptis vagus Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 306: (Cotype $ 31 mm.; ^ 38 mm.) Usambara Mountains, Tanganyika Territory. 1939. Arthroleptis iikamiensis Ahl, Sitzb. Ges. Naturf. Freunde Berlin, p. 308, fig. 3: (Type $ 33 nun.) Ukami, i.e. Uluguru Mountains, Tanganyika Territory. Range. Mrgin forests of Usambara ; Magrotto ; Uluguru ; Ron- do and Rungwe Mountains, *Tanganyika Territory ; also Misuku ; Nchisi; Mlanje and Cliolo Mountains, *Nyasaland; Mozambique and C'hirinda Mountain, *Southern Rhodesia.^" Arthroleptis stenodactylus stenodactylus Pfeffer''' 1893. Arthroleptis stenodactylus Pfeffer, Jahrb. Hamburg. Wiss. Anst., 1892, 10. p. 93, pi. i, fig. 2: Kihcngo, Tanganyika Territory. 192-1. Arthroleptis methneri Ahl, Arch. Naturg., 90, Abt. A, p. 251: Xaugoma Cave, Matumbi near Kilwa, Tanganyika Territory. 1927. '*: Arthroleptis elegans Calabresi, Atti Soc. Ital. Sci. Xat. (Milano), 66, pp. 16, 35: Hongolo, Uadi Hoor, Somalia. Range. Dry savanna, chiefly on the coastal plain, of *Kenya Colony ; *Tanganyika Territory ; Mozambique ; *Nyasaland : *Northern Rhodesia ; *Belgian Congo."^ Genus CACOSTERNUM Boulenger 1887. Cacosternum Boulenger, Ann. Mag. Nat. Hist., (5) 20, p. 51. Type by monotj'py : C. namim Boulenger. -"■•Credibly reported from the Belgian Congo by de Witte (iyo4), though possibly typical stenodactylus of which we have examples from Kan.vama. B. C. 275 A., elegans, described from four 19 mm. frogs, is tentatively listed in the synonymy in the hope that someone, bearing in mind the possibility they may be juvenile s. stcnodtirtylus, will restudy the material. 276 Also reported from French Cameroon and Sierra Leone by Angel (1940), possibly owing to confusion with raridbili.'i Matschie. 354 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Cacosternum boettgeri boettgeri (Boulenger)^" 1882. Arthroleptis ho.ettgeri Boulenger, Cat. Batr. Sal. Brit. Mus., p. 118, pi. xi, fig. 6: Vleis, Kaffraria, Bechuanaland Protectorate. 1887. Cacosternum nanum Boulenger, Ann. Mag. Nat. Hist., (5) 20, p. 52: Vleis, Kaffraria, Bechuanaland Protectorate. 1914. Arthroleptis schebeni Nieden, Sitzb. Ges. Naturf. Freunde Berlin, 1913, p. 438: Klein Nauas, Southwest Africa. Range. Occurs in both upland savanna and arid regions of East Africa, viz. Somalia ; *Uganda ; *Kenya Colony ; Bechuana- land Protectorate ; *Transvaal ; Natal ( ? and/or the race albi- venter Hewitt); *Orange River; Basutoland ; *Cape Province; *Southwest Africa. Genus HEMISUS Glinther 1858. Hemisus Giinther, Cat. Batr. Sal. Brit. Mus., p. 47. Type by monotypy: Engystoma guttatum Rapp. 1863. Eakophrymis Steindachner, Sitzb. Akad. Wiss. Wien, 48, Abt. 1, p. 191. Type by monotypy: K. sudanesis (sic) Steindachner = Engystoma nuirmoratum Peters. Hemisus marmoratus marmorcrtus (Peters) Eastern Sharp-snouted Frog 18.')4. Engystoma marmoratum Peters, Monatsb. Akad. Wiss. Berlin, p. 628 : Cabaeeira, Mozambique. 1863. Kakoplirynus sudanesis (sic) Steindachner, Sitzb. Akad. Wiss. Wien, 48. Abt. 1, p. 191, pi. i, figs. 10-13: Sudan. 1882. Hemisus taitanus Peters, Reise nach Mossambique, 3. p. 175: "Taita, " i.e. Teita, Kenya Colony. Range. *Sudan ; Eritrea ; *Kenya Colony ; *Tanganyika Ter- ritory ; Zanzibar Island ; *Mozambique ; *Nyasaland ; Southern Rhodesia; Bechuanaland Protectorate; *Transvaal; southeast Belgian Congo (Angola records are presumably referable to the western race). 277 From Kundeluugru Plateau research station at 1750 metres, in Katanga, southeast Belgian Congo. Laurent (1950, Revue Zool. Bot. Afr., 44, p. 138) has described a single 14 mm. frog as C. lelcupL It is said to differ from boettgeri by (1) lacking nuixillary teeth; (2) by the absence of markings on its hindlimbs. It is true that limb markings are present in our extensive representation of boettgeri from southern Africa. In East Africa, however, markings are both present (M.C.Z. 2.5478 — 3 ex.) and absent (M.C.Z. 25479 — 2 ex.) in a series from Kinangop Plateau (11,000 feet). In another Kinangop series markings are present in 17 frogs, absent in 3 (M.C.Z. 16104, 16107, 10109). My colleague Dr. E. E. Williams finds maxillary teeth present in all sizes of those he has examined. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 355 Hemisus marmoratus guineensis Cope Western Sharp-nosed Frog 1865. Hemisus guineensis Cope, Nat. Hist. Re\aew, 5, p. 100, footnote. Type in Vienna Museum: Guinea (by inference). 1882. Ejigysloma vermiculatum "Schlegel" Peters, Reise nach Mossam- bique, 3. p. 175: Ashanti, Gold Coast. Range. * Western Tanganyika Territory; Angola; *Belgian Congo ; French Congo ; French Equatorial Africa ; *French Cameroon; Nigeria; Dahomey; *Gold Coast; *Liberia ; *Sierra Leone ; French Guinea ; Portuguese Guinea ; Gambia ; Senegal. Family PPIRYNOMERIDAE'^' Genus PHRYNOMERUS Noble 1847. Brachymerus A. Smith (not of Dahlbom:1845), Illus. Zool. S. Africa, Rept., text to pi. Ixiii. Type by monotypy: B. bifascia- fus A. Smith. 1926. Phrynomerus Noble, Amer. Mus. Novit., No. 237, p. 20. New name for Brachymerus A. Smith (preoccupied). Phrynomerus bifasciatus bifasciatus (Smith) 1847.-"^ Brachymerus bifasciatus A. Smith, Illus. Zool. S. Africa, Rept., pi. Ixiii: "Country to the east and north-east of Cape Colony," i.e. east of Cape Province, South Africa. 1849.2^*^ Dendrobates inhambanensis Bianconi (1848), Nuovi Ann. Sci. Nat., (2) 10, p. 107, ].l. V, litcs. 4-4a ; also 1850, Spec. Zool. Mosamb., Rept., p. 26, pi. v. figs. 4-4a : Inhambane, Mozambique. 1944. Phrynomerus bifasciatus nyasalandensis Hoffman, Soolog. Navors. Xas. Mus. Bloemfontein, 1. p. 181, fig. 9: "Chitiala," i.e. Chitala River, Nyasaland. -78 The status of the African genus Phrynomerus has been a chequered one. In 1858 (as Brachymerus) it was made the type of a family (BRACHYMERI- DAE) by Giinther (Cat. Batr. Sal. Brit. Mus., p. 124), onlv (as Phrynomantis) to be merged in the family ENGYSTOMATIDAE by Boulenger (1882, Cat. Batr. Sal. Ecaud. Brit. Mus., p. 172). In 19:51 (as Phrynomerus) it was given sub- family rank (PHRYNOMERINAE) in the BRBVICIPITIDAE by Noble (Biologv of the Amphibia, p. 538). later to be raised to full family status (PHRYNOMERI- DAE) by Parker (1934: Monograph of the Frogs of the Family MICROHYLI- DAE, p. 9) principally on account of its intercalary phalanges, a character it shares with the RHACOPHORIDAE. For other ways in which PHRYNOMERIDAE differs or agrees with MICRO- HYLIDAE, together with a lucid tabulation of the characteristics of all African firmisternal families, see Parker {loc. cit. pp. 8-9). 279 Actually appeared in December, 1847, fide Waterhouse (1880, Proc. Zool Soc. London, p. 490). 280 Bianconi subsequently claimed that this description first appeared on February 4, 1847. in Mem. Accad. Sei. 1st. Bologna, 1, p. 498 (a publication not available to me). 356 BtTLLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Range. Kenya Colony; Tanganyika Territory; Zanzibar and Mafia Islands ; south to Zululand ; west to Cape Province ; northwest through Southwest Africa to Angola and the Belgian Congo. 281 *o^ Family MICROHYLIDAE'"*' Genus CALLULINA Nieden 1910. Callulina Nieden, Sitzb. Ges. Naturf. Freunde Berlin, p. 449. Type by nionotypy: C. Jcreffti Nieden. Callulina kreffti Nieden 1910. Calbilina kreffti Nieden, Sitzb. Ges. Naturf. Freunde Berlin, p. 449 : Amani, Usambara Mountains ; and Tanga, Tanganyika Territory. Range. Usambara, Magrotto and Uluguru Mountains, Tan- ganyika Territory. Genus SPELAEOPHRYNE A hi 1924. Spelacophryne Ahl, Zool. Anz., 61. p. 99. Type by original designation: S. methneri Ahl. Spelaeophryne methneri Ahl Scarlet-snouted Black Frog 1924. Spelaeophryne methneri Ahl, Zool. Anz., 61, p. 99: Nangoma Cave, Matumbi near Kilwa, Tanganyika Territory. Range. Eastern Tanganyika Territory. Genus PROBREVICEPS Parker 1931. Prohreviccps Parker, Ann. Mag. Nat. Hist., (10) 8. p. 262. Type by original designation: Breviceps macrodactylus Nieden. Probreviceps macrodactylus macrodactylus (Nieden) 1926. Breviceps macrodactylus Nieden, Das Tierreich, 49, Anura, 2. p. 281 The western race microps Peters, has been reported from the Belgian Congo, Sudan and the *Gold Coast (M.C.Z.). 282 This name, originally spelled MICRHYLIDAE by Giinther (1S5S, Cat. Batr. Sal. Brit. Mus., p. 121), was amended and used by Parker (1943, Monograph of the Frogs of the Family MICROHYLIDAE) for the family that, following Noble, I have hitherto consistently called BREVICIPITIDAE of Cope (part). It is typified by Rana gibbosa Linnaeus = Breviceps oibbosus (Linnaeus), the only member of the family known to him in 1758. The argument for retaining BREVICIPITIDAE advanced by Dunn (1935. Copeia, p. 108) appears to be fallacious, being based on a misconception of Parker's reasons for selecting MICROHYLIDAE. It is necessary only to add that the sole species included by Giinther in his MICRHYLIDAE was Microhyla achatina Tschudi of Java. L0\T:RIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 357 G: Usambara Mountains, Tanganyika Territory. 1928. Breviceps usamharicu3 Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 251 : Amani, Usambara Mountains, Tanganyika Territory. Range. Usambara Mountains, Tanganyika Territory. Probreviceps macrodactylus loveridgei Parker 1931. Probreviceps macrodactylus loveridgei Parker, Ann. Mag. Nat. Hist., (10) 8. p. 263: Bagilo, Uluguru Mountains, Tanganyika Territory. Range. Ulugnru Mountains, Tanganyika Territory. Probreviceps macrodactylus rungwensis Loveridge 1932. Probreviceps macrodactylus rungwensis Loveridge, Bull. Mus. Couip Zool., 72. p. 387: Nkuka Forest, Eungwe Mountain, Tan- ganyika Territory. Range. Rungwe Mountain, Tanganyika Territory. Probreviceps uluguruensis (Loveridge) 1925. Breviceps iduguruensis Loveridge, Proc. Zool. Soc. London, p. 789, pi. i, fig. 3 : Bagilo, Uluguru Mountains, Tanganyika Terri- tory. Range. Uluguru Mountains, Tanganyika Territory. Genus BREVICEPS Merrem 1820. Breviceps Merrem, Vers. Syst. Amphib., p. 177. Type by mono- typy: Pana gibbosa Linnaeus. 1826. Engystoma Fitzinger, Neue Class. Eept., pp. 39, 65. Type by original designation: Sana gibbosa Linnaeus. 1830. Systoma Wagler, Nat. Syst. Amphib., p. 205. Type by monotypy: Breviceps gibbosus Merrem = Sana gibbosa Linnaeus. Breviceps mossambicus Peters Mozambique Short-headed Frog 1829. Engystoma grcunosum Cuvier (nomen nudum), Eegne Animal, ed. 2, 2, p. 112: "Cap," i.e. South Africa; probably Natal. 1854. Breviceps mossambicus Peters, Monatsb. Akad. Wiss. Berlin, p. 628: Mozambique Island, and Sena, Mozambique. 1944. Breviceps mitchelli Hoffman, Soolog. Navors. Nas. Mus. Bloem- fonteiu 1, p. 182, fig. 10: "Chitiala," i.e. Chitala Eiver, Nyasa- land. Range. Tanganyika Territory, south to Natal, northwest through Southern Rhodesia to Angola and the Belgian Congo. 358 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Genus PARHOPLOPHRYNE Barbour & Loveridge 1928. Parlioplopliryne Barbour & Loveridge, Mem. Mus. Comp, Zool., 50. p. 260. Type by original designation: P. vsambaricus (sic) Barbour & Loveridge. Parhoplophryne usamborica Barbour & Loveridge 1928. Parhoplophryne usambaricus Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 260: Amani, Usambara Mountains, Tan- ganyika Territory. Range. Usambara Mountains, Tanganyika Territory. Genus HOPLOPHRYNE Barbour & Loveridge 1928. Hoplophryne Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 253. Type by original designation : H. %dugyru.ensis Barbour & Loveridge. Hoplophryne rogersi Barbour & Loveridge Usambara Banana-Frog 1928. Hoplophryne rogersi Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 258, pi. ii, fig. 5 : Mount Bomoli, Amani, Usambara Mountains, Tanganyika Territory. Range. Usambara and Magrotto Mountains, Tanganyika Ter- ritory. Hoplophryne uluguruensis Barbour & Loveridge Uluguru Banana-Prog 1928. Hoplophryne uluguruensis Barbour & Loveridge, Mem. Mus. Comp. Zool., 50, p. 254, pi. ii, figs. 3-4: Mount Mbova, Nyange, Ulu- guru Mountains, Tanganyika Territory. Range. Uluguru Mountains, Tanganyika Territorj'. LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS 359 LIST OF SPECIES OR RACES STILL UNREPRESENTED IN THE MUSEUM OF COMPARATIVE ZOOLOGY, CAMBRIDGE LIZARDS *Hemidactylus alhopunctaUis Loveridge Ebennvia sp. Chamaeleo hitae7uS7 collaris, Miodon e., 283 LOVERIDGK: E. AFRICAN' REPTILES AND AMPHIBIAXS IX Colobopus, 191 colonorum, Agama, 192, 194 Colpochelys, 168 Coluber, 2.")7, 270 COLUBRIDAE, 249, 268 colubrina, Dendrophis, 273 COLUBRIXAE, 249 eoniorensis, Euprepes, 210 comorensis, Mabuya in., 209, 210 r-omorensis, Tvphlops, 244 nompactilis, Bufo, 309, 310 complanatus, Crocodilus, 177 compressicauda, Lygosoma, 217 coucolor, Aparallactus 1., 286 concolor, Choristodon, 281 concolor, Hyperolius, 332 I'oncolor, Leptopelis, 318 concolor, Uriechis, 280 condanarus, Coluber, 278, 279 congestus, Onycliocephalus, 242 congica, Agama c, 194 congica, Atractaspis, 296 congicus, Aparallactus, 286 congicus, MegaJixalus f., 322 congicus, Scincodipus, 220 congicus, Tvphlops, 242 conica, Boa, 248 conjuncta, Leptotyphlops e., 246 conjunctum, Stenostoma, 246 conuali, Rhamphiophis, 277 conradsi, Atractaspis, 297 conradti, Lygodactylus, 187 Constrictor, 248 I'onstrictor, Coluber, 257 controversa, Thalassochelys, 168 Copeglossum, 208 Coracodichus, 350, 351 CORDYLIDAE, 223 Cordylus, 223 cordylus, Laeerta, 223 coriacea, Dermoehelys, 164 coriacea, Testudo, 164 Coriudo, 164 coronata, Calainaria, 258, 259 coronatus, Meizodon, 259 coronis, Mesocoronis, 300 corpulentum, Brachycrauion, 296 corticata, Thalassochelys, 167 Coryphodon, 257 cotti, Rana, 338, 341 cowani, Microscalabotes, 1S6 cowieii. Alligator, 177 Cranopsis, 310 Crassonota, 197 crawshayi, Naja n., 292 crepidianus, Phelsuma, 190 Orepidius, 310 Cricochalcis, 226 cristatus, Chamaeico, 197 croeea, Laeerta, 229 Crocodilus, 176 crocodilus, Laeerta, 176, 177 CROCODYLIA, 176 CROCODYLIDAE, 176 Crocodylus, 176 Crotaphitis, 337 Crotaphopeltis, 271 cruciger, Gekko, 185 cruentatus, Bufo, 308 Crypsidonius, 264 Cryptoblephaiiun, 218 Cryptoblepharus, 218 CRYPTODIRA, 163 Cryptopodus, 172 Cryptopus, 172 Cryptothylax, 316 cryptotis, Rana d., 345 cubitoalbus, Hylanibates, 318 cucullatus, Chamaeleo, 197 cuneiformis, Cadmus, 266 currori, Feylinia, 222 Cursoria, 249 curtus. Python, 248 Cusoria, 249 cutipora, Rana, 336 cyanogaster, Agama, 195 eyanogaster, Stellio, 193, 195 cyanophlyctis, Rana, 33<) BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Cyclanorbis, 172 Qv-clodei-ma, 172 CVclosaurus, 192 Cylindrurus, 235 Cyneosaura, 197 C.>Tiisca, 237 cynnamomeus, Hylambates, 317 cynodou, Dipsaa, 268 Cynodontophis, 283 Gyoclosaurus, 192 dabagae, Chamaeleon w., 205 Daboia, 300 Dactylethra, 307 damaranus, Euprepes, 212 damarensis, PelomtMiusa g., 174 darlingi, Honiopus, 170 DASYPELTINAE, 2S7 Dasypcltis, 162, 287, 288 (leeipiens, Hyperolius, 330 decosteri, Elapsoidea, 290 deeosteri, Elapsoidea s., 290 dofilippii, Caiisus, 299 defilippii, Heterodon, 299 degeni, CrotaphopeltLs, 272 degeni, Leptodira, 272 degrijisi, Gonionotophis, 255 Deirodon, 288 delalandii, Euprepes, 208 delalandii, Lacerta, 230 delalandii, Pyxicephalus, 337, 344 delalandii, Eana d., 344, 345 Dendraspis, 293 Dendroaspis, 293 Dendrobates, 308 dendrobates, Arthroleptis, 348 dendrobates, Phrj-nobatrachus, 348 Dendroeehis, 293 dendrophila, Dipsas, 268 Dendrovaranus, 235 denhardtii, Boulengerula, 305 dentatus, Sternotherus, 175 depressa, Chelonia, 166 depressus, Hemidactylus, 181 derbianus, Sternotherus, 175 deremensis, Chamaeleo, 205 deremeusis, Cliamaeleon, 205 Dermatochelys, 164 DERMOCHELYIDAE, 163 Dermochelys, 163 deserticola, Kassina, 321 devilliersi, Pelomedusa g., 174 dewittei, Siaphos, 216, 217 dhara, Coluber, 270 dhara, Telescopus d., 270 Diaphorotyphlops, 240 Diceros, 196 dickersoni, Gonatodes, 180 Dicranosaura, 197 Dicroglossus, 337 diesneri, Mabuia, 211 Dilepis, 197 dilepis, Chamaeleo, 197, 200 dilepis, Chamaleo d., 199 dimidiatus, Rhinocalamus, 282 Dinodipsas, 298 Dinophis, 293 Dioptroblepharis, 232 Diphalus, 237 DIPLASIOCOELA, 314 Diplodactylus, 189 Dipsadoboa, 269, 271 DIPSADOMORPHINAE, 268 Dipsadomorphus, 268 dipsas, Herpetodryas, 257 Dipsina, 276 Dipsoglyphophis, 271 Dirodon, 288 Discodaetylus, 189 discodactylus, Hyperolius, 326 discodaetylus, Phrynobatraehus, 347 DISCOGLOSSIDAE, 307 disjuncta, Pelomedusa g., 173, 174 disparilis, Diaphorotyphlops, 240 Dispholidus, 273 distanti, Agama h., 193 distanti, Leptotyphlops, 246 distanti, Leptotyphlops e., 246 Distichurus, 298 LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS XI Docidophryne, 309 dodomae, Agama a., 195 dodomae, Agama 1., 195 Dolichophis, 257 dolloi, Aparallactua, 286 dominiceusis, Mabuya, 208 dorbignyi, Bufo, 309 doriae, Ceramodactylus, 179 doriai, Latastia, 237 dorsalis, Afrixalus f., 322 dorsalis, Eremias, 232 dorsalis, Hyperolius, 322 dorsalis, Leptophis, 262 dorsalis, Philotliamnus s., 262 dorsalis, A'ipera, 298 dorsimaculatus, MegalLxalus, 322 Doryura, 181 Dracaena, 234 dracaena, Lacerta, 235 drapiezii, Dipsas, 268 Dromophis, 159, 277, 278 Dromoplectus, 310 Dryomedusa, 273 Duberria, 266 duberria. Coluber, 266 dubia, Clielonia, 168 dubia, Phelsuma d., 190 dubius, Pachydaetylus, 190 dubius, Typhlops, 240, 242 dugesii, Lacerta, 229 dulcis, E«na, 246 Dumerilia, 220 dumerili, Euprepes, 215 dumerilii, Coronella, 256 dumerilii. Monitor, 235 dussumieri, Chelonia, 168 dutoiti, Arthroleptides, 336 Ebeuavia, 190, 359 Echidna, 301 Echidnoides, 300 ochinata, Lacerta, 230 Eehis, 303 echis, Vipera, 303 edulis, Pyxicephalus, 337, 344 edulis, Rana, 344 cdwarsiana, Lacerta, 228 eggeli, Scelotes, 219, 220 Elabites, 208 Elapechis, 290 Elaphe, 257 Elaphrodytes, 295 Elaphromantis, 31C Elaphropus, 229 ELAPIDAE, 290 Elapomorphus, 284 Elaposchema, 282 Elaps, 291 Elapsoidea, 290, 291 Elasmodaetylus, 191 elegans, Acontias, 222, 223 elegans, Arthroleptis, 353 elegans, Coluber, 278 elegans, Coronella, 259 elegans, Cusoria, 249 elegans, Euprepes, 214 elegans, Feylinia c, 223 elegans, Miculia, 218 elegans, Pachydaetylus, 191 elegans, Stenodactylus, 179 elegans, Testiido, 169 elegans, Tupinambis, 236 elegans, Vipera, 300 elegantissimus, Coluber, 266 elegantissimus, MLke, 279 elgonensis, Cnemaspis a., 180 elgonensis, Rhamnophis a., 264 elgonensis, Thrasops a., 264 elgonis, Agama, 195 elgonis, Agama a., 195 ellenbergeri, Chamaetortus a., 272 ellenbergeri, Mabuya s., 213, 272 ellioti, Chamaeleo b., 200 ellioti, Chamaeleon, 200 elongata, Caouana, 167 elongata, Dasypeltis, 289 elongata, Eremonia, 167 embuensis, Chamaeleon a., 203 emini, Ahaetulla, 261 xn BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY enimi, Glaueonia, 247 emini, Leptotyphlops e., 247 eniini, Nucras, 230 Eminophis, 159 Empagusia, 235 empusa, Peltaphryne, 309 eiiantiodaetylus, Hylambates, 315 Endodroniius, 228 Engystoma, 357 ENGYSTOMATIDAE, 355 Ensirostris, 197 Epictia, 246 Epidalea, 309 cpioticus, Crepidius, 310 Eremias, 232 Eremioplanis, 192 Ereniioscopus, 232 Eremonia, 167 Eretmoclielys, 166 Erizia, 197 erlangeri, Hemidactylus, 185 erlaiigcri, Eana, 341 orosa, Kinixys, 171 crosa, Testudo, 170, 171 erythraea, Hyla, 336 Eryx, 162, 248 esehiichtii, Typhlops, 239, 241 esculenta, Caretta, 165 esculenta, Rana, 336, 337 etiennei, Chamaeleo g., 199 EUBLEPHARIDAE, 178 Eucephalus, 245 Euchelonia, 165 Euehelys, 165 Eudioptra, 232 Eudipsas, 268 Eugnathus, 250 Eugong3''lus, 214 Euleptes, 189 Euniecia, 213, 214 Euphlyctis, 336 Euprepiosaurus, 235 Euprepis, 208 I'uprootus, Typhlops, 244 Eurhina, 309 Eurhous, 182 europaeus, Phyllodactyhis, 189 Eurystephus, 296 Eutropis, 208 Evoluticauda, 206 ewerbeeki, Amphisbaeua, 238 ewerbecki, Cliirindia, 238 exanthematiea, Lacerta, 23.5 t'xanthematicus, Varanus e., 236 oxcentricus, Typhlops, 241 oxceutrieus, Typhlops s., 241 oxcubitor, Chamaeleo f., 203 t'xcubitor, Chamaeleon t., 203 I'allax, Atractaspis, 296, 297 fallax, Taibophis, 270 fasciata, Dasypeltis, 288, 28!i fasciata, Dipsas, 268 fasciata, Grayia, 268 fasciata, Hoimonota, 182 fasciata, Pelamis p., 296 fasciata, Platypholis, 191 fasciata, Platypholis p., 191 fasciatus, Chiromantis, 316 fasciatus, Dasypeltis, 289 fasciatus, Hemidactylus f., 181 fasciatus, Strongylopus, 336 fasciolata, Dasypeltis s., 289 fasciolatus, Coluber, 258 fastidiosus, Cranopsis, 310 fenestratus, Diphalus, 237 feniandi, Riopa, 213, 214 fernandi, Tiliqua, 214 fevniquei, Eappia, 329 ferrandii, Lygosoma, 21;") Feylinia, 222 fiechteri, Glaueonia, 247 fiechteri, Leptotyphlops, 247, 3.")0 fiechteri, Rana, 338 FIRMISTERXIA, 314 fischeri, Chamaeleo f., 204, 359 fischeri, Chamaeleon, 204 fischeri, Dipsas, 269 fischeri, Zamenis, 259 LOVERIDGE: E. AFRICAN REPTILES AND AMPHIBIANS xiii fitzingeri, Notopholis, 228 fitzsimonsi, Gerrhosaurus f ., 225 flaveseens, Monitor, 235 flavigula, Pyxicephalus, 344 flavigularis, Dendrophis, 263 flavigularis, Gerrhosaurus, 224 flavitorques, Aparallactus, 285 flavoguttatus, Hyperolius, 328, 335 flavomaculatus, Hyperolius, 318 flavomaculatus, Leptopelis, 318 flavoviridis, Hyperolius, 332 fletcheri, Typhlops, 245 florulentulus, Coluber, 258 florulentulus, Coluber f ., 258 floweri, Eana, 338, 341 Fordia, 171 for mesa, Ohelonia, 166 fornasini, Afrixalus f., 322, 325 fornasini, Euchnemis, 322 fragilis, Heniidactylus, 184 francei, Arthroleptis a., 352 franklini, Oriychocephalus, 240 frenatus, Cycloderma, 172 frenatus, Hemidactylus, 181, 183 frenatus, Xonurus, 223 f riedericianus, Adolfus, 228 frontatus, Crocodilus, 178 fiilleborni, Chamaeleo, 205 fiilleborni, Chamaeleon, 205 fuelleborni, Hyperolius, 328 fiilleborni, Eana, 343 fiilleborni, Eana f., 343 fuliginosa, Amphisbaena, 237 fuliginosus, Boaedon f ., 251 fuliginosus, Lycodon, 250, 251 fulvicollis, Microsonia, 283 fulvicollis, IVIiodon, 283 fulvovittatus, Afrixalus f ., 323 funereus, Bufo, 311 Furcifer, 196 fuscigula, Eana, 340 fuscigula f., 389, 340 fuscorosea, Coronella s., 258 fuscum, Trachischum, 159 fuscum, Triglyphodon, 268 fuscus, Dipsas, 268 gabonensis, Elapomorphus, 283 gabonensis, Miodon, 283 gabonica, Bitis g., 302 gabonica. Echidna, 302 galamensis, Eana, 339 galamensis, Eana g., 338 galeata, Testudo, 173 galloti, Lacerta, 229, 230 Gallotia, 229, 230 gambiense, Lyeophidion c, 253 gardineri, Hemidactylus, 186 garmani, Bufo, 310 garnotii, Hemidactylus, 181 gasconi, Pelomedusa, 173 Gastropholis, 226 Gastropyxis, 263 gecko. Gecko, 181 gehafie, Pentonyx, 173 geitje, Lacerta, 191 GEKKONIDAE, 178, 179 gemmatus. Coluber, 279 gemonensis, Natrix, 257 Geocalamus, 238 Geochelone, 168, 169 Geodipsas, 255, 268 geometrica, Testudo, 169 geometricus, Lycodon, 250 geiardi, Apostolepis, 284 Gerrhopilus, 239 Gerrhopygus, 189 gerrhopygus, Diplodactylus, 189 GEEEHOSAUEIDAE, 223 Gerrhosaurus, 224 giardi, Grayia, 256 gibbosa, Eana, 356, 357 gibbosus, Breviceps, 356 gierrai, Typhlops, 242 gierrai, Typhlops p., 242 Gigantorana, 338 gigas, Caretta, 168 glaber, Hetevolepis, 253 glaber, Tropidonotus, 267 x\y BULLETIN : MUSEUM OP^ COMPARATIVE ZOOLOGY ^landicolor, Hyperolius, 334, 359 Glaniolestes, 267 Glauconia, 246 Globiceps, Dipsas, 269 slobosa, Sehoutedenella, 351 glohulosus, Chaimus, 309 (ilypliolycus, 250 ftootzei, Chainaeleo g., 202 goetzei, Cliamaeleon, 202 goetzei, Hj'perolius, 325 ffoldii, Naia, 293 goldii, Pseudohaje, 293 Roliath, Raiia, 338 gollmeri, Rana, 337 Goiigvloniorphus, 219, 220 (jongylophis, 248 gouldii, Hydrosaurus, 235 gracilis, Aspidorhinus, 232 gracilis, Chamaeleo, 198 gracilis, Chamaeleo g., 198 gracilis, Chlorophis, 262 gracilis, Leptophis, 263 gracilis, Phrynobatrachus n., 348 gracilis, Typhlops, 243, 245 gracillima, Ahaetulla, 261 graeca, Testudo, 16S'. atascadera, S. mexicana, S. grandis). The genitalia of 8. atascadera seem more variable in that part of its range where no other species of the S. bipunctata group are found, than in the northern part of its range where S. hespera is also found. The male palpi of S. grandis are quite uniform in structure, but the epigyna are ex- tremely variable. Both the palpi and epigyna of 8. mexicana are variable. Where collected together, 8. grandis and 8. mexicana always are quite distinct; however females from many localities may be difficult to place. It is assumed that in these latter local- ities only one or the other species occurs. The question may be raised here whether 8. horealis and 8. hespera are two species or subspecies. There is no overlap in the two groups and the distribution would indicate that only one species is involved. However males and females of the two species in the Rocky Mountain region are always easy to place in one species or another, while some female 8. hespera from the West Coast appear much more like 8. horealis. In other words the species seem more distinct along the common border 412 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY of their range than in other parts of it, which would indicate that we are dealing with two species. Natural History. The species of Steatoda live under bark, in crevices of rocks, sometimes under stones. They are common in buildings wdiere they build small webs in corners ; however, during daytime the spiders stay hidden in crevices, coming out to the web in the evening. Females and males may inhabit the same web. In captivity slight vibration of flies in the web, or sometimes only the presence of a dead fly in the web, brought the spiders out of their retreats. Kaston (1948) described the egg sac of S. horealis as being "loosely woven, so that the ivory to pale pink eggs show through the brownish white threads. The shape is approximately spherical 5-6.5 mm. in diameter and I counted in four egg sacs, 37, 47, 76, and 95 eggs respectively." Distribution. This group may be mainly holarctic in distri- bution. The male Steatoda hrasiliana Keyserling from Brazil may be an introduced form. Key to females of the Steatoda bipunctata group Note that figures were drawn with the epigynum lying flat, thus from a slightly anterior position of the abdomen. ]. Raised anterior median lobe of epigynum wider than long (Fig. 113) 2 1. Raised anterior median lobe as long as wide, or longer than wide (Fig. 93, 105, 109) 4 2. Anterior arms of depression pointing straight anterior (Figs. 107-109), California coast range (Map 17) atascadera 2. Arms of depression pointing slightly to sides (Figs. Ill, 113, 115) ... .3 3. Posterior portion of depression swollen and wider than arms (Figs. 117, 118). East of Rocky Mountains to Alaska (Map 19) . . .horealis 3. Posterior portion of depression as wide as arms (Figs. Ill, 113, 115) Western United States to southern British Columbia (Map 19) .... hespera 4. Raised anterior median lobe as wide as long (Figs. 107, 109). Cali- fornia coast range (Map 17) atascadera 4. Raised anterior median lobe longer than wide 5 5. Found in southern California (Map 17) palomara 5. Not found in southern California . 6 r>. With a large swollen posterior lip and swollen areas on each side of depression (Fig. 89) bipunctata (i. Lacking swollen areas on each side of depression 7 LEVI : SPIDER GENERA CRUSTULINA AND STEATODA 413 7. Openings in a deep depression on sides of a very narrow septum (Figs. 93-97). Depression often with material difficult or impossible to remove. Western United States (Map 17) grandis 7. Openings in a shallower depression, septum wider (Figs. 99-103). Material in depression easily removed. Western United States and Mexico (Map 18) mexicana Key to males of the Steatoda bipunctata group Note that figures of the palpus in mesal view were drawn with the radix lying flat, thus from a slightly anterior view. 1. Radix with a well-developed median lobe (Figs. 148, 153). East of Rocky Mountains to Alaska (Map 19) borealis 1. Radix lacking median lobe (Figs. 119, 129, 142, 155) 2 2. Radix enclosing an area, the diameter of which is less, equal or only slightly more than the greatest width of radix (Figs. 119, 155) . . .3 2. Radix enclosing an area, the diameter of which is much greater than the width of the radix (Figs. 129, 145) 4 3. Embolus short and thick (Fig. 156) bipunctata 3. Embolus long and thin (Fig. 123). Western United States (Map 17) grandis 4. Radix with a distinct ectal lobe below tip (Figs. 130-132, 135, 137, 139, 141) 5 4. Radix with no lobe near tip 6 5. California coast range south to Riverside County (Map 17) atascadera 5. Southern California (Map 17) palomara 6. Radix with tip pointing eetally, above a notch, (Figs. 130-132) Southern California (Map 17) palomara 6. End of radix pointing ectal but without notch below (Figs. 125, 146) 7 7. Base of embolus almost circular with rim curved outward (Figs. 144, 147) Western United States, southern British Columbia (Map 19) hespera 7. Base of embolus oval with rim not curved outward (Fig. 128) Western United States and Mexico (Map 17) mexicana Steatoda bipunctata (Linnaeus) Figures 86-89, 155-156 ; Map 16. Aranea bipunctata Linnaeus, 1758, Systema Naturae, 10th ed., p. 620. Aranea 4-punctata, Fabrieus, 1775, Systema Entomologiae, p. 434. Aratnea punctata, De Geer, 1778, Memoires pour servir h. I'histoire des Insectes, vol. 7, p. 255, pi. 15, fig. 1. Theridion 4-punctatum, Walckenaer, 1805, Tableau des Aran6ides, p. 73, pi. 7, figs. 69, 70. 414 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Steatoda 4-punctatum, Sundevall, 1833, Conspectus Arachnidum, p. 17. Phrurolithns ornatu.'i C.L. Koch, 1839, Die Arachniden, vol. 6, p. 114, fig 515. Eiicharia hipunctata, C.L. Koch, 1845, ibid., vol. 12, p. 99, fig. 1027. Theridhnn cruciatvm Giebel, 1869, Zeitschr. Ges. Naturwiss., vol. 34, p. 303 Steatoda hipunctata, Thorell, 1856, Nova Acta reg. Soc. sci. Upsaliensis ser. 3, vol. 2, p. 140. Keyserling, 1884, Die Spinnen Amerikas, vol. 2 pt. 1, p. 116, pi. 6, fig. 76,9,5. Petrunkevitch, 1911, Bull. Amer. Mus Nat. Hist., vol. 29, p. 187. Wiehle, 1937, in Dahl, Die Tierwelt Deutsch lauds, pt. 33, p. 193, figs. 200-204,9,(5. Kurata, 1939, Canadian Ent. vol. 53, p. 81. Roewer, 1942, Katalog der Araneae, vol. 1, p. 412 Gertsch, 1946, in Procter, Biological Survey of the Mount Desert Eegion, pt. 7; Gertsch, 1949, American spiders, p. 258. Locket and Millidge, 1953, British spiders, vol. 2, p. 56, figs. 38,9.5. Hackman, 1954, Acta Zool. Fennica, vol. 79, p. 4. Types. The type locality is presumably Sweden. Diagnosis. This species may have the legs very slightly shorter Map 16. American distribution of Steatoda bipunctata (Linnaeus). than in other species of Steatoda. It is readily distinguished by comparison of the genitalia (Figs. 89, 155, 156). The total length of females is 4.8-7.3 mm., of males, 4.4-6.0 mm. Natural History. This species may have been introduced in recent times. It seems to have survived mainly along the coast or the shores of Lake Ontario. C. Dondale has collected speci- mens on a barn. Distribution. Europe, Siberia, Kamchatka (Wiehle, 1937). Introduced in French Guiana and Venezuela (Keyserling, 1884). LEVI : SPIDER GENERA CRUSTULINA AND STEATODA 415 Probably introduced on the coast of Maine to Newfoundland, north shore of Lake Ontario (Map 16). Records. Newfoundland: (liackman, 1954) ; Nicholsville. iVoyc Scotia: Weymouth; Grand Pre; Baddeck; Barrington; Lequille. New Brunswick: Grand Manan Isl. Quebec: Bonaventura Isl. Ontario: Toronto (W. J. Gertsch) ; Lakeport. Maine: Cumber- land Co. : Sebago Lake ; Hancock Co. : Mt. Desert Isl. ; Lincoln Co.: Jefferson; York Co.: Saeo. New Hampshire: Coos Co.: Gorham, 1946, (E. L. Bell). Steatoda grandis Banks Figures 92-97, 119-123; Map 17. Steatoda grandis Banks, 1901, Proe. Acad. Nat. Sci. Philadelphia, p. 578; 1910, Bull. U.S. Natl. Mus., no. 72, p. 21, Petiunkevitch, 1911, Bull. Anier. Mus. Nat. Hist., vol. 2l», p. 18S. Eoewer, 1942, Katalog der Araneae, vol. 1, p. 413. Steatoda zionis Chamberlin and Tvie, 1935, Bull. TTuir. Utah, bid. ser. vol. 2, no. 8, p. 12, figs. 3G, 42.9,c5. Roewer, 1942, Katalog der Araneae, vol. 1, p. 414. NEW SYNONYMY. Steatoda mcrala I. Fox 1940, Proc. Biol. Soc. Washington, vol. 53, p. 41, fig. 3, 9 . NEW SYNONYMY. Types. Female type of S. grandis from Albuquerque, New Mexico, in the Museum of Comparative Zoology. Male holotype, female allotype of 8. zionis from Zion National Park, Utah, in the American IMuseum of Natural History. Female type of 8. merula from ]\It. Lemmon, Arizona, in the United States Na- tional Museum (no. 1368). Diagnosis and Variation. The epigyna of this species are ex- tremely variable, particularly in the southern portion of its range (Figs. 93-97). The epigynum of the type of 8. grandis resembles Figure 93. Only in the narrower septum and slightly deeper depression on each side of the epigynum does this species differ from S. americana; furthermore the depression of the epigynum is frequently filled with a material difficult or impos- sible to remove. The epigyna of the types of both 8. grandis and *S'. merula are partly filled in this way. The palpus, in contrast to the female genitalia, shows little variation (Figs. 119, 120, 123). Only two males from central Colorado had the tip of the radix slightly different (Figs. 121, 122). Total length of females 4.9-9.0 mm. ; that of males 4.3-7.4 mm. 416 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Natural History. In Colorado this species has been collected under stones, rocks and in cliffs from lower elevations (7000- 8000 ft.) than S. hespera. Dist7-ihution. Western United States (Map 17). Records. South Dakota: Custer Co.: Blue Bell, 4900 ft., $ . Pennington Co. : Horsethief Lk., 9 ; S. of Rapid City, 9 . Wyom- ing: Devil's Tower Natl. Mon., Aug. 5, 1952 (B. Malkin), 9 . S. atascadera " S grandis • S palomaro a 1^4 Map 17. Distribution of Steatoda grandis Banks, S. palomara Chamber- ]in and Ivie and S. atascadera Chamberlin and Ivie. Colorado: Archuleta Co. : Piedra, 9 . Boulder, 9 . Denver, 9 . Douglas Co. : Parker, 9 . El Paso Co. : Cascade, 9 . Gunnison Co. : Taylor River, 9 . La Plata Co. : Durango, 9 . Larimer Co. : Estes Park, 7800 ft., S ; Fort Collins, 9 . Mineral Co. : West Fork of Wolf Creek, 7800 ft., 9 . Rio Grande Co. : Monte Vista, 9 . Routt Co. : Steamboat Springs, 9 . Teller Co. : Pikes Peak, 10,000 ft., 9 . New Mexico: Colfax Co.: Raton, <5 . Ber- nalillo Co. : Sandia Mts., 9 . Catron Co. : 25 mi. N. of Alma, 9 . LEVI : SPIDER GENERA CRLTSTULINA AND STEATODA 417 Lincoln Co., 9 . Santa Fe Co. : 10 mi. S. of Santa Fe ; Little Tesugue Canyon, S . Valencia Co. : Bluewater. Utah : Carbon Co. : Helper, $ ; Price, 9 , £ . Iron Co. : Parowan, $ S . Emery Co. : Ferron, 9 , $ . Garfield Co. : Esealante, S ; Red Canyon, 9 ,S . Millard Co. : Scipio, 9 . San Juan Co. : Monticello, 9 . Sevier Co. : Fish Lk., 9 ; Richfield, 9 ,$ . Uintah Co. : Vernal, 9 ; Watson, 9 ; White Riv. nr. Evacuation Cr., 5000 ft., 9 . Utah Co. : Levan, 9 . Wayne Co. : Fruita, 9 ; Notom, 9 , $ . Ari- zona: Apache Co. : Wide Ruin, $ , ,J ; 17 mi. NE. of Whiteriver, 9 , S ; Hannagan, ? . Cochise Co.: Chiricahua Mts. (many col- lections) 9 , S ; Carr Canyon, Huachuca Mts., $ . Coconino Co.: Oak Cr. Canyon, 9 . Gila Co. : 7 mi. N. of Payson, 2 ; Middle Pioneer Cr., Pinal Mts., 9 , S . Graham Co. : Graham Mt. (many collections) 9 , S . Pima Co. : Santa Catalina Mts. (many col- lections) 9 ,S . Santa Cruz Co. : Santa Rita Mts. (many collec- tions), 9 , $ . Yavapai Co. : Prescott, 9 . Oregon: Wheeler Co. : Richmond, Fall 1948 (Donnely),? ; 3 Aug. 1951 (V. Roth), 9,3. Steatoda mexicana, new name Figures 98-103, 124-128 ; Map 18. Stearodea amerioana F.O.P. Cambridge, 1902, Biologia Centrali-Americana, Araneidea, vol. 2, p. 377, pi. 35, fig. 12,$. Not Steatoda americana (Emerton). Steatoda americana, Petruukevitch, 1911, Bull. Amer. Mus. Nat. Hist., vol. 29, p. 187. Eoewer, 1942, Katalog der Araneae, vol. 1, p. 413. Not Steatoda americana (Emerton). Not Banks, 1909, Proc. Acad. Nat. Sci., Philadelphia, vol. 61, p. 205. Steatoda c/randis, Chamberlin and Ivie, 1935, Bull. Univ. Utah, bid. ser., vol. 2, no. 8, figs. 37, 40, 41, $,5. Not S. grandis Banks. Types. Two female syntypes from Omiltemi, Guerrero in the British Museum (Natural History). Diagnosis and Variation. The epigyna of this species are very variable particularly in the northern and eastern portion of its range (Figs. 99-103). The epigynum of a syntype examined resembles that from St. George, Utah (Fig. 99) but the scape is slightly wider. However, it is slightly narrower than that from Chiricahua Mountains (Fig. 103). Most females can readily be told from S. grandis by the wider scape of the epigy- 418 BULLETIN : MUSEUM OP COMPARATH^ ZOOLOGY num. The shape of the radix of the palpus is variable. Some almost resemble that of S. grandis, in others the radix is long and stretched. The shape of the base of the embolus (Fig. 128) readily distinguishes this species from *S'. hespera. Total length of females 5.1-9.0 mm. ; of males 4.8-5.7 mm. Distribution. Idaho to Texas, Arizona south to Guerrero (Map 18). Records. Texas: Big Bend Natl. Pk., Chisos Mts., $ . Walker Co.: Huntsville, 2 Oct. 1950 (M. A. Cazier), 9 . Colorado: Lari- mer Co.: Eed Feather Lakes, June 14, 1933 (A. B. Klots) 5. New Mexico: Otero Co. : Camp Mary White, 9 , $ . Idaho: Bear Lake Co. : Bloomington, 9,5. Utah : Bryce Canyon Natl. Pk., $ ; Garfield Co. ; 15 mi. N. of Boulder, ? ; Henry Mts., 9 ; 10 mi. SE. Map 18. Distribution of Steatoda mexicana, new name. of Panguiteh, 9 . Salt Lake City, 9 . Sevier Co. : Fish Lk., 9 . Utah Co. : Lehi, 9 ; west shore of Utah Lk., 9 , ^ . Washington Co. : St. George, 9 ; Beaver Dam Wash., 9 . Arizona: Cochise Co.: Rustler's Camp. Chiricahua Mts., 9 , c? . Coconino Co.: Flagstaff, 9 . Grand Canyon Natl. Pk., 9 . Graham Co. : Gra- liara Mt., Pinecrest, 9 , cJ . Santa Cruz Co. : 4 mi. SE. of Ruliy, 9 . LEVI : SPIDER GENERA CRUSTULINA AND STEATODA 419 Yavapai Co.: Prescott, 9 . Chihuahua: SW. of Santa Barbara. 8400 ft., 9. Durango: " Providencia, " 7700 ft., $ . Steatoda palomara Chamberlin and Ivie Figures 104, 105, 129-133; Map 17. steatoda palomara Chamberlin and Ivie, 1935, Bull. Univ. Utah, biol. ser., vol. 2, no. 8, p. 13, fig. 43, $ . Roewer, 1942, Katalog der Araneae, vol. 1, p. 414. Type. Female type from Mt. Palomar, California in the American Museum of Natural History. Diagnosis. The epigynum (Fig. 105) and the tip of the radix of the palpus (Figs. 130-132) distinguish this species. Total length of females 5.0-6.8 mm. ; males, 5.3 mm. Distribution. Known only from San Diego County, California (Map 17). Records. California: San Diego Co. : Mt. Palomar, July 25, 1931 (W. Ivie), 9 ; July 13, 1953 (W. J. and J. W. Gertsch), 9 ; 3000-5000 ft., June 30, 1956 (W. J. Gertsch and V. Roth), 9 , S ; Descanso, June 25, 1947 (W. M. Pearce). Steatoda atascadera Chamberlin and Ivie Figures 106-109, 134-141 ; Map 17. Steatoda atascadera Chamberlin and Ivie, 1942, Bull. Univ. Utah, biol. ser., vol. 7, no. 1, p. 39, figs. 83, 84, 9, o. Types. Male holotype, female allotype from Atascadero Lake, San Luis Obispo County, California, from under concrete bridge at south end of lake, in the American Museum of Natural History. Diagnosis and Variation. The epigyna (Figs. 107-109) and palpi (Figs. 134-141) of this species are extremely variable. Hardly two look alike. The tip of the radix of the male palp as well as the width of the anterior raised lobe of the epigynum separate this species from 8. palomara and 8. hespera. Total length of females 4.8-6.0 mm., of males, 3.8-4.4 mm. Distribution. Coast range of California from Santa Cruz to Riverside County (Map 17). Records. California: Los Angeles Co.: nr. Claremont; (V. Roth. W. J. Gertsch), 9. San Gabriel Mts. (many records). 420 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 9,$ ; Santa Monica Mts. (R. X. Schick), 9. Glendale (E. I. Schlinger), 9 . Monterey Co.: Hastings Nat. Hist. Reserve (J. Linsdale ) , 5 . Orange Co.: Silverado Canyon, Santa Ana Mts. (R. X. Schick) 9 . Riverside Co. : San Jacinto Mts. (R. X. Schick), 9 . Santa Cruz Co. : Big Trees Park (A. F. Archer), $ : Brookdale (W. H. Irvin),5 . Yolo Co.: Davis (E. I. Schlinger). 9. Steatoda hespeba Chamberlin and Ivie Figures 90, 91, 110-115, 142-147; Map 19. Steatoda hespera Chamberlin and Ivie, 1933, Bull. Univ. Utah, biol. ser., vol. 2, no. 2, p. 9, figs. 4-6, 9,5. Gertseh, 1939, Amer. Mus. Novitates. no. 1032, p. 4. Chamberlin and Ivie, 1941, Bull. Univ. Utah, biol, ser., vol. 6, no. 3, p. 12. Roewer, 1942, Katalog der Araneae vol. 1, p. 414. Levi and Levi, 1951, Zoologica, vol. 36, p. 220; 1955, Canadian Field Nat., vol. 69, p. 69. Lowrie and Gertseh, 1955, Amer. Mus. Novitates, no. 1736, p. 7. Types. Syntypes from Clear Creek, Raft River Mountains, Utah, in the American Museum of Natural History. Diagnosis and Variation. The proportions of the anterior raised lobe as well as the even width of the depressed area of the epigynum (Figs. 111-115) distinguish this species from S. hore- alis and S. atascadera. The radix of the palpus lacks the median lobe typical of S. horealis and lacks the lobe below the tip typi- cal of 8. atascadera (Figs. 143, 146). Specimens from the western part of the range seem slightly larger in size than those of the east. Total length of females 4.2-7.5 mm. ; of males, 3.6-5.4 mm. Natural History. In Colorado this species has been collected under bark of dead conifers and in cabins at elevations between 8500 ft. and timberline. Distribution. Western United States (Map 19). Records. British Columhia: Salmon Arm; Vernon. Montana: Helena ; Flathead Co. ; Gallatin Co. ; Glacier Natl. Pk. ; Ravalli Co. Wyomi7ig : Grand Teton Natl. Pk. ; Sublette Co. ; Teton Co. ; Yellowstone Natl. Pk. Colorado: Gunnison Co.; La Plata Co.; Pitkin Co. ; Routt Co. Idaho: Adams Co. ; Bear Lake Co. : Frank- lin Co.; Idaho Co.; Payette Co.; Teton Co.; Twin Falls Co.; LEVI : SPIDER GENERA CRUSTULINA AND STEATODA 421 Valley Co. Utah: Box Elder Co.; Cache Co.; Garfield Co.; Millard Co. ; Rich Co. ; Salt Lake Co. ; Sevier Co. ; Summit Co. ; Utah Co.; Wasatch Co. Nevada: Clark Co.; Reno. Washington: Ferry Co.; Mt. Ranier Natl. Pk. ; Spokane Co.; Walla Walla; Whitman Co.; Yakima. Oregon: Baker Co.; Crater Lk. Natl. Pk. ; Deschutes Co. ; Grant Co. ; Jackson Co. ; Klamath Co. ; Lake Co. ; Lane Co. ; Morrow Co. ; Three Sisters Primitive Area ; -T-'^ -^^.J U Map. 19. Distribution of Steatoda hespera Chamberlin and Ivie and 6'. borealis (Hentz). Wheeler Co. California: Calaveras Co.; Contra Costa Co.; Eldorado Co. ; Fresno Co. ; Kings Canyon Natl. Pk. ; Los Angeles Co. ; Madera Co. ; Mariposa Co. ; Mendocino Co. ; Modoc Co. ; Mono Co. ; Monterey Co. ; Nevada Co. ; Placer Co. ; Sacramento Co. ; Santa Clara Co. ; Santa Cruz Co. ; Sequoia Natl. Pk. ; Shasta Co. ; Sierra Co. ; Siskiyou Co. ; Tulare Co. ; Yolo Co. ; Yosemite Natl. Pk.; Yuba Co. 422 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Steatoda borealis (Hentz) Figures 116-118, 148-154; Map 19. Theridion boreale Hentz, 1850, Jour. Boston Soc. Nat. Hist., vol. 6, p. 274, pi. 9, fig. 4,5 ; 1875, Spiders of the United States, p. 145, pi. 16, fig. 4,5 pl.21, fig. 13,5. Steatoda horcalis, Emerton, 18S2, Trans. Connecticut Acad. Sci., vol. 6, p. 19, pi. 4, fig. 1,$,5. Keyserling, 1884. Die Spinnen Amerikas, vol. 2, no. 1, p. 119, pi. 6, fig. 77,?, 5. McCook, 1890, American Spiders, vol. 2, p. 55. Marx, 1890, Proc. U.S. Natl. Mus., vol. 12, p. 521; 1892, Proc. Ent. Soc. Washington, vol. 2, p. 156. Fox, W., op. cit., p. 268. Banks, 1892, Proc. Acad. Nat. Sci. Philadelphia, p. 31. Emerton, 1894, Trans. Connecticut Acad. Sci., vol. 9, p. 406. Baker, 1894, Ent. News, vol. 5, p. 164. Banks, lS95a, Jour. New York Ent. Soc, vol. 3, p. 84; 1895b, Ent. News, vol. 6, p. 182; 1899, Proc. Ent. Soc. Wash- ington, vol. 4, p. 189. Emerton, 1902, The common spiders, p. 119, figs. 276-279, 9,6 . Scheffer, 1905, Trans. Kansas Acad. Sci., vol. 19, p. 192. Banks, 1907, Ann. Eept. Indiana Dept. Geol. and Nat. Res. no. 31, p. 739. Bryant, 1908, Oceas. Papers Boston Soc. Nat. Hist., no. 9, p. 16. Banks, 1910, Bull. U.S. Natl. Mus., no. 72, p. 21. Petrunkevitch, 1911, Bull. Amer. Mus. Nat. Hist., vol. 29, p. 188. Banks, 1911, Proc. Acad. Nat. Sci. Philadelphia, vol. 63, p. 445. Comstock, 1912, The spider book, p. 360, fig. 363,9. Emerton, 1913, Appalachia, vol. 12, p. 155. Barrows, 1918, Ohio Jour. Sci., vol. 18, p. 304. Bishop, 1924, Bull. New York State Mus. no. 251, p. 21. Barrows, 1925, Ann. Ent. Soc. Amer., vol. 18, p. 506. Emerton, 1928, Univ. Toronto Studies, Biol., vol. 32, p. 45. Crosby and Bishop, 1928, Mem. Cornell Univ. Agr. Exp. Sta., no. 101, p. 1040. Emerton, 1930, Publ. Nantucket Maria Mitchell Assoc, vol. 3, p. 164. Worley, 1931, Studies Univ. Nebraska, vol. 27, p. 28. Chickering, 1931, Papers Michigan Acad. Sci., vol. 15, p. 351. Elliott, 1932, Proc. Indiana Acad. Sci., vol. 41, p. 424. Chamberlin and Ivie, 1933, Bull. Univ. Utah, biol. ser., vol. 2, no. 2, fig. 7, 9 . Stiles and Detwiler, 1938, Proc. Iowa Acad. Sci., vol. 45, p. 286. Kurata, 1939, Canadian Nat., vol. 53, p. 81. Comstock, 1940, The spider book, rev. ed. p. 375; fig. 363,9. Kurata, 1941, Univ. Toronto Studies, biol. ser., no. 48, p. 109. Truman, 1942, Bull. Univ. Pittsburgh, vol. 38, no. 2, p. 4. Roewer, 1942, Katalog der Araneae, vol. 1, p. 413. Lowrie, 1942, Bull. Chicago Acad. Sci., vol. 6, p. 169. Kurata, 1943, Canadian Ent., vol. 57. p. 10. Muma, 1943, Common spiders of Maryland, p. 65, pi. 4, fig. 2, pi. 12, fig. 20. Chamberlin and Ivie, 1944, Bull. Univ. Utah, biol. ser., vol. 8, no. 5, p. 46. Muma, 1945, Bull. Agr. Exp. Sta. Univ. Maryland, no. A38, p. 27. Gertsch, 1946, in Procter, Biol. Surv. of the Mount Desert Reg., pt. 7, p. 519. Rapp, 1946, Bull. Brooklyn Ent. Soc.,- vol. LEVI : SPIDER GENERA CRUSTULINA AND STEATODA 423 41, p. 4. Chamberlin and Ivie, 1947, Bull. Univ. Utah, biol. ser., vol. 10, no. 3, p. 27. Lowrie, 1948. Ecology, vol. 29, p. 350. Kaston, 1948, Bull. Connectifut Geol. Nat. Hist. Surv., no. 70, p. 85, figs. 51-56, 2014, $,,5. Gertseh, 1949, American spiders, p. 258. Kurata, 1949, Canadian Ent., vol. 81, p. 127. Elliott, 1953, Proc. Indiana Acad. Sci., vol. 62, p. 309. Kaston, 1953, How to know the spiders; p. 162. Levi and Field, 1954, Amer. Midland Nat., vol. 51, p. 444. Kaston, 1955, Trans. Illinois Acad. Sci., vol. 47, p. 166. Type locality. "United States." The type specimens are lost. Diagnosis. The wide raised median lobe, the swollen posterior area of the depression of the epigynum (Figs. 117-118), and the median lobe on the palpal radix (Fig. 148) distinguish this species from S. hespera. Total length of females: 3.8-7.0 mm. ; of males, 4.3-5.9 mm. Natural History. This is the common Steatoda of the eastern United States, which is found on buildings. Distribution. East of the Rocky Mountains south to Texas and North Carolina; north to Alaska (Map 19). There are more old records from the southern states than recent ones. Shreve- port, Louisiana (Banks, 1899), Alabama (Hentz, 1875) and Georgia (Keyserling, 1884), however, this species has only rarely been collected in these states during the last fifty years. One record probably an accidental one, is that of a male and female collected at Lakeside, San Diego Count}', California, by N. Banks, another from Eugene, Oregon, April 1947 (B. Malkin), $ , probably introduced. Records. Alaska: (Chamberlin and Ivie. 1947) ; Fox; Tanana: Mt. McKiiiley Natl. Pk. Nova Scotia: Coldbrook. Quebec: Montreal; Matapedia. Ontario: (Emerton, 1928; Kurata, 1939, 1941, 1943) ; 36 mi. N. of North Bay; St. Thomas; Lake Tema- gami; Newmarket; Haliburton. Manitoba: Riding Mountain. Northwest Territory: (Kurata. 1949), Outpost Island, Great Slave Lk. Alberta: Medicine Hat; MacMurray; Seba Beach; Mt. Sentinel. Maine: (Bishop, 1924) ; Cumberland Co.; Piscata- quis Co. New Hampshire: Carroll Co.; Cheshire Co.; Coos Co.; Grafton Co. Vermont: Addison Co. Massachusetts: (Emerton, 1930); Es.sex Co.; Middlesex Co.; Norfolk Co.; Plymouth Co.; Worcester Co. Rhode Island : Vvoxidence. Connecticut: (Kaston, 1948). New York: (Banks, 1892; Bishop and Crosby, 1928); Clinton Co. ; Fulton Co. ; Jetferson Co. ; Nassau Co. ; Rockland 424 BULLETIN : MUSEUM OF COMPABATIVE ZOOLOGY Co.; St. Lawrence Co.; Steuben Co. New Jersey: (Rapp. 1946) ; Bergen Co. ; Cape May Co. ; Mercer Co. ; Morris Co. Pennsyl- vania: Bradford Co.; Montgomery Co.; Philadelphia Co.; Pitts- burgh; Schuykill Co. Delaware: Wilmington. Maryland: (Muma, 1943, 1945). Virginia: Fairfax Co.; Pulaski Co. Wesi Virginia: Ohio Co. North Carolina: (Banks, 1911) ; Avery Co.; Buncombe Co.; Guilford Co.; Swaine Co. Mississippi: Rankin Co. Ohio: Athens Co.; Columbiana Co.; Cuyahoga Co. Michigan: (Hentz, 1875; Baker, 1894; Chickering, 1931; Lowrie, 1948); Chippewa Co.; Ogemaw Co.; Washtenaw Co. Indiana: (Banks, 1907; Lowrie, 1948; Elliott, 1953). Kentucky: Hardin Co. Wis- consin: (Levi and Field, 1954); Crawford Co.; Grant Co.; Marathon Co.; Milwaukee Co.; Racine Co. Illinois: (Lowrie, 1948, Kaston, 1955); Champaign Co.; Cook Co.; Winnebago Co. Minnesota: Clay Co.; Clearwater Co.; Freeborn Co.; Mar- shall Co.; Mille Lacs Co.; Minneapolis; Saint Louis Co. Iowa: (Stiles and Detwiler, 1938) ; Ames; Cerro Gordo Co.; Woodbury Co. Missouri: (Banks, 1895b). North Dakota: Divide Co.; Cass Co. South Dakota: Custer Co. Nebraska: (Worley, 1931) ; Saline Co. Kansas: Riley Co. Texas: McLellan Co.: Moody, Sept. 1941 (C. and M. Goodnight),?. Montana: Cascade Co.; Custer Co.; Daniel Co.; Prairie Co. Colorado: Denver; El Paso Co. ; Larimer Co. ; Morgan Co. Figs. 1-3. Crustulina sticta (O.P. Cambridge). 1. Female genitalia, ventral view. 2. Epigynum. 3. Abdomen of female, dorsal view. Figs. 4-6. C. altera Gertseh and Archer. 4. Female genitalia, ventral view. 5. Epigynum. 6. Colulus and spinnerets. Fig. 7. C. sticta (O.P. Cambridge), left palpus. Figs. 8-10. C. altera Gertseh and Archer. 8. Palpus. 9. Palpus expanded, subventral view. 10. Female, dorsal view. Abbreviations: C, conductor; E, embolus; M, median apophysis; R, radix; S, subtegulum ; T, tegulum; Y, cymbium. 17 12 / Figs. 11-13. Stcatoda nigrofemorata (Keyserlmg). 11. Left palpus. 12. Female genitalia, ventral view. 13. Epigynum. Fig. 14. S. quaesifa (O.P. Cambridge), palpus. Figs. 15-18. S. moesfa (O.P. Cambridge). 15. Abdomen of female, dorsal view. 16. Female genitalia, ventral view. 17. Epigynum. 18. Palpus. Figs. 19, 20. S. saltensis, new species. 19. Female genitalia, ventral view. 20. Epigynum. Figs. 21, 22. (S. autumnalis (Banks). 21. Female genitalia, ventral view. 22. Epigynum. 24 28 • v^ #':' '■>^ jiffia^pgg^^^M^^ 29 Figs. 23-27. Steatoda transversa (Banks). 23. Female genitalia, ventral view. 24. Epigynum. 25. Abdomen of female, dorsal view. 26, 27. left palpus. 26. Mesal view. 27. Ventral view. Figs. 28-31. S. quadrimaciilata (O.P. Cambridge). 28. Female genitalia, ventral view. 29. Epigynum. 30, 31. Palpus. 30. (Tamaulipas). 31. (Guerrero). 36 Figs. '62, 33. Steatoda fulva ( Keyserling j , left palpus. 32. (California). 33. (Florida). Figs. 34-36. S. meflialis (Banks), palpus. 3-1. (Texas). 35. (Arizona). 36. (California). Figs. 37-41. S. piilcher (Keyserling), palpus. 37. Mesal view (Cali- fornia. 38-41. Ventral view. 38. (California K 39. (Texas). 40. (Coa- huila). 41. (Morelos). 46 44 ■."?r*^-i.' 47 ^^ St '-•^f^ », -^ : Ifl^* > ^ 1 5.^'' 52 53 Figs. 42-44. Steatoda mcdiaU.s (Banks). 42. Female genitalia, ventral view. 43, 44. Epigynum. Figs. 45-47. S. fulva (Keyserling). 45. Female genitalia, ventral view. 46, 47. Epigynum. Figs. 48-51. S. ixdclicr (Keyserling). 48. Female genitalia, ventral view. 49, 50. Epigynmn. 51. Abdomen of a female, dorsal view. Fig. 52. S. fulva (Keyserling), abdomen of female, dorsal view. Figs. 53-55. iS'. mcdialis (Banks). 53. Carapace of female. 54, 55. Alidoinen of female, dorsal view. Figs. 56-65. Steatoda alboinaculata (De Geer). 56. Carapace of female. 57. Female genitalia, ventral view. 58. Epigynum. 59-61. Left palpus. 60. Expanded, suljventral view. 61. Expanded, subeetal view. 62. Colulus and spinnerets. 63. Left ehelieerae of male, inside view. 64-65. Abdomen of female. 64. Ventral view (Colorado). 65. Dorsal view (Massachusetts). Ahhrcriations: C, conductor; E, embolus; M, median apophysis; P, paracymbial hook; R, radix; T, tegulum. 66 ^ 71 67 72 Figs. 66-69. Steatoda americana (Enierton). 6Q. Female genitalia, ventral view. 67. Epigynuni. 68. Left Palpus. 69. Palpus, expanded, subventral view. Figs. 70-73. ti. ticptemmaculalu (Keyserling ). 7U. Abdomen of female, dorsal view. 71. Female genitalia, ventral view. 72. Epigynuni. 73. Palpus. Abbreviations: C, t-onductoi-; E, embolus; M, median apophysis; E, radix; T, tegulutn. .-i^. ^, -■-— '^S?>^''*^'^^'^^^:ii-"'^- .---■ 78 81 85 Fig. 74. Steaioda grossa (C.L. Koi-h), left palpus. Figs. 75, 7^1. alpus. 7."). Expanded, ectal view. 7(5. Ventral view. Figs. 77-79. S. caNtanea (Clerek). 77. I'alpus. 78. Female genitalia ventral view. 79. Epigyiuiiii. Figs. 80-82. jS'. triangulo.sa (Walekenaer). 80. Female genitalia, ventral view. 81. Epigynum. 82. Abdomen of female, dorsal view. Figs. 83-85. S. grossa (C.L. Koch). 83. Female genitalia, venti-al view. 84. Epigynum (California). 85. Epigynum (Michoacan). Abbreviations : C, conductor; E, eniliolus; M, median apophysis; R, radix. Figs. 86-89. Steatoda bipunctata (Linnaeus). 86. Colulus and spinnerets. 87. Left cheliceia of female, inside view. 8S. Female genitalia, dorsal view. 89. Epigynum. Figs. 90, 9L S. Jtespera Cliamberlin and Ivie, female. 90. Abdomen, ventral view. 91. Dorsal view. 'k.' ■ 94 95 96 97 100 101 Figs. 91! 97. Stratuda graiidis Banks. 92. Female genitalia, dorsal view (Zion Natl. Pk.). 93-97. Epigynuni. 93. (Zion Natl. Pk.). 94. (Richfield, Utah). 95. (Graham Mts., Arizona). 96, (Hannagan. Arizona). 97. (Alma, New Mexico). Figs. 98-103. S. nic.ricuHd, new name. 98. Female genitalia, dorsal view (Chiricahiia Mts., Arizona). 99-103. Epigynum. 99. (St. George, Utah). 100. (Big Bend, Texas). 101. (Otero Co.. New Mexico). 102. (Utah I>k.. Utah). 103. (Chiricahua Mts., Arizona). a.fi 104 106 107 108 109 10 114 Figs. lU-t, lUo. Steatoda palomara Chamberlin and Ivie. 104. Female genitalia, dorsal view. 105. Epigymun. Figs. 106-109. S. atascadera Chamberlin and Ivie. 106. Female genitalia, dorsal view. 107-109. Epigynum (all from San Gabriel Mts.). Figs. 110-115. S. hespera Chamberlin and Ivie. 110, 112, 114. Female genitalia, dorsal view. Ill, 113, 115. Epigynum. 110, 111. (Montana). 112, 113. (Colorado). 114, 115. (California). Figs 110-118. .*^. hnrealis (Hentz). 116. Female genitalia, dorsal view. 117, 118. Epigynum. 116, 117. (Wisconsin). 118. (North Carolina). Figs. 119-123. Steatuda gramlis Banks, left palpus. 119, 120, 123. (Zioii Natl. Pk.). 119. Mesal view. 120. Ventral view. 121, 122. Radix, ventral view. 121. (Denver, Colorado). 122. (Estes Park, Colorado). 123. Ectal view. Figs. 124-128. -S'. nu.ricana, new name, palpus. 124, 12;'), 128. (Chirica- hua Mts., Arizona). 124. Mesal view. 125. Ventral view. 126, 127. Radix, ventral view. 126. (Utah Lk., Utah), 127. (Blooniington, Idaho). 128. Mesal view. Figs. 129-133. .S. palomara Clianilierlin and Ivie, palpus (all from Mt. Palomar, California). 129. Mesal view. 130. Ventral view. 131, 132. Radix of two different specimens. 133. Ectal view. Figs. 134-141. Steatoda ntascadera Chamberliii and Ivie, Left palpus. 134, 135, 136. (San Gabriel Mts., California). 134. Mesal view. 135. Ventral view. 136. Ec'tal view. 137. Radix, ventral view (Fish Canyon). 13S, 139. (San Gabriel Mts.). 138. Mesal view. 140, 141. (Santa Cruz Co.). 140. Mesal view. 141. Radix, ventral view. Figs. 142-147. S. hcspera Clianiljerlin and Ivie, palpus. 142-144. (Mon- tana). 145-147. (California). 142, 145. Mesal view. 143, 146. Ventral view. 144, 147. Eetal view. Figs. 148-154. Steafoda horcaUs (Heiitz), left palpus. 148, 149, 152. (Wisconsin). 148. Mesal view. 149. Ventral view. 152. Eetal view. 150, 151. Radix, ventral view. 150. (South Dakota). 151. (Massachusetts). 153, 154. Expanded. 153. Mesal view. 154. Ectal view. Figs. 155-156. S. hipunctata (Linnaeus), palpus. 155. Mesal view. 15(3. Ectal view. Abbreviations: C, conductor; E, embolus; M, median apophysis; P, paracymbial hook; R, radix; T, tegulum; Y, cymbium. Bulletin of the Museum of Comparative Zoology AT HARVARD COLLEGE Vol. 117, No. 4 CONTRIBUTION TO A REVISION OF THE EARTHWORM FAMILY OCNERODRILIDAE. THE GENUS NEMA TOGENIA . By G. E. Gates CAMBRIDGE, MASS., U.S.A. PRINTED FOR THE MUSEUM September, 1957 Publications Issued by or in Connection WITH THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE Bulletin (octavo) 1863 — The current volume is Vol. 117. Breviora (octavo) 1952 — No. 81 is current. Memoirs (quarto) 1864-1938 — Pul)lication was terminated with Vol. 55. JoHNSONiA (quarto) 1941 — A publication of the Department of Mollusks. Vol. 3, no. 35 is current. Occasional Papers of the Department op Mollusks (octavo) 1945 — Vol. 2, no. 21 is current. Proceedings of the New England Zoological Club (octavo) 1899-1948 — Published in connection wath the Museum. Publication terminated with Vol. 24. The continuing publications are issued at irregular intervals in num- bers which may be purchased sei^arately. Prices and lists may be obtained on application to the Director of the Museum of Comparative Zoology, Cambridge 38, Massachusetts. Of the Peters "Check List of Birds of the World," volumes 1-3 are out of print ; volumes 4 and 6 may be obtained from the Harvard Uni- versity Press ; volumes 5 and 7 are sold by the Museum, and future volumes will be published under Museum auspices. Bulletin of the Museum of Comparative Zoology AT nAEVARD COLLEGE Vol. 117, No. 4 CONTRIBUTION TO A REVISION OF THE EARTHWORM FAMILY OCNERODRILIDAE. THE GENUS NEMA TOGENIA . By G. E. Gates CAMBl{]])(iE, .MASS., U.S.A. !■ R i X T E U FOR T HE M L' S E U -M September, 1957 No.4 — Contribution to a revision of the earthworm family Ocnerodrilidae. The genus Nematogenia. By G. E. Gates INTRODUCTION Since 1900, three supposedly endemic ocnerodrilids have been added to the long list (cf. Gates, 1942a) of North American species known only from types or original material. Three exotic species have been recorded (Gates, 1954, pp. 245-246) from the mainland or a neighboring island. The casual spoil of other collecting (as so often in the past) now has made possible, for the first time in nearly six decades, some advance in our understanding of an important constituent of the oligochaete fauna of the continent. The author's thanks are extended to Dr. P. J. Darlington. Jr., and to Dr. G. E. Pickford, for the opportunity of studying these interesting earthworms. Genus NeMATOGENIA Eisen, 1900 Ooiierodrilm {Nematogenia) Eisen 1900, Piot-. California Ac. Sci., (3) 2, \>. 112. (Type species, Pygmaeodrilus lacuum Beddard, 1893.) Xematogenia, Michaelseu, 1900, Das Tierieicli, 10, p. 376. Xcmaiogpnia, Stephenson, 1930, The Oligochaeta, p. 862. The genus has been defined (Stephenson, 1930) essentially as follows : Dorsal pores present. Male pores, united with aper- tures of the single pair of prostates, on xvii. Spermathecae, adiverticulate, one pair, with pores on or behind 8/9. Gizzards, two, in vi-vii. Calciferous sacs, one pair, in ix. Metandric. Accordingly, two, three or four characters contraindicate inclu- sion of the Hispaniolan species, described below. Admission to all other ocnerodrilid genera is, by definition, also denied. Erection of at least two new genera is required by previous prac- tice. Re-examination of the basis of classification of the family, how^ever, first seems preferable. Ocnerodrilid genera have been defined (cf. Stephenson, 1930, pp. 856-863) by the following characters; closeness of pairing 428 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY of setae, presence or absence of dorsal pores, presence or absence of ectal union of male deferent ducts with ducts of one pair of prostates, number of spermathecae and location of the sperma- thecal pores, number and location of ^zzards, number and loca- tion of ealciferous glands, presence or absence of diverticula on the sperraatliecal duct, numljer and location of prostates, andry (number and location of testes). However, little import- ance seems to have been attached to somatic structure. Calci- ferous glands and gizzards are not mentioned in the definition of Pygmaeodrilus, and species of Gordiodrilus may have a pair of lateral sacs or a single, median and ventral one, as well as one, two or no gizzards. Several genera have the same gizzard or calciferous-gland characters. Dorsal pores are present, in the Afro-American portion of the family (that with extramural eal- ciferous glands), only in two species. Setal pairing is not mentioned in definitions of five of the genera recognized in 1930. Wide setal pairing, in anterior and middle regions of the body, has been reported from one species (Seychelles), in middle and posterior regions of the body, from two species (South America). Obviously, then, division of the family into genera has been based mainly on the genital organs. The reproductive system, in all earthworm families, always has provided most of the characters for distinguishing and de- fining species. It is now being found that this system is "most susceptible to evolutionary change when reproduction is par- thenogenetic" (Gates, 1956a, p. 227) for parthenogenesis permits more rapid accumulation of mutations. Such "evolutionary plasticity of the reproductive system necessitates considerable caution in use of its structures for taxonomic purposes at genus or higher level" (Gates, 1956b, p. 30). Uniquely diagnostic somatic characters, in the Oligochaeta as known today, are very rare. None has been found in the Ocnero- (Irilidae. All species herein considered do, however, share a pair of somatic characters which are unknown in that combination elsewhere in the family. A combination of three genital char- acters, also found in each of the species herein considered is, however, shared with other groups that now have subgeneric or generic status. Obviously, then, since other characters are un- available, the genus must be defined, at least primarily, by GATES: THE GENUS NEMATOGENIA 429 somatic structure, as below. For the present, non-diagnostic but universally descriptive genital characters are retained in the definition. Latero-oesophageal hearts in x-xi appear to be charac- teristic in the family and need no mention. Defmition. Bithecal, pores at or behind 8/9. Biprostatic, pores on xvii. Male pores in or close to xvii. Gizzards, two, in vi-vii. Calciferous glands, lateral, one pair, in ix. Spermathecae adiverticulate. Distribiviion. Dominican Republic (Hispaniola). The genus is known to be represented elsewhere only by a peregrine species formerly thought to be African but now ap- parently of American origin. This peregrine species extends the generic range to include the following : Panama, Costa Rica, Cameroons, Fernando Po, Belgian Congo, Ceylon. Key to species of Nematogenia 1. a. Body small, less than three mm thick, dorsal pores present 2 b. Body larger, more than three mm thick, dorsal pores lacking 3 '1. a. Spermathecal pores on 8/9, male deferent and prostatic ducts of a side unite in parietes to open by a common pore laciium h. Spermathecal pores midway between 8/9 and eq/ix, male pores discrete and behind prostatic pores panamaensis o. a. Typhlosole present 4 b. Typhlosole lacking 5 4. a. Holandric alba b. Metandric montana 5. a. Intestinal origin in xii magna b. Intestinal origin in xiv dominicana External characteristics that might provide a more useful key for preliminary identification are liable to individual varia- tion. Until more is known about such differences, reliance on characters less liable to intraspecific variation is preferable. Deficiencies in our knowledge of species erected in the previous century complicate key construction and obviate any attempt to show phylogenetic relationships. 430 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Nematogenia magna sp. nov. Loiiia Kucilla, Cordillera Central, Dominican Republic, under grass in open pine woods at 8-10,000 feet, June 193S, 0-0-1. Dr. P. J. Darlington Jr. (Loma Rucilla is said to be the highest mountain in the West Indies.) External characteristics. Length, 190 -|- mm {-{- 70 mm?). Diameter, 8 mm. Segments, ca. 400 + (+ 129?). Pigmentation, unrecognizable (alcoholic preservation). Prostomium, prolobous. Secondary furrows; one well marked and postequatorial on each of iii-vii, one less distinct and pre-equatorial on each of v-vi, one pre-equatorial and one postequatorial deep furrow on each of viii-xiii. The posteriormost secondary annulus on each of ix-xi has a tertiary furrow. Nephropores and dorsal pores unrecog- nizable, the latter probably lacking. Setae deeply retracted on preclitellar and clitellar segments, often unrecognizable, apparently small and not closely paired. Shortly behind the clitellum, AB=CD CD, AA appears to be consistently > BC throughout though the difference may be small. In xvii, a setae probably present but h setae lacking. Clitellum red (preservation artefact?), annular, on xiv-xxviii, the posteriormost portion of xiii with a slight red coloration. Sphermathecal pores, one pair, at or slightly median to B, on 8/9. Female pores widety separated, equatorial, at approximate sites of apertures of a follicles, on xiv. Male and prostatic pores discrete, the former diagonally slit-like, the others smaller, circular and anterior, all at or just lateral to B and in a single, transverse, smooth, glistening area with distinct anterior and posterior boundaries about at levels of 16/17 and 17/18. Lateral margins of the area indistinct but in BC and nearer C. The pores of each side are in a slight depression within an area of special tumescence that is most obvious in AB. No other genital markings. Internal anatomy. Septum 6/7 slightly muscular, 7/8-10/11 thickly muscular, 11/12 strengthened but translucent. Gizzards large, in vi and vii. Calciferous glands one pair, in ix, of solid type (no lumen), large, lateromesially flattened, thickly discoidal, slightly notched ventrally. Duct fairly thick, GATES : THE GENUS NEMATOGENIA 431 soft, from pointed dorsal apex of each gland to lateral face of gut shortly in front of 9/10, with rather large aperture into gut lumen. Longitudinal lamellae on inner wall of gut in front of gland apertures gradually disappear anteriorly in ix. In- testinal origin in xii. Typhlosole lacking but at mD on roof of gut for some distance behind Ivi there is a very low ridge, round to triangular in cross section. Supra-intestinal gland, in several segments around civ, an obvious yellowish thickening of gut roof where the tissues have a honey-comb texture. Dorsal blood vessel single, disappearing from view within the pharyngeal glands in iii. Ventral trunk unrecognizable anterior to vii. Subneural lacking or unrecognizable, extra-oesophageals not distinguishable. Supra-oesophageal recognizable only in ix where a large branch on each side passes laterally on dorsal face of calciferous gland stalk and then down into the gland. Commis- sures of vi-ix obviously connect dorsal and ventral trunks. Com- missures, or hearts, of x-xi probably latero-oesophageal but there is no blood in either of the apparent dorsal bifurcations of each vessel. A small transversely placed body resting on the dorsal blood vessel but attached to the anterior face of each septum, from 50/51 posteriorly, resembles the "lymph" glands present in similar locations of certain species of the megascolecine genus, Pheretima Kinberg 1866. Metandric. A longitudinal membrane on each side of the body between 10/11 and 11/12 just lateral to the testes and male funnel looks much like the wall of a testis sac but a roof was not found (testicular coagulum extending well upwards). Sem- inal vesicles, one pair, in xii, large, lobed. Prostates ca. 20 mm long, would reach into xliv at least if straight. A duct is not clearly marked oif but may be about 3 mm long. The male deferent duct comes into contact with lateral face of the prostatic duct in xvii but is continued on slightly, somewhat thickened, before turning mesially to pass into the parietes just behind the prostatic duct. Spermathecae, without diverticula or seminal chambers, ca. 6 mm long, flattened against the parietes back to 9/10 and with a slightly wider portion vertically flattened against anterior face of 9/10 to which the spermatheca is bound by connective tissue. Duct and ampulla are not distinctly demarcated ex- 432 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY ternally. If an ectal portion with thick wall and very small lumen is to be regarded as the duct, it then is much shorter than the ampulla. A terminal ental portion of the spermatheca is distended by a white coagulum. Remarks. A tail portion had been recently broken o&, pre- sumably during collection. The hind end that was in the same tube may well be of the same worm though the jagged edges of the breaks do not exactly fit together. The clitellum is cracked in several places as often happens when worms are pre- served with that organ in maximal tumescence. Except for local macerations of intestinal wall the specimen otherwise is in good condition. A supra-intestinal gland of any kind has been unknown hitherto in any ocnerodrilid. Its anatomy in N. magna needs characterization from microtome rather than freehand sections but now appears to be similar to that of the "grid" at end of the typhlosole in the octochaetine Barogaster Gates 1939. Supra- intestinal glands have not as yet been looked for in many earthworm genera. Hitherto the glands usually have been associ- ated with the posterior end of a median typhlosole. In Nemato- genia, however, the glands are present only when there is no typhlosole. N. magna is distinguishable from all ocnerodrilids, except two species of Kerriona Michaelsen 1924 and one of Paulistus Michaelsen 1925, by its size as well as segment number, and from the species of Kerriona and Paulistus by the presence of two gizzards (instead of none). A size of 260 x 8 mm, in a family where all species hitherto known, except those just mentioned, are 15-60 x 1-3 mm certainly is significant. Recognition of addi- tional differences from other species of Nemafogcnia awaits more adequate characterization of those previously erected. A species, when understood to be an interbreeding population of organisms with more or less individual if not also geographic variation, obviously cannot be defined from a single individual. Nevertheless, some sort of a concise characterization is needed. In such a precis it is also convenient to list characters, presum- ably of taxonomic importance, about which information is de- sired. GATES : THE GENUS NEMATOGENIA 433 Precis. Spermathecal pores minute, superficial, slightly median to B, at 8/9. Female pores at sites of missing a setae. Male pores at or just lateral to B and shortly behind prostatic pores, both pores of a side in an area of special tumescence within a median field reaching laterally into BC. Clitelium, annular, on xiv-xxviii. Setae, not closely paired. (Nephropores? No dorsal pores.) Prostomium prolobous. (Pigmentation?) Segments, ca. 529. Size, 260 x 8 mm. Intestinal origin in xii. (No typhlosole.) Supra-intestinal gland in region around 155th segment. Lymph glands present posteriorly. Metandric. Spermathecal duct shorter than am- pulla. Nematogenia dominicana sp. nov. San Cristobal, Santo Domingo. From earth of damp forested hillside, July 19, 1930, 1-0-0. Damp soil in hills to west, July 20, 1930, 0-1-0. R. Bond per Dr. G. E. Pickford. External characteristics. Length, ca. 195 mm. Diameter, 5+ mm. Segments, ca. 675 (No. 2). Pigmentation lacking ("gray- ish pink or pink, paler posteriorly," collector's note). Pro- stomium, broadly prolobous (No. 1) or indenting i (No. 2) and then slightly proepilobous. Secondary furrows; one deep and postsetal on each of vii-xiii, one deep and presetal on each of viii-xiii, less obvious pre- and postsetal secondary furrows recog- nizable back to xl. Each pre- and postsetal secondary annulus of ix-xi has a well marked tertiary furrow\ Nephropores un- recognizable. Dorsal pores lacking. Setae first certainly recognizable on ii or iii, very closely paired, AB ca.=C'Z>, BC muchY2^, in xxv, BC : AA : :5 :7 gradually changing to 1 :2 posteriorly, and thence on constant. Spermathecal pores, one pair, at B, on 8/9. Female pores un- recognizable but possibly equatorial and in AB (a present) on xiv. Male and prostatic pores discrete, the former minute, circu- lar, the others transversely slit-like and anterior, all about at B on xvii. Porophores transversely elliptical, flat but markedly protuberant, with center about at B, reaching into AA and to 17/18 but not to 16/17. A distinctly demarcated, smooth, small. 434 BULLETIN : MUSEUM OF COMPAKATIVE ZOOLOGY circular portion bearing the prostatic pore appears to be slightly retractile and the immediate margin of the pore is raised in a teat-like protuberance. The male pore is near the posterior margin of the porophore. No other genital markings. Internal anatomy. Septum 5/6 membranous, 6/7 slightly mus- cular, 7/8-11/12 thickly muscular and funnel-shaped. Gizzards well developed, in vi-vii. Calciferous glands one pair, in ix, of solid type (no lumen), rather heart-shaped, with slight ventral indentation (not disc-shaped). Intestinal origin in xiv (2). Typhlosole lacking and no slightest rudiment recog- nizable from xiv-xl but from xli to Ixvi there is a low, band-like area the wrinkled surface of which contrasts strikingly with the smoothness of other portions of the gut wall. The band gradually disappears but at niD a fine ridging still is recognizable and gradually passes into the supra-intestinal gland, a wide thicken- ing of gut roof also ridged on veuti-al side. The gland extends from 180/181 to 194/195 (No. 1, ca. 365 segments) or from 187/188 to 208/209 (No. 2). The gland is slightly constricted at septal insertions (the juvenile) and has an appearance of being marked off into segmental glands. Dorsal blood vessel single, disappearing from view in tissue masses of the pharyngeal bulb that reach to 5/6, in intestinal region receiving in each segment two pairs of large vessels from the gut, one pair near each septum. Supra-oesophageal trunk recognizable only in ix-xi, apparently ending with hearts of xi. Extra-oesophageals filled with blood and recognizable only in viii-ix where they are median to the hearts and free from the gut. A large vessel passes out from the supra-oesopliageal in ix, on each side, and shortly bifurcates, one branch passing along calciferous duct to and into the gland, the other branch passing down to join the extra-oesophageal. Subneural quite unrecognizable (probably lacking?). Segmental commissures connecting dorsal and ventral trunks quite slender in v-vii, some- what larger in viii-ix (lateral hearts?). Hearts of x-xi appar ently latero-oesophageal but blood present only in the bifurca- tions that pass to the supra-oesophageal. Lymph glands present and apparently as in magna. GATES: THE GENUS NEMATOGENIA 435 Metandrie. Seminal vesicles, one pair, in xii. Prostates 18-20 mm long, would reach at least into xliv if straight, pink (for- malin developed artefact?). An ectal white portion, slightly slenderer, 1-2 mm long, with marked muscular sheen constitutes the duct. The male deferent duct is thicker behind 16/17, in contact with lateral face of prostatic duct but continued on as in other species and is narrowed again within the parietes. Spermathecae, 4+ nim long, without diverticula or seminal chambers, but with distinctly demarcated duct and ampulla. The latter, ca. 1 mm long, elliptical in cross section, ca. twice as thick as ental portion of duct, filled with a white coagulum. Duct, circular in cross section, very shortly and closely looped eetally. Reproduction. Brilliant iridescence on male funnels and in spermatheeal ducts eetally (but not in coagulum of spermathecal ampullae) shows that sperm had been matured and that copula- tion had been completed, in the larger specimen. Reproduction presumably is sexual and biparental. Regeneration. Segment length, at 395/396, abruptly becomes very much shorter and so continues posteriorly, as in a tail re- generate. Remarks. The aelitellate worm had been ruptured in the in- testinal region. No signs of postsexual regression were recog- nized in that worm which may have been preserved prior to development of the clitellum. In that case copulation takes place, as has been suggested for other species, before the clitel- lum develops. Absence of a tj^phlosole and presence of a supra-intestinal gland indicate relationship to magna from which dominicana is distinguished as follows : By constant close pairing of setae throughout the Ijody as well as other setal characters. Presence of paired male porophores. Intestinal origin in xiv (rather than xii), an important distinction in a family where the origin so very frequentl}' is in xii. Greater length of spermathecal duct relative to that of the ampulla. Further differences may be recognizable when clitellate specimens become available. Precis. Spermathecal pores minute, superficial, at B, on 8/9. Female pores in ABl Male pores about at B, shortly behind prostatic pores, in paired, transversely elliptical porophores. 436 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY (Clitellum?) Setae very closely paired, BCy2C. (Nephropores? No dorsal pores.) Prostomium prolobous. Un- pigmented (?). Segments, 675. Size, 195 x 5 mm. Intestinal origin in xiv. (No typhlosole.) Supra-intestinal gland in ten or more segments behind 180/181. Lymph glands present posteriorly. Metandric. Spermatheeal duct longer than ampulla, ectally with short loops. NeMATOGENIA MONTANA Sp. nOV. Loma Rucilla, Cordillera Central, Dominican Republic, under grass in open pine woods at 8-10,000 feet, June 1938, 0-0-2. Dr. P. J. Darlington Jr. External characteristics. Length, 85 and 108 mm. Diameter, 4 mm (clitellum). Segments, 178 and 144 (±1, a metameric anomaly). Pigmentation, unrecognizable (alcoholic preserva- tion). Prostomium, prolobous (2 specimens). Secondary fur- rows, one presetal and one postsetal each on vii-ix. Nephropores and dorsal pores unrecognizable, the latter probably lacking. Setae closelj^ paired anterior to clitellum, more widely paired behind clitellum, farther posteriorly AB ca.=or < CD Z>=or. a. Calciferous sacs paired, lateral 4 h. Califerous sacs unpaired, median and ven- tral 8 * Malabarin, Thutonin. Deccania arc known only from India and Burma Maheina onlv from the Seychelles (to which it may have Ix-en transported recently), ciirgimm only from India (to which it may have l)een introduced recently). Gordiodriliiii (as of Michaelsen & Stephenson) is African. Other genera are Arnerican or American and African. 444 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY 4. a. Calcif erous sacs in ix and x 5 b. Calcif erous sacs in ix only 6 5. a. One gizzard, in vi Maheina. b. No gizzard Paulistus, Quechuona. (i. a. One or more gizzards 7 b. No gizzards Oonerodrilus, including Liodrilus, Ka-plodrilus, Ilyogenia, as well as Pygmaeodrilus, Eu- Icerria (part), Bia- phorodrilus, Kerriona. 7. a. One gizzard, in vii Euherria (as it must be restricted). b. Two gizzards, in vi-vii Nematogenia. 8. n. Calcif erous sacs in ix-x, (gizzard in vii) . . . Curgiona. b. Calcif erous sacs in ix 9 9. a. One or more gizzards 10 b. No gizzard Gordiodrilus (part), as it must be restricted. 10. a. One gizzard, in viii Gordiodrilus, a "robustus" group of species. b. Two gizzards, in vii-viii Gordiodrilus, a ' ' nannodrihis ' ' group of species. DISCUSSION A taxonomy based primarily on genital structure culminated, in the Oenerodrilidae, in the congeries known as Gordiodrilus (cf. Gates, 1942b, pp. 75-85). That taxonomie waste-basket which defeated Steplienson, "I can do nothing with this hetero- geneous group" (1930, p. 863), probably could have been avoided (CTates, 1942b, p. 76) if somatic structure had been taken into consideration, even as little as previously. Collec- tions that would justify a real attempt at revision of Gordio- drilus, unfortunately, are no more available now than in 1942. Errors of typography, observation and interpretation, so frequent in the literature on Gordiodrilus, may well be much less common in the literature on other ocnerodrilid genera. Neverthe- less, more adequate characterization of somatic systems still is GATES: THE GENUS NEMATOGENIA 445 prerequisite to a revision of any portion of the family. Also needed is information as to individual and geographical varia- tion about which next to nothing is now known. Some of the taxonomic changes that will be made in th(? future can be deduced from a key constructed entirely on characters supplied by just one somatic system. That is the digestive system. The organs involved are gizzards and calci- ferous glands, with the glands having the greater importance. Genera that key out together are presently distinguishable from each other only by means of characters furnished by the system usually most liable to rapid evolutionary change. The genital characters by which those genera are defined and dis- tinguished, elsewhere in the family and/or in other families, do not, and sometimes cannot, have that taxonomic value. Relationships suggested by somatic systems are of course often different from those that were deduced in the past from reproductive structure. Thus DiapJiorodrilus now seems to be closer to Ocnerodrilus than to Gordiodrilus with which it w^as mistakenly united. Kerriona may be closer to Ocnerodrilus than to its supposed ancestor Kerria. REFEEENCES Gates, G. E. 1942a. Check list and bibliography of North American earthworms. Am. Midland Nat., 27: 86-108. 1942b. Notes on various peregrine earthworms. Bull. Mus. Comp. Zool. Harvard, 89: 63-144. 1954. Exotic earthworms of the United States. Bull. Mus. Comp. Zool. Harvard, 111: 219-258. 1956a. Reproductive organ polymorphism in earthworms of the oriental megaseoleoine genus Phcretima Kinberg 1867. Evolution, 10: 213-227. 1956b. Notes on American eavthworms of the family Lumbricidae. III-VIT. Bull. Mus. Comp. Zool. Harvard, 115: 1-46. Stephenson, J. 1930. The Oligochaeta. Oxford. Bulletin of the Museum of Comparative Zoology AT HARVARD COLLEGE Vol. 117, No. 5 TRIASSIC REPTILE FOOTPRINT FAUNULES FROM MILFORD, NEW JERSEY By Donald Baird With Four Plates CAMBRIDGE, MASS., U.S.A. PRINTED FOR THE MUSEUM November, 1957 Publications Issued by or in Connection WITH THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE Bulletin (octavo) 1863 — The current volume is Vol. 117. Breviora (octavo) 1952 — No. 81 is current. Memoirs (quarto) 1864-1938 — Publication was terminated with Vol. 55. Johnsonia (quarto) 1941 — ^A publication of the Department of Molliisks. Vol. 3, no. 35 is current. Occasional Papers of the Department of Mollusks (octavo) 1945 — Vol. 2, no. 21 is current. Proceedings of the New England Zoological Club (octavo) 1899-1948 — Published in connection with the Museum. Publication terminated with Vol. 24. The continuing publications are issued at irregular intervals in num- bers which may be purchased separately. Prices and lists may be obtained on application to the Director of the Museum of Comparative Zoology, Cambridge 38, Massachusetts. Of the Peters ' ' Check List of Birds of the World, ' ' volumes 1-3 are out of print ; volumes 4 and 6 may be obtained from the Harvard Uni- versity Press ; volumes 5 and 7 are sold by the Museum, and future volumes will be published under Museum auspices. Bulletin of the Museum of Comparative Zoology AT HARVARD COLLEGE Vol. 117, No. 5 TRIASSIC REPTILE FOOTPRINT FAUNULES FROM MILFORD, NEW JERSEY By Donald Baird With Four Plates CAMBRIDGE, MASS., U.S.A. printed for the museum November, 1957 No. 5 — Triassic Reptile Footprint Faunules from Milford, New Jersey By Donald Baied CONTENTS Page FOREWORD 449 GEOLOGICAL OCCURRENCE 451 SYSTEMATIC DESCRIPTIONS 452 THEROPODA Grallator sulcatus Baird, n. sp 453 Anchisauripus milfordetisis (Bock) 462 Anchisauriprts parallelus (Hitchcock) 464 Genus incertum 465 Taxonomy of theropod footprints 467 PSEUDOSUCHIA Chirotherium parvum (G. H. Hitchcock) 473 Chirotherium ey.ermani Baird, n. sp. 479 PHYTOSAURIA Apatopus lineatus (Bock), n. gen. 486 LACERTOIDEA INCERTAE SEDIS Rhynchosauroides hyperiates Baird, n. sp 494 Rhynchosauroides sp. 497 CORRELATIVE AGE OF THE MILFORD FAUNULES 501 REPTILES OF THE BRUNSWICK FORMATION 503 APPENDIX I : The nature of Otouphepm 507 APPENDIX II : Brontozoum 509 SUMMARY 511 REFERENCES 514 FOREWORD In a recent paper Wilhelm Bock (1952) has described a num- ber of important faunules of fossil footprints from various horizons and localities in the Upper Triassic (Middle Keuper) of the Newark Basin. One of these faunules, found in a gray sandstone layer of the upper Brunswick formation near Milford, New Jersey, had been discussed in the literature as early as 1886 ; but until 1952 only its dominant species, Chirotherium parvum, had been adequately described. In the same paper 450 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Bock described a new species of pseudosuchian footprints, Chiro- therium lulli, found in a reddish-brown siltstone bed at the same locality. More recently a slab bearing the counterpart trackway of C. lulli and two forms of dinosaur tracks has come to light and, with Bock's generous permission, has been described (Baird, 1954). Most of the Milford footprints discussed by Bock had been collected by John Eyerman in 1885 and deposited in the geologi- cal museum of Lafayette College at Easton, Pennsylvania (here- after abbreviated LC). One slab, privately owned, had been acquired by Bock for the Academy of Natural Sciences of Phila- delphia (ANS). "While Bock's paper was in press, Eyerman 's second and much larger collection from the same quarries came to light in the Museum of Comparative Zoology (MCZ), and received independent preliminary study. This material, though described in part by Eyerman in 1889, had been overlooked by subsequent authors. A search of the American Museum of Natu- ral History collections (AMNH) has produced eight additional footprint-bearing sial)s collected at Milford by Eyerman in 1886. Besides increasing the Chirotherium parvum faunule by sev- eral species and permitting morphological re-interpretation of those already described, the additional material includes a new species of dinosaur footprint which represents a third Milford faunal horizon. As this material also explains certain 19th- century misidentifications which have distorted the faunal pic- ture, and the geological correlations which have been based on it, I have undertaken a comprehensive re-examination of the Milford reptile footprints and their zoological and stratigraphic significance. I am particularly indebted to Wilhelm Bock for photographs and for permission to mold the type of Chirotherium copei, to J. L. Dyson for the loan of the Lafayette College types, and to E. H. Colbert for the loan of American Museum specimens and for generously making available his unpublished data on Coelo- physis. Special thanks for the privilege of studying the collec- tions under their care are due to G. W. Bain, A. E. Wood, and R. E. Schortmann at Amherst, R. G. Chaffee at Dartmouth, E. H. Colbert and Mrs. R. H. Nichols at the American Museum, D. H. Dunkle at the U. S. National Museum, J. T. Gregory at Yale, J. L. Kay at the Carnegie Museum, and C. W. Waldron, TRIASSIC REPTILE FOOTPRINTS 451 Jr. at the Museum of Science, Boston. For their helpful and stimulating susrgfestions 1 am pleasantly indebted to C. L. Camp, E. H. Colbert, the late R. S. Lull, the late J. H. McGregor, F. E. Peabody, and D. M. S. Watson. The continued support and encouragement of Dr. Alfred S. Romer, as well as his critical reading of the manuscript, are most gratefully acknowl- edged. GEOLOGICAL OCCURRENCE Two distinct footprint horizons bearing mutually exclusive faunules occur in the gray sandstones of the Milford quarries. Eyerman nowhere indicates the relative stratigraphic positions of these horizons, but they are probably not very widely sep- arated : both are referred to the upper Brunswick formation, about 5,100 meters above the Triassic base. For clarity they are here arbitrarily designated Levels A and B. Although this can- not be stated as certain, Eyerman 's discussion of the MCZ material (1889, p. 32) implies that the Smith Clark quarry was the source of the Level B or Chirotherium parvum faunule, while another quarry about one-half mile to the east furnished all the material from Level A. At Level A, the footprints were impressed in gray shale and are preserved as natural casts on the under surface of a gray sandstone 4 to 6 cm. thick. The upper surface of this sandstone is irregularly excavated into shallow pockets in which adhere pieces of the overlying gray shale. All the footprints are referred to a single species, Grallator sulcatus. Most are deeply impressed, between 1 and 2 cm. below surface level at their deep- est points; many are obscured by slippage, overlapping, or slumping of the extruded clay. None shows impressions of the plantar scales and very few show fringe-scale striations. Among the tracks on MCZ 216 is a striated drag-mark 25 mm. wide by 2 mm. deep, tlanked on either side by intermittent shallow scratches. This might be interpreted as a tail-trace but is more probably the drag-mark of a Triassic conifer or horsetail rush. Very similar marks in the Chirotherium sandstone of Thiiringeu have been described as drag-marks of Equisetites by Riihle von Lilienstern (1939, p. 370, pi. 12). Narrow shrinkage cracks, rarely more than 5 mm. wide, characterize Level A. 452 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Level B, which carries the larger faunule, somewhat resembles A in lithology. The sandstone overlying the recording surface, however, quarries out in layered flagstones 1.5 to 3.5 em. thick; the flat upper surfaces typically bear shallow ripples and run-off marks. Shrinkage cracks of variable width are common. Rain- drops have pitted both ground surface and footprints, obscur- ing the plantar surfaces; but the sides of several deeper tracks show striated claw-marks and fringe-scale furrows. Certain parallel series of sharp, shallow scratches may be of arthropod origin, but no trackway sequences are preserved. Annelids have perforated the layers after deposition ; their sand-filled burrows loop down into the shale or course along the sandstone-shale interface, looking superficially like dasycladacean algae. To the Milford flora recorded by Bock (1952, 1952 A) may be added the genus Neocalaniites (MCZ 211). Several slabs bear branch- ing or reticular patches, presumably algal. The third Milford footprint horizon, discussed in my previous paper (1954), lies in a reddish-brown siltstoue which crops out in the Smith Clark quarry some 15 meters above Level B. Rep- tile footprints of pseudosuchian appearance occur at another horizon in that quarry but are unfortunately too vague and fragmentary for adequate characterization (see Bock, 1952, pi 49, fig. 8). The stratigraphic correlation of these beds with those of southeastern Pennsylvania, the Connecticut Valley, and central Europe will be discussed after an analysis of the fauna. SYSTEMATIC DESCRIPTIONS Nineteenth-century publications on the Milford footprints are so vague, confused, and contradictory that little faith can be placed in them. The foundations of Triassic ichnology had been so monumentally laid by Edward Hitchcock that footprints from any part of the Newark series were customarily identified by reference to Hitchcock's figures of species from the Connecticut River Valley. This pi'actice led inevitably to confusion, for the classic Connecticut Valley faunas actually have little in com- mon with those of Pennsylvania and New Jersey. As the nomen- clature of Triassic footprints evolved, faunal revisers changed the names of Milford species without checking the determina- TRIASSIC REPTILE FOOTPRINTS 453 tions. For brevity, the terminologies of earlier authors and of this paper are summarized in Table 1. Before Bock's restudy, only C. H. Hitchcock (1889) had erected new species for Milford footprints. His descriptions, unaccompanied by figures, are so vague that of the species mentioned only one, Chirotherium ["Otozoum"] parvum, can be identified positively. His new species "Chiniaerichnus in- gens," characterized only as "two relief tracks of considerable size and representing only one-half of the foot," is a nomen nudum. Any nomenclatural questions arising from C. H. Hitch- cock 's work have, fortunately, been settled by Bock 's designation of types. Order SAURISCHIA Suborder THEROPODA Infraorder COELUROSAUBIA Form-family GRALLATORIDAE Lull, 1904 (==ANCHISAURIPODIDAE Lull, 1904) Genus CtRALLATOR E. Hitchcock, 1858 Type species G. cursorius E. Hitchcock (1858, p. 72) designated by Lull (1904, p. 494). Grallator sulcatus Baird, n. sp. Figure 1 ; Plate 1, figure 1. Type. MCZ 215, a left pes imprint. Hypodigm. MCZ 215-228 inclusive, slabs bearing 37 pes im- prints; AMNH 1982 and 1983, 6 imprints; LC S487, parts of 3 overlapping imprints. All collected by John Eyerman, 1885- 1887. Horizon. Upper Triassic (Middle Keuper), Newark series, upper Brunswick formation. Level A. Locality {-fide Eyerman, 1889, p. 32). Quarry of the Messrs. Clark about one-half mile east of the Smith Clark quarry, near Milford, Hunterdon County, New Jersey. 454 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY TABLE 1 : ATTEMPTED SYNONYMY OF MILFORD FOOTPRINTS Dashes indicate derivation from a previous author. EYERMAN, 1886 C. H. HITCHCOCK, 1889 EYERMAN, 1889 LEVEL A : aff. Anomoepus major Brontozoum isodaetylum Grallator cuneatus [ ?] Grallator parallelus [?] Grallator cuneatus [ ?] Grallator cuneatus [?] Grallator gracilis LEVEL B : aff. Anomoepus major Otozoum parvum, n. sp. [ ?] Grallator tenuis Anomoepus minor [?] Chimaeriehnus ingens, n. sp. [?] Polemarchus gigas Argozoum dispari-digitatum Diagnosis. A moderately large Grallator, differing from other species in the anterior position and relative shortness of digit III and in having the bases of II and III closely united, while IV is separated from III by a pronounced sulcus which extends back to the metatarso-phalangeal pad of III. Ovoid metatarso- phalangeal pads of II and IV regularly and equally impressing, that of II more anterior. Second phalangeal pad of II opposite first pad of III. Trackway characters unknown. Measurements. Defects of impression and preservation make impossible a valid and consistent suite of measurements. Dimen- sions of the best-preserved specimens, however, suffice for de- scriptive if not for statistical treatment. Measurements of the type, adjusted for slippage and a transverse crack-filling, are as follows : TRIASSIC REPTILE FOOTPRINTS 455 LULL, 1915 BOCK, 1952 BAIRD, 1957 Anchisauripus parallelus Grallator cuneatus — — Grallator ' Grallator sulcatus, n.sp. Grallator gracilis ♦Chirotherium parvum Sauropus barrattii Sauropus ingens Polemarchus polemarchius Argoides macrodactj'lus Gigandipus ? (Anchisauri- pus) milfordensis, n.sp. Grallator gracilis Chirotherium parvum Chirotherium copei, n.sp. Otozoum (?) lineatus, n.sp. Sauropus barrattii . Anchisauripus milfordensis Anchisauripus parallelus Genus incertum Chirotherium parvum Chirotherium eyermani, n.sp. > Apatopus lineatus, n.gen. (Unidentifiable) Ehynchosauroides hyperbates, n.sp. Pes, digit no. H HI IV Length, mm. 54 72 59 Length including metatarso-phalangeal pad 78 98 84 Length of pes, 100 mm.; including metatarso-phalangeal pads, 123 mm. ; width, 60 mm. Measurements of digit divarication will vary greatly in a single footprint depending on one 's reference points. In the type specimen the angles between axes drawn through the centers of claw base and metatarso-phalangeal pad are : II-7°-III-19°-IV. Grallator sulcatus footprints, however, show quite a range of divarication values. In my experience linear and angular meas- urements such as these have a very limited value for the diag- nosis and differentiation of theropod footprints. 456 BULLETIN : MUSEUM OF COMPAKATIVE ZOOLOGY Morphology. The abundance of deeply-impressed footprints of this species affords an unusually clear picture of its foot structure, including the relationship of the distal ends of the metatarsals which is rarely recorded in Grallator trackways. As in other theropod dinosaurs and in most cursorial birds the Fig. 1. Grallator sulcatus n. sp., x %. Composite outline based on the four best-preserved specimens. Section A- A' across the inverted natural cast shows the close association of digits II and III and the sulcus separating digit IV. Proximal phalanges of the restored skeleton appear foreshortened. gait was digitigrade with most of the body weight falling on the joints between the first and second phalanges of digits II, III, and IV. Nevertheless, the pads which underlay the joints be- tween metatarsals and phalanges of digits II and IV impressed TRIASSIC REPTILE FOOTPRINTS 457 regularly, though less deeply than the digital pads, even in shallow imprints. Apparently this Brunswick species had not achieved the advanced degree of digitigrady shown by later in- habitants of the Connecticut Valley. Unfortunately the lack of trackway measurements precludes any useful comparison be- tween gait patterns of Brunswick and Connecticut Valley Grallatoridae. The claws of Grallator sulcatus are acuminate in some indi- viduals but blunted by wear in others. Regardless of depth of impression the claws were always extended, rather than flexed as in the other Milf ord theropods ; in particularly deep tracks the exaggerated depth of claw-marks is due not to flexure but to pivotal down-turning as the posterior part of the foot was raised. The slightly domed phalangeal pads coalesce to form a sole callus w^hich appears to have been slightly wider than the toe proper, as in many cursorial birds. The three metatarso-phalan- geal pads, though they impressed less deeply, form an integral part of the sole and are separated from the phalangeal pads only by shallow sulci. The close union of digits II and III and the broad separation of IV differentiate Grallator sulcatus strikingly from its congeners, in which the digits are either laterally equidistant or III and IV are more closely appressed than II and III. Variation. Footprints from Level A reveal a sobering degree of variability both in apparent form and in manner of impres- sion. An increased divergence of the lateral toes is less common than an exaggerated out-turning of the claws alone. Whether this out-turning reflects some lateral flexibility at the ungual joint, or merely individual variation in the plane of a hori- zontally immobile joint, cannot be determined without trackways showing successive imprints of the same foot. Distorted deep impressions reveal that the lateral toes were spread apart as the foot sank into the mud and were contracted again in birdlike fashion as the foot was withdrawn. Many of the footprints are so deformed by accidents of im- pression that they give a decidedly erroneous picture of the foot structure. Such anomalies serve to emphasize the dangers involved in any attempt to characterize and interpret footprint species without adequate quantities of well-preserved material. 458 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY A footprint is not an organism but the by-product of dynamic contact between an organism and its environment. The all-too- common typological analysis of isolated examples cannot be expected to yield zoologically significant information. Osteology. Heilmann (1927, p. 179 fP.) and Peabody (1948, p. 399 ff.) have independently concluded that in the Grallator- idae the phalangeal pads must have underlain the joints. This is the situation in many cursorial birds, in which only the ungual joint may coincide with a transverse crease. The osteological restoration of Grallator sulcatus in Figure 1 conforms to this principle but differs enough from the examples cited by Heil- mann and Peabody to require explanation. In my opinion the pes of Procompsognathus which Heilmann (1927, fig. 130 K) has superimposed on the footprint of Grallator tenuis would fit a great deal better if the foot were advanced until the ungual joints of digits II and III corresponded with the distal pads in- stead of their creases. All the joints would then line up neatly opposite pads, with both ends of the fourth phalanx of digit IV included in the distal pad. Except for the over-long claws of Procoynpsognathus, foot and footprint correspond well in struc- tural pattern. The otherwise convincing skeletal restoration of Grallator cur- sorius in Peabody 's figure 37 B shows a crease coinciding with the joint between phalanges 3 and 4 of digit IV. Here again I would prefer to enclose both ends of phalanx 4 in the distal pad: and this is readily done if we use Hitchcock's original figure of the footprint (1858, pi. 13, fig. 3), in which the distal pad appears longer and more square-shouldered like that of G. tenuis. Other Triassic theropod footprints such as Anchisauri- pus and Eubrontes definitely have a long distal pad on digit IV, and an occasional median constriction of this pad suggests that two joints are actually represented. When the bones of any of these footprint types are restored on the plan advocated here, the phalangeal pattern is precisely that of Triassic and later theropods : the phalanges of digit IV are short, stout-shafted, and successively decreasing slightly in size. No trace of a hallux is visible in any Grallator sulcatus foot- print even where the metatarso-phalangeal pad of digit II is impressed as deeply as 14 mm. Obviously the hallux, while pre- sumably present, was vestigial and completely non-functional. TBIASSIC REPTILE FOOTPRINTS 459 The lengths of the proximal phalanges, usually indeterminate in more digitigrade coelurosaur footprints, can be estimated with fair accuracy from the positions of the regularly impressed metatarso-phalangeal pads. Pads II and IV are impressed to equal depths, differing only as one side of the footprint is deeper than the other. This situation contrasts sharply with that in other species of Grallaior and in Anchisanripns and some species of Euhrontes, where the metatarso-phalangeal pad of digit II impresses less deeply or not at all. From this anomaly we may obtain some clue to the probable structure of the metatarsus. If a Newarkian coelurosaur such as Podokesaurus is restored in a walking pose based on Grallator trackways (cf. Von Engeln and Caster, 1952, fig. 277), the metatarsus slants forward and downward at an angle of some 115° to the ground plane as the foot is implanted. As the metatarso-phalangeal pads were evidently developed to cushion the sole against the thrust trans- mitted down the metatarsus, and as they served this function only during implantation and the beginning of the propulsion phase, it follows that the position of their imprints is (broadly speaking) that of projections of the cross-sectional areas of their respective metatarsals when in the implantation position. If this be the case, as comparison with cursorial birds suggests, then the type of transverse arching of the metatarsal bundle at its distal end determines whether pad II or pad IV, or neither, makes the more posterior impression. Similarly, the relative length of the lateral metatarsals plus the thickness of their underlying pads determines the relative depth of the pad im- pressions. Of course the relative thickness of the pads is un- knowable, but where the imprint of pad II is shallow or absent (as in most Newarkian theropod footprints) metatarsal II was probably shorter than IV. Conversely, the dinosaur represented by Grallator sulcatus appears to have had lateral metatarsals of approximately equal length. Among Upper Triassic coelurosaurs metatarsal II is decidedly shorter than IV in Procompsognathus , but the lateral metatarsals are equal or subequal in Ammosaurus, Halticosaurus and Coelo- physis. For comparison with Grallator sulcatus let us restore from their pedal skeletons the footprints that would have been made by two of these coelurosaurs. 460 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Halticosaurus liliensterni, a massive species from the KnoUen- mergel of Thiiringen, is described by Huene (1934) as having lengths of 20.5, 22, and 20 cm. for metatarsals II, III, and IV; digit I is unknown. To show the relative positions of the digits in walking pose we must make a three-dimensional reconstruc- tion of the pes; when this is projected onto a horizontal plane •- i Fig. 2. Eeconstructed right pedes of Keuper coelurosaurs, with restored footprints. A. Ealtioosaurus, x %,. B. Prccompsognathus, x 1. Data from Huene. (analogous to a footprint-recording surface) the metatarsals and proximal phalanges are properly foreshortened (Figure 2 A). Structures in the metatarso-phalangeal region indicate a foot with a moderately compact base and strongly divergent digits; the bulk of the animal and the proportions of the meta- tarsus suggest that the posterior pads impressed regularly and to nearly equal depths. The positions of the plantar pads are deduced from the skeleton and a plausible reconstruction of their TRIASSIC REPTILE FOOTPRINTS 461 original form can be derived from Connecticut Valley footprints of equivalent size. The footprint thus restored is certainly neither a Grallator nor a Gigandipus, and apparently not an Anchisauripm sensu stricto. In Eiihrontes giganteus and Anchisauripus minusculus we find the closest similarities in size, relative lengths of digits and phalanges, and regular and subequal imprints of the meta- tarso-phalangeal pads. Without venturing any positive correla- tion we can say that these footprints may well have been made by coelurosaurs of the Halticosmirus type. Procompsognathus friassicus from the Stubensandstein of Wiirttemberg is, in contrast to the hulking Halticosaurus, a graceful and diminutive species. When reconstructed (from Huene, 1921) and projected as before (Figure 2 B), its foot ap- pears strikingly slender and laterally compressed. Neither the hallux nor metatarso-phalangeal pad II normally made contact with the ground. The restored footprint is obviously that of a Grallator closely comparable to a small G. cursorius but having a relatively shorter central digit. If the divarication were larger and the lateral digits somewhat shorter, correlation with a small G. tenuis would be indicated. A comparison of Figures 1 and 2 shows clearly that, even dis- regarding size, Grallator sulcatus has much stronger resem- blances to Procompsognathus than to Halticosaurus. So far as osteology can be inferred from the footprint, only size, minor differences in digital proportion, and the shortness of metatarsal II distinguish Procompsognathus from the Milford trackmaker. Unpublished data on skeletons of Coelophysis from the Upper Triassic of New Mexico, made available for this study through the courtesy of E. H. Colbert, show that the foot of that genus is essentially similar to that of Procompsognathus except that the lateral metatarsals are equal in length. Coelophysis thus closely approximates in structure the reconstructed skeleton of Grallator sulcatus. While no positive correlation can be made, the Level A footprints attest the presence of a Coelophy sis-like coelurosaur in the Newark basin of deposition. 462 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Form-Family GRALLATORIDAE Lull, 1904 Genus AnCHISAIIRIPUS Lull, 1904 Type species A. sillimani (E. Hitchcock), 1843. Anchisauripus milfordensis (Bock) Figure 3 A Gigandipus ? (AncMsaiiripiis) milfordensis Bock, 1952, p. 403 ; pi. 43, fig. 3 (in legend, for "No. 15210" read S488). Type. LC S488, a left pes imprint slightly distorted by the paratype pes of Chirotherium parvum. Hypodigm. The type and two other imprints on the same slab; MCZ 135 and 229; AMNH 1981 and 1984. These seven imprints collected by John Eyerman, 1885-1887. Horizon. Upper Triassic (Middle Keuper), Newark series, up- per Brunswick formation about 5100 meters above the Triassic base. Level B. Locality. Smith Clark quarry near Milford, Hunterdon County, New Jersey, 1.25 km. north of Delaware River bridge. Tentatively assigned. Certain tracks from the Gettysburg shale (a Brunswick equivalent) near Goldsboro, York County, Pennsyl- vania (Wanner, 1889, pis. 6, 7, 10). This material was acces- sioned by the U. S. National Museum in 1888 but is now lost, so that precise comparisons are impossible. Diagnosis. A small, short-footed, broad-based species. The base of claw II lies opposite the crease between the first and second phalangeal pads of digit III, and the tip of claw IV opposite the distal part of the second pad. Metatarso-phalangeal pads II and IV circular and strongly domed, IV nearly twice the size of II. Claws habitually flexed, hastate in outline with basal recesses. Morphology. Collation of the new hypodigm permits some am- plification and emendation of Bock's description. Removal of some of the obscuring underclay reveals that the structure which Bock interpreted as a semi-rotated hallux is actually a filled shrinkage-crack which runs from the margin of the slab through metatarso-phalangeal pad II of the type, narrows abruptly, and passes under the label and into the manus of Chirotherium parvum. Nowhere is there evidence of a hallux imprint, even TRIASSIC REPTILE FOOTPRINTS 463 in MCZ 135 where the first phalanoeal pad of digit II impressed to a depth of 11 mm. This fact and the osteological differences listed in Table 2 preclude assignment of the species to Gigandi- pus. Individual variation is apparent in even this small sample. MCZ 135 in particular differs from the others in its relatively shorter digit IV, its large size (103 mm. long vs. 93 mm. in Cm. I— 5 H Fig. 3. A. Anchisauripus milfordensis (Bock), x 3/4. Composite outline based on all seven Milford specimens, drawn the size of S488 or MCZ 229. Section A-A' shows the domed lateral metatarso-phalangeal pads. B. Anchi- sauripus parallelus (Hitchcock), AjMNH 1789, x %. the smallest individual, the type), and its angles of digit divari- cation (II-13°-III-10°-IV vs. II-15°-III-9°-IV in the type). The outstanding features of the species — the broad-based foot, the broad, rounded phalangeal pads, and the flexed, hastate claws — have been adequately discussed by Bock. Metatarso- phalangeal pad III is ill-defined, but the dome-like lateral pads are impressed to approximately equal depths with reference to 464 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY the rest of the foot. The lack of creases between the phalangeal pads suggests the thick, relatively inflexible plantar padding of a foot which is so specialized for cursorial habits that it retains very little power of grasping. Fringe scales on the median side of digit III are represented by parallel striae, somewhat irregu- larly spaced, which average 10 to the centimeter. They tell little about the squamation but at least prove that the full width of the digit is recorded. Osteology. A close comparison of Figures 1 and 3 A reveals significant differences in the restored skeletal structure. The individual phalanges of Anchisaurijnis milfordensis are rela- tively shorter than those of Grallator sulcatus, and the unguals may have been, like the claws, shorter and blunter. More im- portant is the dift'ereuee in the relative positions of the digits. While in G. sulcatus the joint of ungual IV lies opposite the joint between phalanges 1 and 2 of diigt III, in A. milfordensis it lies opposite the waist of phalanx III-2, and the joint between phalanges I and 2 of digit III is aligned with the waist of phalanx IV-3. Clearly the central digit of A. milfordensis is less advanced with respect to the lateral digits. Additional osteo- logical criteria for comparing this species to other dinosaur foot- prints will be discussed in a subsequent section. Anchisauripus parallelus (E. Hitchcock) Figure 3 B Material. AMNH 1789, a sharp impression of a right pes associated with a footprint of Ajjatopus Uneatus. Collected by John Eyerman, evidently in the Smith Clark quarry. Discussion. In its form and proportions, especially in the slenderness and lack of divergence of digit IV and the relative positions of the articular pads, this specimen is indistinguish- able from footprints of A. parallelus found in the Portland formation of Massachusetts and Connecticut. Its reconstructed skeleton corresponds exactly to that of the type specimen as figured by Lull (1953, fig. 42). The Milford footprint, however, is only 6/11 the size of the Massachusetts type. This striking difference in size is associated with no significant difference in proportions : one specimen is a miniature of the other. Despite TRIASSIC REPTILE FOOTPRINTS 465 our ignorance of trackway characters and range of variation in the Milford form, its form-specific identity is assured. This is the only footprint species common to the Milford and Connecticut Valley faunules. Whether it represents the same reptilian genus in both areas is of course unknown. Although no hallux impression is present on the slab, hallux function within populations (or even individual trackways) of Anchisau7-ipus is too variable to be of significance in an isolated footprint. Deep creases separate the pads of digits II and III and mark the bases of the strongly flexed lateral claws. Evi- dently the foot of this dinosaur was rather flexible — perhaps prehensile — at least more so than those of A. milfordensis and Grallator sulcatus. The fourth digit of A. parallelus is longer but clearly less robust than the second, and (at least in the Milford form) appears to be closely joined to the strong third digit. In this respect A. parallelus differs from A. milfordensis, in which the lateral digits are about equal in robustness and in independence of the third, and differs even more from G. sulcatus, in which the second and third digits are joined and the fourth exceptionally independent. The adaptive significance and evolution of these structural patterns can, unfortunately, only be surmised until stratigraphic sequences have been estab- lished and skeletal material is available. GENUS INCERTUM Figure 4 D A fourth and most peculiar form of dinosaur is represented by a single small, deep imprint of a left pes, MCZ 214. This footprint might be dismissed as an anomaly of impression if a similar specimen had not been found by Wanner (1889) in the Gettysburg shale of York County, Pennsylvania. Wanner 's drawing is reproduced as Figure 4 E ; the loss of his original at the U. S. National Museum precludes further comparison. This animal had a broad, compact foot with the central digit projecting only a toe-width beyond the lateral ones. The im- print is deepest on the lateral side (where it is obscured by worm burrows) and in the free lengths of the second and third digits; the rest of the broad "sole" is two to three millimeters shallower. Little rotational or lateral slippage seems to have 4(36 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY Fig. 4. A-C, three dinosaur footprints drawn to common size and orienta- tion, with phalanges restored. A. So-called Grallator gracilis C. H. Hitch- cock, Dartmouth 5023. B. Type of Anchisauripus hitchcocki Lull, Amherst 56/1. C. Plesiotype of Grallator tenuis E. Hitc:hcock, Amherst 17/4. B and C from Lull. One centimeter scale. D. Tentative reconstruction of genus incertum from Milford, MCZ 214, .X 1. E. Similar footprint from Goldsboro, Pennsylvania, x 1 (from Wanner). Apparent sharpness of claw tips may be caused by shrinkage cracks. TRIASSIC REPTILE FOOTPRINTS 467 occurred: the peculiar proportions of the foot are real. But the material available is insufficient for proper analysis and certainly inadequate for taxonomic treatment. TAXONOMY OF THEROPOD FOOTPRINTS Up to this point we have assigned dinosaur footprints from the Delaware Valley to genera best represented in the Con- necticut Valley, employing a necessarily typological classifica- tion based on similarities and differences in footprint form. This is the traditional approach pioneered by Edward Hitchcock in his "Ornithichnology" of 1836. In Hitchcock's subsequent papers an increasing number of form-taxa were variously grouped and regrouped, with unfortunate wholesale renaming of many units. In this century Lull has achieved a logical taxo- nomic and nomenclatural reorganization of the Connecticut Val- ley footprints, a classification which is crystallized in his recent revision (1953). The three determinable species of theropod footprints from Milford are here assigned to two New England genera which Lull has made the types of two families. Lull's family diagnoses are not, however, mutually exclusive : Grallatoridae : typically small, footprint tridactyl, limbs very long ; with or without tail trace. Anchisauripodidae : bipedal, tetradactyl; hallux when im- pressed rotated to the rear; well marked phalangeal pads; an- terior claws acuminate but not strongly raptorial ; no caudal impression. In current practice GraUator is distinguished from Anchi- sauripus "by greater relative length of stride, smallness of track, and the absence of a hallux impression" (Lull, 1953, p. 153). Within each genus the smaller species are distinguished by size, relative length of digits and of stride, and divarication of the lateral digits. But intermediate forms are common, and isolated or incomplete footprints may be difficult to assign to genus and species even when they are clearly impressed. Let us examine a specific example of close resemblance. A trackway identified by C. H. Hitchcock as his GraUator gracilis (Dartmouth 5023) and the type specimen of Anchisauripus hitchcocki Lull (Amherst 56/1) are morphologically so similar 468 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY that the differences between them fall within the range of in- dividual variation. Lull's plesiotype of Grallator tenuis E. Hitchcock (Amherst 17/4) differs from them chiefly in having digit III slightly farther in advance of the lateral toes (see Figure 4 A-C). All are from the Portland formation of the Turners Falls area, Massachusetts. Their differences in size, gait, and incidence of hallux impression are tabulated below. (Data on the Amherst specimens from Lull.) "G. gracilis" Dartmouth 5023 A. hitchcocki Amherst .56/1 G. tenuis Amherst 17/4 Pes length, mm. 51 119 73 Pace length 299 360 to 500 195 Pes : pace length 1:5.9 av. 1:3.6 1:2.7 (1:3.3 in type) Hallux imprint absent ( ? ) present absent What do these differences mean? Size difference may be genetic or ontogenetic. Biometric analyses of true populations of Connecticut Valley footprints have not yet been made, but useful series of measurements from footprints of varied ages and sources have been compiled by Lull (1953). In general, dimensions in each species tend to cluster about a mean, with "occasional gradational footprints which bridge the dimensional gaps between the species. ' ' This has been interpreted as evidence of several species of reptiles; but as Lull observes, differences caused by sexual dimetry or by selective representation of age groups (perhaps caused by seasonal migration) are difficult if not impossible to evaluate. The presence or absence of a hallux imprint appears to be a valid distinction in the huge, broad-soled dinosaur tracks of the Connecticut Valley : in Euhrontes the hallux is almost never in evidence, while in Gigandipus it usually impresses its entire length. In the Grallator-Anchisauripus group, however, the dis- tinction is less clear. Rarely does more than the tip of the hallux claw impress, and even this may be totally absent, as in a four- imprint trackway of A. sillimani from Turners Falls (MCZ 252. ex Dartmouth 14, cited in Grier, 1927). In current taxonomic practice the absence of a hallux impression does not debar a specimen from Anchisauripus but the presence of one debars it fi'ora Grallator. TRIASSIC REPTILE FOOTPRINTS 469 This distinction is unconvincing, especially when we consider that the hallnx imprint is typically absent in the smaller and present in the larger members of a morphologically homogeneous series. The foot of a large individual bears relatively more weight than that of a small one, as body bulk is proportional to the cube of the linear dimensions: in walking this greater weight on the metatarsus might well cause the pendant hallux claw to impress more fretiuently. Similarly, a juvenile dinosaur might be expected to take a relatively longer stride than its more ponderous parent. For these reasons it seem unrealistic to maintain generic, much less familial, distinctions between dinosaur footprint species solely on the basis of such largely extramorphologic characters as size, incidence of hallux imprint, and length of stride. / subynit that the characters most diagnostic for the classifica- tion of footprints as such, as well as most useful for comparison with skeletal remains, are those which reflect the bony structure of the foot. In most adequately-known varieties of dinosaur footprints the presence of articular swellings and pads permits a reasonably accurate analysis of the skeletal pattern. To clarify the relationships of the Milford species, therefore, let us compare their restored pedal skeletons with those of other Upper Triassic theropod tracks. For this purpose we must add the phalanges to published figures of the various species, extrapolating to find the length of phalanx III-l where the posi- tion of its proximal end is not indicated by a pad. Because of possible errors in the interpretation of joint position in published figures, this analysis should be considered provisional. When the skeletons thus reconstructed are compared, the main differences between them are seen to lie in (1) the relative positions of the metatarso-phalangeal joints;^ (2, 3) the projec- tion of the central digit beyond the lateral ones; and (4) whether digit II or IV projects farther forward. These factors are of course to some extent interrelated. They are more specifically keyed out below, and their distribution among a number of foot- print species is shown in Table 2. 1 This character, the most useful for classification and for skeletal correlations, Is also the most difficult to determine in published drawings and faintly-impressed tracks. Close analysis of original matorial will doubtless necessitate refinements In the subdivision attempted below. 470 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY This method of analysis parallels in part the highly instructive comparisons by means of Cartesian diagrams which Lull has ap- pended to his 1953 monograph. (The new insights provided by these diagrams have not, curiously enough, caused Lull to modify his classification of 1915.) My approach, however, concentrates on differences in presumed skeletal structure. By comparing the species of several genera on a single objective basis, rather than matching the previously assigned species of each genus against the type species, it brings out striking inter-generic similarities which might otherwise be overlooked. DIAGNOSTIC CHARACTERS 1. A. Metatarso-plialangeal joint III about opposite waist of phalanx II-l and joint 1-2 of IV. B. Metatarso-plialangeal joint III about opposite proximal end of II-I and waist of IV-1. C. Metatarso-phalangeal joint III about opposite that of II and joint 1-2 of IV. 2. A. Joint of ungual II about opposite joint 1-2 of III. B. Joint of ungual II about opposite waist of III-2. C. Joint of ungual II about opposite joint 2-3 of III. 3. A. Joint 1-2 of III about opposite joint of ungual IV. B. Joint 1-2 of III about opposite joint 3-4 of IV. (\ Joint 1-2 of III about opposite joint 2-3 of IV. 4. A. Digit II projects farther anterior than IV. B. Digits II and IV project about equally far. C. Digit IV projects farther anterior than II. Let US now attempt a generic regrouping of these species : I. Grallator, type species G. ciirsorius. G. sulcatus clearly belongs to this group despite its atypical forward-set central digit and subequally projecting lateral digits. G. tenuis appears to be intermediate between groups I and II, resembling the latter especially in the relative positions of the metatarso-phalangeal joints. A surprise here is that Otouphepus minor, though super- ficially similar to 0. niagnificus, is otherwise quite unlike it and shows itself to be a Grallator- in size and morphology. This species, based on a single isolated footprint, is of questionable validity.' G. forynosus, included doubtfully by Lull because of its long stride and lack of a hallux imprint, belongs rather to Group III. - TliP nature of Otoiiiihc/iiix is discussed in Appendix 1. TRIASSIC REI'TILE FOOTPRINTS 471 TABLE 2: COMPARISON OF SOME UPPER TRIASSIC THEROPOD FOOTPRINTS 1. 2_ 3. 4. Grallator cursorius A A-B B A O G. cuneatus A *A B A ~ G. gracilis-'' A A B f 2 Otouphepus minor A A B B - G. sulcatus (Figure 1) A A A B G. tenuis (Figure 4 C) B A-B B B O Anehisauripus sillimani B B B A O "A. exsertus" plesiotype, Amherst 54/1 B B B A i2 A. hitchcoeki (Figure 4 B) B A-B B B - Dartmouth 5023 (Figure 4 A) B A-B B B Anehisauripus tuberosus B A-B C C A. exsertus B B C C - A. parallelus (Figure 3 B) B B C C S A. australis (Argentina) B B C C 2 Jeholosauripus ssatoi (Manchuria) B B C B-C S, Grallator formosus B A-B C C - Otouphepus magnificus ?B B B-C B A. minusculus B A-B C B A. milfordensis (Figure 3 A) B A B-C B Eubrontes giganteus B B C B Gigandipus caudatus C C B A II. Anchisauripus (sensu stricto). A. sillimani, the type species, shows closest affinities with A. hitchcoeki, G. tenuis, and the Dartmouth trackway assigned to G. gracilis. Amherst speci- men 54/1, figured by Lull (1904, fig. 7; 1915, fig. 41; 1953, pi. 10, fig. 11 — but not fig. 41 which is mis-captioned Amherst 54/1 but actually shows 16/6, the type) as Anchisauripus exsertus proves instead to be a large A. sillimani. (The same is true of Amherst 34/14 in Lull's pi. 10, fig. 12.) Much original material of these forms must be compared before taxonomic decisions can be reached. ii Lull's figure of the O. gracilis plesiotype appears rather different from C. H. Hitchcock's original tigure of the type (in E. Hitchcock. 1805, pi. 9, fig. 7). Perhaps the latter figure is inaccurate or the plesiotype is misassigned. 472 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY III. Anchisaurepus (sensu lato). Quite a distinct aspect is shown by the larger species of Anchisauripus (earlier assigned to Brontozoum) , in which digit IV is set well forward. Within this group A. tuberosus, A. exsertus and A. parallelus are so similar that they might be considered a single variable species. A. australis Lull (1942) from the Pagazano beds of Argentina is a typical representative of the group. Jeholosauripus ssatoi from the ''Eo-Mesozoic" of Manchuria (Shikama, 1942) clearly belongs here, as do Grallator formosus and Otouphepus mag- nificus. Suppression of the latter genus and species (Baird, 1956; see Appendix I) will remove them from consideration. A. minusculus is, as Lull notes, a sharply defined species ; it seems to be intermediate between the other members of this group and Euhrontes giganteus. A. milfordensis is also some- what transitional but is referred to Group III because of its forward-set central digit and its small size. Strict consistency would require that Group III be dis- tinguished as a separate genus of the Grallatoridae. Unfortu- nately there is no Hitchcockian name available for these forms,^ and rather than make major nomenclatural changes on the basis of this preliminary survey, I will for the present follow Lull's terminology. The dinosaurian footprint associated with Chirotherium lulli at Milford (Baird, 1954, fig. 2 C) and the specimen from York County, Pennsylvania, which Hickok and Willard (1933, fig. 6 B) assigned to Anchisauripus sillimani are so different from the forms in Table 2 (and from the Anomoepodidae as well) in the position of their lateral digits that they cannot be classi- fied in the key used here. They apparently belong to a distinct family, which is represented in Europe by Coelurosaurichnus (sensu stricto: see Baird, 1954, p. 182). This family is appar- ently not represented in the Meriden and Portland formations of the Connecticut Valley. 4 The appropriate ami previously used name Brontozoum is excluded by nomen- clatural technicalities which are discussed iu Appendi.x IT. TRIASSIC REPTILE FOOTPRINTS 473 Order THECODONTIA Suborder PSEUDOSUCHIA Form-family CHIROTHERIIDAE Abel, 1935 Genus ChirOTHEKIUM Kaup, 1835 Brachychirotherian Group Diagnosis. Specialized Upper Triassie chirotheriids lacking a thumb-like fifth phalangeal segment distinct from the metatarso- phalangeal pad. Chirotherium parvum (C. H. Hitchcock), 1889 Figures 5, 6 A ; Plate 1, figure 2. Otozoum parvum C. H. Hitchcock, 1889, pp. 122, 123, 127. Oiozoum parvum. Lesley, J. P., 1889, pp. 571-573, 3 figs. ()tn~oum parvum. Lyman, B. S., 1894, p. 214; 1895, pi. 608, figs. 1-3. '.Oinznum parvum. Lull, E. S., 1904, p. 515. '.rhirnfherium parvum. Lull, E. S., 1915, p. 226, fig. 77. Clieiroiherimn (?) parvum. Lull, E. S., 1917, p. 119. rhirollnnium parvum. Peabody, P. E., 1948, p. 346. nnroilurinm parvum. Bock, W., 1952, pp. 410-414; pi. 41 with plate title 42; pi. 42 with plate title 41; pi. 43, fig. 2 (for Paratype No. S488 read Ilolotype No. S490). Chirotherium copei Bock, 1952, pp. 414-415; pi. 43, fig. 1 (for x 1 . . . No. S491 read x 1/2 . . . ANS 15210). Chirotherium [copeil. Eichards, H. G., 1953, fig. 176. Chirotherium parvum, C. copei. Baird, D., 1954, pp. 174, 175. Type. LC S490, designated by Bock, a large right manus-pes set; the missing posterior half of metatarso-phalangeal pad V is preserved on MCZ 212. Hypodigm. The type, LC S488 and MCZ 211 probably rep- resent two individuals of similar size. ANS 15210 (type of C. copei), MCZ 209, MCZ 210 and AMNH 2257 appear to represent one individual. Horizon, locality and collector as for Anchisauri- pus milfordensis (collector of ANS 15210 is unknown). Diagnosis. Phalanges of pes digit V reduced and included in the metatarso-phalangeal pad ; pes digits IV and V clawless ; narrow, curved claws on pes digits I to III borne high above the thickly padded plantar surface and divergent laterally; meta- tarso-phalangeal pads III and IV coalesced. 474 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY History. The type of this remarkable species has been validly designated by Bock, yet curiously enough this may not be the pri- mary specimen on which Hitchcock's species concept was based. The original description, fragmented between two sections of a notoriously imprecise paper, designated no type and was not illustrated. Outline figures of the species, however, appeared the same year in Lesley's hodgepodge "Dictionary of Fossils" wnth this notation : "Otozoum parvum, n. sp. C. H. Hitchcock, Fig. 1, hind foot track; fig. 2, fore foot track (both natural size) ; fig. 3, reduction of the two foot prints to show in what relation they stand to each other on the slab of Trias sandstone in the quarries at Milford, on the New Jersey side of the Delaware river, about thirty miles above Trenton. Discovered and traced iu outline by Prof. C. H. Hitchcock. (See MS. letter, Dec. 1888.) Upper Trias. — Note. When first seen it was thought to be a Cheirotherium track, but it has only four toe marks on each foot." (Lesley, 1889, pp. 571-572.) These figures, though reprinted by Lyman (loc. cit.), have hitherto escaped the attention of bibliographers and researchers alike. A second outline drawing by Hitchcock, first published in 1915 by Lull, is not identical with the first. Both, however, show the cast of a right manus-pes set, and they are so similar in proportions and particularly in the relative positions of the manus and pes (seldom duplicated exactly in a chirotheriid trackway) that both probably represent the same specimen. Only one of the existing specimens, the type of C. copei Bock, closely resembles these drawings in the features cited. Hitch- cock's statement that the pes "is 5 inches long besides 2y^ inches of heel" applies to this specimen but cannot refer to the Lafayette College types. Now the early history of the C. copei slab, which Bock obtained (mislabeled as to locality) from the efl^ects of an amateur collector of fossils, is obscure ; yet the in- ternal evidence indicates that this specimen or one remarkably like it constituted the undesignated, conceptual type of Hitch- cock's species. Thus Chirotherium copei may be an objective as well as a subjective junior synonym of C. parvum. Variation. The three individuals (or so) represented in the population show a good deal of variation and (if the isolated TRIASSIC REPTILE FOOTPRINTS 475 tracks are correctly assigned) some intra-individual variation as well. These irregularities are, of course, in addition to those caused by differences in impression and preservation. The type and paratype may represent the left and right feet of the same individual. Pes digit I of the type, however, is relatively longer : its claw-mark extends nearly to the tip of digit II instead of to Fig. 5. Chirotherium parvum (C. H. Hitchcock), right manus-pes sets, X 1/3. A. Type of C. copei, ANS 15210. B. Type of C. parvum, 8490 + MCZ 212, showing striated digit-tip impressions. 476 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY the base of the distal pad on II, as in S488, MCZ 210, and ANS 15210. MCZ 209 is intermediate in the length of digit I but is otherwise nearly a mirror image of ANS 15210. In relative width of the digit I-TV group and in robustness of manus digit I the type and paratype resemble each other and differ from the smaller, slenderer specimens which are referred to the "C. copei" individual (cf. Figure 5). In my opinion these differences are insufficient to distinguish two species. Individual, age, and sexual variation are to be ex- pected in the Milford reptile population. Differences observed within the hypodigm are minor and do not appear to be corre- lated. All the specimens exhibit a syndrome of specializations which is unique within the form-genus Chirotherium and almost certainly represents a single reptilian genus. Sympatric species of this unusual genus are of course possible, but since similari- ties greatly outweigh differences, the burden of proof would seem to lie with the splitter. Morphology. Most of the salient features of this species have been described by Lull and Bock. Curiously enough the pes claws, which are present in every specimen and constitute one of the most distinctive features of the species, remain unde- scribed and their existence has only been surmised (Bock, 1952, p. 412). These claws were carried high above the plantar surface so that only their tips ordinarily impressed. Spalling of the natural casts obscures the relationships, but where measure- ments can be made the base of the claw lies at least 6 mm. above the sole of the digit tip. Instead of forming a linear extension of the digit these claws are turned strongly outward, so much so that the first and second must have nearly touched the sides of the digits lateral to them. The claws were carried in the normal vertical posi- tion; their lateral divergence of some 35° seems to be the effect of an oblique ungual articulation, for the digits themselves are otherwise nearly straight. Claw I is long and narrow and slightly curved laterally; its distal half forms a convex ventral keel rather than a point. Claw II, the heaviest, is carinate along its concave ventral profile but ends in a conical point, sometimes blunted by wear. Claw III is similarly curved and pointed. Of course the depth of the claw- tip impressions has been exaggerated when the toes dug in at TBIASSIC REPTILE FOOTPRINTS 477 the end of the propulsive phase. Digit IV appears to be claw- less in all five specimens, even where the digit-tip impression is 8 mm. deep — an anomalous condition, for a well-developed I'law IV characterizes most other chirotheriids. The vestigial fifth pes digit, so different from the well-de- veloped "thumb" of most chirotheriids, is a striking feature of Chirothcrium parvum. A single ovoid pad underlies the pha- langes and the metatarso-phalangeal joint. As in several other Keuper species (Baird, 1954, p. 174) this pad has migrated somewhat medially from the primitive position, so that the distal end of metatarsal V must have underlain the shaft of IV. Onl.v slight marginal indentations and a suggestion of separate plan- tar thickenings distinguish the phalangeal section of the pad, which is scarcely longer than its width and narrows abruptly to an ogival, clawless tip. Almost as distinctive as this digital "heel" is the pattern of plantar pads underlying the metatarsal cross-axis. The meta- tarso-phalangeal pads form a straight line of low bosses across the posterior edge of the sole ; a single flat, subeircular pad underlies the bases of digits III and IV. This condition is quite unusual among chirotheriids. A pad common to two digits occurs in both large-manus and small-manus species but this pad is always central, joining the bases of digits II and III, while the pads of I and IV are distinct. Only in Otozoum do we find a single pad for digits III and IV. Unlike the pes, the manus is typically chirotherioid in form and shows little specialization except that claws are apparently absent and digit V is only slightly offset. (Digits IV and V are definitely clawless but evidence on the others is incon- clusive.) The fifth digit is not small and abortive, as Bock terms it, but normal in length and robustness. In the exception- ally deep paratype imprint, moreover, its pad is revealed as merely the slenderer distal part of a large ovoid "heel" which extends postero-medially to a point in line with the axis of digit I. (See Bock, 1952, plate 41 with plate title 42.) The imprint made by this "heel" pad is 13 mm, deep and slightl}^ undercut laterally. Its steep posterior and postero- medial margins are in strong contrast to the gently sloping mar- gins of the pedal "heel" and suggest that the fifth metacarpal 478 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY cm. 10 I ''■■'■■'* I cm. 10 1 1 1 1 1 1 1 1 1 1 1 TRIASSIC REPTILE FOOTPRINTS 479 was held nearly vertical as the manus was implanted. The other metatarsals apparently sloped less steeply than the fifth but con- siderably more steeply than the metatarsal bundle. This evidence of a more digitigrade manus supports the correlation of Chiro- therium parvum with a reptile of bipedal ancestry in which the forelimbs were shorter than the hind, and in which such a func- tional lengthening of the forelimb pendulum would have made for more efficient quadrupedal locomotion. The paratype manus imprint is divided by a large shrinkage crack which, oddly enough, does not exaggerate its width but rather compresses it laterally so that the third and fourth digit impressions are coalesced. The type imprint of Anchisauripus milfoi'densis is also compressed where it is crossed transversely by the same crack. My explanation for this phenomenon is that the shrinkage crack antedated the footprints (see Soergel, 1925, figs. 45-48 for other examples). Before the sand which now forms the natural casts was washed over the recording surface the clay was flooded and consequently expanded, narrowing the shrinkage cracks and thus compressing the footprints. Ich- nologist, take warning: things are not necessarily what they seem. Chirotherium eyermani Baird, n. sp. Figure 6 B ; Plate 2 Eyerman's chirotherium. Baird, 1954, pp. 174, 175-176. Type. MCZ 134, an isolated left pes imprint collected by John Eyerman in 1887 ; the only known specimen. Horizon and locality as for Anchisauripus niilfordensis. Diagnosis. A moderately large species with slender digit I, short, robust pes digits II-IV bearing heavy claws at sole level, and greatly shortened, hoof -like digit V enclosed in a single pad. Digit IV shorter than II. Metatarsal cross-axis oblique; plantar padding undifferentiated. Manus and trackway unknown. Fig. 6. Brachychirotherian (A, B), small-manus (C), and large-manus (D) chirotheriids with attempted skeletal restorations. A. C. parvum, com- posite of entire hypodigm. B, C. eyermani n. sp., type (MCZ 134). C. C. lomasi, Keuper of Storeton, Cheshire (Yale Peabody Museum 3762, a small individual). D, C. barthii, Moenkopi of Cameron, Arizona (University of California Museum of Paleontology 37315). A-C x 1/3, D x 1/5. 4v0 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY MorpJiology. The single footprint of this species, which is im- pressed to a maximum depth of 40 mm. below ground level, illustrates a fact often overlooked : a footprint is not a cast of the underside of a foot but a composite record of the foot in motion. The effects of structure, function, and preservation are combined in the specimen and must be separated by analysis. If we eliminate the functional distortions of the imprint (which are revealed by slippage marks and variously inclined groups of striations produced by the claws and scales) the foot itself appears to have the form shown in Figure 6 B. A tendency toward functional tridactyly is evidenced by a strengthening of the three central digits at the expense of the first and fifth. Indeed, the soft mud was resistant enough to bend the feeble first digit upward at a 20° angle to the second. The inward-turned imprint of claw I may be atypical, for such a deflection occurs sporadically in chirotheriid traclcAvays. The undulant outlines and almost undifferentiated pads of digits II- IV are reminiscent of C. parvuni, but in strong contrast are the straight, massive claws with their slightly convex soles forming an unbroken continuation of the general plantar surface. Also distinctive is the shorter, more divaricate digit I-IV group with its more oblique metatarsal cross-axis and flat, subrectangular sole. From the bases of digits I and IV a pair of ridges which appear to represent the flcxores digitorum hreves extend prox- imally, converging toward the tarsus. The slope of the longi- tudinal arch, a feature rarely revealed in chirotheriid footprints, indicates that the first four metatarsals formed an angle of 155° with the proximal phalanges as the pes was implanted. Digit V, revealed in unusual detail in this footprint, closely resembles that of C. parvum except that its phalangeal segment is less pointed and is quite undifferentiated from the metatarso- phalangeal pad. This "heel" pad is relatively flat instead of ovoid as in C. parvum; it is impressed about 9 mm. deeper than the ventral surface of the longitudinal arch. A definite bridge connecting the "heel" to the base of digit IV suggests a liga- mentous connection between the distal ends of the metatarsals. From the sole at the tip of digit V a small falciform web curves upward and joins the side of digit IV, well above its sole. TRIASSIC REPTILE FOOTPRINTS 481 The distal margin of digit V shows a striation of flutings which average 2.5 mm. between crests. This fluting resembles the striated 'impressions of claws III and IV and is quite distinct from the scalloped profile produced by fringe scales 3 mm. wide on the medial border of digit II. The greatly reduced fifth digit thus appears to have been tipped with a broad nail or hoof. Such a structure has been reported in no other species of Chiro- therium. Osteology of Level B Chirotheriids The thick, little-difiPerentiated plantar padding in Chiro- therium parvum and C. eyermani makes it impossible to locate all the joints and thus to determine precisely the phalangeal formula. Representative phalangeal patterns may, however, be deduced in other species where articular swellings are well de- veloped. Figure 6 D shows the arrangement in C. harthii, the common large-manus species of the European and American Lower (or Middle) Triassic. Here the formula indicated is manus 2-3-4- ?4-?3, pes 2-3-4-5- ?4. An Upper Triassic member of the same group, C. lulli, is basically similar but appears to have only three phalanges in the thumb-like fifth digit (Baird, 1954, fig. 2 A). In the best-known small-manus species, C. lomasi of the English Keuper (Figure 6 C), a formula of 2-3-4-5- ?4 is evident in the pes but no skeletal pattern is discernible in the hoof-like manus. In preparing the skeletal restorations of C. parvum and C. eyermani I have indicated joints at localized thickenings in the sole wherever possible, and have extrapolated as little as pos- sible from the structure of other species. In C. parvum a slight differentiation in the coalesced pad of pes digit V suggests the presence of three phalanges, so by analogy the same number is hypothesized for C. eyermani. Classification of Chirotheriids Peabody, working chiefly with Lower Triassic species, has sep- arated the better-known chirotheriids into a typical or large- manus group and a specialized small-manus group. Ratios of manus to pes area in representative members of each group are 482 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY tabulated below, in order of decreasing relative manus size. (Areas include only the digits and metapodio-phalangeal pads. These ratios, based on single manus-pes sets, are not necessarily typical of the species listed.) Large-Manus Small-Manus C. lulli Bock 1 :2 C. coltoni Peabody 1 :5.7 C. storetonense Morton C. marskolli Peabody 1 :5.7 ca. 1 :3 C. lomasi Baird 1 :6.6 C. minus Sickler 1 :3 C. wondrai Heller 1 :3.4: C. harthii Kaup 1 :3.6 In Chirotherium parvum the type manus is too incomplete for areal measurement but ANS 15210 furnishes reliable figures of 25 cm. 2 for the manus and 112 cm.^ for the pes. (Bock's meas- urement of 85 cm.- for the pes apparently omits the "thumb" pad.) The resulting manus-pes ratio of 1:4.6 falls between the ranges of the two groups. Except for the closeness of digit V to its fellows the manus of C. parvum is more similar to that of the large-manus species than to the compact, hoof-like forefoot of the small-manus group. Unfortunately, nothing is known of the manus in C. eyermani. The shortening of pes digit IV, moderate in C. eyermani but extreme in C. parvum, would seem on typological grounds to ally these species with the small-manus group, in which such a shortening is characteristic. In terms of function, however, this condition is obviously essential to the maintenance of symmetry in any walking foot in which the fifth digit has undergone reduction. Other examples are found in the feet of dinosaurs, crocodiles and birds (where apparent exceptions such as Gavia, Pelecanus and Hesperornis prove the rule, since walking is not the primary function of the foot). Thus the shortened fourth digit in the Milford species is functionally correlated with the vestigial fifth digit and indicates parallelism rather than affinity with the small-manus group. The Level B chirotheriids thus appear to represent a lineage distinct from the small-manus group although paralleling it in adaptive modifications and similarly derived from a primitive large-manus stock. This lineage has been progressively modified TRIASSIC REPTILE FOOTPRINTS 483 to a point where it must be considered a distinct sub-generic group, characterized by the short, oval, heel-like fifth pes digit. This group, here recognized in North America for the first time, is represented by specimens from several horizons in the Middle Keuper of Germany (Heller, 1952). The oldest of these, an imperfect pes imprint from the Blasensandstein of Franken, was described by 0. Kuhn (1936) as Chirotherium sp. In size and proportions it shows a decided resemblance to the Level B species, particularly C. eyermani, differing chiefly in the greater length of digit V. A rather similar but slightly younger form is Chirotherium thiiringiacuni Rlihle von Lilienstern (1938) from the Platten- sandstein (km 4 zeta) of Thiiringen. This species recalls C. eyermani in its short, divergent pes digit I-IV group. It re- sembles C. parvum in its round, apparently clawless digit tips; and the manus with its plump, clawless toes and close-set, medi- ally shifted digit V is strikingly similar. Slightly higher in the Middle Keuper, in the Semionoten- Sandstein of Franken, occurs a group of chirotherioid footprints which Beurlen (1950) has named Brachychirotherium hassfur- tense. Though the specimens figured are somewhat heterogeneous they all show a foot structure like that described above, with apparently clawless digits and an abbreviated "thumb." The latter appears so inconsistent in its size, position, and orienta- tion, and the relative lengths of the other digits are so variable from specimen to specimen, that definitive diagnosis of the species (singular or plural) is impossible. Unfortunately, all these German forms have been described from individual foot- prints rather than analyzed on the basis of a composite of each population. Specific distinctions are by no means clear, particu- larly as no types have been designated, and the differences care- fully tabulated by Heller may or may not be significant. Never- theless these Middle Keuper species from Germany are clearly referable to the same group as C. parvum and C. eyermani. For this group Beurlen has established the separate genus Brachychirotherium. Since the differences between this and the large-manus group are little greater than those which distinguish the small-manus group from both, I prefer to broaden slightly the scope of the form-genus Chirotherium to include a third 484 BULLETIN: MUSEUM OP COMPARATIVE ZOOLOGY sub-generic group for which Beurlen's name may appropriately be retained. Following the example of Peabody I assign this group no formal Linnaean status. Evolutionary Tendencies The modifications seen in this brachychirotherian group are clearly consistent with the general evolutionary trend of Keuper chirotheriids which I have previously outlined (1954, p. 188) : the tendency of an osteologically pentadactyl stock to become functionally tetradactyl through elimination of the lateral prop- ping function of the fifth pes digit. In the brachychirotherian group this was being accomplished by several means : by shifting the distal end of the unshortened metatarsal V more or less medially, beneath and behind the shaft of metatarsal IV; by progressively shortening and straightening the fifth digit to minimize its lateral projection; and ultimately by incorporating the metatarsal and phalanges into a single, inflexible heel structure. During this process a gradual shortening of digit IV tended to maintain the symmetry of the foot. Concomitantly the manus became smaller and more compact. Although — most unfortunately — we have no trackways from which to determine the locomotor habits of the group, it seems inescapable that these modifications promoted more efficient locomotion, presumably cursorial and (as Riihle von Lilienstern points out) showing bipedal tendencies in some members. In- creasing competition from the expanding and diversifying stocks of saurischian dinosaurs may well have channeled chirotherian evolution along these lines. Changes in foot structure and locomotor pattern are of course only a part of the overall adaptive modification which must have taken place in the brachychirotherian group. Only a part, yes : but surely an important one, for, as I have shown (1954, p. 174), convergent modifications were independently taking place in Keuper representatives of both the large-manus and small- manus groups. Apparently the chirotheriid locomotor equipment of early Triassic time had been found inadequate for the con- ditions of mid-Keuper life. The chirotheriid response to changing conditions was, how- ever, insufficient. We have noted that even in the most special- TRIASSIC REPTILE FOOTPRINTS 48;") ized species, C. parviim and C. eyermani, the once thumb-like fifth digit was still functional in modified form. Dinosaurs and small pseudosuehian or proto-erocodilian types such as Batracho- pus, in which locomotion had become more efficient through the loss or atrophy of the fifth digit, eventually replaced Chirother- iinii. Huge size and bipedality (along with other specializations yet unknown) permitted a lone descendant of the chirotheriid stock, Ofozonni, to survive until the close of Newark time — but that is another story. Osteological Correlations That chirotheriid trackways are the products of various pseudosuehian reptiles is now" generally accepted. Beyond this premise the foregoing discussion has avoided assumptions as to the identity of the Milford trackmakers, attempting instead an objective analysis of the ichnological evidence as such. The possible position of these trackmakers among the Pseudosuchia can now^ be examined. At present the pedal skeletons of most Upper Triassic pseudo- suchians are too inadequately known to permit useful compari- sons with the three Milford species of Chirotheriitm. The rela- tively conservative large-manus species C. lulli has been inter- preted as a small, persistently quadrupedal ornithosuchid not too different from the presumably ancestral form Eupar'keria; but closer comparisons are impossible. Among middle to late Triassic reptiles which might be compared to the brachychiro- therian species only two, Prestosuchus and Typotliorax, are represented by even relatively complete foot material. As restored by Peabody (1948, fig. 39 B-C^) the pes of Presto- suchus is broad and short-toed, suggesting (as Peabody notes) the brachychirotherian C. thuringiacum Riihle. Lack of the tip of digit IV and all phalanges of digit V, however, precludes for the present any correlation of Prestosuchus w^ith one or another of the chirotheriids. Certainly the phalanges are relatively much shorter than those restored for the Milford footprints. As Huene (1944) records only a single phalanx from the manus, Pea- 3 Peabody has followed Huene in reversing the proximal overlap of the meta- tarsals although the natural relationship is preserved in the specimen itself ^ — cf. Huene, 1944, pi. 20, hg. 4 a-b. 486 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY body's very plausible reconstruction of the manus digits cannot be used for comparison. Despite the strong contrast in habitus between the relatively narrow-bodied, lightly-armored, predaceous Prestosnchus and the broad-backed, heavily-armored, necrophagous Typothorax (Sawin 1947), the two are striking^ similar in phalangeal pattern. Thus the general aspect of Typothorax is brachychiro- therian-like but its short phalanges seem to eliminate it from comparison with the Milford footprints. The manus is decidedly chirotherian in form although its fifth digit appears to have been somewhat weaker than that of most chirotheriids. Areas of the digit I-IV groups in manus and pes bear a ratio of about 1 to 3, a ratio found in C. thuringiacum and some large-manus chirotheriids. If brachychirotherian trackways were available the pedal similarities between Prestosuchus and Typothorax should cause little difficulty in correlation, for the swift-striding predator must have left a much narrower trackway than its broad-beamed, ponderous relative. But pending the discovery of such additional evidence the systematic positions of the Level B pseudosuchians must remain uncertain. Order THECODONTIA Suborder PHYTOSAURIA Form-family APATOPODIDAE Baird, nov. Genus APATOPUS Baird, nov. Type speeioM A. Imeatus (Bock), 19."2; family and genus monotypie. Diagnosis. Quadrupedal trackway wdth pace angulation about 108°, pes but not manus toed-out. Feet pentadactyl with well- developed articular swellings and sharp claws ; manus short and symmetrically radiate, pes long and narrow^ with digits in order of increasing length V-I-II-III-IV. ApATOPUS LINEATUS (Bock) Figures 7, 8 ; Plates 3, 4 Otosoum (?; lineatus Bock, 1952, pp. 408-409, pis. 48, fig. 1 and 46, fig. 1 (niistitled "Type No. 15230" for LC S489). Probable earlier syn- TRIASSIC REPTILE FOOTPRINTS 487 onyniy in Table 1 ; the variety of past assignments suggested the generic name. Diagnostic characters are those of the genus. Type. LC S490. a left maniis-pes set (Bock's type) plus MCZ 212, the succeeding right set of the same trackway. Hypodigm. The type, S489, and MCZ 213, presumably three adults ; AMNH 1789 and 2258, smaller individuals ; AMNH 2259 and MCZ 214, juvenile (same individual?). Source data as for A nch isa u rip u s ni ilfo rclensis. Trackway. By a remarkable and fortunate coincidence a tiny contact area between corners of the Lafayette and Harvard slabs permits the reconstruction of the type trackway. (This contact crosses the "thumb" of the Chirotherium parvum type, cf. Figure 5 B.) Trackway parameters determined by direct measurement or by extrapolation (*) are tabulated below. In the first three the reference point is the center of metatarso- phalangeal pad III : Pace, angular 42.5 cm. Stride *70 cm. Pace angulation *108° Gleno-acetabular length (approximate) *52 cm. Interpes width (between median margins) 13.7 cm. Pedal morphology. The additional material now available permits analysis which was impossible with the incomplete, ob- scure, single manus-pes set on which the species was based. A composite restoration derived from all this material is presented in Figure 7. The digits of both manus and pes are shod with domed, oval articular pads separated by slender waists, so that joint positions are apparent in profile as well as in section. The claws are slender and rather crocodilian in the manus but appear heavier and more triangular in the pes, where rotation during withdrawal has obscured their exact form. As the plates show, laterally flexible ungual articulations permitted much lateral movement in both manus and pes claws. The manus is nearly symmetrical around digit III, the meta- carpo-phalangeal pads forming a regular semicircle, so that in form and proportions it recalls the manus of Mesosaurus — an aquatic form. A phalangeal formula of 2-3-4- ?5-?3 is indi- cated. In normal function the manus was digitigrade and only in the deepest imprint did the palm register (Plate 4, figure 1). The pes with its long fourth digit and peculiar sole is quite 488 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY A CM.-r 5- 10- 15-^ Fig. 7. Apatopus lineatus (Bock), composite restoration based on type (S490 + MCZ 212) with parts of digits IV and V restored from other specimens, x 1/2. Arrow represents midline of trackway. TRIASSIC REPTILE FOOTPRINTS 489 distinct from other Triassic footprint types. At first glance it appears to show affinities with primitive reptilian pedes of the pelycosaur or captorhinomorph type, but the fact that meta- tarsal IV is slightly shorter than III (rather than the reverse) indicates that digit IV has been secondarily lengthened by hyper- trophy of its phalanges. A similar secondary lengthening of the phalanges of digit IV occurs in the manus of the marine turtle Chelonia mydns (cf. Abel, 1912, fig. 111). Aquatic adaptation is the most obvious explanation for such a modification, and this interpretation is supported by the evidence of interdigital webs. Although the length and position of the short fifth digit are evi- dent its structural details are obscure ; my conjectural restora- tion shows four phalanges, making the pes formula 2-3-4-5- ?4. The extent of interdigital webbing in Apatopus is uncertain. ])articularly because the pes in walking was often laterally com- pressed so that web margins would have left no mark. The best- represented web connects the bases of pes claws III and IV, an arrangement quite in contrast to that in living crocodilians, in which the deeply recessed webs occupy only part of the inter- digital area. Additional (though circumstantial) evidence for the presence of webbing is the fact that digits IV and V are adequately recorded only in the deepest footprint, which sug- gests that distribution of the animal's weight over a w^ebbed area effectively reduced the depth of impression. Here as in other fossil footprints, unfortunately, the evidence is less than satisfactory. Traces of the squamation are preserved only in the juvenile MCZ 214 (Plate 4, figure 1). Here the digits of manus and pes show a fine beaded or pustulose texture which compares closely with the skin of corresponding areas in a juvenile Gavialis gangeticus. Except in the deepest impressions the sole is very indistinctly recorded. Some understanding of it can nevertheless be obtained by collating outlines printed from latex molds of several speci- mens. On the medial margin a bulge marks the position of the first metatarsal's proximal head, and opposite this is a circular pad which must have underlain the fifth metatarsal. On this basis the first four metatarsals have been restored in normal alignment. Its distinct individual pad suggests that the proximal end of metatarsal V was less elevated than that of metatarsal IV ; 490 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY if SO, then the fifth metatarsal probably articulated with the tarsus by a hooked, dorso-medially directed process. Postero- medial to this area a large, subcircular calcanear pad forms the heel of the footprint. The possible presence of a tuber eal- canei is suggested by the relatively deep impression of this pad. Though plausible, this interpretation of the tarso-metatarsal area rests on admittedly tenuous evidence. Variation. Aside from striking variations in the manner of impression, which are apparent in the plates, the Apatopxis footprints reveal only minor differences among the six (or so) individuals of the population. Chief of these is size : the small- est individual, AMNH 2259, has a pes length just half that of the type (as restored). Osteological analysis reveals the struc- tural unity of these footprints which might otherwise be as- signed to a number of form-species. Affinities of Apatopus Examination of Triassic footprints from every continent yields nothing similar enough to Apatopus to justify comparison in terms of footprint taxonomy. We may more profitably in- vestigate the nature and systematic position of the trackmaker. In its foot structure and body proportions Apatopus shows closest affinities to quadrupedal members of the Subclass Archosauria — i.e. the Phytosauria, Pseudosuchia and Crocodilia. These groups will be considered in reverse sequence, beginning with the best known. Locomotion in crocodiles, which has been investigated by Abel (1912, p. 217), Huene (1913), and Schaeffer (1941, p. 443 ff.), is of three types: a slow, lizard-like dragging of the body, a spring in which all four feet thrust simultaneously, and a strid- ing gait in which the belly is well above the ground and only the tail-tip leaves a mark. Of these gaits only the last would produce trackways comparable to that of Apatopus; thus neither the "Gehspur" nor the "Laufspur" which Huene figures is pertinent to this discussion. An example of the true striding gait in a juvenile Alligator mississipiensis has been recorded on film by Schaeffer. The resulting trackway shows a pace angu- lation of only 80°, i.e. the stride is little longer than the pace. In evaluating this record the smallness of the alligator and the TRIASSIC REPTILE FOOTPRINTS 491 o CO C o o PI bo o3 CO M bo ^3 492 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY fact that it was walking on glass must be considered. For ex- ample, the 20° to 30° outward rotation of the pes between implantation and withdrawal can hardly be normal : no such rotation is evident in the alligator trackway figured by Jaeger (1948, pi. 170 c), and only in very muddy footing did the pes of Apatopus rotate a few degrees outward (cf. Plate 4, figure 2). I believe that on a fairly firm surface a crocodilian in rapid motion would produce a trackway rather similar in proportions to that of Apatopus. We have already noted the crocodilian habitus indicated by the aquatic adaptations in the feet of the Milford traclanaker. Despite, these similarities in habitus Apatopus can hardly have been a crocodilian by heritage. The ancestral crocodilian Protosuchus from Upper Triassic or Lower Jurassic beds of Ari- zona had already developed an essentially modern type of foot with an elongate metatarsus, reduced fourth and rudimentary fifth digits, and a formula of 2-3-4-4-0 (Colbert and Mook, 1951). These characteristics were probably inherited without much modification from pseudosuchian ancestors and have been trans- mitted without much modification to the living crocodilian.s. Even the aberrant pes structure in the marine thalattosuchians can be more readily derived from the Protosuchus pattern than could that of Apatopus. Another point of distinction is the out- turned manus which characterizes crocodilian trackways. Thus the Milford trackmaker cannot be referred to any crocodilian lineage. Among trackways which have been ascribed to pseudosuchians certain members of the Batrachopns group approach the Apato- pus pattern. Nevertheless the reconstructed feet of these forms — and the few known pseudosuchian foot skeletons — are dis- tinctly different, particularly in the fourth and fifth pes digits. In the Phytosauria we find the logical correlative for a trackmaker which is crocodilian in body form and presumed habitus but not in skeletal structure. The comparative propor- tions of the crocodilians Gavialis, Tomistoma and Alligator, and the phytosaurs Eutiodon, Macliaeroprosopus and Mystriosuchus have been analyzed by Colbert (1947), who attests to the re- markable parallelism noted by earlier authors. Very little articu- lated phytosaurian foot material, unfortunately, is available for comparison with the footprints. The manus is best represented TRIASSIC REPTILE FOOTPRINTS 493 in Mystriosuchus (McGregor, 1!)06, fig. 10), in which — like Apatopus — the metacarpal bundle is nearly bilaterally sym- metrical and digit V bears three phalanges which are abont twice as long as they are wide. Further comparisons cannot be made. McGregor's reconstruction of the pes in Ridiodon, based on incomplete, scattered material of several individuals, is remarkably similar to my restoration of Apatopus — which (ex- cept where drawn in dashed lines) was quite independently derived from the footprints alone. If Apatopus is indeed a phytosaur, then Dr. McGregor proves to have been indeed a prophet worthy of honor. Correlation with the Phytosauria is supported by all the evidence and contradicted by none; it may therefore be taken as a working assumption. Possible equivalents exist in the two pln-tosaurs of the Newark series, Eutiodon and Clepsysaurus {'i=Machaeroprosopus) . The significant differences in skeletal proportions between these genera would presumably be reflected in their trackways and may someday permit positive assignment of Apatopus lineatus to one or the other. Present knowledge, however, does not justify comparison of the single, partial Apatopus trackway with the composite skeleton of Rutiodon from North Carolina (Colbert, 1947) as opposed to incomplete skeletons from Arizona which may be congeneric with Clepsysau- rus (Camp, 1930). Geographic distribution of the Newark phytosaurs offers a possible guide to the identity of A. lineatus, for Colbert and Chaffee (1941) concur with Camp's conclusion that all the phytosaur remains from Pennsylvania, New Jersey, and Connecticut represent Clepsysaurus while the North Caro- lina finds all pertain to Rutiodon.^ Apatopus lineatus would thus appear to be a correlative of Clepsysaurus. So far as I know these Milford specimens are the only phyto- saur footprints yet described.^ Paradoxically, phytosaur bones are the commonest skeletal remains of both the Chinle formation of the Southwest and the Newark series of the East (excluding. 6 H. E. Wanner's (1926) specimens of "Rutiodon caroUnensis" from the Newark series of Yorlj County, Pennsylvania — overloolied by the above-mentioned authors and by the Hay-Camp bibliographies — merit further study. 7 An isolated pes imprint, considered by Bock (19o2, pi. 46, fig. 2) to be probably that of a large parasuchian of the Clepsysaurus type, bears no resem- blance to Apatopus. So far as its preservation permits comparisons, it seems more similar to ornithoid footprints of the Plectropterna type from the Con- necticut Valley. 494 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY significantly, the Meriden and Portland formations). In the ecology lies the probable explanation for this discrepancy : Triassic footprint assemblages as a rule represent terrestrial faunas while the fossil deposits contain a disproportionate num- ber of aquatic forms. Thus the presence of several phytosaurs in a dinosaur-pseudosuchian-lacertoid footprint faunule at Mil- ford is rather anomalous. REPTILIA LACERTOIDEA INCERTAE SEDIS Genus RhYNCHOSAUROIDES Beasley w Maidwell, 1911 Fontopus Nopcsa, 1923, p. 141. Keiiper of England. Akropus Riihle von Lilicnstern, 1939, p. 298. Bunter of Germany; Moenkopi of Arizona (Peabody, 1948, p. 319 ff.). Hamatopus Riihle von Lilienstern, 1939, p. 319. Bunter of Germany. TJiyncliocephalichnus Huene, 1941, p. 14. Keiiper (Verrucano) of Italy; Keuper of Germany (Heller, 1956). Eurichnus Lull, 1942A, p. 502; Branson, 1947, p. 590. Lower Triassie (Chugwater) of Wyoming. The type species is here designated as R. rectipes Beasley in Maidvrell, the type specimen of which is a manus-pes set from the Keuper of Dares- bury, England (Dr. Ricketts' collection, University of Liverpool Geology Department). RHYNCHOSAUROIDES HYPERBATES Baird, n. Sp. Figures 9 A, 10 Type. ANS 15210, a trackway of one pes and three manus imprints associated with the type of Chirotherium "copei." Hypodigm. The type ; an isolated left pes on the same slab ; MCZ 210, a partial left pes associated with C. parvum. Diagnosis. Manus i}road with little disparity in digit lengths : gait digitigrade with hallux non-functional and proximal ends of proximal phalanges normally carried clear of the ground. The specific name alludes to the overstepping of manus by pes. Morphology. The trackway is rather narrow, showing a stride of 44 cm., forelimb paces of 23 and 25.5 cm., and a pace angula- tion of 130° between manus imprints. (In these measurements a 3-cm. fissure filling which crosses the trackway is ignored, as the relative ages of trackway and fissure are unknown.) As in TBIASSIC REPTILE FOOTPRINTS 495 Fig. 9. A. Bhynehosauroides hyperbates n. sp., composite restoration of manus (at left) and pes in normal position. B-C. Small Bhynehosauroides footprints from Milford: B, (ANS 15210); C. manus imprint (MCZ 213). D. Manus imprint from Trostle quarry near York Springs, Adams Co., Pa. (Carnegie Museum 12082). E. Manus of Monjurosuchus reconstructed from Endo's photograph. F. Manus and pes of Polysphenodon miilleri Jaekel, modified from Huene. All x 4/5. 496 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY a «l-l o on S3 Oi ■1^ t-l o a> o o &< >-, $^ '*" ^ ai lO be '-C CO ii =c JS HJ Oi ® c .5 =0 g ^ ft s §•5 . « c3 ^ - C3 o ffl rt I a f^ ■: e •- a- . O S" ^ i ^ ^ ;i ft * TRIASSIC REPTILE FOOTPRINTS 497 Other Triassic lacertoid trackways the pes is strongly out-turned and evidently overstepped the manus as it implanted.^ The trackmaker, which is restored in walking pose in Figure 10, had a gleno-acetabnlar length of about 18 cm. and a total length of perhaps 75 cm. The foot structure, revealed by comparing outlines printed from latex molds of all the imprints, is decidedly lacertoid. Well-developed articular swellings indicate an apparent pha- langeal formula of 1-2-3-4- ? in the manus and ?-2-3-4-? in the pes. This formula must be incomplete, for the obvious reptilian character of the feet connotes a normal formula of 2-3-4-5-3(4). Like the structurally similar Moenkopi genus Rotodactylus (Pea- body, 1948, p. 329), the Milford form must have been digitigrade to such an extent that each proximal phalanx was elevated with its metapodial ; normalh^ only the padded distal articulation of phalanx 1 touched the ground. All the digits recorded bear sharp claws, either extended or curved toward the trackway midline. The pollex is represented only by claw impressions. Nothing which can be identified as the imprint of a fifth manus digit is visible. The absence of any hallux imprint and the disparity in length of the pes digits indicate that the hallux was non-func- tional. Pes digit V has left no unequivocal record, but obscure marks on MCZ 210 and ANS 15210 may have been made by a lacertoid, laterally projecting minimus. Both these marks are shown, for what they may be worth, in Figure 9 A. Rhynchosauroides sp. Figure 9 B, C ; Plate 3, Figure 2 A pair of minute footprints on the ANS slab, unfortunately too incomplete for reliable analysis, and indistinct trackways of comparable size on a slab with Apatopus lineatus (MCZ 213) prove the presence of another species of Rhynchosauroides in the Level B faunule. Smaller size and greater disparity in the lengths of the manus digits distinguish it from B. hyperhates. 8 1 have seen no evidence in Pennsylvanian, Permian, or Triassic lacertoid trackways that the pes rotated nearly 90° outward during the propulsive phase as Schaeffer (1941, p. 440) has observed in experiments with Anolis. Except tor an occasional rotational slippage of a few degrees in unusually soft mud the pes remained in implantation position until withdrawn. 498 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Relationships Lacertoid trackways of this sort are familiar companions of Chiroiherium in the Lower and Upper Triassic of both Europe and America. As indicated in the synonymy, a number of generic names have been applied to them. The differences between these nominal genera are, however, slight : for example, Riihle von Lilienstern distinguished Ilamatopus from Akropus largely by the more hooked appearance of its digit tips. A more practical treatment here proposed is to consider these forms as species of a single form-genus Rhyncliosaur aides which is comparable to Chirotherium in its range of variation. This form-genus, of cos- mopolitan distribution in continental Triassic beds, undoubtedly represents several reptilian genera; but until more is known of Triassic lacertoid feet and footprints there seems little advantage in proliferating form-genera on differences of uncertain diag- nostic value. Our lack of data on the fifth digits of manus and pes hinders comparison of Ehynchosauroides hyfjerhatcs to other species. Its closest similarities, however, are to the footprints from the English Keuper which were described by Beasley (1905) and comprehensively reviewed by Maidwell (1911, 1914). The smaller Milford lacertoid is less similar to the larger one than to Rhynchosauroides [Rhynchocejyhalichnus] franconicus (Heller, 1956; see also Haarlander, 1938) from the Keuper of Germany. Such trackways have never been found in the well-known footprint faunas of Portland and Meriden age in the Connecticut valley, though they are not uncommon in the Brunswick and Lockatong formations of the Pennsylvania-New Jersey basin. The Milford forms cannot adequately be compared with Rhyn- chosauroides ['^ Orthodactylus"Y howelli (Bock) from an earlier horizon of the Brunswick near Schwenkville, Pennsylvania, until better-preserved material is available. An isolated Rhyncho- 9 This form has nothing to do with Hitchcock's Orthodactylus from the Port- land formation of Gill, Massachusetts. The type specimen of the type species O. florij^rus is an obscurely recorded trackway of pseudosuehian or proto- crocodilian tvpe, having the manus anterior to the pes and strongly out-turned as in the contemporary (Tit HOi/ieroide«. _ Superficial similarity of the digital gouge- ma " ■ - - -" dactylh.. -_ — . - .„,,-. and equally indeterminate were made the type of a new species, Procolophoinpus conhuenti 'Bock (p. 419), ascribed to a plump, slow-moving reptile which jumped quadrupedally without leaving manus imprints. TRIASSIC REPTILE FOOTPRINTS 499 sauroides maniis imprint of Brunswick age from the Trostle stone quarry near York Springs, Adams County, Pennsylvania, is shown in Figure 9 D ; and another of Lockatong age from Phoenixville, Pennsylvania, has been figured without name ])y Bock (1952, pi. 48, fig. 2). Aside from some undescribed track- ways from the Doclmm group of the Cimarron Valley in north- eastern New Mexico (AMNH 6338), these are the only records of Rhynchosauroides in the Upper Triassic of North America. The systematic position of the trackmakers is debatable. Foot- prints of this sort have been referred by various authors to the Protorosauria, Eosuchia, Pseudosuchia, Rhynchocephalia and Lacertilia: and indeed each of these groups has Triassic repre- sentatives which would have made lacertoid footprints. Among the Protorosauria the tanj'stropheid Macrocnemus from the Al- pine Middle Triassic (Pe^^er, 1937) is similar in size and pro- portions. Its reduced fifth digits of manus and pes exclude it from correlation with the European species of Rhynchosauroides, and its greater disparity in the lengths of the manus digits distinguishes it from the larger Milford form. A similar dis- parity characterizes the manus of Trilophosauriis from the Upper Triassic (Dockum) of Texas (Gregory, 1945) ; but its pes digits are much less unequal in length. The proximal end of ungual II lies opposite those of phalanges III-3 and IV-2, while in the footprints it lies opposite the proximal end of III-2 and the waist of lV-1, much as in the Permian Protorosaurus. But as the known Triassic protorosaurian genera undoubtedly consti- tute only a fraction of the number that once lived, the group cannot be disregarded as a possible source of Keuper lacertoid trackways. Too little is known of limb structure in the Eosuchia to permit useful comparison. Peabody (1948, p. 337) has referred his Lower Triassic foot- print genus Rotodactylus to the Pseudosuchia on the basis of its long-striding gait with pendulum limb movement, its semi- bipedal concentration of the body weight on the long hind limbs, and its digitigrade pes with posteriorly rotated, prop-like fifth digit. Isolated mauus-pes sets of Rotodactylus appear decep- tively lacertoid (especially where the fifth digit has failed to impress), but the whole organization of the animal is significantly 500 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY different from that of lizard-like reptiles/" Aside from its over- stepping gait and similar digitigrade specialization the Milford trackmaker appears to be basically unlike Rotodactylus. No comparable pseudosuchian skeleton is known. The case for the Rhynchocephalia as a source for Triassic lacertoid footprints, reviewed by Huene (1938, 1941) and Pea- body (1948), dates historically from Richard Owen, who in 1842 assigned the small footprints of the Grinsill quarries near Shrewsbury to Bhynchosaurus articeps on the grounds that the bones and footprints correspond in size and occur in the same beds, to the exclusion of other bones and footprints. This correla tion was accepted by subsequent students of British Keuper foot- prints and indeed motivated Beasley to coin the name Bhyncho- saur aides. A basic similarity certainly exists between Rhynchosauroides footprints and the feet of Bhynchosaurus as restored by Smith Woodward (1907). But closer inspection shows that nearly all the footprints were made by long-legged, slender-toed reptiles, whereas Bhynchosaurus is comparatively short-legged and broad- toed as a modification for aquatic life. At this point it may be illuminating to note that in the same year in which he proposed the Bhynchosaurus correlation Owen was asserting, on essentially similar grounds, that Chirotheriuni footprints represented Laby- rinthodon. The great contrast (now understood) between the feeble-limbed, bottom-crawling " Lahyrinthodon" and the up- right-limbed, semi-cursorial Chirotheriuni points up the ecologi- cal unsoundness of both correlations: the genera common in the Keuper bonebeds represent an aquatic facies-fauna while the footprints record a terrestrial facies-fauna. Among terrestrial rhynchocephalians we may eliminate the Rhynchosauridae because of their un-lacertoid proportions and short, hea\y phalanges. The Sphenodontidae, however, offer closer comparisons. Proportions similar to those of the Milford trackmaker are found in the Upper Jurassic Homoeosaurus; the Manchurian sphenodontid Monjurosuchus (Endo, 1940) is rather similar in manus structure (Figure 9 E) although its body pro- portions suggest that the pes probably did not overstep the 10 Bock's attempt (1952, p. 422) to synonymize Rotodactylus with the araeosceloid Lower Permian footprint •Ichnium gampsodactijhnn" — more cor- rectly termed Dromopus [Saurichnites] lacertoides (Geiuitz), n. comb. — over- values the apparent similarities at the expense of fundamental ditTorences in foot structure and tracljway pattern. TRIASSIC REPTILE FOOTPRINTS 501 manus as much as in Rhynehosanroidcs hyperhates. Upper Tri- assic genera are (at this writing) less completely known, but the limbs of Polysphenodon compare favorably (as Huene points out) with footprints of the Rhynchosauroides type from the Keuper of Italy. As shown in Figure 9 F, the phalangeal pattern of Polysphenodon also corresponds approximately to that of the larger Milford lacertoid. The publication of current research on several skeletons of Glevosaurus, a sphenodontid from Keuper fissure fillings in Gloucestershire (Robinson, 1955), may well provide the means to a better understanding of Rhynchosau- roides. Triassic Lacertilia, a group in which major advances in knowl- edge are in progress, may similarly prove to correlate with some of the lacertoid trackways which I have brigaded under the name Rhynchosau7'oides. On the evidence now available R. hyperhates is assigned tentatively to the sphenodontid Rhyncho- cephalia, but laeertilian origin remains a distinct possibility. At present the various trackways cited above constitute the only evidence of rhynchocephalians or lacertilians in the North Amer- ican Triassic. CORRELATIVE AGE OF THE MILFORD FAUNULES Research on several problems of Pennsylvanian and Permian ichnology, still unpublished, indicates that fossil footprints — especially footprint faunas — can be used in a limited way for stratigraphic correlation. As footprint species can at best be equated only with generic units of tetrapods they cannot ap- proach the diagnostic precision of skeletal remains. Assemblages of footprints are of course subject to the same ecological bias as assemblages of skeletons, and may be equally deceptive in suggesting the non-contemporaneity of beds which are actually equivalent in age. Despite such limitations ichnological evidence can be a valid basis for correlation; indeed, it is sometimes the only basis available. One pertinent example is the presence of similar Chirotherium harthii faunas in the Bunter of Germany and the Moenkopi of Arizona, which (with other evidence) .establishes the equivalence of these strata. Consequently when a specimen of C. barthii which is indistinguishable from the Arizona ma- 502 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY terial appears (under the name ''C. higuerensis") in so-called Upper Permian beds of Argentina (Rusconi, 1951) we need feel no doubt that the beds are in fact Triassie/^ The amount of material now available and the size and diver- sity of the combined Milford faunules justifies an attempt at correlating them with faunas of other areas. In this assay foot- prints will be compared with footprints and bones with bones : for in the past, uncritical herpetological interpretations of mis- identified and uuanalyzed footprints have too often proved misleading in stratigraphic work. Two views on the correlative age of the Milford footprints have been advanced : one, that the footprint assemblages of Mil- ford and Whitehall, New Jersey, York Springs, Pennsylvania, and Aldie, Virginia, compare closely with the youngest foot- print fauna of the Portland formation in Massachusetts (Bock, 1952, pp. 429-430) ; the other, that the Milford faunules correlate neither with the Portland nor the underlying Meriden but with the New Haven, lowermost of the Connecticut Valley formations (Baird, 1954, pp. 184-187). Let us examine these faunules, be- ginning with the southernmost. The only specimen from Aldie available to me (MCZ 236) is a well-preserved pes imprint, 35 cm. long including the hallux, of Anchisauripus tuherosiis (Hitchcock). This species ranges from the lower Meriden to the upper Portland formation. Of the Aldie fauna Gilmore (1924) says: "Three-toed imprints predominate though they vary in size from a length of three to fourteen inches. A few tracks were noticed having four toes, evidently terminated with wide, flat unguals. All of these are probably of dinosaurian origin, but a few small 4- or 5-toed tracks with traces of sharp claws per- haps pertain to some other group. ' ' Until this important assemblage has been thoroughly studied no valid comparisons can be made. Specimens from the Gettysburg shale of the Trostle stone quarry near York Springs, Adams County, Pennsylvania, are figured by Stose and Jonas (1939, pi. 22) ; others are preserved at the Carnegie Museum. In addition to invertebrate trackways 11 Siuce the foregoing was written Peabody (19.55) has published the same observation and noted the importance of this evidence on Triassic faunal distribu- tion. TRIASSIC REPTILE FOOTPRINTS 503 and some indeterminate vertebrate tracks the faunule contains three genera: (1) a uniquely quadrupedal dinosaurian type re- sembling Anchisauripus gwyneddensis Bock from the Lockatong formation, (2) a Blujnchosauroides (Figure 9 D) comparable to those from the Brunswick of Schwenkville and Milford and the Lockatong of Phoenixville, and (3) large footprints consisting of a sub-circular plantar pad close behind an arc of broad ungual depressions, rather suggestive of a dicynodont foot. Nothing like any of these forms occurs in the Connecticut Valley. The Whitehall footprints are known to me only from the thoroughly unreliable faunal list of C. H. Hitchcock (1889). Further discussion is useless until the material can be restudied by modern methods. Only one of the Milford species, Anchisauripus parallelus, also occurs in the Connecticut Valley. The other dinosaurian tracks are distinctly different from those of the Meriden and Portland formations. Chirotherium footprints, which at Milford record three genera of pseudosuchian reptiles, are unknown in New England ; so are trackways of the Apatopus and Rhyrichosau- roides types. Their absence there is real and cannot be ex- ])laiued as the result of inadequate collecting or superficial study — far from it 1 In the absence of any evidence of barriers be- tween the New Jersey-Pennsylvania basin and the ancient Con- necticut Valley trough, the conclusion presented in m.y previous j)aper seems inescapable : the footprint beds of the upper Bruns- wick formation at Milford, New Jersey, antedate the Portland and Meriden formations and more probably correlate with the New Haven arkose of Connecticut. The most comparable foot- prints in the European sequence occur in formations of the mid- dle to upper Middle Keuper of Germany. REPTILES OF THE BRUNSWICK FORMATION To recapitulate, the Milford footprints comprise three mu- tually exclusive faunules from three horizons of the Brunswick formation. What these footprints signify in herpetological terms is summarized, so far as known, in Table 3. These footprints probably represent only a few of the different types of reptiles which inhabited the area, and the reptiles themselves only a .segment of the biota. Nevertheless a partial account of the 504 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY CO W. > I— I <1 K Q m P^ <) H ^^ m "^ P p^ Ph H O O P P5 o P VI hJ <5 , — , & c5 Eh + KO + Q r-f l-H T— 1 <1> CC 1— 1 f— 1 CO 1-1 CO ►— t t* 10 > l-H 02 im" ft S5 t— < •S ^ § § c« 1- •3 o O f^ S 1^ Z s" a> 0 O c3 • r-t ft CIS <; s -u 00 o 'fH &4 c CO 13 ft 0 0 p< ^ _o (3 fl 0-, H 'S o ■ o c3 • rH a •rH 1 0 3 5<; ^ *"* t-< O J3 ,G ^ f"^ k^ o 3 0 u 0 M s cs h ^ t-, 3 3 tH 0!] o o 51 o S 0 00 0 =h' 13 ^ Eh CO (h" u Ih c3 02 ■TS -^ 0 •t-i • iH 3 O S 3 P 0 3 3 u 3 fl % g rt 03 c« a 03 P. ft 3 0 o o o ^ o O O • (-( CO 13 0 r— ( ^ Eh Eh ^ ^ a be Eh £ 3 IS "rt IS 'o 0^ Qi 03 I>1 ^ s g s o o O s C^ 03 ^ a m CQ zc O O U CO iJ a Oh CO =c •ei s< i '« •2 CO 0 d ^ «0 p *** ?»i t« ■^ s &I g CM m S 2 Si. V. 5 5 o e 5; s S 5^ to |p4 £ ^ 5^ 0 ^ s. =0 =c « _S rSi fl (D to S 0 •- p "^ o o ^ oa 0 ■Si 0 e a- O V-" ^ ^ ^ C 0 '^ fc; %. « O o a> □D < > cS i-i 0 oj o (D a> M hq >^ TRIASSIC REPTILE FOOTPRINTS 505 Brunswick fauna and its ecological relationships can be formu- lated from the known footprints and bones. Faunal content. In the preceding attempts to determine the zoological positions of the trackmakers, failure to achieve a correlation between footprint and reptile has been due less often to inadequacy of the footprint material than to lack of com- parable skeletal material. This lack is twofold in origin. First, the Triassic reptiles as known are largely a spectral crew of footless bodies and disembodied heads which can be made whole only by persistent and lucky collecting. As our knowledge of these forms increases, closer comparisons with the tracks will no doubt become possible. Second, and more fundamental than the fi-agmentary preservation of individual taxa, is the fragmentary and selective preservation of the original fauna. At present we know more types of Brunswick tetrapods from their footprints than from skeletal material. Aside from a dinosaur leg excavated by laborers but never collected (Colbert, 1946, p. 231), the only genera known are the metoposaurid amphibian Eupelor, the procolophonid cotylosaur Ilypsorjnathus, the phytosaur Clepsy- saurus, and the aetosaurid pseudosuchian Stegomus. Other Brunswick footprint assemblages — e.g. from Sanatoga and Collegeville, Pennsylvania, and Tumble Falls and White- hall, New Jersey — which would augment the fauna have been omitted here because of the unreliability of old identifications. Figured specimens of Anchisauripus [" Otouphepus"] poolei (Bock) and RhyncJwsauroides ["Orthodactylus"] howelli (Bock) from Schwenkville, Pennsylvania, which Wilhelm Bock has kindly made available for study are specifically indeter- minate but appear to record a eoelurosaur and a lacertoid rep- tile of sphenodontid aspect. The Schwenkville " Grallator cnnea- tus" (Bock, 1952, pi. 45, fig. 1) is surely not that genus but instead has the phalangeal pattern of Anchisauripus; it evi- dently represents another variety of eoelurosaur. Ecology. The Brunswick fauna is predominantly terrestrial but includes representatives of the aquatic biotope in the form of metoposaurs and phytosaurs. These were certainly fish-eaters and presumably fed on the coelacanths, semionotids, dictyopy- gids and palaeoniscids of the Newark lakes and streams. The presumed sphenodontid rhynchocephalians of Level B may (like 506 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Homoeosaurus) also have fed occasionally on fish, but the bulk of their diet (like that of Sphenodon) probably consisted of insects, other arthropods, grubs, and small tetrapods. Except for the cotylosaur Hypsognathus, a presumed herbivore, the other known members of the fauna were carnivorous. Lacertoids and procolophonids furnished food for small, agile pseudo- suchians of the Chirothcrium lulli type, and all three groups formed the prey of the raptorial coelurosaurs. Whether the larger Milford pseudosuchians were predators or carrion-feed- ers, or both, cannot yet be determined. Any assessment of the Brunswick ecology must allow for a disproportionate percentage of aquatic animals in the skeletal record ; the footprint horizons, on the other hand, record a more representative proportion of terrestrial genera. More important, the footprint faunules constitute biocoenoses in a strict sense, while the skeletal remains occur as isolated finds or (at best) as bonebed thanatocoenoses of animals which may or may not have formed natural communities in life. Thus we can say with confidence that the reptiles of Level B, for example, constituted an animal community which occupied the Milford mudflat dur- ing a period measured in days. Our sample is too small, un- fortunately, to furnish significant data on the numerical com- position of the faunules. Survey of the Newark tetrapods. Undoubtedly the assemblage described above is only a fraction of the actual tetrapod fauna of Brunswick time. To afford a clearer perspective, the known terrestrial vertebrates of the entire Newark pro^'ince are listed with the Milford trackmakers in Table 4. How incomplete this faunal picture remains may be deduced from the mowj reptiles and amphibians, both genera and higher groups, which are known from the Upper Triassic of Arizona, Utah, New Mexico, Texas and Wyoming. Among these are the metoposaurid amphibian Eupelor [^Bucttneria, etc.] (Colbert and Imbrie, 1956), the large protorosaur Trilophosaurus, a vari- ety of phytosaurs, the armor-plated pseudosuchians Desmatosu- chus and Typothorax, the little ornithosuchid pseudosuchian Hes- perosuchus, the coelurosaur Coelophysis, and the dicynodont Pla- cerias. Many of these forms (or their relatives) must have in- habited the Newark depositional area. Still other types of TRIASSIC REPTILE FOOTPRINTS 507 tetrapods yet unknown, small forms in particular, completed the fauna. It would obviously be premature to attempt here a reconstruc- tion of late Triassie tetrapod faunas in space and time on the basis of skeletal and ichnological data now available. This is a major project to which several researchers are individually con- tributing, and in which the collaboration of invertebrate paleon- tologists, paleobotanists, and geologists is essential. Nevertheless the present study may serve to illustrate the contribution which one field of investigation, ichnology, can make to the solution of the general problem. APPENDIX I: THE NATURE OF OTOUPHEPUS A study of the type specimen of Otouphepus magnificus Cush- man, lent through the courtesy of the Museum of Science, Bos- ton, suggests that it is not the direct imprint of a foot but rather an impression transmitted to a surface some millimeters below the one on which the animal walked. This circumstance has softened the outlines and produced the illusion of a compact foot in which the digits were firmly united and surrounded by a web. The dark-colored web outline described by Cushman (1904, p. 155) proves to be a thin coat of gum which was readily removed with soap and water. Analysis of the pattern of articu- lar pads reveals a pedal structure which is well within the limits of Anchisaiiripus. A previously undescribed claw-mark occupies exactly the position of an Anchisauripus hallux imprint. The only plesiotype cited by number (Lull, 1953, p. 177) is morphologically indistinguishable from Anchisaurus tuberosus — a fact which Lull has pointed out in recent correspondence. In size and skeletal structure the unique specimen of 0. minor Lull (1915, p. 190) is a typical Grallator; its peculiarities are readily explained as artifacts of impression. (Latex molds of the two preceding specimens were kindly furnished by the Yale Peabody Museum.) A third species, 0. poolei Bock (1952, p. 407) has been based on an obscure and isolated footprint from the Brunswick forma- tion near Schwenkville, Pennsylvania. With Wilhelm Bock's gracious permission this specimen has been latex-molded and 508 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY MILFORD TAXA SKELETAL REMAINS TRACKMAKERS (as genera) (B) AMPHIBIA Stereospondyli: Metoposauridae Eupelor [Calamo'ps'] (S, G) Dictyoceplialus (DR) REPTILIA Cotylosauria : Procoloplionidae Eypsognathus (B) Rhynchocephalia (?) 2 fsphenodontids Pseudosuehia: 2 large pseudosuchians Ornithosuchidae Stegomosuchus (P) 1 ?+ ornithosuchid Aetosauridae Stegomus (B, NH) Phytosauria Clepsysaurus (L, B. G, PD, NH) Eutiodon (DR. ?G) 1 cf . Clepsysaurus Coelurosauria : 3+ coelurosaurs Ammosauridae Ammosaurus (P) Podokesauridae Podokesaurus (P) 1 cf . Coelophysis Prosauropoda: Thecodontosauridae Yaleosaurm (P) Thecodontosaurus (L. P) Cynodontia Dromatherium (DR) Microconodon (DR) INCEETAE SEDIS Gwyneddosaurus (L) Lysorocephalus (L) TRIASSIC REPTILE FOOTPRINTS 509 analyzed. An apparent structural formula of B, B, C, C allies this species with the Anchisauripus tuherosus-exsertus-parallelus species group. The species is probably distinct, but in the absence of adequate material its small size is the only characteristic which cannot be attributed to impression phenomena. If Ofouphepus Cushman (March, 1904) and Anchisauripus Lull (April, 1904) are synonymous the former takes priority. To abandon a well-established name which has been applied to several species comprising thousands of specimens, in favor of a name which has been applied to four specimens whose chief point of similarity is their obscurity, would be unreasonably legalistic. I have therefore (Baird, 1956) petitioned that the International Commission for Zoological Nomenclature exercise its plenary powers to suppress Ofouphepus magniflcus as a nomcn duhium ; this action automatically entails the suppression of Lull 's family Otouphepodidae. At the same time Anchisauri- pus sillimani and the family name Anchisauripodidae are to be placed on the official lists of names in zoology as nomina con- servanda. APPENDIX II: BRONTOZOUM The generic name Broniozoum, formerly applied to a number of species of theropod footprints from the Newark series, was erroneously reduced to synonymy in Hay's 1902 bibliography. In his monograph of 1904 Lull followed Hay's usage, and since that time the name has been in disuse. Broniozoum nevertheless remains a valid prior generic name for the type species of Anchi- sauripus Lull, and could legally be revived to replace that well- known generic name. The facts are as follows : Broniozoum Hitchcock, 1847, was proposed to include "five" species of which only three were cited by name : the new species TABLE 4. SURVEY OF THE NEWARK TETRAPODS Formations: Deep River coal measures (DR) of North Caro- lina; Stockton (S), Lockatong (L), Brunswick (B) of Pennsyl- vania-New Jersey basin; Gettysburg shales (G) of York County, Pennsylvania (Wanner, 1926) ; bed below Palisades diabase (PD) of New Jersey; New Haven (NH), Meriden, Portland (P) of Connecticut Valley. 510 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY parallelum,^^ sillimmiium (an invalid emendation of silUmani, 1843) and giganteuni. No type species was designated. No sub- sequent author, to my knowledge, has designated a type species, so all three remain available. A case of sorts could be made for the selection of parallelum through a broad construction of Article 30 A(a)(ii) of the International Rules of Zoological Nomenclature, which provides that the expression "n.g., n.sp.," can constitute a designation of type species. In this case such a construction is highly dubi- ous, particularly as it cannot be claimed to represent the intent of the original author : for Hitchcock evidently had no intention of designating a type species for any of his footprint taxa. Furthermore, parallelum is based on two specimens — neither designated as type — which are not conspecific. My analysis assigns one of these (Hitchcock's fig. 2 b) to Anchisauripus silU- mani and the other (fig. 2 a) to Grallator tenuis. To avoid confusion I designate the specimen represented by fig. 2 b as the type of parallelum, and reduce this species to subjective syn- onymy with sillimani. The second species, Brontozoum sillimani, is the type species of AnchisaMripus Lull, 1904 (see Baird, 1956, paragraph 6). Lull clearly erred in erecting a new genus with a type species for which the name Brontozoum was available. The third species originally assigned to Brontozoum, B. gigan- teum, was first described by Hitchcock (1836) as a species of Ornithichnites and was later transferred by him to Euhrontes (1845) and to Brontozoum (1847). In 1902 Hay designated it as the type of Euhrontes. Since according to Article 30 B(g)(i) of the International Rules the designation of a species as the type of one genus does not in itself preclude the selection of that species as the type of another genus, either sillimani or giganteum is eligible to be the type species of Brontozoum. If sillimani be selected then Anchisauripus (1904), an important name accepted for more than fifty years, must be replaced by Brontozoum (1847), a name generally rejected or ignored during the same period. If giganteum be selected, however, then Brontozoum becomes a 12 This species B. parallelum is not to be confused with Anchisauripus [Gralla- tor] parallelus (Hitchcock, 1865) although Hay (1902, p. 545) did so confuse it. TBIASSIC REPTILE FOOTPRINTS 511 junior objective synonym of Eubrontes (1845) through the iden- tity of their type species. I therefore designate Ornithichnites giganteus Hitchcock, 1836, as type species of the nominal genus Brontozoum, thereby syn- onymizing that genus with Eubrontes. This suppression of Brontozoum serves to maintain the familiar nomenclature of Lull's classic monographs of 1915 and 1953 which have done so much to bring nomenclatural stability into a confused field. SUMMARY Problem and method. Quarries in the Delaware valley near Milford, New Jersey, have yielded three faunules of reptile footprints at three horizons of the upper Brunswick formation (Newark series, Upper Triassic). Each determinable species was analyzed by comparison of outlines printed from latex molds of all adequately preserved specimens. Body proportions and locomotor habits of the traekmakers were deduced from track- ways (where available) and the pedal skeletons were recon- structed from the arrangement of articular swellings in the footprints. Comparison with other Triassic footprints furnished data on faunal distribution and stratigraphic correlation; com- parison of the restored skeletons with those of Triassic reptiles served to elucidate and enlarge the known Brunswick fauna. Reddish-brown siltstone layer. The youngest faunule (dis- cussed in a previous paper) records three reptilian genera as footprint species : an indeterminate small dinosaur, another re- sembling Coelurosaurichnus (sensu stricto) of the German Mid- dle Keuper and the Gettysburg shale of Pennsylvania, and an ornithosuchid pseudosuchian, Chirotherium lulli Bock, the last- known member of its genus. This faunule and that of Level B are the only American Chirotherium-dinossinr associations known. Level A (gray sandstone over shale). Grallator sulcatus, n. sp., a small dinosaur with digits II-III united and IV inde- pendent, is less digitigrade than its Connecticut Valley con- geners. Among eoelurosaurs the massive German Halticosaurus (as shown by pedal reconstruction) made footprints resembling Eubrontes; little Procompsognathus compares more closely with 512 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Grallator, especially G. cursorius; and Coelophysis correlates approximately with G. stdcatus. Level B (gray sandstone over shale). Three dinosaur genera occur: Anchisauripus milfordensis (Bock), a broad -based foot- print without functional hallux, found also in the Gettysburg shale of York County, Pennsylvania; A. parallelus (Hitchcock), only 6/11 the size of conspecific New England footprints; and an enigmatic small form of York County affinities. Taxonomie placement follows a new type of key l)ased on relative posi- tions of phalangeal and metatarso-phalangeal joints. Re-analysis of various TTpper Triassic theropod tracks by this method necessitates reassignment of several species and emphasizes the similarities between Grallator and Anchisauripus (sensu stricto), genera previously considered familially distinct but here united as Grallatoridae and correlated with coelurosaurian dinosaurs. Two large chirotheriids with hoof-like fifth pes digits represent a lineage previously known only from the German Middle Keu- per, here designated the brachychirotherian group. In Chiro- therium parvuni (C. II. Hitchcock) (=C. copei Bock), narrow claws on pes digits I-III were carried well above the distal pad and directed obliquely laterad ; pes digit I was robust. In the new species C. eyerniani straight, hea\y claws I-IV lay at sole level and digit I was weak. Skeletal reconstructions of the large- manus C. harthii (cosmopolitan. Lower Triassic) and the small- manus C. lomasi (Keuper of England) contrast with those of the Milford brachychirotherians. All three Milford chirotheriids show parallel evolutionary tendencies — characteristic of late Triassic species — toward functional tetradactyly through elim- ination of lateral propping function in pes digit V. Inefficient locomotion and increased dinosaurian competition probably con- tributed to the extinction of Chirotherium. Pseudosuchians cor- relative with the large Milford species are unknown. Discovery of additional material permits redescription of "Otozoiim" lineatus Bock as a new type of footprint, Apatopus n. gen., correlated with a phytosaur such as Clepsysaurus — the first phytosaur footprints known. Crocodilian in trackway pat- tern except for the forward-turned manus, Apatopus differs from all crocodilians in its long fourth and functional fifth pes digits. Webs apparently connected the claw baises. Both adults TRIASSIC REPTILE FOOTPRINTS 513 and juveniles are represented, one adult having a gleno-acetabu- lar length of 52 cm. Lacertoid footprints of the genus Rhynchosauroides [Akro- pus'\ occur in the Brunswick, Gettysburg, and Lockatong forma- tions of Pennsylvania-New Jersey, the Dockum of New Mexico, and the Keuper of England and Italy. Rhynchosauroides hyper- hates, n. sp., represents an agile reptile some 75 cm. long with a gleno-acetabular length of 18 cm. Pes regularly overstepped manus, and both were digitigrade with the metapodio-phalangeal joints carried clear of the ground. A smaller Milford species is inadequately known. Affinities appear to lie with the spheno- dontid Rhynchocephalia or the Lacertilia. Correlative age. The Milford faunules are akin to those of the Pennsylvania-New Jersey basin and the German middle to late Middle Keuper but not (as sometimes held) to the familiar foot- print assemblages of the Portland and Meriden formations of New England. Only one of the dinosaurian species also occurs in the Connecticut Valley; while Chirotherium, Apatopus, and Rhyncliosauroides are definitely absent from the abundant and thoroughly studied Portland and Meriden faunas. No topo- graphic or ecologic barriers would seem to have prevented inter- regional migration. Presence of the phytosaur Clepsysaurus and the pseudosuchian Stegomus in the lowest formation (New Haven arkose) of the Connecticut Trias as well as in the Bruns- wick formation, and the possible correlation of the three episodes of igneous extrusion in New Jersey with the three lava flows of Meriden time, suggest that the Brunswick antedates the Port- land. The Milford footprint beds of the upper Brunswick ap- pear to correlate best with the New Haven arkose. Reptiles of the Brunswick formation. Known skeletal remains include undetermined dinosaurs, the cotylosaur Hypsognathus, the aetosaurid pseudosuchian Stegomus, and the phytosaur Clepsysaurus. To this fauna the Milford footprints add three or four types of coelurosaur, one small ornithosuchid and two large undetermined pseudosuchians, a phytosaur, and two lacertoids. From these combined faunal lists the ecological picture may at least be glimpsed, though only a fraction of the Brunswick fauna is yet known. The present study serves to illustrate the possible contribution of ichnology to the solution of Triassic faunal and stratigraphic problems. 514 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Appendix I. The nominal genus Otouphejms proves to be a catch-all for obscurely preserved dinosaur footprints. Its sup- pression as a nonien dubium is proposed in order to prevent possible displacement of Anchisauripus as a junior synonym. Appendix II. The long-disused name Brontozoum also threat- ens the stability of the later name Anchisauripus. Brontozoum is therefore cashiered by the designation of a type species which is also the type species of the earlier genus Euhrontes, so that the genera become objectively synonymous. REFEKENCES Abel, Othenio 1912. Grundziige fler Palaeobiologie der Wirbeltiere. Stuttgart: Schweizerbart, xv + 708 pp. B.\iRD, Donald 19r)4. Chiroihcrium lulli, a pseudosuchian reptile from New Jersej'. Mus. Comp. Zool. Bull., vol. Ill, pp. 1(33-192, 2 pis. 1956. Proposed use of the Plenary Powers (i) to protect the generic name Anchisuuripus Lull, 1904, by supjiressing the generic name Otouphepus Cushman, 1904, and (ii) to suppress the specific name magnificus Cushman, 1904, published in combination with the latter generic name (Class Reptilia: Theropoda [Ichnites]). Bull. Zool. Nomenclature, vol. 12, pp. 221-224. Beasley, H. C. 1905. Report on footprints from the Trias, Part II. British Assoc. Adv. Sci. Rept. for 1904, pp. 275-282, pis. 3-5. Beurlen, Karl 1950. Neue Fahrtenfunde aus der Frankischen Trias. Neues Jahrb., Monatshefte (B), pp. 308-320. Bock, Wilhelm 1952. Triassic reptilian tracks and trends of locomotive evolution. Jour. Paleontology, vol. 26, pp. 395-433, pis. 41-50. 1952A. New eastern Triassic ginkgos. Wagner Free Inst. Sci. Bull., vol. 27, pp. 9-14, pi. 1. Branson, E. B. 1947. Triassic (Chugwater) footprints from Wyoming. Jour. Paleon- tology, vol. 21, pp. 588-590. TRIASSIC REPTILE FOOTPRINTS 515 Camp, C. L. 1930. A study of the phytosaurs, with description of new material from western North America. Univ. California !Mem.. vol. 10, 161 pp., 6 pis. Colbert, E. H. 1046. Hi/psoffnatJius, a Triassic reptile from New Jersey. Am. Mus. Nat. Hist. Bull., vol. 86, pp. 225-274, pis. 25-33. 1947. Studies of the phytosaurs Machacroprosopus and Futiodon. Am. Mus. Nat. Hist. Bull., vol. 88, pp. 53-96, pis. 3-10. Colbert, E. H., and R. G. Chaffee 1941. The type of Clepsi/sauru/i peniisylvanicns and its bearing upon the genus Eutiodoti. Acad. Nat. Sci. Philadelphia Notulae Naturae, no. 90, 19 pp. Colbert, E. H., and John Imbrie 1956. Triassic metoposaurid amphibians. Am. Mus. Nat. Hist. Bull., vol. 110, pp. 399-452, pis. 25-28. Colbert, E. H., and C. C. Mook 1951. The ancestral crocodilian rrotosuchus. Am. Mus. Nat. Hist. Bull., vol. 97, pp. 143-182, pis. 9-16. CUSHMAN, J. A. 1904. A new foot-print from the Connecticut valley. Am. Geologist, vol. 33, pp. 154-156, pi. 6. Endo, Riuji 1940. A new genus of Thecodontia from the Lycoptera beds in Man- choukuo. Central Nat. Mus. ^lanehoukuo Bull. 2, pp. 1-14, 2 pis. Eyerman, John 1886. Footprints on the Triassic sandstone (Jura-Trias) of New Jersey. Am. Jour. Sci., vol. 131, p. 72. 1889. Fossil foot-prints from the Jura( ?)-Trias of New Jersey. Acad. Nat. Sci. Philadelphia Proc, pp. 32-33. GU.M0RE, C. W. 1924. Collecting fossil footprints in Virginia. Smithsonian Misc. Coll., vol. 76, pp. 16-18. Gregory, J. T. 1945. Osteology and relationships of Trilophosaurus. Univ. Texas Pub. 4401, pp. 273-359, pis. 18-33. 516 BULLETIN: MUSEUM OP COMPARATIVE ZOOLOGY Grier, N. M. 1927. The Hitchcock lecture upon ichnology, and the Dartmouth College ichnological collection. Am. Midland Nat., vol. 10, pp. 161-197. Haarlander, Wilhelm 1938. Bericht iiber einige Funde aus dem mittelfrankischen Keuper. Oberrhein. geol. Ver. Jahresber. (N. T.), Bd. 27, pp. 1-8, pis. 1-2. Hat, O. p. 1902. Bibliography and catalogue of the fossil vertebrata of North America. U. S. Geol. Survey Bull. 179, 868 pp. Heilmann, Gerhard 1927. The origin of birds. New York: Appleton, vii + 210 pp. Heller, Florian 1952. Eeptilfahrten-Funde aus dem Ansbacher Sandstein des Mittleren Keupers von Franken. Geol. Blatt. NO-Bayern, vol. 2, pp. 129-141, pi. 9. 1956. Ehynchocephalen-Fahrten aus dem Mittleren Keuper bei Hass- furt am Main. Ibid., vol. 6, pp. 50-55, pi. 2. HicKOK, W. 0. 4th, and Bradford Willard 1933. Dinosaur foot tracks near Yocumtovvn, York County, Pennsyl- vania. Pennsylvania Acad. Sci. Proc, vol. 7, pp. 55-58. Hitchcock, C. H. 1889. Eecent progress in ichnology. Boston Soc. Nat. Hist. Proc, vol. 24, pp. 117-127. Hitchcock, Edward 1836. Oruithichnology. Description of the footmarks of birds {Orni- thichnites) on New Red sandstone in Massachusetts. Am. Jour. Sci., vol. 29, pp. 307-340, 3 pis. 1843. Description of five new species of fossil footmarks, from the red sandstone of the valley of Connecticut River. Assoc. Am. Geologists & Naturalists Trans., pp. 254-264, pi. 11. 1845. An attempt to name, classify and describe the animals that made the fossil footmarks of New England. Assoc. Am. Geol- ogists & Naturalists Proc. 6th Ann. Meeting, pp. 23-25. 1847. Description of two new species of fossil footmarks found in Massachusetts and Connecticut, or of the animals that made them. Am. Jour. Sci. (2), vol. 4, pp. 46-57. TRIASSIC REPTILE FOOTPRINTS 517 ISfiS. Ichnology of New England. A report on the sandstone of the Connecticut valley, especially its fossil footmarks. Boston : Wni. White, 220 pp., 60 pis. 186,"). Supplement to the ichnology of New England. Boston: Wright & Potter, X 4- 96 pp., 20 pis. HUHNE, FRIEDRICH VON 1913. Beobachtungen iiber die Bewegungsart der Extremitaten bei Krokodilen. Biol. Centralbl., vol. 33, pp. 468-472. 1921. Neue Pseudosuchier und Coelurosaurier aus dem Wiirttemberg- ischen Keuper. Acta Zoologica, vol. 2, pp. 329-403, pis. 1-4. 1934. Ein neuer Coelurosaurier in der thiiringischen Trias. Palaeont. Zeitschr, vol. 16, pp. 145-170, pis. 13-16. 1938. Zur Bestimmung von Fussspuren der Protorosauriden und Khynehosauriden. Zentralbl. Mineralogie (B), no. 2, pp. 58-64. 1941. Die Tetrapoden-Fahrten im toskanischen Verrucano und ihre Bedeutuug. Neues Jahrb., Beilage-Bd. 86 (B), pp. 1-34, pis. 1-8. 1944. Die fossilen Reptilien des siidamerikanischen Gondwanalandes. Miinchen: C. H. Beck, viii -f- 332 pp., 38 pis. Jaegek, Ellsv^torth 1948. Tracks and trailcraft. New York: Macmillan, x -f- 381 pp. Kaup, J. J. 1835. Thier-Fahrten von Hildburghausen : Chirotherium oder Chiro- saurus. Neues Jahrb., pp. 327-328. KUHN, OSKAR 1936. Weitere Parasuchier und Labyrinthodonten aus dem Blasen- stein des mittleren Keuper von Ebra«h. Palaeontographiea, vol. 83 (A), pp. 61-98, pis. 8-13. Lesley, J. P. 1889. Dictionary of the fossils of Pennsylvania and neighboring states, vol. 2. Pennsylvania (2d) Geol. Survey Eept. P 4, pp. 439-914 -I- i-x. Lull, R. S. 1904. Fossil footprints of the Jura-Trias of North America. Boston Soc. Nat. Hist. Mem., vol. 5, pp. 461-557, pi. 72. 1915. Triassic life of the Connecticut valley. Connecticut Geol. Nat. Hist. Survey Bull. 24, 285 pp., 3 maps, 12 pis. 1917. The Triassic fauna and flora of the Connecticut Valley. U. S. Geol. Survey Bull. 597, pp. 105-127, pis. 3-4. 1942. Triassic footprints from Argentina. Am. Jour. Sci., vol. 240, pp. 421-425, pi. 1. 518 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1942A. Chugwater footprints from Wyoming. Am. Jour. Sci., vol. 240, pp. 500-504. 1953. Triassic life of the Connecticut Valley (revised). Connecticut Geol. Nat. Hist. Survey Bull. 81, 331 pp. Lyman, B. S. 1894. Some New Bed horizons. Am. Plulos. Soc. Proc, vol. 33, pp. 192-215. 1895. Eeport on the New Red of Bucks and Montgomery Counties. Pennsylvania (2d) Geol. Survey Summary Pinal Eept., vol. 3, pt. 2, pp. i-iii -f 2589-2638, pis. 596-611. Maidwell, F. T. 1911. Notes on footprints from the Keuper of Runcorn Hill. Liverpool Geol. Soe. Proc, vol. 11, pp. 140-152, pis. 4-6. 1914. Notes on footprints from the Keuper, II. Liverpool Geol. Soc. Proc, vol. 12, pp. 53 71, pis. 3-5. McGregor, J. H. 1906. The Phytosauria, with especial reference to Mystriosuchu^ and Rhyticlodon. Am. Mus. Nat. Hist. Mem., vol. 9, pp. 27-101, pis. 6-11. NorCSA, FUANZ 1923. Die Familien der Reptilien. Fortschr. Geologie u. Palaeontol- ogie, vol, 2, pp. 1-210, pis. 1-6. Owen, Richard 1842. Description of an extinct lacertian reptile, Ehynchosaurus articeps (Owen), of which the bones and foot-prints character- ize the upper New Red sandstone at Grinsill, near Shrewsbury. Cambridge Philos. Soc Trans., vol. 7, pp. 355-369, pis. 5-6. 1842 A. Description of parts of the skeleton and teeth of five species of the genus LabyrintJiodoti . . . with remarks on the probable identity of tlie Cheirotheriuni with this genus of extinct Batrachians. Geol. Soc London Trans. (2), vol. 6, pp. 515-543, pis. 43-47. Peabody, F. E. 1948. Reptile and amphibian trackways from the Lower Triassic Moenkopi formation of Arizona and Utah. Univ. Calif. Dept. Geol. Sci. Bull., vol. 27, pp. 295-468, pis. 23-45. 19r)5. Occurrence of Chirotherium in South America. Geol. Soc Am. Bull., vol. 66, pp. 239-240. TRIASSIC REPTILE FOOTPRINTS 519 Pf.yer, Berxhard 1937. Die Triasfauua dcr Tessiiier Kalkalpou, XII. Macrocnemvs hassanii Nopcsa. Schweizer. Palaeont. Gesell. Abh., vol. 59, pp. 1-140, pis. 55-63. RlCH.\RDS, H. G. 1953. Record of the rocks. New York: Rouald Press, xiii + 413 pp. Robin-son, P. L. 1955. Specimens of Triassic reptiles obtained from fissures in the Slickstones quarry, Gloucestershire. Geo!. Soc. London Proc, Xo. 1527, pp. 113-115. KlTHLB VON LnjKNSTERN, HUGO 1938. Fahrten ans dem Blasensandstein (km 4 zeta) des mittlereu Keupers von Siidthiiringen. Neues .Tahrbuch, Beilage-Bd. 80 (B), pp. 63-71, pis. 2-3. 1939. Fahrten und Spuren un Chirotheriurasandstein von Siidthiirin- gen. Fortsehr. Geologie u. Palaeontologie, vol. 12, pp. i-iv + 293-387, pis. 1-12. RrscoNi, Carlos 1951. Rastros de patas de reptiles permicos de Mendoza. Soc. Hist. Geog. Cuyo Rev., vol. 3, no. 3, pp. 1-14. Sawin, H. J. 1947. The pseudosuchian reptile Typothorax meadi, new species. Jour. Paleontology, vol. 21, pp. 201-238, pi. 34. SCHAEFFER, BOBB 1941. The morphological and functional evolution of the tarsus in amphibians and reptiles. Am. Mus, Nat. Hist. Bull., vol. 78, pp. 395-472. SCHMIDTGEN, OtTO 1927. Tierfahrten im oberen Rotliegenden bei Mainz. Palaeont. Zeitschr., vol. 9, pp. 101-107. Shikama, Tokio 1942. Footprints from Chinchou, Manchoukuo, of Jeholosauripus, the Eo-Mesozoic dinosaur. Central Nat. Mus. Manchoukuo Bull., no. 3, pp. 21-31, pis. 10-15. SOEBGEL, WOLFOANG 1925. Die Fahrten der Chirotheria. Eine palaobiologische Studie. Jena: G. Fischer, vii + 92 pp. 520 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Stose, G. W., and A. I. Jonas 1939. Geology and mineral resources of York County, Pennsylvania. Pennsylvania Geol. Survey (4), Bull. C 67, 199 pp., map. Von Engeln, O. D., and K. E. Caster 1952. Geology. New York: McGraw-Hill, xii + 730 pp. Wanner, Ateeus 1889. The discovery of fossil tracks, algae, etc., in the Triassic of York County, Pennsylvania. Pennsylvania Geol. Survey Ann. Kept, for 1887, pp. 21-35, pis. 3-13. Wanner, H. E. 1926. Some additional faunal remains from the Trias of York County, Pennsylvania. Acad. Nat. Sci. Philadelphia Proc, vol. 78, 14 pp., pi. 3. Woodward, A. S. 1907. On Ehynchosaurus articeps (Owen). British Assoc. Adv. Sci. Kept, for 1906, pp. 293-299, pi. 2. PLATES PLATE 1 Fig. 1. Grallator sulcctus Baird, n. sp. Type, MCZ 215, a left pes imprint. (As all specimens are natural casts, left and right appear reversed.) Fig. 2. Chirotherium i)arvum (C. H. Hitchcock). Left manus-pes set, a composite of MCZ 209 (pes) and MCZ 210 (manus). b' i'i-f cm. 10 _!*;- «> PLATE 1 PLATE 2 Chiro1]n liiim eiitrmani Baird, ii. sp. Three views of the type, MCZ 134, a left pes ini])rint. Xote the fnnows along the lateral Ijorder of hoof-like digit V. r < TLATE Apdtopii.s liiirdtiis (Boek). Fig. 1. Eight mnnus-pes set of the type indi- vidual, MCZ •212. Pes digits IV and V failed to impress. Fig. 2. Well- preserved left pes imprint, MCZ 213. Obliquely left of claw II is a manus print of Ithyncliosaurnidrs sp. (cf. Figure 9 C). PLATE i Apatopus Uneatus (Boek). Fig. 1. Juvenile left iiianus-pes set, MCZ 214; note the beaded squaniation. Fig. 2. Juvenile left pes imprint, AMNH 2259, rotated after implantation. Fig. 3. Small right pes imprint, AMNH 2258. lJ^f!l?fi.fliiiiliii¥r[Wb^iy 4::a.^^ -.. ■'■, ..^ PLATE 4