^^ US HARVARD UNIVERSITY i ^ Library of the Museum of Comparative Zoology BULLETIN OP THE MUSEUM OF COMPAKATIVE ZOOLOGY AT HARVARD COLLEGE, IN CAMBRIDGE VOL. 119 CAMBRIDGE, MASS., U. S. A. 1958-1959 TiiK Cosmos Pkkss, Inc. CAMHHincK. Mass., U.S.A. 1^ CONTENTS PAGE No. 1. — Revision of Five Aerioan Snake Genera. By Ai-tliur TjOvoridpT. July, 1!)5(S .... 1 Xo. '2. — Fi)V\i New Ra.iids ekom the Gulf of Mexico. By Henry l>. Bifjclow and William G. Schroeder. July, l!)."i8 1!)!) No. 3. — The General Histology and Topographic jMicro- anatomy of Attstijalorius glabrattts. By Gliia- Tung Pan. (18 plates). July, 1958 . ' . . 285 No. 4. — Stttdtes on the Ant Fattna of Melanesia TIT. Kiiytidoponera in Western Melanesia and the Moluccas. IV. The Trihe Ponehini. By E. 0. Wilson. August, 1958 .801 No. 5. — A Ne\v Species of Ghelid Turtle, Phrynops (Batrachemys) datilt, from Golombia. By Rainer Zangerl and l^'rcd Modem. (2 plates). August, 1958 ;17;] No. G. — Taractes asper and tiii: Systematic Relation- SHU's OK THE Steinegertidai-: and Trachybery- ciDAE. By Giles \V. Mead and G. E. Maul. (1 plate). Geto))er, 1958 ;W1 No. 7. — Additions to the Pleistocene Mammalian Fauna FROM Melboitrne, FLORIDA, liy Glaytou E. Ray. November, 1958 4P.) No. 8. — Stitdies on the MoRPHOLO(iY and Function of the Skull in the Boidae (Serpentes). Part 1. Granial Differences between Python sebae and Epicrates cenchris. By T. II. Frazzetta. Jan- uary, 1959 . . . " 451 Xo. 9. — The Genus Tetragnatha (Aranbae, Argiopidae) IN Michigan. By Arthur M. Ghickering. Februai-v, 1959 . . ' . . . . . . ' . 473 Bulletin of the Museum oi Comparative Zoology AT HARVAED COLLEGE Vol. 119, No. 1 REVISION OF FIVE AFRICAN SNAKE GENERA Bv Arthur Loveridge CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM July, 1958 Publications Issued by or in Connection WITH THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE Bulletin (octavo) 1863 — The current volume is Vol. 119. Breviora (octavo) 1952 — No. 89 is current. Memoirs (quarto) 1864-1938 — Publication was terminated with Vol. 55. JoHNSONiA (quarto) 1941 — -A publication of the Department of MoUusks. 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 oi Comparative Zoology AT H A E V A R D C O L Tj E G E Vol. 119, No. 1 REVISION OF FIVE AFRICAN SNAKE GENERA By Arthur Loveridge CAMBEIDGE, MASS., U. S. A. printed for the museum July, 1958 No. 1 — Revision of Five African Snake Genera By Arthur Loveridge CONTENTS Page Tntvodiu'fioii iind Acknowledgments 1 (lenus Lyeodonoinorphu!^ Fitzinger (key on page 7) 6 Genus Kaivicitcrcs Loveridge (key on page 26) 25 (ienus P]iilflt]iaiu7iu.s Smith (key on page 52) 49 (ienus Frosymna Gray (key on page 129) 127 Genus Chilorliinopliis Werner (key on page 169) 168 Bil)liography 179 INTRODUCTION AND ACKNOWLEDGMENTS This paper represents the author's last attempt at revis- ing snake genera. The five dealt with here were selected because of the contused status of their component species, and this re- (fuired elucidating in connection Avith the report on Nyasaland reptiles. The work was undertaken about 1950. The typescript was then laid aside with the intention of borrowing certain snakes on which (juestionable records were based, also in the faint hope of completing the African COLUBRIDAE, to which family these five genera belong. Eleven other colubrid genera have been dealt with already in three contrilnitions to this Bulletin published in 1939, 1940 and 1944. The procedure and format of the current contribution is essen- tially that of its predecessors, i.e. an attempt to synopsize all pertinent information appertaining to every species assigned to the genera under revicAv. In the present instance the period covered is from 1880 to 1956, though the last year is possibly incomplete. However, in the case of Lijcodonomorphus Fitzinger, with which I synonymize (Hypliolycus Giinther, I have treated the 4 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY little-known^ hicolor of Lake Tanganyika in greater detail than usual. This is due to the collaboration of C. J. P. lonides, Esq., who generously procured more than 100 specimens of this aquatic reptile for study. The astonishing variation displayed by this fine series is published here for the first time. One might add that Mr. lonides also submitted comparably large quantities of at least one species for each of the four remaining genera. Others in Africa who have aided this investigation by sending in material recently, are Messrs. D. C. Broadley (Southern Rho- desia), D. Vesey-FitzGerald (Northern Rhodesia), M. C. Lesage (Ghana) and J. S. Owen (Sudan). During, and subsequent to, 1950 my esteemed colleague Dr. R. F. Laurent of the Musee Royal du Congo Beige has added several interesting species to the genera Lycodononiorphus and Prosymna. AVith the latter genus, Laurent's extensive AYest African material enabled him to clarify the situation respecting the races of ambigua occurring in the Belgian Congo. Though time and the accumulation of more material may demonstrate that two of these forms are untenable, I have accepted his ar- rangement except in one instance. Fortunately for me Laurent's researches prevented my merging a. hocagii with a. ainbigiia, something I was preparing to do because of the confusing mis- identifications of hocagii in the literature, as maj^ be seen from its synonymy. In acknowledgement of this, and as a tribute to Laurent's contributions to African herpetology, I take pleasure in naming the only novelty described in this paper : Prosymna meleagris laurenti subsp. nov. Circumstances have made it (juite impossible for me to borrow paratypes of Laurent's new forms as I should have done if not so pressed for time. In other respects I have endeavoured to bring these revisions up to date as of 1956. At the time {ca. 1950) I first did them 1 borrowed much material or pestered herpetological correspondents for information respecting speci- mens in their care. For either one or the other I am indebted to the undermentioned colleagues and gladly avail myself of this opportunity of expressing my thanks. 1 This was written before seeing Laurent's 19.56 paper listing 133 specimens I LOVERIDGE: AFRICAN SNAKE GENERA 5 J. R. Bailey (Duke University) ; J. C. Battersby (British Museum) ; C. M. Bogert (American Museum of Natural His- tory) ; G. S. Cansdale ( Lynda! e Avenue, London) ; B. S. Chau- han (Zoological Survey of India) ; D. M. Cochran (United States National Museum) ; R. Conant (Zoological Society of Philadelphia) ; J. Eiselt (Naturhistorische Museum, Vienna) ; V. F. FitzSimons (Transvaal Museum) ; J. Guibe (Paris Mu- seum) ; N. Ilartweg (Michigan University Museum) ; A. Holm (Zoological Museum, Uppsala) ; R. F. Laurent (Musee du Congo Beige) ; R. Mertens (Senckenberg Museum) ; M. G. Netting (Carnegie Museum, Pittsburgh) ; H. W. Parker (British Mu- seum) ; J. A. Peters (Brown University, Providence) ; C. R. S. Pitman (Chelsea, London) ; M. Poll (Musee du Congo Beige, Tervueren) ; C. H. Pope (when at Chicago Nat. Hist. Mus.) ; H. Rendahl (Naturhistoriska Riksmuseet) ; W. Rose (South African Museum, Capetown) ; K. P. Schmidt (Chicago Natural History Museum) ; B. Shreve (Museum of Comparative Zoology) ; R. H. Smithers (National Museum, Bulawayo) ; 0. van Straelen (Li- stitut Royal de Belgique) ; P. E. Vanzolini (Departmento de Zoologia, Sao Paulo) ; H. Wermuth (Zoologisches Museum, Ber- lin) ; 0. Wettstein (Osterreichische Akademie, Vienna) ; E. E. Williams (Museum of Comparative Zoology) ; G. F. de Witte (Listitut Royal de Belgique). When necessity arose to refer to this borrowed, or other, ma- terial, the institution where it may be found is indicated by the following abbreviations : A.M.N.H., American Museum of Natural History, New York; B.M., British Museum (Natural History), London; CM., Car- negie Museum, Pittsburgh; C.N.H.M., Chicago Natural History Museum. Chicago; I, lonides field number; I.R.B., Institut Royal de Belgique; M.C.Z., Museum of Comparative Zoology, Cam- bridge; P.M., Museum National d'Histoire naturelle, Paris; R.G.M.C, Musee Royal du Congo Beige, Tervueren; S.M.F., Senckenbergische Naturforschende Gesellschaft, Frankfurt, a.M. ; S.R.M., Southern Rhodesia Museum, Bulawayo; T.M., Transvaal Museum, Pretoria; U.S.N.M., United States National Museum, Washington. Localities. An asterisk (*) in front of a place name implies that a specimen or specimens from said locality is either in the collection of the Museum of Comparative Zoology or has been 6 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY examined for this revision — as in the ease of types studied at the British Museum in 1951. For one species only {Lycodono- morphus r. rufulus) the asterisk is placed after the locality to indicate that the material from these places is in the Transvaal Museum and was examined for me by Dr. V. F. FitzSimons with his customary kindness. Bihliography. Owing to the spate of semi-popular books on snakes that have appeared in recent years, it is becoming in- creasingly difficult to know what should, or should not, be in- cluded. Where a date is followed by a letter of the alphabet it indicates that during the year cited the author in question pub- lished more than one paper on African herpetology. The letter has chronological significance in a more comprehensive bibliog- raphy of African Herpetology (1880-1953) which it is hoped may be published in the not too-distant future. Family COLUBRIDAE Genus LyCODONOMORPHUS Fitzinger 1843. Lycodunoinuipltus Fitzinger, Syst. Eept., p. 27. Type by original designation: Coronella rufula Schlegel =: Coluber rufulus Lichten- stein. 1848. Lycodontomorphiis Agassiz, Nomen. Zool. Index Universalis, p. 628. Emend, for Lycodoiiomorphus Fitzinger. 1863. NeusteropMs Giinther, Proc. Zool. Soc. London, p. 16, footnote. Type by monotypy : Natrix laevissima Giinther. 18931j. Ablahophi.s Boulenger, Cat. Snakes Brit. Mus., 1. p. 318. Type by monotypy : Coluber rufulus Lichtenstein. 1894b. Glyphohjcus Guntlier, Proe. Zool. Soc. London, p. 629. Type l)y monotypy : G. bicolor Giinther. 1924b. Xerophidion. Werner, Sitzb. Akad. Wiss. Wien, 133, Abt. 1, p. 53. Type by monotypy: A. hypsirhinoides Werner = Glyphohjcus bicolor Giinther. Definition. Maxillary teeth 18-25," small, sube(iual ; mandibular teeth longest anteriorly. Head slightly distinct from neck; eye 2 III siviuK lS-19 fur bicolor, possibly Cott ('1935, p. 965) failed to allow for a mlssinfe' tooth ; there are clearly 19 in a ? sluill (M.C.Z. 54091) and appear to have l)eeii 'JO in another (M.C.Z. ;;i>07t!). Giinther said "about 21" for the type, while l{oulenj;er (lS9(;d. p. <)15) merely stated that the dentition of Gli/pliolinus was similar to that of Laiiiinoiihix and Jiotliroli/cus. However, Laurent (19541), P 43) jiives 19-20 teeth for liotli his .s. snbtaciiiutii)^ and s. upcmhac. LOVERIDGE : AFRICAN SNAKE GENERA < moderate, with round or vertically subelliptic pupil; nostril directed upwards, in a semi-divided or divided nasal; a loreal; preoculars 1, rarely 2; lateral head shields separated by an in- conspicuous groove from the upper labials. Body cylindrical ; scales smooth, with or without apical pits, in 19-25 rows ; ventrals rounded; anal entire, rarely divided. Tail moderate, tapering; subcaudals paired. Hemipenis of male not (hicolor and siibtaeniatus) or distally bifurcate (rufulus) ; sulcus spermaticus forked. Hypapophyses present posteriorly in the vertebral column. Range. African lakes and rivers south of 3° S. Remarks. For earlier comments on the status of this genus and the number of pterygoid teeth, see Loveridge (1953e, p. 253) and Laurent (1954b, pp. 38-43). As Laurent's admirably described species and its race bridge the gap between hicolor and laevissi- mus in the west, as do r. mlanjensis and r. tvhytii between hicolor and rufulus in the east, there no longer remains any valid reason for retaining Glypholycus as a distinct genus. Bogert, on whose work (1940, pp. 18-19) I have depended in respect to dentition, hemipeneal and hypapophj-seal characters, concurs in this view. It is true that in hicolor the 4th labial is the only one to normally enter the orbit, whereas in all the other species both 4th and 5th are normal, and the 4th only is very exceptional. Such a character, however, cannot be considered of generic sig- nificance. Key to the Species Our knowledge of the full range of ventral and subcaudal counts is so incomplete for certain forms, while in others the known counts display so much overlapping, that it appears next to impossible to devise a key that reflects relationships. However, if ranges are taken into account, and the table of statistical data that follows the key is consulted, the identification of specimens should be greatly simplified. 1. Midbody scale-rows 23, rarely 25 (25 present in only 2 of the 110 snakes examined) ; range: only Lake Tanganyika hicolor (p. 9) Midbody scale-rows 21 to 23 (21 chiefly in ,J <5 , 23 in 9 $); range: western Angola and western Belgian Congo 2 Midbody scale-rows 19 to 21 ; range : eastern Belgian Congo and southern Nyasaland 3 8 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Midbody scale-rows 1'.), rarely 'Jl (21 present in only 2 of about 20 records of Inevi^sinuis) ; range: southern Tanganyika Territory; northern Nyasaland and Africa south of the Zambezi 4 2. Subcaudals 29-40 (9 9 29-34; c?40); ventrals 172-188 ($ 172; 9 9 17.')-188) ; range: southeastern Belgian Congo s. wpemhac (p. 12) Subcaudals 41-58 (9 9 41-48; $i 52-58); ventrals 175-193 {$ i 175-180; 9 9 189-193); range: western Belgian Congo and western Angola .s. suhtaeniaius (p. 13) 3. Throat of adult whitish to gray ; belly mostly black sparsely flecked with cream; tail below entirely, or almost entirely, black (a young one under 230 mm. in total length, pure white; juveniles from 250-600 mm. gradually assume adult coloring, that is of 600-800 mm. speci- mens) ; range: montane streams of eastern Belgian Congo from north end of Lake Tanganyika to region of Lake Upemba r. Icleupi (p. 14) Throat of adult to anus immaculate white ; tail from anus to near its tip white -with a dusky median line; range: montane streams of Zomba, Mlanje and Cholo Mountains, southern Nyasaland r. vtlanjensiit (p. 16) 4. Subcaudals less than 50 (but that of S $ unknown); range: southern Tanganyika Territory (where thej- approach leleupi in color) to north- ern Nyasaland ; possibly also Charre, Mozambique r. ivhytii (p. 17 ) Subcaudals more than 50 (54 to 81; those reported as having less than 54, which I have examined, all had regenerated tail tips); range: south of the Zambezi 5 5. Loreal well separated from first labial; ventrals 160 to 178; subcaudals 54 to 78 ; pupil subcircular, rarely round ; no light line from behind eye to above conier of mouth; upper labials immaculate; underside immaculate or, at most and that but rarely, a few dusky marks ; range: streams of Southern Rhodesia and Bechuanaland, south to Natal, west to the Cape r. rufulus (p. 19) Loreal in contact with first labial or, if separated, only by a granule; ventrals 172 to 183; subcaudals 58 to 81; pupil round; a more or less light line from behind eye to above corner of mouth; upper labials spotted ; underside from throat to end of tail with a dusky median line or occasionally a median series of dusky markings; range: Transvaal (1 record), south to Natal (1 record), east through Cape Province laevissimus (p. 23) LO\^RIDGE : AFRICAN SNAKE GENERA STATISTICAL DATA FOR THE SPECIES OF LYCODONOMORPHUS (Abnormal variations are given in parentheses) Sppcips Midbody soale-rows Malo vt'iitrals Foinalo \('iitrals Feinalo subcaudals Malo subcaudals bioolor 23 (-2.-,) 154-164 152-166 52-59^ 63-71* s. upeinbac 21-23 172 175-188 29-34 40 s. subtaeniatus 21-23 175-180 189-193 41-48 52-58 r. leleupi 19-21 165-171 164-174 52-61 65-67 r. mlanjensis 21 163-167 169 51 60-71 r. whytii 19 ? 159-172 37-47 ? r. rufulus 19 162-17;*) 160-178 54-79 72-78 laevissim us 19(-21) 172-181 172-183 58-69 76-81 LYCODONOMORPHUS BicoLOR (Gimther) 1894b. Glypholycus bicnJor Giinther, Proe. Zool. Soc. London for 1893, p. 629, fig. 1: Shores of Lake Tanganyika. 1896d. Boulenger, p. 61.1. 1910a. Sternfeld, p. 15. 1915a. Boulenger, p. 201. 1915c. Boulenger, p. 619. 1924b. Loveridge, p. 4. 1929a. Werner, p. 50. 1933h. Loveridge, p. 232. 1933m. Witte, p. 86. 1937f . Loveridge, pp. 489, 496. 1941a. Uthmoller, p. 40 (as Glypholicm). 1947c. Laurent, p. 10. 1952. Witte, p. 18. 1953. Witte, p. 164, lig. 42. 1956. Laurent, p. 88, pi. ix, fig. 1. 1942b. Nerophidion liypsirliinoidis Werner, Sitzb. Akad. Wiss. Wien., 133, Abt. 1, p. 54, fig. 7: No locality. 1928. Malcolm Smith, p. 496 (synonymizcs with bicolor). 3 50-59 flde Laurent (1954b, p. 43). 4 59-71 fide Laurent (1954b, p. 43). 10 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Common Name. Tanganyika White-bellied Water-Snake. Description. Rostral l)roader than deep, well visible from above ; nostril valvular, directed upwards in a semidivided nasal that is sometimes completely divided, rarely entire' ; internasals narrow anteriorly, as long; as, or longer or shorter than, broad, shorter than the prefrontals ; prefrontals moderate ; frontal 1% (usually 11/2 ) to 2" times as long as broad, rarely' as long as its distance from the end of the snout, usually as long as its distance from the rostral, much shorter than the parietals; loreal longer than deep, in contact with, or separate from, the first labial, sometimes by the presence of a small scale*; eye small, pupil usually round, sometimes subelliptic" ; preocular 1 (on both sides of all 111 snakes) ; postoculars 2 (110 ex.), very rarely 1 (M.C.Z. 54936) ; temporals 1+ 2 (both sides of 103 ex.). rarely 1 +1 (on one side only of M.C.Z. 54886) or 1 + 3 (M.C.Z. 54916, and on one side only of 6 others) ; lateral head shields separated by a deep groove from the upper labials ; upper labials 7, 8 (eighth usually posterior to the gape) or 9 (left side of M.C.Z. 54879), the fourth, rarely fourth and fifth (M.C.Z. 54837, 54934, etc.) entering the orbit (67 ex.) or separated from it (at least on one side) by 1 or 2 minute sul)oculars, or by a suboeular and an extension of the lower postoeular (left side of M.C.Z. 54879), or by an extension of the preocular contacting the lower postoeular (right side of M.C.Z. 54857) ; lower labials 8 or 9 (ninth usually posterior to the gape), first 4 (first 3 only on left side of M.C.Z. 54861 and 54921 ; or first 5 in M.C.Z. 54935 and on one side only in M.C.Z. 54921 and 54942) in contact with the anterior sublinguals, which are much longer than the pos- terior. Midbody scale-rows'" 23-25 (23 in 109 snakes; 24 in 3, viz. M.C.Z. 54908 and 54940-1; 25 in 2, viz. M.C.Z. 54853 ( $ ) s Entire in M.C.Z. 54851, 54919, or on left side only of 54852. 6 Twice in M.C.Z. 54897. " As long as its distance in M.C.Z. 26929, 54858, 54871. 8 Small scale given off by 1st labial in M.C.Z. 30076, 54869, etc. SmaU scale given oflf bv L'nd labial in M.C.Z. 54855, 54881. etc. Second labial transversely divided in M.C.Z. 54853, 54857, etc. 9 Subelliptic in M.C.Z. 30076, 54859, 54916, and on left of 54941. i« 19, given by Boulenger (1915c. p. 619), due to inclusion of tchytii ; 21. stated ill the original description, amended to 23 by Boulenger. LOVERIDGE : AFRICAN SNAKE GENERA 11 and 54924 ( 9 )) ; ventrals 152-166^^ ( S S 154-164; 9 9 152- 166) ; anal entire; subcandals^" 52-71 ( 9 9 52-59; $ $ 63-71) pairs. Color. Above, "'ray or plumbeous, rarely black, uniform except for the two or three outermost rows of scales which are white. Below, throat and belly white, uniform or occasionally some dusky flecking on the anterior third; tail white, sometimes uni- form in young but usually with a dusky longitudinal line along the median sutures of the subcaudals. Size. Largest $ (M.C.Z. 54854), 563 (420 + 143) mm.; largest 9 (M.C.Z. 54902), 778 (615 -f 163) mm.; smallest $ (M.C.Z. 54856), 227 (175 + 52) mm.; smallest 9 (M.C.Z. 54852), 235 (188 + 47) mm. All from Kigoma, Tanganyika Territory. Se.mial dimorphism. This is reflected by their respective tail lengths, tliat of 52 S S being included in total length 3.64 to 4.42 times, with an average of 3.89; in 49 9 9 the tail was included in total length 4.32 to 5.06, with an average of 4.66. For the purpose of these figures snakes with incomplete tails were discarded. The number of subcaudals appear to be an even surer guide to sex, the 9 9 having from 52 to 59, the S S from 63 to 71 pairs. Compare with Laurent's (1956, p. 89) figures expressed as percentages with an average of .255 lor $ $ , .227 for 9 9 . None of his specimens attained the maximum size of those in the Museum of Comparative Zoology. Breeding. A number of the 9 9 taken at Kigoma between April 4 and May 4, 1956, are gravid. They have not been critically examined for precise dates and the number of eggs produced at a time. Diet. I am indebted to Dr. E. Trewavas for determining the species of thirteen fish that Mr. C. J. P. lonides removed from stomachs of these snakes, viz. 1 Boulengerochromis microlepis (Boulenger) from M.C.Z. 54892. 11 184. as given by Bouleuger (lS96d, p. 615) presumably a misprint. Battersby has recounted all British Museum bicolor ami finds them within the range given for M.C.Z. material. 1250 is the lowest given by de Witte (1952, p. 18), and Laurent (1956, p. 89), but the tail tips of all M.C.Z. snakes with less than 52 have been regenerated, though at times this is difficult to see. For dd, Laurent gives 59-71. 12 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1 Haplotaxodoii Iriroii Poll from M.C.Z. 54947. 1 Lamprolorjus callipferns Boiilenger from M.C.Z. 54908. 2 Lamprologua savoryi elotujafiis Trewavas & Poll from M.C.Z. 54872 and 54922. 1 Lamprnlogus sp. from M.C.Z. 54916. 1 Lininochromis pfcffcri (Boulenger) from 1. 677-3. 4 Limnotilapia dardcnii (Boulenger) from M.C.Z. 54917, 54933, 54949. 2 Stolothrissa tanganicae Regan from I. 6773. Witte (1952, p. 18) also mentions recovering a species of Limnothrissa. Except for a couple of cichlids, the stomach con- tents consisted of elupeids (Laurent :1956). Habitat. C. J. P. lonides informs me that he captured these hicolor in the lake at night by the light of a pressure lamp, taking the snakes in a hand net about the size of a shrimping net. Laurent (1956) says they are to be found hiding beneath stones on the lakeshore in the vicinity of rivers. He quotes N. Leleup who captured several in the burrows of gerbils {Tat era sp.) where they presumably pass the hours of daylight. Localities. Tanganyika Territory: *Karema, Mpanda ; Katabi near Bangwe ; *Kigoma; Kirando Bay; Lagosa Bay; *Sumbawa. Northern Rhodesia: *Mpulungu. Belgian Congo — Tanganyika District : Kabimba Bay ; Mtoto Bay ; Pala Bay ; Uvira. Belgian Urundi: Makamba (but considered doubtful by the collector) ; Rumonge. Range. Lake Tanganyika off the coasts of Tanganyika Terri- tory; Northern Rhodesia; Belgian Congo and Ruanda-ITrundi. LYcoDONOAtORPiius suBTAEXiATUs UPEMBAE Laurent IdSdni. Board on lineatns Witte (part: not Diuiieril & Bibron), p. 86. 1953. Witte, p. 168. 1954b. Lycodonomorphus suhtaeniatus upemhae Laurent, Museo do Dundo, No. 23, p. 41, figs. 5-8: Nyonga, Katanga, Belgian Congo. Common Name. Eastern Congo White-bellied Water-Snake. Description.^'^ Tnternasals shorter than the prefrontals; pre- frontals moderate ; frontal as long as, or shorter than, its distance from the end of the snout, much^'* shorter than the parietals; 13 Taken entirely from the original description, where Laurent tabulates the scale-counts tor the d and all 21 9?. LOVERIDGE: AFRICAN SNAKE GENERA 13 preociilar 1 (41 sides), rarely 2 (3 sides); postoculars 2 (42 sides), rarely 1 (2 sides) ; temporals 1 + 2. Midbody scales with apical pits, "in 21-23 rows; ventrals 172-188 (S 172; $ 9 175- 188) ; anal entire; subcaudals 29-40 ( 5 9 29-34; $ 40). Color. Above, coloring lighter than in s. suhtaeniatKs and the dark dorsal zone narrower; the two or three outermost rows of scales white. Size. Only S (ex Nyonga), 454 (387 + (i7) mm.; largest 9 (ex Nyonga), 831 (739 + 92) mm. Sexual dimorphism. The solitary $ and one 9 have 21 mid- body scale-rows, the remaining 20 9 9 have 23 rows. The $ has fewer ventrals and more subcaudals than any of the 9 9 ; also his tail is longer ( £ 157r ; 9 9 10 to 12% of total length). Dentition. Maxillary teeth 19-20. Localities. Belgian Congo: Kina-Mwena ; Nyonga near Lake Upemba. Range. Southeastern Belgian Congo. Lycodonomorpiius subtaeniatus subtaeniatus Laurent 1952b. Boaedon vlrgatus Laurent (not Hallowell), p. 199. 19.")4:b. Lycodonomorpiius iiuhtae7iiatu.t' 11<2'.I, ;uul i-piieated in lit.-)."), ji]). L'N, !•(•, ITS jis I.iicoiJoiki- uiorijhus )-itfHltis. lii S4 fide Boettger (1889). LOVERIDGE: AFRICAN SNAKE GENERA 21 upper and lower lips (li^ht pink in life) lemon to ochre or yellow- ish white, immaculate. Eyes dark brown flecked with reddish brown, the pupil edged with reddish brown. In young snakes the back and sides are a deep yellowish brown, the underside a paler immaculate yellow than in the adults. Below, anteriorly yellowish white becoming l)righter yellow towards the anus; tail yellow, usually immaculate, occasionally with a dusky median line (fide Giinther: 1858; Boettger: 1889; Miiller : 1890b ; their specimens coming from Natal ; Botschabelo near Middleburg, Transvaal; and Harrismith, Cape Province, respectively), or each scale dark-edged. Size. Largest $ (A.M.N.H. 60113), 605 (445 + 160) mm.; largest 9 (M.C.Z. 21306), 827 (635 + 192) mm.; smallest, a $ (M.C.Z. 55143), 269 (205 -f 64) mm. Remarks. The type of rufulus is presumably still in the Berlin Museum, while that of leucophilus is unknown. Rochebrune's (1884a, p. 158) listing of this species from Senegal is omitted as obviously erroneous. The two snakes from Massangulo, Portu- guese Nyasaland, referred to rufulus by Cunha (1935, p. 5; 1937, p. 1779) were actually based on Boaeclon I. lineatus, so Dr. A. A. Themido tells me. The scale counts furnished by Cunha are not those of his specimens but w^ere taken from Boulenger's (1893b, p. 318) catalogue. Dentition. Maxillary teeth 25 on right rami of both Merebank snakes, the six foremost ones slightly less recurved than those following; posterior teeth somewhat smaller (Bogert). Hemipenes. Hemipenis bifurcate at extreme end, extends to the level of the tenth subcaudal; sulcus spermaticus forked at the sixth subcaudal ; spines feebly enlarged at the base, arranged in diagonal rows or, towards distal end, in chevrons, those on the terminus of each lobe decreasing in size. Sexual dimorphism. Verified subcaudal counts for $ 9 is 54-73; in ,J ^ 72-78 (possibly to 84 if Boettger 's Botschabelo snake was a $ and the count correct). Intact tails of 9 9 9 are 19-23% of total length ; those of 3 5 . t- — • f-H I—* ^^ ,^^ ^— - ,^-. ..— »— ( rH .— ( ^H *■ + + + + ^H ?— 1 f— 1 ?— 1 CO fO CO CO CO ■ u ,'~^ ^-^s ^ V ^ s r. 1 CI CI CI «:■ j^ C :3 o I-H t. — p^ ■;) C) 01 CI =1 ~ — ^^ ^H '^-'' I-H 1—4 CJ 1—\ f-H o OJ w cd =0 u 5 c s =c !? !;i _ '^ K & '? 5i ^ ^ > r5 • ^ ^ Ai s 'o ;£' © s> o © c _. J^ a C c c p £ '■/: "^ CH IT f-H O " Ot t^ o ^ ^ r>-' f-^ T~^ I—* .5 + + + + + ^ o c o o o "Zi f-H -t< tr^ ^ f— 1 tl. CI CI c^o -» "~ — :z % y. z ^ _; ^ _' —J — ■—I ^ ^ ~ ^ « f', -f CI ^ 00 l.t o CI O "^ l~ o CO CO CC I-H I— ( f-H rH + + + + + o l-O o CI ■^ ■ — ' ■o o CO t^ CI CI CO CO CO '/, c3 CO 00 '^^ t^ LO "^ to t~ 00 00 Ol rt Oi CO c'l 1-H \n c. ■^ !£> to l-O t^ r- S M X K C 1 CI CI CI rH < »•. 00 t^ Tt< c. I>- -< CJ CO Tt< ■rt< CO c 1-H 1-H 1-H I-H I-H — CO ■* \n 00 c'l c (M C] ] > I-H rH I-H I-H I-H 0) o at d u «c c c S SO - 5~. £ M a, v. 5- 'c •«*< s © ^ c* © s~. c u a, -4- Si'- o c c M b — • IK c< o C.2 o C .a S 0) C3 o ^«5 y: cc c3 > >3rX Cj Oii-ll- fl c; c LO •-,::; 0/1 •a n ~ 2 — ' CCf-H OJ m 00 n 4) ffl e -4H a> > C te X A ■«-> .a a i" o LOVERIDGE: AFRICAN SNAKE GENERA 29 Natriciteres olivacea olivacea (Peters) 1854b. CuiuHclla oUvacea Peters, Monatsb. Akad. Wiss. Berlin, p. ()22 : Tcti", Zambezi Kiver, Mozambique. 185.'). Peters, p. 5:2. 1858e. Giinther, p. 39. 1806a. Boeage, p. 47. 186G)). Boeage, p. 66. 1882a. Boeage, p. 28S. 1882:i. I'eters (part), p. 114, pi. xvii, tig. 1 (omit Angola & Zanzibar eoast). 18821). Miiller, p. 168. 1887b. Mocquard, p. 66. 1888. Moequard, p. 128. ]881t. Pfeffer, p. 9. 18!)3b. Boettger, p. 123. 1893. Pf cf¥er, p. 79. 1899. Moequard, p. 219. 1908b. Moequard, p. 558. 1860a. Coronella {Meisodon) dumcrilii Giinther, Proc. Zool. See. London, p. 429, fig. — : "Sierra Leone" i.e. Gold Coast, fide Boulcngcr. lS63a. E'nicognathus piinctatostriatus Jan, Areh. Zool. Anat. Fisiol., 2. p. 278: No loeality (Type in Hamburg Museum). 1866. Jan, pi. iv, fig. 4. 1877e. Coronella (Misodon) oJiraeea Peters, p. 614. ]8S8a. Boettger, p. 84. 1877e. Neu.'^tcrophis atratiis I'eters, Monatsb. Akad. Wiss. Berlin, p. 614, pi. — , lig. 1 : Chinc-hoxo, Cabinda. 1888a. Giinther, p. 51. 1884a. /aeliolus oUvaceus Eoehebrune, p. 156 (ignored in this revision). 1884a. MthodoH dumcrilii Eoehebrune, p. 158 (ignored in this revision). 1886. Grayia Giardi Dollo, Bull. Mus. Eoy. Hist. Nat. Belgique, 4. p. 158, figs. 1-2: Lake Tanganyika region, Belgian Congo. 18y4a. Boulenger, p. 288. 1910a. Sternfeld, p. 23. 1893b. Tropidonntus olivaceufi Boulenger, p. 227. 1896d. Boulenger, p. 604. 1896e. Boulenger, p. 216. 1896. Peracca, p. 2. 1896. Tornier, p. 67. 1897b. Boulenger, p. 278. 1897e. Boulenger, p. 800. 1897g. Boulenger, p. 279. 30 BULLETIN: MUSEUM Or COMPARATIVE ZOOLOGY 1897. 1897b. 1898. 1898. 1898. 1898. 1900b. 1900. iy02d. 1905c. 1907a. 1907a. 1908a. 1908b. 1908c. 1908. 1908a. 1909b. 1909. 1909a. 1909b. 1910b. 1910. 1910. 1910a. 1911. 1911. 1912. 1912. 1912c. 1913. 1915a. 1915c. 1915(1. 191 Of. lOKJa. 19171). 1917. 1918a. ]919a. liHyg. Tornier, p. 67 (ro])iint of 1896 book). Tornier, p. 65. Boettger, p. 22. Johnston, p. 361a (iiol in 1897 ed.). Sclater, p. 98. Tornier, p. 292. Boulenger, p. 451. Flower, p. 968. Boulenger, p. 446. Boulenger, p. 112. Boulenger, p. 10. Eoux, p. 76. Sternfeld, p. 404. Sternfeld, pp. 211, 228. Sternfeld, p. 243. Werner (1907), p. 1866. Werner, p. 170. Boulenger, p. 303. Peracca, p. 172. Sternfeld, p. 11. Sternfeld, p. 10. Boulenger, p. 503. Peracca, p. 3. Eoux, p. 98. Sternfeld (part), p. 14, tig. 12. Lampe, p. 187. Sternfeld & Niedcn, p. 385. Hoblej', p. 47. FitzSimons, i\ W., p. 82. Sternfeld, p. 265. Boettger, pp. 345, 347, 353. Boulenger, p. 201. Boulenger, p. 619. Boulenger, p. 645. Chabanaud, p. 368. lioveridge, pp. 77, 82. Chabanaud, p. 10. Sternfeld, p. 460. Loveridge, p. 334. lioulenger, p. 276. Boulenger, j). 19. LOVERIDGE: AFRICAN SNAKE GENERA 31 1919d. Chabanaud, p. 567. 1921a. Angel, p. 42. 1922. Aylmer, p. 15. 1923e. Loveridge (part), p. 87(5. 1924b. Loveridge, p. 4. 1927d. Wittc, p. 321. 1929a. Werner, pp. 15, 24. 1933. Schouteden, p. 236. 1933 j. Witte, p. 122. 1933m. Witte, p. 85. 1934a. Schwetz, p. 382. 1937b. Monard, p. 113. 1894e. Coronella olivac.ea var. diniK rilii Giinther, ]). 87. 1897. Johnston, p. 362. 1898. Johnston, p. 361a (reprint of 1897 book). 1895a. Mizodon olivaceu^ Bocage, p. 74. 1896a. Bocage, p. 91. 1897b. Mocquard, p. 8. 1923. Natrix olivacea Schmidt (as olivaceous) , p. .")8. 1925a. Loveridge (part), p. 71. 1928c. Barbour & Loveridge (part), p. 109. 1928. Cott, p. 953. 1928g. Loveridge (part), p. 34. 1928 j. Loveridge (part), p. 71. 1929. Flower, p. 174. 1929h. Loveridge (part), p. 19. 1929. Worthington, p. 124. 1933f . Angel, p. 72, figs. 25-25a. 1933h. Loveridge (part), p. 231. 1934e. Mortons, pp. 78, 171. 1934. Pitman, p. 293. 1935a. Corkill, p. 14. 1935. Cott, p. 964. 1941. Moreau & Pakenham, p. 108. 1943. Scortecei, p. 270. 1951. Monard, p. 147. 1936h. Natrix olivacea olivac.ea Loveridge, p. 21. 1936J. Loveridge, p. 236. 193G. Pitman, p. 224, pi. iii, fig. 4 ; pi. B, fig. 2. 1937c. Loveridge, p. 270. 1937f. Loveridge, pp. 489, 493, 496. 19381). Mortens, p. 46. 32 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1938a. Pitman, pp. 301, 326. 1938b. Pitman (repaged reprint embracing 1936 and 1938a in book form). 1938. Uthmoller, p. 42. 1939a. Sc'ortecci, p. 270. 19-lla. Uthmoller, p. 26. 19411). Uthmoller, p. 230. 1947. Pakenham, p. 138. 1940. XeusteropJiis oliracois nlivaceit.s Bogert, p. 35. 1941. Witte, p. 173. 1942e. Loveridge, p. 261. 1942. Moreau & Pakenham, p. 62. 1949a. Parker, p. 28. 1950f. Laurent, p. 128. 1953. Witte, p. 160, fig. 40. 1945. Ncuatcrophis olivaceus Lceson, p. 3. 1950. lonides, p. 100. 1950. Leeson, pp. ix, 36, 38, 66, 136. 1950. Rose, pp. 272, 320. 1950a. Villiers, p. 39. 19501). Villiers, p. 69, figs. 71-72. 1954. Dekayser & Villiers, p. 963. 1955. Rose, pp. 92, 178, fig. 1. 1953e. Xatriciterp.s olivacca ulivacca Loveridge, iip. 251, 317. 19541). Laurent, p. 44 (as olivaceus). 1955. Witte, p. 214. 1956. Laurent, p. 132, fig. 21. 1954d. Natriciteres olivaceus Laurent, p. 305. Further references to "olivacca" or "olivaceus'' will be found under its subspecies ulugtirucnsis and pemhana. Common Names. Olive Marsh-Snake (preferred A. L.), many names have been proposed beginning with Black-backed Grass- Snake (Flower, 1929), Grass-Snake; Olive Grass-Snake; Olive Water-Snake ; Black-backed AVater-Snake ; and Couuuou AVater- Snake; inkubayoka (Nyarwanda :/!'de Laurent). Description. Preoculars 1, rarely 2; postoculars 3, rarely 1 or 2 ; temporals 1 + 2, rarely 1 + 1 or 1 + 3 ; upper labials 8, rarely 7 or 9, the fourth and fifth, rarely the third, fourth and fifth, or fourth, fifth aud sixth, or fifth and sixth, entering the orbit ; lower labials S-IO, the first four, five, or six in contact with the LOVERIDGE: AFRICAN SNAKE GENERA 33 anterior sublinguals, which are much shorter than the posterior. Midbody scale-rows 19, rarely 18 or 17; ventrals 128-149 (130- 147 in M.C.Z.) ; anal divided; subcaudals ( ? 51, 52) 57-87. For characters common to all species, and tabulated data, see pp. 25-26, 28. Color. Above, olive brown, brown, gray, or bluish black ; upper lip yellowish, the labial sutures black; juvenile specimens sometimes exhibit a pale spot on the parietals ; usually a more or less distinct darker, sometimes chocolate, vertebral band four or five scales in width, bordered on either side by a narrow buffy one or, more rarely, by a series of light dashes or dots ; the color- ing of the flanks impinging as olive, gray, pale blue, red, or bright mauve on the ends of the ventrals, which are otherwise yellowish (sometimes orange in life) occasionally edged witli l)lack. Pitman (1936, p. 225) connnents on the handsome coloring of many of the snakes taken around Lakes Bunyonyi (6,400 feet) , Chahafi and Mutanda (6,000 feet), whose black-edged, dark brown, verte])ral band is flanked by narrower bands of bright chestnut with paler edges ; beyond this again is a narrow dark line below which is a broad band of mauve or reddish chestnut occupying most of the lower flank. His book should be consulted for further notes on coloration. The wide range of ventral coloring is (piite irrespective of sex. Laurent (1956, p. 134) mentions a red-eyed albino juvenile from Bukavu ; also a Kitutu specimen that is black l)elow. Size. Length of S (M.C.Z. 40329), 464 (332 + 132) mm., of largest perfect 9 (M.C.Z. 40334), 553 (409 + 144) mm., both from Ngatana, but surpassed in head and body length by a 9 (M.C.Z. 40317) of 585+ (460 + 125+) mm. with regenerated tail-tip. Remarks. The type oi Coronella dumerilii Giinthev (1860) was said to have a single anal and to have come from Sierra Leone. I mention this as N. fuliginoides has a single anal and does occur in Sierra Leone. However, Boulenger (1893, pp. 227-228) syn- onymized duinerilii with olivacea (which has a divided anal) and changed its locale to Gold Coast, donor II. H. Evans. Only :U BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY one speeiinen is listed. J. C Battersby informs me (29. v. 50) that it was one of five snakes received from H. Evans of the Gohl Coast (now Ghana). Sternf eld's (1917) record of seven olivacea from Duma with entire anals is surely a lapsus, for in both ventral and midbody scale-rows they agree with olivacea. The scale-rows on an Albert- ville snake (M.C.Z. 30074) reported as 17 by me (1933h), are actually 19 at mathematical midbody. The four Magrotto Moun- tain snakes with 19 roAvs brought to me (1942e) by natives, may well have come from the cultivated areas lower down the moun- tain as the half-dozen snakes I personally captured at the forest- edge all have 17 rows. Sexual dimorphism. This is so slight as to be negligible, for example in M. C. Z. material examined in 1950 : ventrals of 26 ^ c5 ranged from 130-141, giving an average of 134.9 38 $ $ " " 131-147, " " " " 137.7 subcaudals of 17 <5 6 " " 57-80, " " " " 69.1 " 19 9 9 " " 51-80, " " " " 76.2 tail lengths 17 ^ <5 " " 43-o67o of the total length 47.8^0 36-51% " " " 43.8% For these last two characters fewer snakes were available owing to the large number whose tails were truncated or re- generated. Breeding. Typical olivacea produces more eggs than its mon- tane race. On February 14, at Kaimosi, a 9 held 6 eggs measuring 23 x 7 nnn. On February 19, at Kaimosi, a 9 held 6 eggs measuring 23 x 7 mm. Between June 11-20, at Xgatana. a 9 held G eggs measuring "small" Between June 11-20, at Xgatana, a 9 held 6 eggs measuring 15 x 6 mm. Between June 11-20, at Ngatana, a 9 held 6 eggs measuring 19 x 7 mm. Between June 11-20, at Ngatana, a 9 held 6 eggs measuring 19 x 11 mm. Between June 11-20, at Ngatana, a 9 held 6 eggs measuring 22 x 10 mm. LOVERIDGE: AFRICAN SNAKE GENERA 35 Between June 11-20, at Ngatana, a 9 held 6 eggs measuring 23 x 11 mm. On June 15 at Dar es Salaam, a 9 laid 8 eggs measuring 17 x 8 mm., this last lot possibly prematurely as it was shortly after capture. Diet. This captive snake swallowed two Rana a. fuscigiila hindlegs first on June 22, and another headfirst on June 24. llemains of a Rana sp. in a Witu snake; Phrynohafrachus sp. in a Duem specimen (Werner, 1908) ; PhrijnohatrachHS mhiutits in both a Kaimosi and a Mwaya snake ; HyperoJius milnei at Golbanti; Leptopelis argent eus at ? Morogoro ; Biifo steindach- neri on four occasions at Laini and Ngatana ; fish in an Ngatana snake also. According to Pitman (1934) an Olive Marsh-Snake has been seen swallowing locust hoppers, while Corkill (1935a) mentions caterpillars in the Sudan. Parasites. Cestodes and Pentastomida (now U.S.N.M. 47029) were present in an jMtimbuka snake. Defensive tactics. A gentle snake, never biting, but relying on a nauseous odor (Laurent :1956). When captured, according to Pitman (1934), this small snake exhibits considerable trucu- lenee but soon becomes tame and confiding in captivity. My own impression is of violent struggling to get free which they often achieve by the loss of their tails. Only once was I personally responsible for such a loss. Having grabbed a snake by the tail, 1 continued holding it while getting a bag in which to drop the reptile, but its violent wriggling resulted in its breaking away, leaving the tail in mj- grasp. As it is also quite usual to find members of the genus Grayia with truncated tails, it would seem probable that aquatic snakes are more liable to seizure by waders or other waterfowl than are more terrestrial types, and by active wriggling they frequently achieve liberty by sacrificing their tails. In some localities they doubtless suffer from attacks by other predators such as turtles or fish. Hahitat. Though principally a lowland reptile, typical olivacea occurs u]) to 6400 feet in the undulating uplands of Central Africa. Schmidt (1923) has suggested this is a savanna species that has recently invaded the Congo f orCvSts ; actually the Olive Marsh-Snake is a forest-edge, rather than a sylvieoline, form, oceurfing in clearings and open glades. More usually, however, 36 BULLETIN : MUSEUJI OF COMPARATIVE ZOOLOGY olivacea is foiiud in moist situations such as, for example, be- neath grass strewn on the floor of a watcher's hut in a rice swamp ; beneath grass and reeds piled around the edges of native plots adjacent to a river; one was hoed up in grassland near tbc lakeshore, another had sought shelter beneath a bundle of thatch- ing beside an unfinished hut (Loveridge). Two were found be- neath the movable gangway of a pier by Pitman, who remarks that this marsh-snake will readily dive to hide beneath sub- aqueous vegetation. I once observed a marsh-snake swimming across the Tana River at a point where it was fifty yards across ; on reaching tlie north bank the reptile paused to rest. Localities. Sudan: Bahr el Jebel; *Bo River; Duem ; Lake No ; *Mongalla ; Wau ; White Nile south of Khartoum. Ethiopia: Lake Rudolf. ? French Somaliland: Cote des Somalis (Paris Mus.).'" Somalia: Mogadiscio Uganda: *Bukatakata ; Busu Hill; Damba Island; Entebbe; Fort Portal; *Gulu; Ibanda, Toro ; Kaianja, Lake Edward ; *Katwe ; Kitala ; Lake Bunyonyi ; Lake Chahafi ; Lake Edward; Lake Kyoga (Kioga) ; *Lake Mu- tanda ; Lake Victoria ; Lalle, Lake Kyoga ; Lugasi near Kampala ; Mitiana; Mjanji; *Mushongero; Nansere, Masaka; Sesse Islands. Kenya Colony: *Belazoni ; *Golbanti ; *Kaimosi ; Kakamega ; Kenya Mountain; *Laini ; Lake Rudolf; *Mkonuml)i; *Ngatana; Peccatoni ; Pokomonie ; *Witu ; Yala River. Tanganyika Terri- tory: Arusha ; Bukoba ; *Dar es Salaam ; *Ikiju, Musoma ; Ipiana ; *Kabare ; *Kigoma ; Kilimanjaro Mtn. ; Kingani ; *Ki])angate, Lake Rukwa ; Klein Popo ; Kome Id. ; Madimula, Usaramo ; *Ma- grotto Mtn. (foot of ?) ; Makindo River; *Morogoro ; *Mwaya; *TTjiji; *Ukerewe Id. Zanzibar Island: Kokotoni ; Zanzibar. Maiia Island. Mozambique: Angoche ; Fambani ; Quilimane ; Rikatla; Tete ; Zambezi Expedition. Nyasaland: Fort Hill; Karongwa to Kondowe ; Lake Nyasa ; *Mtimbuka ; Nkata Bay to Ruarwe; "Nyika Plateau." Northern Rhodesia: *Abercorn ; Kazungula; Lealui ; Mpulungu; Petauke ; Shambanza's. Nau- senga River; *Wantipa. Lake IMweru ; Zambezi Plains. Southern Rhodesia: Salisbury. Angola: Dundo; Malange (Malanji) ; 2y The incltisioii of SdiiialilaiKl in the ran>,'e b.v Boctt^tT (ISSSa) was apparently due to I he title of Morquard's (ISSS) i)ai)er. l)iit ^Mociiiiard clearly states tliat his sinjrle siieeinien came from Zaiizihar. The listiiij; of this species from Somalilaiiil liy r.oiileiiirer (ISUTj^) was hased on his (ISiKJe) own identitication of a snake from l.aUe Ivndolf. LOVERIDGE : AFRICAN SNAKE GENERA 37 Muita River; north of Kwanza River; Puiigo Ndongo (Adungo; Andongo). Possibly some are referable to o. ulugur}iensis. Ca- binda: Cliinchoxo. Belgian Congo:^° *Albertville ; Banana ; Beni ; Bitshumbi (Vitsehuml)i) ; Boma ; Bugasia, Lake Edward; Dika ; Elisabethville ; *Idjwi Island ; Kabare, Lake Edward ; Kaben- gere ; Kabiinda ; Kak^^elo ; Kamanda, Lake Edward ; Kando ; Kanonga River ; Kapolowe ; Kasai River ; Katanda ; Kiambi ; Kikondja ; Kuniingn ; Lake Tanganyika ; Lofoi ; Lnknla ; Mabwe River; Maji (May) ya Moto ; Makaia Ntete; Manda; Mangbatta (Monbuttn) ; j\Iati ; Medje ; Mokabe-Kasari ; Musosa ; Mutsora ; Ni- angara ; Nyonga ; Pare Albert ; Poko, Upper ITele ; Pweto ; Rnngu ; Ruwindi; Stanleyville; Teml)we; Zaire = Congo River; Zaml)i. Ruanda-Urundi:'^^ Bukavu ; Kissenji; Miileria region. French Congo: Brazzaville; confluence of Dougon and Kibali Rivers; Gabon ; Lambarene ; Loango ; Passa River region, affluent of Ogooue River; Yakoma-Abiras, Upper Oubanghi. French Cam- eroon: Garua. British Cameroon: Isongo (specimen re-examined by Mertens :1938b). Nigeria: Brass; *Lagos. Dahomey: Widab. Togoland: Kete ; Misabohe. Ghana:^- Achimota School (prob- ably). Ivory Coast: Bouafle. French Guinea: Beyla. West Africa: Abadafi (not located. Midler :1885b). Range. Sudan, east to Somalia, south to Mozambique, Avest through Southern Rhodesia to Angola, north and west to French Guinea. Natriciteres olivacea uluguruensis (Loveridge) 1882a. Coronella olivacea Peters (part), p. 115 (specimens with 17 scale- rows). 1884a. Fischer, p. 6 (Masailand snake with 17 scale-rows). 1894a. Coronella olivacea var. dumerilii Giinther, p. 618 (Zomba Mtn. ; material now in Brit. Mus. exhibits 17 scale-rows). 1896d. Tripodonotus olivaceus Boulenger (part), p. 604 (Zomba and Mlanje Mtns.). 30 For precise localization of Congo localities, see the respectiye articles by Witte and Ijaurent. 31 For 30 additional localities, mostly in Ruanda-Urundi, see Laurent (1956, pp. i:i3-134). 32 The type of dumerilii Giinther, originally said to come from Sierra Leone (hence its listing from there by Aylmer :1922) was later stated by Boulenger (1893b) to have come from the Gold Coast = Ghana. 38 BULLETIN : MUSELTM OF COMPARATIVE ZOOLOGY 1910a. Sternfeld (part), p. 14 (Aniani, though possibly 19 scale-rows). 1911b. Nieden, p. 442 (Aniani). 191311. Werner, p. 717 (Amani). 1923o. Loveridgp (part), p. 876 (Bagilo, tliough with 19 scale-rows). ]92.")n. Xntrix olivacea Loveridge (part), p. 71 (Bagilo). 1928c-. Barbour & Loveridge (part), p. 109 (all except Dar es Salaam). 1933h. Loveridge (part), p. 231 (Rungwe Mtn. only). 19391). FitzSimons, Y. F., p. 20 (Silinda Mtn.). 1935. Trnpidonolus sp., Cunha, p. 4 (Massangulo, Niassa, with 15 scale- rows). 1935. TropidonoUis fuliginoides Cunhii (not Giinther), p. 3 (Massangulo). 1937. Cimha (1935), p. 1778. 1941. Theniido, p. 16 (Massangulo). 1937b. Monard, pp. 113, 115 (Kalukembe, Angola). 1935e. Natrix olivacea idugurnensis Loveridge, Bull. Mus. Comp. Zool., 79, p. 7 : Nyange, Uluguru Mountains, Tanganyika Territory. 1937f. Loveridge, p. 502. 1941a. Uthmoller,- p. 42. 1947. Pakenham, p. 138. 1940. Xcusfcrophis olivaceitm uluguruensis Bogert, p. 35. 1942a. Loveridge, p. 261. 1953e. Natriciteres olivacea uluguruensis, Loveridge, pp. 252, 318. 1953h. Loveridge, p. 143. 1955e. Loveridge, p. 181. 1955e. Loveridge, p. 181. 1956c. Loveridge, p. 42. Common Names. Montane Marsh-Snake (English) ; nyoka usamhia (Sambara, but not specific). Description. Preocular 1, rarely 2; postoculars 3, rarely 2; temporals 1 -|- 2, rarely 1 + 1, 1 + 2 or 2 -j- 2 ; upper labials 8, very rarely 9, the fourth and fifth, or very rarely the third, fourth and fifth, or fourth, fifth and sixth, enterinof the orbit ; lower labials 8-10, the first 4 or 5 in contact with the anterior sublinguals, which are much shorter than the posterior. Midbody scale-rows 17, rarely 15 or 19; ventrals 125-144 '=* ( S S 125-140, $ 9 130-144); anal divided; subcaudals 62-84 (9 9 (i2-77 ; $ S 66-84). 33 144 for a Zomba snake (B.M. 48.1.1.87) contirmed by J. C. Battersby. If Hscher's (1884a) Masailanrl snake really had 17 midbody scale-rows, presumabl.v it was an aberrant o. olivacea, judgring by its 148 ventrals and .j4 subcaudals ( ? intact). Cunha's (19.35) counts are rejected as probably not being those of the Massangulo snake, which I have examined. Its ventrals were 135, not 153 as stated. LOVERIDGE: AFRICAN SNAKE GENERA 39 For characters common to all species, and tabulated data, see pp. 25-26, 28. Color. Above, olive brown, brown, gray, or bluish black; upper lip yellowish, the labial sutures black ; usuallj^ a more or less distinct darker, sometimes chocolate, vertebral band, bordered on either side by a narrower buffy one two scales wide, or rarely by a series of light dashes or dots; the coloring of the flanks impinging as olive, gray, pale blue, red, or bright mauve on the ends of the ventrals, which are otherwise yellowish (in life creamy white, bright yellow, deep orange) though sometimes edged with black. The wide range of ventral coloring is quite irrespective of sex. Size. Length of S (M.C.Z. 23129), 430 (300 + 130) mm., from Nyingwa, ca. 7500 feet, the highest point at which any were taken; on the other hand, the largest 9 (M.C.Z. 23118), 490 (350 + 140) mm., was from Nyange, only 2500 feet. Remarks. The snake from Massangulo referred to Tropidono- tus sp. by Cunha (1935 and 1937) agrees in all respects with knoAvn variants of this race except in its ventral count of 153 (a transposition for 135). For example M.C.Z. 44110 from Ugano, Matengo Mountains, just north of Massangulo, has 15 midbody scale-rows, 2 preoeulars, and 1 + 2 temporals, while others from the same locality have 17 scale-rows, 1 preocular, and 1 + 1 temporals. Moreover, Cunha (1935 and 1937) records three other snakes from Massangulo as fuliginoides because they have 17 scale-rows. Through the courtesy of Dr. Themido, who reprinted Cunha 's list in 1941, I have been able to examine one of these snakes and find it is unquestionably an o. iduguriiensis with di- vided anal. It might be as well to add that the scale-counts and color de- scriptions given by Cunha (1935) are not those of his specimens but translations from Boulenger's catalogues into Portuguese, which later (1937) he retranslated back into English! Barbour & Loveridge (1928c) furnished scale-counts around neck, midbody, and preanal region, for sixty-two montane snakes. These counts were made by a student ; the subcaudal range, given as 63 to 87, I have since amended to 62-84. 40 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Dentition. See Bogert (1940). Anatomy. The hemipenis is described by Bogert (1940). Sexual dimorphism. This is so slight as to be negligil)le, as the following figures will show : viMitrals of 40 $ S range from 125-140, giving an average of 134.9 " 58$$ " " 130-148,3* >i i( a i( i3g_5 subcaudals 21 5 5 " " 66-84, " " " " 74.8 " 28 9 5 " " 62-77, " " " " 69.6 taillength 2\ $ $ " " 48-61%, " " " " 53.3% 28 9 9 " " 43-54%, " " " " 50.0% The last, or tail length, character, is expressed as a percentage of total length, for which only half the total numl^er of snakes were available as exactly half of these reptiles had truncated or regenerated tail tips. Breeding. Apparently these montane snakes lay fewer eggs than the lowland form. On July 22, at Bagilo, a 9 held 3 eggs measuring 22 x 6 mm. October 8, at Nyange, a 9 held slighth" developed eggs 5 mm. in diam. October 9, at Nyange, a 9 held 2 eggs measuring 12 x 6 mm. November 19, at Amani, a 9 held 4 eggs measuring 20 x 8 mm. November 25, at Amani, a 9 held 4 eggs measuring 16 x 7 mm. November 29, at Amani, a 9 held 4 eggs measuring 21 x 8 mm. But during November the majority of females held unde- veloped ova. Diet. Three snakes held frogs {Arthroleptis .9. lonnhergi and A. .renodactylus) ; another eight frog's eggs; a fifth stomach con- tained three round white eggs, apparently a frog's but 4.5 mm. in diameter; there was a caterpillar in a sixth. Parasites. Nematodes {KalicepJialus micrurus) in a Nyange snake, while in one Amani 9 were Kalicephalns sp., Ophidascaris sp., and Amplicaecnm africaniim. A Magrotto snake held an indeterminate nematode but it is unusual for snakes of this race to be infected. Enemies. A Cape File-Snake {Mekelya c. capensis) captured near Lujeri River, Mlanje Mountain, disgorged the tail of a Mon- 34 Thpse flgurt's (180-148) are m.v own counts; 148 occurs in a snake (M.C.Z. :,:',(\iV2 ) from Achimota, Ghana, as well as in one (M.C.Z. 54057) from Aberconi, -Northern Rhodesia. LOVERIDGE : AFRICAN SNAKE GENERA 41 tane Marsh-Snake. A Vine-Snake (Thelotornis k. kirtlandii) was observed swallowing another in a vivarium from which several marsh-snakes had already disappeared. Defensive tactics. When seized by the tail this little snake wriggles violently nntil its eandal appendage breaks off. The tail-tip is missing from 41 of the 108 snakes I have examined, and the suggestion it is not only predators that are responsible is scarcely warranted by the incidence of loss as between the sexes. Habitat. Though not primarily a forest species, this snake is frequently to be found beneath fallen bark or logs lying at the forest edge where congenially moist conditions are present. In native plots they conceal themselves in piles of rubbish or under stones. Marshy spots and the banks of streams are visited in search of small frogs, and after heavy rain these reptiles are sometimes found on paths. I should have said that in tropical East Africa this race was essentially a montane form characterized by 17 midbody scale- rows, though at Amani (3000 feet) and on adjacent Magrotto Mountain examples with 19 scales are not uncommon. Recently, however, at Liwale (2100 feet), Mr. C. J. P. lonides has col- lected abundant material of this race and found that 19 rows do not occur there, but 10% of those examined (i.e. 3 out of 28) possess 15 midbody scale-rows ; such individuals also crop up in series from Songea and the Matengo Mountains. Localities. Tanganyika Territory: *Liwale ; Magrotto Mtn. — *Magrotto ; Matengo Mtns. — *Ugano ; *Mbeya ; *Rovuma River ; Rungwe Mtn. — *Nkuka Forest ; *Songea ; Tukuyu ; Ulu- guru Mtns. — *Bagilo, *Mkangazi, *Nyange, *Nyingwa, *Vituri; Usambara Mtns. — *Amani ; *Bumbuli. Mozambique: Massan- gulo. Nyasaland: *Cholo Mtn.; Mlanje (Milaugi) Mtn.; Ten- gadzi River ; Zomba Mtn. Northern Rhodesia; Fwambo. South* em Rhodesia: Selinda Mtn. — *Chirinda Forest. Angola: Kalu- kembe. Range. Chiefly montane forests and marshes of Tanganyika Territory, south through northeast Mozambique to Southern Rhodesia, west to Angola. 42 BULLETIN: MISEIJAI OF COMPARATIVE ZOOLOGY Natkiciteres olivacea pembana (Loveridge) 1925a. Xatrix olivacea I^overidge (part), p. 71. 1935c. Natrix olivacea pemhana Loveridge, Bull. Miis. Conip., 79, p. 8: Chakechakc, Peniba Island, north of Zanzibar Island. 1937f. Lov(M-idgo, p. 493. 1041. Moreau & Pakcnham, p. 108. l<)4L-i. Uthmoller, p. 42. 1947. Pakenham, p. 138. Common Name. Dwarf Marsh-Snake. Description. Preoculars 1-2; postoeulars 3; temporals 1 -|- 2 ; upper labials 8, third, fourth and fifth, or fourth and fifth, entering- the orbit ; lower labials 8, the first 4 in contact with the anterior sublinguals, which are much shorter than the posterior.^^ Midbody scale-rows 15-17 ; ventrals 123-128 (for $ $ alone, consequently for both sexes); anal divided; snbcaudals 49-63. For characters common to all species, and tabulated data, see pp. 25-26, 28. Color. Above, uniform brownish olive; upper lip yellowish, the labial sutures black ; flanks olive only slightly impinging on the ends of the ventrals, which are otherwise yellowish. Size. Length of S (M.C.Z. 46125), 278 (200 + 78) mm., from Wete; length of holotype 9 (M.C.Z. 19112), 285 (210 + 75) mm., from Chakechake. Sexual dimorphism. There is no difference between the sexes in ventral counts, both ranging from 123-128, and only an aver- age difference ( 9 9 49-56; $ $ 52-63) in the number of sub- caudals, more mai'ked in tail length expressed as a percentage of total lengtli, viz. 32-35% for three 9 9 , 36-43% for three $ £ possessing apparently intact tails. Remarks. For a detailed study of midbody lepidosis consult Pakenham (1947) where he points out that of eleven snakes he collected, only three had 15 midbody scale-rows, the rest having 17. However, he finds Pemba snakes are eonsistenth' recognizable in possessing 15 rows in the preanal region, while Zanzibar rep- tiles (which he regards as o. olivacea, with some intermediates with 0. uluguriiensis) invariably display 17, or an occasional 16, in this region. -o' 3'> Except in M.C.Z. 46126, which is abnormal on one side. LOVERIDGE: AFRICAN SNAKE GENERA 43 Jn mainland populations of o. uluguruensis from southeast Tanganyika Territory I have encountered five examples with 15 midbody seale-rows ; these were from Liwale (3), Songea (1), and Ugano (1), in every instance the snake being one of a much larger number of typical uluguruensis. Habitat. Of five snakes captured at Wete by Mr. R. H. W. Pakenham, three were taken beneath coconut husks in shade, one under a tussock in a grassy field, and one beneath a piece of wood in swampy ground. Localities. Pemba Island: *Chakechake; *Wete. Range. Pemba Island, Indian Ocean, is separated from the African mainland by a channel of exceptional depth — 400 fathoms, i.e. eight times as deep as the shallow sea between Zanzi- bar and the adjacent coast of Tanganyika with which it was united in geologically recent times. However, the habits of this marsh-snake make transportation with agricultural produce possible. Natriciteres variegata (Peters) 1861a. Misodon variegatus Peters, Monatsb. Akad. Wiss. Berlin, p. 358: Pel, Ghana. 1902b. Mocquard, p. 415. 1893b. Tropidonotus variegatus Boulenger, p. 217. 1893c. Matschie, p. 211. 1895f. Boulenger, p. 33. 1896d. Boulenger, p. 603. 1898. Boettger, p. 19. 1901. Schenkel, p. 156. 1908a. Sternfeld, pp. 40-4, 421. 1908b. Sternfeld, pp. 211, 218. 1909a. Sternfeld, p. 10. 1909b. Sternfeld, p. 10. 1916f. Chabanaud, p. 368. 1919a. Boulenger, p. 276. 1921a. Chabanaud, p. 467. 1921b. Chabanaud, p. 523. 1922. Aylmer, p. 18. 1929a. Werner, p. 21. 1897b. Tropidonotus fuUginoides Werner ( ?part, with 15 scales), p. 402. 1930a. Natrix fuUginoide.^ Barbour & I.ovcridge (not Giinther), p. 772. 44 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY ]i).'-! ( •_' snakes), i:'.T (li) and 140 (1). The type was said to have 143. possibly due to a transposition of figures'.' 41 Verified counts of M.C.Z. material are 64-74 ; both 63 (for a $ ) and 73 (for two dd) are based on Mt. Nimha snakes (Angel et al:1954). LOVERIDGE: AFRICAN SNAKE GENERA 45 Size. Leiifrth of S (M.C.Z. 49692), 400 (265 + 135) mm., from Mampon<-; leno-th of a 9 (M.C.Z. 22502), 341 (240 + 101 ) nun. The nnsexed type (Leyden Mns.) was only 328 (222 + 106) mm. Remarks. My (1936h) aetioii in synonymizing variegatus with fuliginoides because both, with identical coloring, occur at Bitye, was ill-advised in view of the fact that 15 scale-rows is correlated with a divided anal in all fifteen variegatus subsequently studied by me, whereas 17 scale-rows and a single anal characterize the eight fnliginokles in the Museum of Comparative Zoology. Later 1 (1942e) pointed out they should be regarded as full species because of the considerable overlap in their distribution. Lee- son's (1945) key errs in assuming the preoculars always number 2 ; this is the case in eight of our snakes, but only 1 preocular is present in six others. Sccual dimorphisfu. There is no difference in ventral counts as between the sexes and only an average difference ( 9 9 64-76 ; S S 75-78) in the number of subcaudals; the difference is more marked in tail length expressed as a percentage of the total length, viz. 50-56% for six 9 9 , 60'/v for all four S S with tails intact. Diet. A frog in one Paiata snake. Parasites. A linguatulid {PorocephaUis suhidifer) from a Tafo snake was identified for me by Dr. II. R. Hill (5.ix.50). Enemies. Chabanaud (1921a) comments on the tail being in- tact in only 1 of 8 snakes from French Guinea. It is truncated in only 6 of the 17 specimens (chiefly Ghanian) in the Museum of Comparative Zoology. Localities. French Cameroon: *Bitye; Yaunde. Nigeria: Asaba. Togoland: Kete ; Misahohe ; *Worawora. Ghana: Aburi ; *Achimota; *Khong, Volta River; *Kumasi; *Mampong; *Oda ; Pel; *Somanya; *Tafo. Ivory Coast: Banco Reserve; Bossou ; Gama; Keoulenta; Nzo (17 scales!) ; Serengbara; Tonkoui Mtn. ; Ziela; Zouguepo. Liberia: *Bandaja; *Bolahun; Bromley; Gibi; *Piata (Paiata; Peahtah) ; *Suacoco. French Guinea: Beyla; Kerouane; Kouroussa; Macenta; Nimba Mtn.; Nzebela; Nzere- kore; Sampouyara. Sierra Leone: (fide Aylmer). Range. French Cameroon west to French Guinea. 40 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Natriciteres fuliginoides (Giinther) 1858c. Coronella fuliginoides Giinther, Cat. Snakes Brit. Mus., p. 39: West Africa. 1889. Mocquard, p. 143. 1896. Giinther, p. 264. 1863d. Meisodon longicauda Giiutlier, Ann. Mag. Nat. Hist., (3) 12. p. 352, pi. V, fig. A : Fernando Po. 1875a. Peters, p. 198. 1884a. Eochebrune, p. 158 (ignored). 1887a. Socage, p. 184. 1884a. Zacholu^ fuliginoides 'Rochehrune, -p- 156 (ignored). 1887b. Coronella longicauda sp. u. Mocquard, Bull. Soc. Philom. Paris, (7) II, p. 69: Franceville, French Congo (restricted by A.L.). 1890. Coronella (Meizodon) variegata Miiller (not Peters), p. 692. 1893b. Tropidonotus fuliginoides Boulenger, p. 217. 1896d. Boulenger, p. 603. 1897. Sjostedt, p. 24. 1897b. Werner (part), p. 402. 1898. Boettger, p. 19. 1898a. Werner, p. 212. 1899a. Werner, p. 135. 1900b. Boulenger, p. 451. 1902a. Werner, p. 343. 1905f. Boulenger, p. 185. 19061. Boulenger, p. 211. 1908a. Sternfeld, pp. 404, 421. 1908b. Sternfeld, pp. 211, 227. 1909a. Sternfeld, p. 10, figs. 12-13. 1909b. Sternfeld, p. 10, figs. 10-11. 1909d. Werner, p. 247. 1910. Muller, p. 595. 1911. Lampe, p. 186. 1915a. Boulenger, p. 201. 1917a. PhisalLx, p. 335. 1917. Sternfeld, p. 459. 1919a. Boulenger, p. 276. 1921a. Cliabanaud, p. 467. 1921b. Chabanaud, p. 523. 1922. Aylmer, pp. 15, 18. 1927(1. Witte, p. 321. 1929a. Werner, p. 21. 19:i3iii. Witte, p. 85. LOVERIDGE : AFRICAN SNAKE GENERA 47 1895a. Mizodon fuliginoidcs BocaRc, p. 75. 1895c. Boeage, p. 13. 1896b. Mocquard, p. 45. 1897b. Mocquard, p. 8. 1903a. Boeage, p. 42. 1929. Natrix fidiginoides Flower, p. 174. 1933f. Angel (part), p. 74 (omit Yakoina Abiras specimens). 1940a. Mertens, p. 241. 1951. Monard, pp. 147, 157. 1940. Neusterophis fulifiinoidcn Bogcrt, p. 33, fig. 3. 1945. Leeson, p. 2. 1950. Leeson, pp. ix, 36, 38, (Mj, 135. 19501). Villiers, p. 69. 1953. Witte, p. 159, fig. 39. Further citations of "fuliginuide.s" will be found under oli- racea nlKgiiruciisis and variccjata. Common Names. Collared Marsh-Snake (Loveridge) ; Smoky Snake (Flower). Description. Preoculars 1-3; postoculars 2-3; temporals 1 + 2, rarely + 2 or 1 + l'"; upper labials 8, rarely 7, the fourth and fifth entering the orbit; lower labials 8-9,^^ the first 4 or 5 in contact with the anterior sublinguals, which are much shorter than the posterior. Midbody scale-rows 17"; ventrals 122-137 •*"; anal entire; subcaudals 75-95.'*'"' For characters common to all species, and tabulated data, see pp. 25-26, 28. Color. Above, brown ; end of snout and most of parietals paler; upper lip yellowish, the labial sutures usually black; nuchal collar white to buff, more or less heavily edged with black ; dorsum very rarely exhibiting indistinct traces of a darker ver- tebral band ; flanks brown with occasionally a dorsolateral series ■i.i{) +2 iu an aberrant Metet snake (Bof,'ert :1940) ; 1 -h 1 on left side only of a Kribi reptih' (M.C.Z. 7845). 43 Allegedly 11 in a cotype of longicauda Mocquard (1887b). 44 Allegedly I'J in a Congo snake (Boeage :1887a) ; possibly an olivaceaT « Allegedly 117 (fide Angel :193;{f ), or 119 for a Franceville snake and 137 for a Kerouane specimen I fide Guibe in litt. 8.i.\.r)0). 46 Allegedly 57 iu a Congo snake (Boeage :1887a) ; possibly an olivaceat Sexed material too scanty to justify giving sexual ranges. 48 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY of wliitish dots, the brown impinging on the ends of the other- wise white ventrals that are sometimes edged with brown or black. Size. Length of c^ (M.C.Z. 9265), 468 (282 + 186) mm. from Lolodorf, but surpassed in snout to anal length by a S ( A.M.N.H. 45927) with a head and body measurement of 370 mm., its tail, like that of the largest recorded 9 (A.M.N.H. 50552) of 310 mm., being truncate. Remarks. Mocquard (1887b) comments on the fact that the tails are intact in only 2 of his 6 cotypes, and later (1897b) in only 2 out of 8 snakes. Boulenger (1893b; 1896d) records intact tails in onlj^ 5 of the 12 specimens in the British Museum. The tails are intact in 2 of the 8 examples in the Museum of Com- parative Zoology. Sexual dimorphi.wi. There is no difference in ventral counts as between the sexes, and truncated tails preclude speculation about the subcaudals or relative tail lengths. Dentition. See Bogert (1940). Anatomy. Parotids are present according to Phisalix (1917a). The hemipenis is well figured, described, and discussed bv Bo- gert (1940). Diet. A frog {Rana mascareniensis) in stomach of a Cameroon frog (Werner, 1897b). Habitat. Fre(|uently seen on bush i)aths in Sierra Leone ac- cording to Aylmer (i;)22). Sternfeld (1908b), however, is mis- taken in saying these snakes are confined to the coast. This author comments on the occurrence of all three species in the Cameroons, but whereas fuliginoides is there the most abundant and rare in Togo, the situation in Togo is reversed for varicgatus (21 speci- mens) is plentiful and fiilifjiiioides scarce. Localities.^' Belgian Congo: Albertville ; Arebi ; Butu-Polo ; Ganda-Sundi ; Kabwe ; Kai-Bumba ; Kapanga ; Kenia Stream, af- fluent of Lusinga River ; Kisala ; Kunungu ; Lofoi ; Lower Congo ; Lukolela; Lukonzolwe; *Makaia Ntete ; Mubale-IMunte ; Sandoa; Tembwe ; Temvo ; ITpoto. French Congo: x\biras ; Alaima Leketi ; Franceville ; Gabon; Lambarene ; Loudinia-Niari; Mayombe (Mai'jombe or Majumba) ; Ogooue. Rio Muni: (as Spanish Guinea). Fernando Po: Musola; Natividad ; Sao Carlos. French ■*" Angolan records of snakes with 17 nii(ll)od.v scale-rows liuve divided anals and are consetiuentl.v referalile lo o. iihi!it(riicnftin. LOVERIDQE: AFRICAN SNAKE GENERA 49 Cameroon: Baga near Konn; Bipindi; Dehane; Dibongo near Edea ; Dipikar ; Itoki ; Jengwe ; Josplatte ; *Kribi ; Limbe ; *Lol()- dorf; Lonji; Metet ; Mohinda (Molundu) ; Pungo Songo ; *Sak- bayeme. British Cameroon: Bibundi ; Buea ; Buenga-Vorwerk, Moliwe Plantation; 1 songo ; Jobann Albrecbtsbohe ; Ossidinge ; Victoria. Nigeria: Lagos; Old Calabar. Togoland: Misabobe. Ghana: Ancober River. Sierra Leone: {fide Aylmer). French Guinea: Bey la. Range. Belgian Congo nortbwest to French Gninea. Genus PhILOTHAMNIJS Smith*- 1840. Philothamnus A. Smith, 111. Zool. S. Africa, Eept., footnote to text for pi. lix. Type by monotypy: Dendrophis (Philothamnus) semi- variegata A. Smith. 1857. Chlorophis Hallowell, Proc. Acad. Nat. Sei. Philadelphia, p. 52. Type by monotypy: C. heterodermus Hallowell. 1866a. Eerpetacthiops Giinther, Ann. Mag. Nat. Hist., (3) 18. p. 27. Type by monotypy: H. bellii Giinther = C. heterodermus Hallowell. 1868. Chlorophis Theobald, Cat. Eept. Asiatic Soc. Mus., p. 49. Type by monotypy: C. oldhami Theobald = Ahaetulla lioplogaster Giinther. ]895h. OUgolepis Boulenger, Ann. Mag. Nat. Hist., (6) 16. p. 171. Type h\ monotypy: 0. macrops Boulenger. In earlier days members of this genus were referred to Voluhei- (in its Linnaean sense), or to the Asiatic genera Ahaetulla (or its synonyms Dendrupliis and Lepiophis) and Cyclophis (now a synonym for Opheoclrys). Definition. Maxillary teeth 20-40, subequal or posterior long- est ; mandibular teeth subequal. Head more or less elongate, dis- tinct from neck ; eye moderately large, with round pupil ; nasal divided; a loreal (rarely fused with prefrontal); a preocular (rarely divided). Body cylindrical, elongate ; scales smooth, with apical pits, in 11-15 rows, narrow and disposed obliquely on an- terior part of body ; ventrals rounded or more or less sharply keeled on either side, usually a noteb corresponding to the keel in •»8 The revision of this genus iuul those preceding it were ready for press when I received a copy of I>aureiit's monuniental work on Central African herpetology (ISoG). With tiic preceding genera I was able to incorporate most ni'w oliserva- tions, but in tlie case of I'liilotliuminiK Laurent's wealtii of material enabled him to observe aberrations almost as numerous as my own. For these and his com- ments on them, Laurent's paper slioulil be consulted. 50 BULLETIN: MUSEUM OP COMPARATIVE ZOOLOGY strongly keeled species. Tail long ; subcaudals paired, not or but rarely keeled and notched (subgenus Chlorophis), or else nor- mally keeled and notched (subgenus Philoiliamnus) . Range. All Africa (where suitable forest, savanna, bush or swamp conditions occur) south of 16° N., except in the extreme southwest (see distributional chart for the range of individual species). Remarks. Proportions which are subject to change with growth {vide Parker: 1949) or other factors, do not provide a stable foundation for taxonomic purposes. Nevertheless, I have pains- takingly examined all the available material for them — except the very lengthy series of i. hattershyi, i. irregularis and s. semi- variegatus. In those forms where an abundance of material was available for study, a wide variation in scale proportions was noted. Naturally the bulk of the species fall within a mean, a few overlapping from this mean to extremes on one side or the other. A summary of these chai-acters that are more or less common to all species in this rather homogeneous assemblage, follows : Rostral broader than deep, the portion visible from above measuring about one quarter {natalensis and semivariegatus subspp.) or a third to a half (almost all species) to two-fifths {heierodermus, Iwplogaster and i. irregularis) its distance from the frontal ; internasals slightly shorter than, or as long as. the prefrontals (both conditions in all species where a good series was available, and in Jioplogaster even longer than the pre- frontal) ; frontal one-and-a-third to one-and-a-half times as long as broad (in most of the species) but one-and-a-quarter to twice as long as broad in hcterodermus, as long as, or longer than, its distance from the end of the snout (both conditions in every form except /. ornatus, s. girardi and s. thomcnsis of which material is lacking or meagre), shorter than (in all species except s. girardi and s. thomensis) or as long as (in 7 forms) the parietals; loreal (rarely fused with a prefrontal in hoplogasfcr, i. hattershyi and jmffl^fH.s/.s) varies from one-and- a-quarter to twice as long as deep (in /. hattershyi and heterolepi- dotus) to one-and-two-thirds to three times as long as deep (in s. semivariegatus) ; a useful summary of loreal/nasal ratio in all British Museum material of semivariegatus was recently LOVEBIDGE : AFRICAN SNAKE GENERA 51 published by Parker (1949a, pp. 58-61) but the loreals (and subcaudals) of our Pemba snakes are well within the range of variation exhibited l)y mainland specimens ; the preocular is in contact with, or separated from, the frontal (both conditions in all ten adecjuately represented forms) ; the anterior sublinguals are shorter than, subequal to, or longer than, the posterior (all three conditions in hdcrodcrynn^, only the two former in most of the others). Assuming that the evolution of these snakes has proceeded from the generalized iiiacrops to the specialized scmivaricgatus, we observe an increase in temporals, midbody scale rows, ventrals and subcaudals is taking place. The development of lateral keels (though these may be present or absent in several of the swamp or bush-dwelling intermediate forms) on both ventrals and sub- caudals attains its highest perfection in s. semivariegaftis, the s])ecies best adapted to an arlioreal life and enjoying the widest distribution of any member of the genus. Yet it seems strange that the slight sexual difference in subcaudal counts that may be noted in macrops and its allies is lost in s. semivariegatus. If, on the other hand, we postulate that the widespread scinlroricga- tus is the oldest species and developed when Africa supported more forest than is the case today, then it might be argued that the loss of keels is correlated Avitli the departure to a more ter- restrial existence in which frogs, instead of lizards, form the principal ol)ject of pursuit. Due to the difficulty of deciding whether or not the subcaudals of an occasional /'. /. iyvvgnlaris are keeled, or those of a s. nit id us not, 1 (19r)la:p. 189) suggested Chlorophis be regarded as of subgeneric status only, a view that has found general, though not universal, acceptance among my colleagues. Either as AhaetnUa, Cldorophi^ and/or Philothamnns, cpiite a number of keys or sunnuaries to this genus have been published by Giinther "(1863, pp. 283-287), Socage (1882c, pp. 1-19). Boulenger (1894a, pp. 91-102), Giinther (1895, pp. 527-528), Schmidt (1923, pp. 73-79), Hecht (1929, pp. 329-335), and Loveridge (1951c, pj). 1-12). The key now offered is the last one brought uj) to date. It was principally based on the abundant material in the Museum of Comparative Zoology, augmented by extensive loans of critical material from more than half-a-dozen 52 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY nniseums. The types of thirteen species or their synonyms, pre- served in the British Museum, have been studied in detail. Un- fortunately the more we know of variation in these reptiles, the more difficult it becomes to devise a clear, yet comprehensive, key. Aberrations from the normal are sufficiently frequent as to make it wise to use the key with the utmost caution, and take into consideration the ranges as set forth in the chart. It is as well to admit that I am not satisfied about a few species and races {i. irreg\daris, heterolepidotus, s. nitidus, s. dorsalis and s. semivariegatus) whose ventral and subcaudal counts (also, in some instances, geographical ranges) overlap. The only way to be confident that no misidentifieations in the literature have been accepted, would be to borrow all the specimens listed in the literature from all the museums concerned — a somewhat overwhelming undertaking which I should like to see done. Key to the Species 49 1. Subcaudals rounded or angular, occasionally faintly notched, but not so angular as to be called keeled; in most species less than 130 pairs of subcaudals''" 2 (Chlorophis) Subcaudals sharply angular (not always so in s. dorsalis), keeled and notched like the ventrals; usually more than 130 pairs of subcaudals. 10 (Philuthamnu.s) 2. Midbody scale-rows 13, very rarely 11 3 Midbody scale-rows 15, very rarely 13 4 3. Anal divided, very rarely entire; usually 2 labials entering orbit; temporals 1 + 1 or 1 + 2, very rarely 2 + 2 ; range : montane forests of eastern Tanganyika Territory and a rice swamp in Zanzibar .... macrops (p. 58) ■m Not eierii individual snake will respond to this key as exceptions to the normal are numerous in this jreinis. Mistakes in identitication may be avoided by taking into account the range, consulting the variational and distributional charts on pi;. .").")-.")T. ano Prior to counting the subcaudals it is equally important to ascertain if the tail tapers yruduallii to a tine point. Tlie presence of a conical point is not necessarily proof that the tail is intact, for, wlien U)st early in life, the terminal point may be regenerated. LOVERIDGE : AFRICAN SNAKE GENERA 53 Anal entire; usually 3 labials entering the orbit; temporals 2 + 2, very rarely 1 + 2, 2 + 1, 2 + 3 or 3 + 2 ; range : virgin forests of western Kenya (Kaimosi), west to the Cameroons and Fernando Po. Also French Guinea ^i- c-armntns (p. 60) 4. Anal entire; subcaudals 79-99; temporals usually 2 + 2, occasionally 1 + 1, 1 + 2, 1 + 3, 2 + 1 or 2 + 3 5 Anal divided, very rnrely entire in i. haiiershyi, i. irregularis and heterolepidofiui) ; subcaudals 73-134 6 5. Ventrals 142-164; adult 9 9 attain 862 mm.; range: virgin forests of Angola, north and west to Portuguese Guinea h. heterodcrmus (p. 63 ) Ventrals 174-181; adult 9 9 attain 962 mm.; range: virgin forests of southwest Uganda, Belgian Ruanda and adjacent Belgian Congo south to Mpala ''• ruandae (p. 68) 6. Usually 2, though occasionally 3, labials entering orbit; range: eastern half of Africa ^ Usually 3, though occasionally 2, labials entering orbit; range: west- ern Africa, penetrating the East only in the Nyasa-Zambezi region 8 7. Subcaudals in 9 9 73-98, in <5 i 87-106; range: eastern Africa south of the Usambara Mountains in Tanganyika Territory, south through Mozambique and the Ehodesias to Port Elizabeth and Durban, South Africa Iwplogastcr (p. 70 ) Subcaudals in 9 9 90-111, in <$ <5 100-120; range: eastern Africa in the Usambara Mountains, north through Kenya to Ethiopia, west to the Nile in the Sudan and Uganda (where it meets with the typical form), south around Lake Victoria, Tanganyika Territory i. hattershyi (p. 76) Subcaudals in 9 9 114-122, uv $ $ 120-128; range: Transvaal, south to Natal and eastern areas of Cape Province, South Africa Tvatalensis (p. 98) 8. A broad brown dorsal stripe ^^ on back and tail; range: Angola, east to Northern Ehodesia (Lealui) i- ornatus (p. 82) No such distinct dorsal stripe (unless occurring occasionally as a recessive character) 9 51 Possibly this stripe is not diagnostic in which event the name ornatus might be applied to all southwestern irregularis which have much lower scale-counts than those in the extreme northwest of Africa. 54 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 9. Subcaudals in 9 9 •'''- 97-116, in i $ 103-126 ; build moderatelj' slender, more so in S S than in 9 9; range : Senegal due east to Metemma on the White Nile, south down the Eift Valley (Lakes Victoria, Tanganyika, and both sides of Nyasa) following the Zam- bezi to the East Coast ; south of the Zambezi it oeeurs from Southeiu Rhodesia west to Damaraland, north and west to Senegal i. irrrf/ularis (p. 8")) Subcaudals in 9 9 109-126, in $ S 115-134; build excessively slender, except in gravid 9 9 , but head noticeably narrower than in i. irregularis; range: Khartoum on the White Nile, southeast to mouth of Bovuma River, Tanganyika Territory (? Mozambique also), west to Angola, northwest to Togo, east to the Nile heierolcpidotuK {\). 100) 10. JNlidbody scale-rows 13; range: Annobon Island, Gulf of Guinea .s. f/irariJi (p. 124) Midbody scale-rows 1.1, rarely 13 11 11. Ventrals 153-217 ; subcaudals 115-161 12 Ventrals 200-220; subcaudals 15G-175; range: Sao Thome Island, Gulf of Guinea >s'. Ihomensis (p. 125) 12. Snout usually pale brown; temporals usually 1 + 1, very rarely 2 + 2 or 2 + 3; range: Angola, north to Cabinda and possibly Ogowe River, French Congo •''■ dorsaJis (p. 121) Snout not distinctively colored ; temporals usually 1 + 2 or 2 + 2 .13 13. Ventrals 153-162; temporals usually 1 + 1 or 1 + 2, sometimes 2 + 2; range: virgin forests of French Congo, north through Cameroons, west to Ghana"" •''. vitidm (p. 119) Ventrals 164-217; temporals usually 2 + 2, sometimes 1 + 1, 1 + 2, 2 + 1 or 2 + 3 ; range : Eritrea south to Natal and adjacent areas of Cape Province, northwest to Gambia, i.e. all Africa south of 16° N. except Cape Peninsula, etc s. semivariegatus (p. 105) 52 This is my lowest count for an intact tail, figures in the eighties occur in the literature, chiefly from Angola. If intact, possibly these should be referred to I. nniatus and less stress on dorsal band. 53 This is a restricted concept of nitidus to that previously published by me (1951c, p. 5), and which was apparently followed by Laurent (1956). See footnote to P. 8. nitidui^ on this matter. LOVERIDGE : AFRICAN SNAKE GENERA 55 STATISTICAL DATA FOR THE SPECIES OF PHI LOTH AMNUS (Parentheses are t'iii]tloyoil to indicate abnormal variations; others are dealt with in footnotes to the specific descriptions) SpiMios MiiU>o(l.v scale- rows Male vt'iiirals Female ventrals Anal single divided Female caudals (l>alred) Male caudals (paired) macrop.s (11)13 135-146 135-148 (S) D 74-88-'^* 84-97 h. carinatus 13 141-157 145-167 S 72-86 78-91 h. heterodermus 15 142-160 152-164 S (D) 79-97 79-94 h. ruandae 15 174 175-181 S 82-93 99 hoplogaster (13)1.3 141-160 143-164 D 73-98 87-106 i. hatterfthi/i 15 147-169 153-177 (S) D 90-11155 100-1205= i. ornatun 15 150-160 160-166 D 85-99 102-106 i. irregularis 15 150-179 153-183 (S) D 87-119^6 103-127 natale7isis 15 150-165 151-168 (S) D 114-122 120-128 heterolepidotus 15 164-193" 169-193 (S) D 109-126 115-134 s. semivariegatus (13)15 164-217 164-200 D 126-161 127-160 s. nitidvs 15 153-162 154-164 D 133-151 136-153 s. dorsalis 15 167-190 172-180 D 115-137 128-143 s. girardi 13 18.1-197 D 143-160 ,v. t]i(ii)i) nsis 15 200-220 D 156-175 54 A re-examination of the Nchingidi $ (M.C.Z. 48272) reported as having 61) subcaudals, suggests the tip was regenerated early in life. 55 Uthmoller's (1938) counts of 118 for a ?, and 12.3 for a cT, require verifica- tion ; unfortunately the specimens were destroyed during World War II. 56 87-93 in Dundo 9 ? iflde Laurent :1950) which may possibl ybe referable to i. ornatus in a new sense; 97 was encountered in an Mtimbuka ? (M.C.Z. 51115) and a Mushongero ? (M.C.Z. 48288). Boulenger (1849a) gives up to 133 for West African snakes, but I suspect some of these may be heterolepidotus. 57 With a variational range in d ventrals of 30, it might be supposed that two races are involved : actually both extremes come from Niangara. 56 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY Geoprnphical Distribution of the African Gentis I'hilothamnuR a, 1 OS c g £ I. Si a. e •a e o 3 1 .2 « O 1 1 Si oj 00 06 •*• £ o 00 s ■a 'S 00 s e i 05 05 Sudan — — — — — « * — « — — — « Eritrea — — ■ — ■ — — — — — — « Ethiopia — — — — — « — — — — British Somaliland — — — — — — — — — * Somalia — — — — — # — — — * Uganda Protectorate « — » — * » — ♦ — — — * Kenya Colony — « — — — » — * — — * Tanganyika Territory * — — — « * » — * — — * Pemba Island — — — — — — — — — « Zanzibar Island * — — — — — — — — * Mafia Island — — — — * — — — — « Mozambique — — — — * # — ? — — — • Nyasaland Protectorate — — — * * — — — — « Northern Rhodesia — — ? — *- — * — — — — « Southern Rhodesia — — — — » * — — — — « Bechuanaland Protectorate * « Transvaal, South Africa — — — — * « — — — » Natal, South Africa — — — — * » — — « Orange River, South Africa * Cape Province, South Africa * ♦ ._ » South-West Africa — — — — ? « — — — — • Angola — — « — — « * — • — » — ♦ Cabinda — * « — — « — « 1 — « — LOVERIDGE : AFRICAN SNAKE GENERA 57 Geographical Distribution of the African Genus Philothnmnii.s 7: ft. on c c e s •e e S a. o 'Si OS C OB t OB 5 u c o s 05 o 'a. 1 1 00 s 1 00 00 5 oo' 90 S *.» 1. 5 a OO Belgian Congo — •M- * « — ? » — « — • « • Belgian Euanda-Urundi — « « — — « — « — — — French Congo — « — — — — « — — — ? French Equatorial Africa # — * Annobon Island — — — — — — » — — — Sao Thome Island — — — — — * — — ♦ — — — Fernando Po Island — » « — — — — — — — * British Cameroon — * * — — » — « — — # — French Cameroon — « » — — — « — » — — * — Nigeria — * — — « — « — — » • Dahomey — — — — — — — — — — — Togo — * — — * — « — — * * Ghana — » — — — « — — « « Ivory Coast — » — — — — * — — — — Liberia — *■ — — — » — — — — « Sierra Leone — * — — — » — — — — — French Guinea — # — — — * — — — — « Portuguese Guinea — * — — — « — — — — « Gambia — — — — — * — — — — • French West Africa — — — — — ♦ — — — — « 58 BULLETIX : MrSEIlIM OF COMPARATIVE ZOOLOGY Pmi.oTifAMXUS MACROPS (Boulenfjer) 18951). CliloropJiis nrplrrfiin Werner (not Peters), p. 193. 1895h. Oligolcpis tnacrops Boulenger, Ann. Mag. Nat. Hist., (6) 16, p. 171: Usambara Mountains, Tanganyika Territory. 1896cl. Boulenger, p. 644. 1896. Tornier, p. 78. 1897. Tornier, p. 78 (reprinted). 18971i. Tornier, p. 65. 1908. Chloropliis macrops Sternfeld, p. 95. 1910a. Sternfeld, p. 19. 1911b. Sternfeld, p. 442. 1913b. Werner, p. 717. 1915c. Boulenger, p. 623. 1924b. Loveridge, p. 5. 1928c. Barbour & Loveridge, p. 116. 1929. Hecht, p. 330. 1929a. Wei-ner, p. 100. 1934c. Mertens, p. 157. 1937f. Loveridge, p. 502. 1941a. Uthmoller, p. 40. 1942e. Loveridge, p. 270. 1947. Pakenham, p. 140. 1951e. Loveridge, pp. 3, 6. Common Names. Usambara Green-Snake (Loveridge) ; nyoka ya mani (Sambaa :Loveridge). Description. Preoculars 1, rarely 2; postoeulars 2; temporals 1 + 1, 1 + 2 (or 2 + 2,'' fide Sternfeld) ; upper labials 8, occa- sionally 9, the fourth and fifth (or fourth, tifth and sixth/^ fide Sternfeld) or fifth and sixth entering- the orbit; lower labials 8-12, the first 4, 5 or 6 in contact with the anterior sublinguals. Midbody scale-rows 13, rarely 11; ventrals 135-148''* ( S S 135- 14(i; 9 9 135-148); anal divided, rarely entire (M.C.Z. 23236 and some Zanzibar snakes only) ; subcaudals 74-97 ( 9 9 74-88; S $ 84-97). For characters common to all species, and synopsis of scale- counts, see pp. 49-52, 55. '■'* If these two aberrations were on the same snake, one wonders if a young PhilothamtiHs s. semicariegatiis might not have been included in Sternfeld's long series ; impossible If he included all scale-counts. ■'■'!' 14S in the ? holotype, confirmed by me, though I find 81, not 75 subcaudals. LOVERIDGE : AFRICAN SNAKE GENERA 59 Color. Above, green or bronzy green to olive or brown, the upper lip white ; sometimes as many as 42 irregular light cross- bars formed by scales witli a light yellow or green, black-edged, basal spot turning to light blue on flanks. Below, chin and throat white ; rest of under surface pale green, yellowish green, or bluish-white, uniform, or each ventral shield with a lateral dull red blotch, or laterally bordered with black, the free edge tinged with bluish-gray. Size. Length of S (M.C.Z. 23236) 810 (560 -f 250 )mm.; of a 9 (M.C.Z. 23239) 950 (690 + 260) mm.; the unsexed juvenile holotype 9 (Brit. Mus.) now measures 222 (165 + 57) mm., the youngest topotype pzi (fide Pitman) ; *Sipi Forest, Mount Elgon. Kenya Colony: *Kaimosi. Cabinda. Belgian Congo: * Arebi ; Avakubi ; Babonde s. of Medje ; Bafwasende ; *Bambesa, Uele ; Batama ; Beni ; Bun- yakiri; *Buta ; [Dika: in error] ; Eala; Faradje; *Ibembo, Uele; *Idjwi Island — Upper Mulinga River; Ituri; Kande River; Kansenia; Kasai; Kaswabilenga ; Katanga; *Kulu River; Ku- nungu; Lissinu ; Lukawe River; Lukolela ; Lusema; Maganga Bay; Manda; Manguretshipa ; Medje (as Madie) ; Mombaka ; *Mpala, Lualaba; *Mpesi-Ukesi, Lower Congo; Munoi; Nala ; Niapu; *Panga, Aruwimi River; Shabunda region; *Stanley- ville; Temvo; *Yangambi, Stanlej^ville ; Walikale; *Zambi, Ubangi; *Zobia, Uele. French Congo: Lambarene (as hetero- dernius). Fernando Po: Bokoko-Garcia. French Cameroon: Di- hongo; EboloAva; Longji; *Metet; Molundu; Mukonje Farm. British Cameroon: Barombi ; Bibundi ; Buea ; Camp ; Isongo ; Johann Albrechtshohe ; Mapanga (type locality). French Guinea: Mt. Nimba — Gouela; Nion. Range. Western Kenya Colony, west through the Belgian Congo to Fernando Po. Also Fi-ench Guinea. Philothamnus heterodermus iieterodermus (Hallowell) 1857. CMorophis heterodermus Hallowell, Proc. Acad. Nat. Sei. Philadel- phia, p. 54: Gabon = French Congo. 18(30. Cope, p. 559. 1884a. Rochebrune, p. 173 (not used). 1894a. Boulenger, pp. 97, 358. 1896d. Boulenger, p. 631. 1897. Sjostedt, p. 35. 1898. Boettger, p. 58. 1898a. Werner, p. 209. 1899a. Werner, p. 137. 1900b. Boulenger, p. 452. ll)02a. Werner, pp. 338, 344. 190(3i. Boulenger, p. 213. 1908a. Sternfeld, pp. 407, 425. 1908b. Sternfeld, pp. 214, 230. 64 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOOY 1909a. 1909b. 1909(i. 1910. 1912c. 1917a. 1917. 1919a. 1921a. 1921b. 1922. 1927(i. 1929. 1929a. 1933f. 1933. 1933J. 1933m. 1934(-. 1934a. 1936h. 1936e. 1938a. 1940. 1940a. 1940b. 1941. 1945. 1948d. 1950. 1950a. 1950b. 1951. 1863c. 1885. 1866a. 1884a. 1875a. 1882c. 1888a. 1891b. Sterafeld, p. 14, fig. 20. Sternfeld, p. 15, fig. 15. Werner, p. 247. Miiller, p. 601. Boulenger, p. 470. Phisalix, p. 334. Sternfeld, p. 469. Boulenger, p. 282. Chabanaud, p. 468. Chabanaud, p. 524. Aylmer, pp. 15, 19. Witte, p. 323. Hecht, p. 331. Werner, p. 101. Angel, p. 105, figs. 38-38a. Schouteden (part ?), p. 236. Witte, p. 123. Witte (part), p. 89. Mertens, p. 1()9. Schwetz, p. 380. Loveridge (part), p. 29. Parker, p. 125. Pitman, pp. 294, 305, pi. xvii, fig. 3, pi. T, fig. 2. Bogert, p. 52. Mertens p. 241. Monard, p. 174. Witt« (part), p. 198. Leeson, p. 1. Cansdale, p. 45, photo. Leeson, pp. ix, 39, 40, 66, 133. Villiers, p. 59. VUliers, p. 80, figs. 94-95. Monard, p. 149. Ahaetulla heteroderma Giinther, p. 285. Miiller, p. 683. Hcrpciaethiops Bellii Giinther, Ann. Mag. Nat. Hist., (3) 18. p. 27, pi. vii, fig. B : Sierra Leone, i.e. Victoria, Sherbro Island, Sierra Leone. Eoehebnnie, p. 171 (not used). Philothamnuji heterodermus Peters, p. 199. Bocage, p. 18. Boettger, p. 59. Matschie, p. 61". LOVERIDGE: AFRICAN SNAKE GENERA 65 1893c'. Matschie, p. 212. 1895a. Bocage, p. 89. 1882a. Ahaciulla sp. Muller, p. 149. 1885. Muller, p. 683. 188-lb. FhilotlMhmis heterodonta {sic) Sauvage, p. 201. 1893. Fhilotamnus eterodermns (sic) Prato, p. 11. 1916f. Cltloroplii-s Jicterodcnniti: Pohcguini Chabanaud, Bull. Mus. Hist. Nat. (Paris), 22. p. 371, fig. 12: French Guinea. 1929. Ilecht, p. 331. 1933f. Angel, p. 106. 1940b. Monard, p. 174. 1929. Chlorophis cxjaneus Heclit, Zool. Anz., 81. pp. 331, 334, figs. 1-2: Ajoshohe, Nyong River, British Cameroon. 1951. Monard, p. 150. 1951t'. PJiilothamnus hetcrodermus lietcrodcrmuN Loveridge, pp. 3, 6. 19r)4. Angel; (iuil)e; Lnmotte it Koy, p. 392. 1954b. Villiers, p. 1240. Further citations of ''heterodennus" -will be found under /(. carinatus. Common Names. Emerald Green-Snake (Aylmer) ; Variable Tree-Snake (Cansdale) ; hoprakala (Temiie : Aylmer ) ; 7idawun- dukali (Mende:Aylmer) ; Pitman supplies the general Ganda and Kiga names for Chlorophis, but unless h. carinatus is a syn- onym it is questionable whether //. lieterodermus occurs in Uganda. Description. Preoculars 1, sometimes 2; postoculars 2, rarely 1 {fide Angel) or 3; temporals occasionally 1 + 1, 1 + 2, 1 + 3 [fide Miiller) or 2 + 1, though normally 2 + 2 or 2 + 3 ; upper labials 8-10, the third, fourth and fifth'S- fourth and fifth (type of cxjaneus) ; fourth, fifth and sixth; fifth and sixth (type of pohequini on one side only) ; or fifth, sixth and seventh, entering the orbit ; lower labials 9-11, the first 5, rarely 4,"- in contact with the anterior sublinguals. Midbody scale-row's 15; ventrals 152- 164 ( 5 5 142-160; 2 2 152-164) ; anal entire, rarely divided"'; subcaudals 79-97 (9 9 80-97; £ $ 79-94). «i On right side of a Kumasi $ (M.C.Z. 53691). on whose left side is the more usual 'J, with fourth, fifth and sixth entering:. «2 On li^ft side of an Achiniota d" (C.N.H.M.). on whose right side are the normal 10 labials, the first 5 of which are in contact with the anterior sublinguals (examined by A.L.). fis Divided in a Somanya ;' (M.C.Z. :."').-)212) . 66 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY For characters common to all species, and synopsis of scale- counts, see pp. 49-52, 55. Color. Above blue-green or olive, the interstitial skin black; anterior part of back in young occasionally shows a longitudinal series of paired black spots that sometimes coalesce to form cross- bars; some scales spotted with white, with or without a dark edge. Below, chin and throat yellowish cream ; body and tail cream to pale or bluish green. In life, writes Cansdale (letter of 23.ii.51), an Oda snake measuring 635 (465 + 170) was: Above, head dark olive brown; body olive brown with irregular dark crossbands and noticeably white between scales. Below, throat and body orange yellow gradualh' merging into dark slate posteriorly; tail uniformly dark slate. However, another Oda snake of 549 -|- 190 mm. (Xo. 183), as also a Kumasi specimen (No. 154), were recorded as "Black." Such melanistic individuals, says Cansdale, might easily be mis- taken for a small black cobra. In this connection it will be noted that the type of &eZ/n was described as : Above, deep black. Below, deep black except for chin, throat and thirty anterior ventrals, which are whitish. I assumed that this was due to formalin preservation until I came across Dr. Ilarley's reference to a Ganta snake being purple with a green head (Bogert), and Cansdale 's comments about a similarity to a small black cobra. Size. Length of <^ (type of poheguini in P.M.) 780 (595 + 185) mm., but a comparison with the measurements of a 9 suggests the tail tip is regenerated; consequently Chabanaud's count of 77 caudals is rejected. Length of 9 (type of bellii in B.M. 46.1.10.27) 867 (650 + 217) mm., but surpassed by another 5 (B.M. 36.8.1.669) of 886 (660 + 226) mm. from Congulu; the youngest, a 9 (M.C.Z. 13230) from Metet, 344 (255 -f 89) mm. Remarks. P. h. heterodermus is probably an offshoot of irregu- laris, from normal specimens of which it differs in having fewer subcaudals (see chart on [). oo), a single anal, and usually 2 -|- 2 temporals — though both these last characters may occur occa- sionally in irregularis. LOVERIDGE: AFRICAN SNAKE GENERA 67 The possibility of a short-tailed western race occurring in Sierra Leone requires investigating; if recognizable, tlie name hellii would be available. Though normal hef erode rmus have been reported from Portuguese and French Guinea, P. h. pobe- (juini was based on a single specimen from the latter country. The type had only 2 upper labials entering the orl)it, and a temporal arrangement of 2 -|- 1 on the right side, 1 (incompletely divided) + 1 on the left side. Monard has recorded a second specimen from Portuguese Guinea, having 9 upper labials, the tifth and sixth entering the orbit, and temporals 1 + 1- Two Cameroon snakes (M.C.Z. 7849; 13233) with the normal 9 upper labials, of which the fourth, fifth and sixth enter the orbit on one side of the head, have the poheguini arrangement on the other side. Further discussion on the synonymy of this snake, and that of cyaneus, will be found in an earlier (1936h) paper by me. Dentition. The type of cyaneus had 30-32 maxillary teeth (Ileclit) ; there were 33 and 35 in two Cameroon hetcrodernnis examined by Bogert (1940), who points out that this species ap- parently has 5 more maxillary teeth than irreguJa)is. See also Leeson (1950). Anatomy. The presence of parotids is discussed by Phisalix (1917a). Ifemipenis. Not bifurcate ; sulcus spermaticus undivided ; basal spines enlarged and followed by 2 or 3 flounces, merging distally into reticulate calyces which grade into numerous fine papillae towards the tip, which extends to the seventh caudal (Bogert). Sexual dimorphism. Negligible, though probably some average differences could be demonstrated with more material. Breeding. In June, at Dika, a 9 held eggs measuring about 23 X 7 mm. (A.L.). In November, at Worawora, a 2 held 5 eggs measuring about 35 x 10 mm. Diet. A frog (Werner :1909d) ; feeds on lizards in Sierra Leone according to Aylmer (1922). Habitat. Frequently found in gardens, according to Ayhner (1922), who may be confusing it with i. irregularis. 68 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Localities. Angola: *Congulii. Belgian Congo: *Banana; Bu- rung-a ; *Dika ; Ganda-Sundi ; Kai-Bumbi ; Kamatembe ; Ki- bimiba; Kisala; *Kisantu; *Lufii; Lukiila; *Makaia Ntete; Medje; Povo Nemlao ; Rugari; *Stanleyville *Temvo* Tshimiba. French Equatorial Airica: Fort Sibut ; Gabon. Fernando Po: Bahia de Sao Carlos. French Cameroon: Bipindi; Bitye ; Ebolo- wa ; *Efulen ; *Kribi ; Longji ; *Metet ; Moluiidii ; Miikonje Farm, Muudame ; *Sakbayeme. British Cameroon: Ajoshohe, Nyong River (type of cyaneus) ; Johann Albrechtshohe ; Buea; Tiko; Victoria. Nigeria: Lagos ; Niger River ; Obaii District ; Oil River. Togo: Adele (Bismarckburg) ; Misahohe; *Worawora. Ghana: Accra; *Achimota; *Adjikpo; Aslianti; *Kumasi ; Oduiuasi; *Somanya. Ivory Coast: Banco Reserve ; Tonkui Mountain. Li- beria: Ganta; *Suacoco. Sierra Leone: Freetown; Victoria, *Sherbro Island (type of hellii). French Guinea: Guelemata; Macenta ; Nimba IMountain ; Nion ; Nzebela ; Nzo ; To Mountain. Portuguese Guinea: Catio ; Madina Boe ; Rio Cassine. Range. Angola north and west to Portuguese Guinea (for Uganda, Ruanda, and some eastern Congo records, see heteroder- nius ruandae). Philothamnus heterodermus ruandae Loveridge 1933J. Clilorophis heterodermus Witte (not Hallowell), p. 123. 1933m. Witte (part), p. 89 (Birunga; Lulenga). 193(51i. LovcM-idge (part), p. 29 (Lake Kivu). 1941. Witte (part), p. 198 (Kamatembe; Lake Kivu; Kugare). 1951('. Philothamnus heterodermus ruandae Loveridge, Inst. roy. Sci. nat. Belgique, Bull. 27, Xo. 37, pp. 2, 7: 9 ex Mulungu, Lake Kivu, Belgian Congo. 1953. Witte, p. 195. 19.16. Laurent, p. 174, fig. 2G, pi. xix, fig. 1. Common Name. Ruanda Emerald-Green-Snake. Description. Preoeulars 1-2*^*; postoculars 2; temporals 1 + 2, 2 + 2 or 2 + 3 ; upper labials 8-10, the third, fourth and fifth ; fourth, fifth and sixth ; or fourth, fiftli, sixth and seventh enter- M Two on the left side ol" ;i Kayonza ^ (C.N.H.M. 6975). LOVERIDGE: AFRICAN SNAKE GENERA 69 ing the orbit ; lower labials 8-11, the first 4, 5 or 6 in contact with the anterior sublinguals. Midbody scale-rows 15; ventrals*''^ 165-181 (6 6 165-174; 9 9 175-180); anal entire; subcaudals 82-100 ( 9 9 82-93; S S 98-100). Color. As in P. h. carinatus, with the snout and throat at least partially drab (Laurent). Size.' Total length of S (C.N.H.M. 6975) 725 (515 + 210) mm.; largest 9 (I.G. 16964) 962 (730 + 232) mm.; holotype 9 (I.G. 15860) 912 (680 -f 242) mm. Remarks. From its nearest relatives (of which I have about 75 and 40 counts respectively, the 7 specimens of ruandac seen by me and an additional 20 or so listed by Laurent (1956), differ in their more numerous ventrals and larger size ; characters best seen in a comparative table. Mid- body scale- Sppcies rows Ventrals i» 6 S Ventrals in 9 9 Caudals in $ 9 Caudals in $ S Largest ? in mm. h. carinatus 13 141-157 145-167 72-86 78-91 815 (600 -f 215) h. heterodermus 15 142-160 152-164 79-97 79-94 886 (660 -1- 266) h. ruandae lo 165-174 "5171-181 82-93 98-100 962 (730 -f- 232) Sexual dimorphism. In addition to the ventral and subcaudal counts tabulated above, Laurent finds the percentage of tail lengths into total lengths of 15 9 9 is 24 to 26.6%, while in 3 $3 it is only 27.5 to 29%. Diet. Frogs recovered from stomachs of ruandae by Laurent, have been identified as Leptopelis karissimhiensis (ex Alini- boiigo) ; ScJwutedenella sp. (ex Kabilombo), and Rana fuscigula (ex Mount Kabobo). Localities. Uganda: Impenetrable Forest, Kayonza, British Ruanda. Belgian Ruanda-Urundi: Bugarama; Kabobo; *Lake Kivu; *Lulenga; Mulungu; Nyongwe. Belgian Congo in Kivu region: Alimbongo ; Burunga; Kabilombo; Kabumbe ; Kama- tembe ; Kibumba ; Rugare ; Tshumba ; Uvira ; also allegedly from Mpala, Kanzenze, Lualaba District.^ 66 65 164 in a Kabobo $ according to Laurent (1956) who considers it aberrant. 66 See Laurent, whose paper is a major contribution to our knowledge of this form, for reasons why this locality is probably erroneous. 70 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Range. Southwestern Uganda, west through Belgian Ruanda- Urundi to the Belgian Congo. PniLOTHAMNUS HOPLOGASTER (Giintlier) « 1863c. AhaetuUa lioplogaster Giinther, Ann. Mag. Nat. Hist., (3) 11, pp. 284, 286: "Port Natal" i.e. Durban, Natal, South Africa. 1890. Boulenger, Fauna British India, Eept. & Batr., p. 305. 1866b. Philothamnus neglectus Peters, Monatsb. Akad. Wiss. Berlin, p. 890, footnote: "Prazo Boror" i.e. Boror, Mozambique. 1882a. Peters, p. 130, pi. xixA, fig. 2. 1893. Pfeffer (part), p. 84 (omit Usambara). 1896a. Socage, p. 92. 1868. Chlorophis Oldhami Theobald, Cat. Eept. Asiatic Soc. Mus., p. 49: "Simla, India" (in error). 1876. Cyclophis oldhami Theobald, Eept. British India, p. 159. 1882c. Philothnmnus hoplogaster Bocage, p. 17. 1887h. Boulenger (part), p. 176 (omit Wynberg and Damaraland). 1951c. Loveridge, pp. 4, 7. 1953. Loveridge, pp. 260, 318. 1894a. AhaetuUa neglecta Giinther, 1893, pp. 618, 620. 1898. Johnston, p. 361a. 1894a. Chlorophis hoplogaster Boulenger (part), p. 93, pi. v, fig. 2 (omit Damaraland and Lake Victoria). 1898. Boettger, p. 58. 1898. Sclater, p. 99. 1898. Werner, 1896-7, p. 143. 1907J. Boulenger, p. 486. 1907c. Eoux, p. 734. 1908b. Boulenger, p. 228. 1908. Gough, p. 24. 1910b. Boulenger (part), p. 507 (omit key, etc.). 1910a. Hewitt, p. 57. 1910a. Sternfeld (part), p. 18. 1912. FitzSimons, F. W. (part), pp. 86, 87, 92 (omit key). 1913. Boettger, p. 363. 1914a. Hewitt (part), pp. 242, 245, 246 (omit counts), 1915a. Boulenger (part), p. 205. 1915c. Boulenger (part), p. 622. 1929. Flower, p. 203. 1929. Hecht (part), p. 332. LOVERIDGE: AFRICAN SNAKE GENERA 71 192^^. Werner, p. 100. 1933h. Loveridge (part), p. 236 (omit Ukerewe Id.). 1934. Pitman, p. 294. 1935. Cott, p. 966. 1935. Power, p. 334. ]937o. Hewitt, p. 59, fig. 1. ]937f. Loveridge, pp. 489, 496, 503. 19391). FitzSimons, V. F., p. 22. 1940. Bogert (part), p. 54 (omit Lukolela). 1950. Rose, pp. 271, 320, fig. 166 (photo) ; p. 314, fig. 8 (head). 1955. Rose, pp. 93, 97 (faulty key), fig. 43 (photo); p. 178, fig. 8 (head). 1894a. Chlorophis neglectus Boulenger (part), p. 94 (omit localities in Kenya and Tanganyika). 1896d. Boulenger (part), p. 631 (omit Witu). 1896. Peracca, p. 2. 1897. Tornier, p. 65. 1898. Tornier, p. 294. 1899a. Mocquard, p. 219. 1907a. Boulenger, p. 10. 1907. Lonnberg, p. 15. 1908b. Mocquard, p. 558. 1908c. Sternfeld, p. 246. 1910b. Boulenger (part), p. 507 (omit key). 1910. Peracca, p. 4. 1910a. Sternfeld (part), p. IS (omit Kagera, etc.). 1912. FitzSimons, F. W. (part), pp. 86, 87 (omit key). 1912. Peracca, p. 5. 1913. Boettger, p. 345. 1913. Hewitt & Power, p. 162. 1915a. Boulenger (part), p. 205. 1915c. Boulenger (part), p. 623. 1915d. Boulenger (part), p. 648. 1928c. Barbour & Loveridge (part), p. 115 (omit Usambara localities). 1928. Cott, p. 953. 1928J. Loveridge, p. 74. 1929. Flower, p. 203 (omit range). 1929. Hecht (part), pp. 332, 333. 1929a. Werner (part), p. 100. 1933h. Loveridge, p. 237. 1934. Pitman, p. 294. 1935. Cott, p. 966. 72 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1937a. Parker, p. 630. 1941. Moreau & Pakenham, p. 108. 1941. Themido, p. 16. 1942e. Loveridge, p. 271. 1943. Scortecci, p. 270. 1950. lonides, p. 101. 1950. Eose, pp. 271, 320. 1955. Eose, pp. 93, 97 (faulty key). 1902b. Chlorophis natalensis Boulenger (not Smith), p. 17. 1910b. Boulenger (part), p. 507 (Kafue Eiver record only). 1948a. FitzSimons, V. F., p. 73, pi. i, fig. 1. 1933111. PhUothamnus dorsaUs Witte (part, not Bocage), p. 89. Further citations of "hoplogaster' and "neghctus" will be found under macrops, i. haftershyi, i. ornatus, i. irregularis and heterolepidotos. Common Names. Southeastern Green-Snake (Loveridge) ; Green Water-Snake (FitzSimons) ; Leaf -Snake (Cott) ; imhindi- pidni (Nyakusa :Loveridge) ; masamandimo (SenarCott); naJ- ivinduwindu (Misuku :Loveridge) ; namalanga (Makonde :Lover- idge) ; namasamha (Nyanja, but not even generic) ; nyaluwina (Hehe :Loveridge) ; nyoka amani (Sambaa :Loveridge) ; nyoka msipu (Chewa; Ngoni :Loveridge). Description. Preoeulars 1, rarely 2; postoculars 2, rarely 3; temporals 1 + 1, rarely 1 + 0, 1 -j- 2, 2 -f 1 or 2 + 2 ; upper labials 7-9, the fourth and fifth, rarely the third and fourth, the third, fourth and fifth, the fourth, fifth and sixth, or the fifth and sixth, entering the orbit ; lower labials 9-11, rarely 13,"' the first 4, 5, or 6 in contact with the anterior sublinguals. Mid- body scale-rows 15, very rarely 13 or 14"*^; ventrals 130°^ or 141- 164 (S $ 141-160'^ 9 9 143-164); anal divided; subcaudals 73-106 ( 9 9 73-98"; $ S 87-106). 6T Thirteen on one side only of a Nyange snake (M.C.Z. 123208). 68 Thirteen on a Fort Johnston ? : 14 on a Zonilia 9 (Loveridge :1953h). 63 130 on an aberrant Rungwe reptile recorded and verified by Bogert (1940 :54) ; 141 is the lowest count for 30 cfd" in the Museum of Comparative Zoology. 70 161-163 (Laurent :1936) and 173 given by de Witte (1953 :183) requires checking, as do his low subcaudal counts (9$ 59-88: cfcT 78-10.5) which may lie due to inclusion of specimens with regenerated tail tips. 7198 in a Mazoe snake (B.M. 02.2.12.87). and 91, not 93, in one 9 cotype. LOVERIDGE: AFRICAN SNAKE GENERA 73 For characters common to all species, and synopsis of scale- counts, see pp. 49-52, 55. Color. Above green, the interstitial skin black, uniform, or nape and anterior part of back with a long-itudinal series of paired black, blue (at IIolo), or brown (in life at Lake Rutamba) spots, that sometimes coalesce to form crossbars ; some scales with a white basal spot. Below, pure white to very pale green. Seven black crossbars on a young snake, found in a heap of debris beneath a mango tree at Mikindaui, gave it some resem- blance to a young night-adder (Causus resimus). F. W. Fitz- Simons (1912) remarks that where they occur together, green- snakes are apt to be mistaken for the young of the Green Mamba (Dendroaspis angusticeps) . Size. Length of cotype S (Brit. Mus.), 650 (432 + 218'-) mm. from Durban; exceeded by a 5 (M.C.Z. 51101), 720 (500 -f 220) mm. from the Misuku Mountains, and, if sex and deter- mination are correct, one of 821 (638 -|- 183) mm. recorded by de Witte. Largest 9 (B.M. 02.2.12.87), 945 (685 + 260) mm. from Mazoe (measured by me). Remarks. C. hoplogaster of Boulenger (1894a) was a com- posite of true hoplogaster and a northern form from which hoplogaster was probably derived — for they are separable only on subcaudal counts and relative tail lengths. C. neglectus of Boulenger (1894a) was likewise a composite of the same two reptiles, for Boulenger separated his "neglectus," whose ventrals were more or less distinctly keeled, from his "hoplogaster," in which there were no trace of keels. Actually every gradation in this character is to be found throughout the range of both snakes. In recent times Bogert (1940) was the first to point out that neglectus Peters is indistinguishable from hoplogaster Giinther. The type of oldJiaiui, whose ventral and subcaudal counts are unknown, is no longer in the collection of the Indian Museum as I am informed by its Director ; neither can it be found in the British Museum, says Mr. J. C. Battersby. Dr. V. FitzSimons concurs (letter of 4.xii.50) in the assignment of his (1948a) "natalensis" from the Drakensberg, to hoplogaster. 72 Not 228 by my measuring (A.L.). 74 BULLETIN : MUSEUM OP^ COMPARATIVE ZOOLOGY Dentition. Maxillary teeth numbered 27 in two Rungwe snakes examined by Bogert (1940) ; apparently 25 and 26 respectively in skulls from Lake Rutamba and Ujiji studied by me, the last few teeth being enlarged. Hemipenis. Not bifurcate ; sulcus spermaticus undivided ; basal spines large, merging on the middle third into reticulate calyces which grade into papillae towards the tip, which extends to the eighth caudal (Bogert). Sexual dimorphism. Negligible, owing to the considerable overlap in the number of subcaudals, though there is undoubtedly an average difference. Breeding. On March 3, at Mwaya, a 9 held 5 eggs about 28 x 8 mm. On October 1, at Nyange, a 9 held 6 eggs about 20 x 7 mm. Between October 9-29, at Vituri, a 9 held 3 eggs about 34 x 10 mm. On December 18, at Chitala, a 9 held 6 eggs about 20 x 28 mm. On February 12, at Chowe, a hatchling measured over 193 (140 + 53) mm. It is possible that the Vituri snake had already laid some of her clutch. Giinther (1894a), without giving a date, mentions 5 eggs as ready for extrusion by a Shire Highlands snake. Diet. Lizards and frogs. More specifically, a gecko {Henii- dactylus persimiJis) in the stomach of a Bagamoyo specimen; at Dar es Salaam I watched a captive snake take a gecko {Lygo- dactyhts p. pictnratus) ; FitzSimons (1939b) notes a Chirinda Forest reptile had eaten a gecko (L. capensis) and frog (Arthro- Jeptis s. stenodactylus) ; several frogs of this same race were in Kitaya snakes, and I caught a Bagilo snake with an A. s. whytii in its mouth, recovered a second whytii from a Vituri specimen ; and a third whytii at Ilolo ; A. reichei at Kigogo; three out of four captive green snakes died after making heavy meals of Rana m. mascareniensis. A Nyange snake held a frog, a buprestid beetle and a grasshopper, with the possibility that digestion might have released the insects from the frog's stomach. LOVERIDGE : AFRICAN SNAKE GENERA 75 Parasites. Immature encysted worms {Acanthocephala sp.) were present on the outer wall of the intestinal tract of a Bagilo snake which held an adult 9 ; several Nyange snakes, infested with subdermal cysts, died a few days after being brought in by natives ; Vituri specimens were also parasitized. Eyiemies. One, at least, was eaten by a Thelotornis k. kirtlandii that at first shared the vivarium with some green snakes. Aestivation. At the height of the dry season Neave dug one from a termitarium on the bank of the Loang^^^a River, at which time the ground "was so hard that it could hardly have entered except during the previous rains" (Boulenger, 1907a). Habitat. Southeastern Green-Snakes are very adaptable, being found on the coastal plain, upland savanna, or montane forest, their distribution being governed by conditions sufficiently moist to support the frogs on which they subsist. Consequently they frequent marshes, ponds, and rivers, being as adept at swimming as they are at climbing. F. W. FitzSimons (1912) remarks that when a frog is captured in the water, the snake swims ashore and glides up the bank carrying its prey well off the ground. I (A.L.) found one on the thatch of a watcher's hut in a rice .swamp, and many on the fringe of, or in clearings in, rain forest. According to Cott (1935) they abound among the boras- sus palms near the mouth of the Zambezi. Localities. Tanganyika Territory: *Amboni near Tanga ; *Ba- gamoyo ; *Dar es Salaam; *Ilolo; Kibongoto (Kibonoto) ; Kili- manjaro Mtn. ; *Kilwa ; *Kitaya ; Kitopeni ; *Lake Rutamba ; *Liwale ; *Magrotto Mtn. ; Marangu ; *Mikindani *Morogoro ; *Mwaya ; *Nehingidi ; Rungwe Mtn. ; *Ruponda ; Tanga ; Tukuyu (as Langenburg) *Tunduru; Uhehe ; *Ujiji; *Uluguru Mtns. — *Bagilo — *Nyange — *Vituri ; Uzungwe Mtns. — *Kigogo ; Wanga (as Wange). Zanzibar: Mafia Id. only. Mozambique: Beira; Boror (as Prazo Boror) ; Caia ; Charre ; Fambani; Jeowesa; Lourenco Marques ; Mozambique ; Quilimane. Nyasaland: *Chi- tala River; *Cholo Mtn.; *Chowe ; *Fort Johnston; Karonga; Mandala Hill, Blantyre ; *Misuku Mtns.; Mlanje Mtn.— *Ruo River ; *Nchenachena ; *Nchisi Mtn. ; Shire Highlands ; *Zomba Mtn. Belgian Congo'"': Bukena ; Elizabethville ; Ganza; Inkan- '3 Lukafu being the only Congo locality from which I have seen a specimen, all liut 2 of the remaining 42 are based on identifications by de Witte (1953), whose report should be consulted for more precise locality data. 76 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY gala ; Kabenga ; Kabwe ; Kakyelo on Luombwa River ; Kalule- North River ; Kamina ; Kande River ; Kankala River ; Kanonga River ; Kanzenze ; Kajiiri ; Kateke River ; Katentania ; Kaswabi- lenge ; Kihvezi River ; Kimiala River ; Lofoi ; Lul)umbasbi ; *Ijiika- fu ; Lukima River ; Lukonzohva ; Liikuoa-Niemba confluence : Lupiala River ; Lusinga ; Mabwe, Lake Upemba ; INIpala ; Miinoi ; Musosa ; Ngayu ; Nyonga ; Nyuiizu ; Panda ; Sakania ; Sandoa ; Sanga River ; Senze River ; Seram Station. Belgian Ruanda- Urundi: Mosso Camp, Rutana Territory. Northern Rhodesia: Boruma (as Boroma, Zambezi :1898) ; Broken Hill, Kat'ue River; Kazungula ; Loangwa River ; Lungasa ; Lnapula River ; Macnbu ; Mashie 's ; Mterize River ; Namaliya 's ; Tapper Zambezi River ; Victoria Falls ; Zambezi Plains. Southern Rhodesia: *Chirinda Forest; *Mazoe; *Rietfontein ; Salisbury'*; Yumba Mtn. Trans- vaal: Barberton ; Lydenburg; Middleburg; *Pretoria; *Pretoria West. Natal: Drakensberg; *Durban; Lower Illovo River; Mere- bank; Mseleni, Zululand ; Port Natal = Durban (type locality) ; Ubombo; *Umvoti River; Vryheid. Cape Province: "Cape Colony"; East London; Knysna; Orange River; Pondoland; Port Elizabeth; Port St. Johns.'^ Range. East Africa south of about 5° S., i.e. south of the Lsambara Mountains, Tanganyika Territory, south to Durban, Natal, Avest to Port Elizabeth, Cape Province, and north through the Rhodesias to the eastern Belgian Congo. Philothamnus irregularis battersbyi Loveridge 1878a. Philothamnus hoplogastcr Peters (not Giinther), p. 206. 1892. Philothamnus neglcctus Matschie (not Peters), p. 110. 1893. Pfeffer (part), p. 84 (omit Quiliniane). 1894a. Chlorophis hoplogaster Boulenger (part; not Giinther), p. 93 (Vic- toria Nyanza only). 189G. Tornier, p. 69. 1897a. Tornier, p. 69 (reprinted). 74 Identification checked by Dr. Walter Rose. More recently D. C. Broadley writes me (20.ix.56) that he has collected examples of both hoplogaater and i. irregularis at Salisbury ; both species have also been reported as occurring in Chirinda Forest. 75 The Wynberg record of Boulenger (1910b) is obviously erroneous, for hoplo- (jaster does not occur in western Cape Province. Evidently Boulenger never saw the specimen on which the record was based, for W. Rose informs me (9.xi.56) that it is an example of Hemirhagerrhis nototaenia. LOVERIDGE: AFRICAN SNAKE GENERA 77 1897b. Tornier, p. 63. 1902(1. Boulenger, p. 446. 1910b. Boulenger (part), p. 507 (iiart key & range). 1910a. 8ternfeld, p. 18 (Bukoba). 1912. FitzSimons, F. W. (part), pp. 86, 87, 92 (part key & range). 191(ia. Loveridge, pp. 78, 84. 1918a. i.overidge, pp. 333. 19241). Loveridge, p. 5. 1929. Heeht (part), p. 332 (part key & range). 192911. Loveridge, p. 22. 1929a. Werner (part), p. 100 (part key & range). 19331). Loveridge (part), p. 236 (omit Tanganyika localities). 1936 j. Loveridge, p. 245. 1936. Pitman, p. 275. 1938a. Pitman, p. 305. 1939a. Scortecci, p. 274. 1942b. Bogert, p. 2. 1942e. Loveridge, p. 271. 1894a. Chlorophis n.eglectus Boulenger (part; not Peters), p. 94 (Lamu; Mkonumbi; and Teita in Kenya; Zanzibar Coast). 1894e. Giinther, p. 88. 1895. Giinther, p. 527. 1896a. Boulenger, p. 553. 1896d. Boulenger (part), p. 631 (omit Nyasaland localities). 1896. Tornier, p. 69. 1897a. Tornier, p. 69 (reprinted). 1908c. Sternfeld, p. 240. 1908e. Sternfeld, p. 243. 1909a. Boulenger, p. 193. 19101). Boulenger (part), p. 507 (part key & range). 1910a. Hternfeld (part), p. 18 (Kagera). 1910. Meek, p. 406. 1911. Lampe, p. 193. 1911. Lonnberg, p. 23. 1911b. Nieden, p. 442. 1912. FitzSimons, F. W. (part), p. 86 (part key & range). 1912. Hobley, p. 48. 1913. Lonnberg & Anderson, p. 4. 1916a. Loveridge, pp. 79, 84. 1916b. Loveridge, p. 118. 1917b. Loveridge, p. 180. 1918a. Loveridge, p. 332. 1923a. Loveridge, p. 26. 78 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1923e. 1924b. 1925a. 1925a. 1927. 1928c. 1929. 192911. 1929a. 1934i'. 1936h. 1936.i. 1936. 1937f. 1937. 1938a. 1938b. 1938. 1941a. 1941b. 1946a. 1918. 1929h. 1946. 1948. 1949a. l!»51a. 1951c. Loveridge, p. 878. Loveridge, p. 5. Calabresi, p. 106 (as Clorophis). Angel, p. 32. Calabresi, p. 53 (as Clorophis). Barbour & Loveridge (part), p. 115 (Usambara Iocs. & data). Hecht (part), pp. 332, 333 (part key & range). Loveridge, p. 22. Werner (part), p. 100 (part key & range). Mertens, p. 169. Loveridge, p. 30. Loveridge, p. 246. Pitman, p. 276, pi. v, fig. 3 ; pi. E., fig. 5. Loveridge, pp. 489, 493, 496, 503. Uthmoller, p. 108 (omit range). Pitman, p. 305. Pitman, pp. 89, 90, 305 (reprinted, pis. as in 1936). Uthmoller, p. 43. Uthmoller, p. 26. Uthmoller, p. 233. Loveridge, pp. 97, 108. Pliilotliamnus semivariegatus Calabresi (not Smith), p. 124. Chlorophu sp. Loveridge, p. 23. Blom-Bjorner, 1945, p. 159. "Green Grass Snake" Kingdon, p. 10. Chlorophis irregularis hoplognster Parker (not Giinther), p. 61. CJiloropliis irregularis hattersbyi Loveridge, Bull. Mus. Comp. ZooL, 106. p. 190: Sipi Forest at 6,000 feet. Mount Elgon, eastern Uganda Loveridge, pp. 4, 7. Common Names. Northeastern Green-Snake (English) ; em^un (Karamojong) ; homhoka (Pokomo) ; naranyase (Gishu) ; new- andegala (Ganda). Description. Preoeiilars 1, rarely 2''''; postoculars 2, rarel}' 3'' ; temporals 1 + 1 or 1 + 2, rarely 1 + S'*" or 2 + 2~^ ; upper "« Two ill two Kainidsi snakes (M.C.Z. 40501; 40004). '"■Three iu a Butamlijra snake (M.C.Z. 40492), while the lower postociilar is fused witli the tifth hiliial on one side of a Naivaslui snake (r.S..\..\l. 41701). '8 On one side of a Sanya snake {fide Uthmoller :1938). TflOn one side of a I5utandi};a snake (M.C.Z. 40489). LOVERIDQE: AFRICAN SNAKE GENERA 79 labials 7-9,''" rarely the third and fourth, usually the fourth and fifth, occasionally the fifth and sixth, entering the orbit; lower labials 8-11, the first four, five or six in contact with the anterior sublin*iuals ; niidbody scale rows 15; ventrals 147-177 ( S $ 147- 169; 9 9 153-177); anal divided, rarely entire*'; subcaudals 90-120 (9 9 90-111; £ $ 100-120). For characters common to all species, and synopsis of scale counts, see pp. 49-52, 55. Color. Above, pale or dark green, the interstitial skin black; some scales with a white basal spot, others with or without black edging. Below, chin and throat white ; remainder white, yellow- ish, or pale green. Size. Length of $ (M.C.Z. 40172) 841 (588 + 253) mm., and 9 (M.C.Z. 40155) 1045 (762 + 283) mm., both from Sipi, Mt. Elgon. Remarks. It will be noted that this northeastern race of irregularis usually attains to a greater length than does the south- ern hoplogasfer, with which it has long been confounded. Thus for a hoplogasfer <5 the maximum is 720 mm., for a 9 945 mm., while for an irregularis hattershyi S the maximum is 841 mm., for a 9 1045 mm. Though the northeastern hattershyi has so long been confused with hoplogaster (Glinther) or its synonym neglectus (Peters), with which it agrees in the number of labials entering the orbit, and to which it presumably gave rise, when sexed it may be dis- tinguished as follows: Subcaudals in 9 9 73-98, in 5 ^ 87-106; range: Eastern Congo and East Africa from the Usambara Mountains, south through the Rho- desias and Mozambique to Natal hoplogaster Subcaudals in 9 9 90-111, in $ $ 100-120; range: East Africa from the Usambara Mountains north through Kenya to Ethiopia, west to the Nile and Uganda where it merges with typical irregularis i. hattershyi ^o 7 Labials, 3 & 4 entering on both sides (Nairobi: U.S.N.M. 49007). 7 labials, 4 & 5, entering on 1 side (Nairobi : U.S.N.M. 40985). 8 labials, 4 & 5 entering on 20.1 sides. 8 labials. 5 & 6 entering on 1 side (Nairobi : U.S.N.M. 40985). 9 labials, 5 & 6 entering on 29 sides. 81 Entire in a Naivasha snake (U.S.N.M. 41701). 80 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY From typical irregularis it is separable only on the upper labials as follows : Two labials (rarely the 3rd aud 4th, normally the 4th and 5th, occa- sionally the i'th and 6th) enter the orbit in all northeastern material i. hattershyi Usually 3 labials (rarely 3rd, 4th and 5th, normally 4th, 5th and 6th, rarely the 5th, 6th and 7th) but occasionally (chiefly among Angolan snakes and not more than 5% of the West African population) only 2 labials (the 4th aud 5th, or 5th and 6th) enter the orbit !. irregularity The snake from Gelib and Margherita referred to by Calabresi (1918) as a 5 " semivariegatus" cannot be that species if the low ventral count is correct. It does agree in both ventrals (155) and subcaudals (90) with 9 i. haitershyi, but if a $ the tail must be truncated. Notes on labial variation, based on other material, will be found in Loveridge (1936j, p. 245). Sexual dimorphisin. Reference to the description {vide supra) reveals only a slight average difference in the number of sub- caudals as between the sexes. Breeding. More than a dozen detailed records will be found under "hoplogasfer" and "neglectus" in Loveridge (1936j, pp. 246-247). These, together with earlier records, would seem to indicate that in Uganda and western Kenya, from 4 to 8 eggs are laid towards the end of the lesser (November-December) or greater (February-March) rains, but that in the coastal belt of Kenya eggs are ready for laying about June. The single record of July 16 from Ukerewe Island, Lake Victoria, where only 2 eggs (measuring 29 x 8 mm.) were in the oviducts of a large female, suggests that other eggs of the series had been deposited already. Diet. At various tims I have recorded stomach contents com- prising the following lizards : Cncmaspis a. elgoncnsis; Chamaeleo h. hoehnelii; Mabnya s. striata; Ahlepharus wahlbergii and La- certa jacksoni, and frogs : Rana a. edulis, B. o. gribinguiensis and Hyperolius rossii, sometimes several examples in one snake, at other times several snakes having swallowed the same species. An interesting account of a struggle between a "vivid green grass snake" and a five-or-six-inch catfish {Amphilius) has been LOVERIDGE: AFRICAN SNAKE GENERA 81 furnished by Major J. Kingdon (1948). He was near Thego Ford, Nyeri Station, \s'hen his attention was attracted by a violent com- motion in a sliallow water furrow above whose surface waved the coils of a snake. As Kingdon approached the spot, the reptile ])ulled the tish from the runnel and started with it across the i-ivcr. On reaching deeper water, however, the fish took control and the combatants were carried down stream to shallower water where the snake regained the advantage and, by entwining among the rocks, succeeded, though only with a great effort, in dragging the Amphilius along. Both appeared exhausted by the struggle, especially the snake, which was periodically submerged by the weight of its prey. Eventually the fish, by diving among sunken twigs, achieved its freedom and swam off down the runnel. Parasites. An immature ascarid in a Nairobi snake (Lover- idge:1923e), others encysted in the stomach wall of a Kaimosi reptile (1936J). Enemies. I recovered one of these green-snakes from the stom- ach of a harrier-eagle {Circaetus g. gallicus) shot at the foot of Mt. Debasien, while two others were taken from the crop of a C. cinerascens by Blom-Bjorner (1946). Tempcramenf. One of a pair of green snakes that I caught bit sufficiently hard to draw blood (1917b). Habits. Two, basking on bushes overhanging Nairobi River, escaped by dropping off into the water (1916b). Hahitat. One was resting on a lily pad far out in a swamp where it was probably hunting frogs; in this same flooded area another was associated with two Variegated Bush-Snakes (Philo- fhamnus s. semivariegatiis) in a small doom palm. While this was on the coastal plain, hattcrshyi is as much at home on upland savanna or montane forest-edge. As a variation of its usual habitat, one put its head out of a leather hairbrush-case on a Nairobi dressing table that it had reached through an open casement. Uthmoller (1941b) believed that this and certain other snakes actually thrive and multiply when land is cleared for the cultiva- tion of bananas, coffee or maize, and disappear when the planta- tions revert to bush. I would suggest that their apparent prev- 82 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY alence in cultivated areas is due to the reptiles being more readily seen and killed by the numerous Africans employed about a plantation. Localities. Sudan: *Torit. Ethiopia: Abbai ( ? Lake Abaya, Tancredi Expedition to Lake Tana) ; Addis Ababa; Gadat, Gofa ; Gondar ; Guder ; Let Marefia, Shoa. Somalia: Belet Amin ; pos- sibly Gelib and Margherita, west of Juba River (if the " semi- variegatus" of Calabresi is hattersbyi) ; Kismayu. Uganda: Ka- gera; Mabira Forest — *Mubango; *Mt. Debasien; *Mt. Elgon — *Butandiga, *Sipi ; *Ntenjeru near Sesibwa River. Kenya Col- ony: *Athi River ; *Fort Hall ; * Juja ; Kabete ; *Kaimosi, Kaka- mega (Kakumega) ; *Kau; Kavirondo; Kenya Mtn. ; *Kenya Province; Kibwezi; *Kijabe; *Kirui's Village; Kitui; *Lake Elmenteita ; Lake Naivasha ; *Lake Peccatoni ; Laniu ; Masai Re- serve ; Meru Boma ; *Mkonumbi ; Mtito Andei ; *Nairolii ; Na- kuru ; *Ngatana ; Nyeri Station ; Teita ; Thika ; Tumutumn ; * Voi ; Wambugu ; *Witu ; Yala River Tanganyika Territory: Bukoba ; Gomberi ; *Ikizu ; *Korogwe ; Lake Victoria ; Marangu ; *Matembo ; Sanya ; *Ukerewe Id. ; Usambara Mtns. — *Amani, Bulwa, *Bumbuli, Derema, Lewa, *Mlalo near Hohenfriedeberg, Nguelo; "Zanzibar Coast" (Boulenger 1894a: probably T^sam- bara). Range. East Africa (east of 33° E., south of 15° N., and north of 5° S.). More specifically, the Sudan (east of the Nile), east through Ethiopia to Somalia, south through Uganda and Kenya Colony to (and including) the Usambara Mountains of north- eastern Tanganyika Territory, west to Lake Victoria. Philothamnus irregularis ornatus Bocage 1872. Philothamnus ornatus Bot-age, Joni. Sei. Lisbon, 4, p. 80: Huilhi, Angola (restricted by Bogert:1940). 1882t-. Bocage, p. 15, fig. G. 18841). Sauvage, p. 201 (as PJiilotJiulmas in French Congo list). 1895a. Bocage, p. 93, pi. xii, figs, lac 1896a. Bocage, p. 78. 1897a. Bocage, p. 200. 1894a. Chlorophis ornatus Boulenger, p. 93. 1896d. Boulenger, p. 631. 1905c. Boulenger, p. 112. LOVERIDQE: AFRICAN SNAKE GENERA 83 1915a. Boulenger, p. 205. 1919a. Boulenger, p. 281. 1921a. Angel, p. 42. 1929. Hecht, p. 332. 1929a. Werner, p. 100. 1931. Monard, p. 103. 1933f. Angel, p. 109. 1937b. Monard, pp. 113, 119. 1940. Bogert, p. 51. 1940b. Monard, p. 148. 1950b. Villiers, p. 80. 1937b. Chlorophis hoplogaster Monard (not Giinther), pp. 114, 120. 19.")lo. Philothamnus irregularis ornatus Loveridge, pp. 4, 8. A further citation of "ornatus" will be found under hetero- lepidoius. Common Name. Ornate Green-Snake (Loveridge). Description. Preocular 1 ; postoculars 2 ; temporals 1 -{- !> rarely 1 + 2^'; upper labials 8-10, the third and fourth, normally the third, fourth and fifth, rarely fourth and fifth, or fourth, fifth and sixth, entering the orbit; lower labials 10, the first 5 or 6 in contact with the anterior sublinguals. Midbody scale-rows 15; ventrals 150-166 {i 150; 9 9 160-161); anal divided; subcaudals 85-106 (9 9 99; i 106). For characters common to all species, and synopsis of scale- counts, see pp. 49-52, 55. Color. Above, iridescent green or bronze green, the interstitial skin black, a few scattered black spots on the nape and a broad, usually yellow-edged, chocolate-brown, vertebral band extending from head to end of tail ; this band occupies the median series of tlorsal scales and encroaches on those on either side ; a few scales may be flecked with white. Below, yellowish cream, uniform, or the posterior margins of some ventrals brownish ; tail greenish. Size. Length of $ (A.M.N.H. 51772) 610 (405 + 205) mm.; hirger 9 (P.M. 20-100) 693 (480 + 213) mm., both surpassed liy an unsexed cotype (Bocage) of 710 (510 -|- 200) mm. Remarks. P. i. ornatus is apparently separable from tj^pieal irregularis only by its striking coloration, which resembles that 82 Only ill the Kimbi snake referred to "hoplogaster" by Monard (1937b). 84 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY of the perfectly distinct Pkilothamnus s. dorsalis Bocage of the same general region. P. i. ornafus occupies a fairly well-defined area in southwestern Angola where it is ]iartly surrounded by typical irregularis as it approaches the limits of its range in the southwest. Possibly undue emphasis has been laid on the striking colora- tion of ornatiis and a more representative southwestern race of irrcgidaris would embrace all Angolan material possessing a low number of ventrals and subcaudals. This point could be settled by someone assembling all available Angolan material of the genus. 1 have seen the specimen of ornatus recorded from neighbor- ing Northern Rhodesia by Angel, but the alleged ''ornatus" from the Congo proved on examination to be a misidentified heterolepi- (lotus. I also would ([uestion Sauvage's listing of it from the French Congo, though it may indicate the recrudescence of a recessive pigmentation. Bocage himself (1872) listed one of his three cotypes as coming from Cacheu (Cacheo), a river and bay in northern Portuguese Guinea. Later (1896a) he gives Bissau as the type locality in his list of types from Portuguese Guinea in the Lisbon Museum. I am inclined to suspect an error in labeling, for in a later list Bocage returns to Cacheu. Probably the specimen originally came from Angola, from which countrj^ 10 specimens are now known, justifying Bogert's (1940) action in restricting the type locality to Huilla, from whence we have a cotype. There is but a single example of oriiatus in the British Museum. Dentition. Maxillary teeth 26, slightly increasing in size pos- teriorly (Bogert). Ilemipenis. Not bifurcate ; sulcus spermaticus undivided ; basal spines large, blunt, merging into 6 flounces which occupy the iiiiddle third ; passing on the terminal third into reticulate calyces which grade into papillae towards the tip, which extends to the eighth caudal (Bogert). Localities. Angola: Benguela to Bihe ; Bimbi ; Caconda; Ca- lu(iuembe (Kalukembe) ; Chimporo ; Cunene ; Cutato (Kutatu) ; Huambo; *Huilla. Northern Rhodesia: *Lealui (Lialui).''^ Range. Southwestern Northern Rhodesia west to southwestern Angola. ''' 83 S(>o Kcmarku reKiirdinj? its listing,' from French Congo tiy Sauvage (18S4) and Portuguese Guinea: Cacheu (Bocajie :1872) . LOVERIDGE: AFRICAN SNAKE GENERA 85 Philothamnus irregularis irregularis (Leach) Series I. Mostly typical, having 3 labials entering the orbit. 1789. C{olub€r) Cacndcus Lacepede (not Linnaeus, 1758, p. 218, No. 189), Hist. nat. Serpens, 2, pp. 100, 276: Cap Vert. ]S03e. Coluber caerulescens Daudin (not Linnaeus, 1758, p. 227, No. 385), Hist. nat. Rept., 7. p. 54: new name for caeruleus Lacepede from "Cap Verd." 1818. Coluber cae.mts Cloquet, Diet. Sci. nat. (Paris), 11, p. 201: new name for (-(uruleus Lacepede (Suppressed: cf. Bemarks). 1819. Coluber irregularis Leach, in Bowdich, Miss. Ashantee, p. 494: Fantee, Gold Coast. 1S4:0. Dendrophis (Philothamnus) albo-variala A. Smith, 111. Zool. S. Africa, Rept., pi. Ixiv, figs. 3-3b; pi. Ixv: South Africa towards Tropic of Capricorn (also Ghana; Sierra Leone; Gambia). 1843. Dendrophis Chenonii Reinhardt, Dansk. Vidensk. Selsk. Skrift., 10, p. 246, pi. i, figs. 13-14: Guinea. 1 854a. Leptophis Chenonii Dumeril & Bibron, 7. p. 545. 1860. Dumeril, A., p. 199. 1866a. Bocage, p. 48. 1881. Jan, pi. i, fig. 2. 1858. Ahaetulla irregularis Giinther (part), p. 152. 1863c. Gunther, p. 285. 1874. Reichenow, p. 292. 1882b. Miiller, p. 169. 1885. Muller, p. 683. 1893. Gunther, p. 555. 1898. Johnston, p. 361a. 1867b. Fhilotlxamnus alhovariatus Peters, p. 236. 1882c. Bocage, p. 19. 1875a. Philothamnus irregularis var. longifrenatus Buchholz & Peters, Monatsb. Akad. Wiss. Berlin, p. 199: Cameroon. 1877c. Philothamnus irregularis Peters, pp. 615, 620. 1881b. Boettger, p. 396. 1881d. Peters, p. 149. 1882c. Bocage, pp. 3-4, fig. 1. 1884a. Rochebrune, p. 172 (ignored). 1887a. Bocage, p. 205. 1888a. Boettger, p. 61. 1890. Biittikofer, p. 446. 1895a. Bocage (? part), p. 85, pi. xii, figs. 2a-2c (1 angolensis) . 1896a. Bocage, pp. 78, 92. 86 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 18!)6b. Bocage, p. 111!. 18960. Bocage, p. 17G. 1905. Ferreira, p. 167. 1952b. Villiers, p. 888. 19.i4d. Laurent, p. 305. 1885. AhaetziUa (Leptophis) fraenata Miiller, p. 684 (f lapsus: not A. frenatus Giinther which = Uromacer freruitius of Haiti). 1888a. Ahaetnlla emini Giinther, Ann. Mag. Nat. Hist., (6) 1. p. 51: ' ' Monbuttu, ' ' = Mangbetu Belgian Congo. 1888b. Gunther, p. 325. 1888b. Ahaetulla shirana Giinther, Ann. Mag. Nat. Hist., (6) 1. p. 326: Blantyre Mission, Shire River, Nyasaland. ]1893. Philothamnus GuntJieri I'feffer, Jahrb. Hamburg. Wiss. Anst., 10. p. 85, pi. i, figs. 3-5 : Quilimane, Mozambique. 1905. Ferreira, p. 168 (Angola). 1895. Chlorophis Giinthcri Gunther, p. 528 (key). 1895. C/iioropMs ^/iiroTia Giinther, p. 528 (key). 1891a. Chlorophis irregularis Boulenger, p. 306. 1894a. Boulenger (part), p. 96 (omit locality Ugogo). 1895. Gunther, p. 528. 1896d. Boulenger, p. 631. 1896b. Mocquard, p. 45. 1896c. Mocquard, p. 59. 1896. Peracca, p. 2. 1896. Tornier, p. 70, fig. D. 1897b. Boulenger, p. 278. 1897e. Boulenger, p. 801. 1897. Sjostedt, p. 35. 1897. Tornier, p. 6.1. 1897b. Werner, p. 39S. 1898. Boettger, p. 58. 1898. Johnston, p. 361a. 1898. Sclater, p. 99. 1898a. Werner, p. 209. 1899a. Werner, pp. 147, 149. 1900b. Boulenger, p. 452. 1902d. Boulenger, p. 446 1902a. Werner, pp. 334, 344. 1903a. Ferreira, p. 10. 1905c. Boulenger, p. 112. 1906i. Boulenger, p. 213. 1908a. Sternfeld (part), p. 425 (omit Bipindi). 1908b. St«rnfeld, pp. 214, 230. LOVERIDGE : AFRICAN SNAKE GENERA 87 1909a. 1909b. 1909. 1909. 1909a. 1909b. 1910b. 1910a. 1910d. 1911c. 1911. 1912. 1912b. 1912c. 1915a. 191.5c. 1916f. 1917b. 1917. 1919a. 1921a. 1921a. 1921b. 1922. 1923e. 1923. 1927. 1927d. 1928. 1929. 1929. 1929a. 1930a. 1931. 1932a. 1933f. 1933. 1933. 1933m. 1934:c. 1934. 1934a. 1934b. Chubb, p. 595. Chubb, p. 35. Gendre, p. cvi. Peracca, p. 172. Sternfeld, p. 14. Sternfeld, p. 15. Boulenger, p. 508. Sternfeld (part), p. 19 (omit Ugogo). Sternfeld, p. 64. Boulenger, p. 165. Lepri, p. 324. FitzSimons, F. W., p. 87. Sternfeld, p. 384. Sternfeld, p. 270. Boulenger, p. 205. Boulenger, p. 623. Chabanaud, p. 371. Chabanaud, p. 11. Sternfeld, p. 468. Boulenger, p. 282. Angel, p. 42. Chabanaud, p. 46S. Chabanaud, p. 524. Aylmer, p. 15. Loveridge, p. 878. Schmidt, p. 76. Calabresi, p. 53. Witte, p. 323. Cott, p. 953. Flower, p. 203. Hecht, pp. 331, 334. Werner, p. 101. Barbour & Loveridge, p. 772. Monard, p. 104. Parker, p. 219. Angel, p. 108. Flower, p. 806. Sehouteden, p. 236. Witte, p. 89. Mertens, p. 169. Pitman, p. 294. Schwetz, pp. 380, 383. Schwetz, supp. p. (24). 88 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 193-13. Corkill (part), p. 17 (part heterolepidottis) . ]93o. Cott, p. 966. ]9361i. Loveridge, p. 30 (part hrtrrnlrpiiToins). 1936.i. Loveridge (part), p. 247 (omit Kaimosi specimens). ]936(". Parker, p. 125. 1936. Pitman, p. 278, pi. v, fig. 5, pi. E. fig. 3. 1937:1. FitzSimons, V. F., p. 273. 1937c. Loveridge (part), p. 273. 1937f. Loveridge, pp. 489, 496, 503. 19371). Monard (part), pp. 114, 121 (tentatively omWiin^ angolensis) . 193SO. Mertens, p. 439. 1938a. Pitman, pp. 306, 328. 19381). Pitman, pp. 23, 39, 87, 91-93, 98, 294, 306, 328 (reprinted). 19391). FitzSimons, V. F., p. 22. 194(1. Bogert (f part), p. 53 (? omit one of Capelongo series). 19401 1. Monard, p. 175. 1941. Themido, pp. 4, 10. 1941. Witte (part), pp. 184-198 (part heterolepidotus). 1942e. Loveridge, p. 272. 194"). Leeson, p. 1 (key). 1950f. Laurent, p. 128. 1950. Leeson, pp. ix, 39, 40, 66, 133, fig. 17. 1950. Eose, p. 272. 1950b. Villiers, p. 81, figs. 9(i-!t7. 1951. Monard, p. 149. 1955. Eose, pp. 93, 97 (faulty key). 1894a. CMoroi)his emini Boulenger (part), p. 92, pi. v, fig. 1 (omit Sudan specimens which are heterolepidotus). 1895. Gunther, p. 528 (key only). 1900. Flower, p. 968. 1902d. Boulenger, p. 446. 1908c. Boulenger, p. 5. 1908a. Sternfeld, pp. 407, 424. 1908. Werner (1907), p. 1871. 19091). Boulenger, p. 303. 1909g. Boulenger, p. 246. 1909b. Sternfeld, p. 15. 1910. Peracca, p. 4. 1910. Eoux, p. 99. 1910a. Sternfeld, p. 18, fig. 17. 1911c. Boulenger, p. 165. 1912c. Sternfeld, p. 268. 1915a. Boulenger, p. 204. LOVERIDGE : AFRICAN SNAKE GENERA 89 1915e. Bouleuger, p. 622. 1915d. Boulenger, p. 648. 1916f. Chabanaud, p. 371. 1917b. Chabanaud, p. 11. 1917a. Phisalix, p. 33i. 1922a. Angel, p. 39. 1929. Hecht, pp. 332, 333. 1929. Pitman, p. 145. 1929a. Werner, p. 100. 1933h. Lovei-idge, p. 236. 1933. Schouteden, p. 236. 1933J. Witte, p. 123. 1933ui. Witte (part), p. S8 (omit Elisabethville and Kanzenze). 1934c. Mortens, p. 155. 1935a. Corkill (part), p. 16. 1939. Brien, p. 130. 1892a. Phtlothamnus irregularis var. lagoensis Bocage (not Giinther), p. 183. 1896a. Philothamnus irregularis var. Giintherii Bocage, p. 92 (Boror). 1902a. Chlorophis lieterolepidotus Werner (part), p. 344. 1933111. Witte (part), p. 89 (Faradje and Mahagi). 19401i. Monard, p. 175 (Portuguese Guinea). 1913. Chlorophis hoplogaster Hewitt & Power (not Giinther), p. 162. 1941. Witte, p. 184. 1923. Chlorophis iequaerti SL-hmidt (part), p. 75 (paratype $). 1932. Chlorophis vcrnayi FitzSimons, Y. F., Ann. Transvaal Mus., 15, p. 38: Maun, Thamalkane River, British Bechuanaland. 1935b. FitzSimons, V. F., p. 132, fig. 1. 193Zm. Chlorophis heterochriuus Witte (part:not Ilallowell), p. 89 (Albert- ville). 1941. Witte (part), p. 198 (Buiunga). 1938d. Philothamnus nifidus Loveridge (not Giinther), p. 57. 1948a. Villiers, p. 202. 1950a. Villiers, p. 62. 19.10b. Villiers (part), p. 82, figs. 98-99. 1951a. Villiers, p. 822. 1954b. Villiers, p. 1240. 1946. Philothamnus semirariegatns nitidus Orton & Morrison (not Giin- ther), p. 16. 1951c. Philotham')ius irregularis irregularis Loveridge, pp. 4, 8. 1951d. Loveridge, p. 88. 1953e. Loveridge, p. 261. 1953. Witte, p. 189, fig. 55, pi. xvii, fig. 1. 90 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 1954. Augel, Guiho, Lamotte & Koy, p. 392. 195-41). Laurent, p. 47. 1954d. Lament, p. 296. 1954b. A^illiers, p. 1239. 1955. Witte, p. 216. 1956. Laurent, p. 160, fig. 26, pi. xriii, fig. 1. 1953. Philothamnns lioplogaMer Witte (not Giinther), fig. 52 only. 1955a. Cldorophis irregularis irregularis Mertens, p. 91. Further citations of "irregularis," " emini," and "bequaerti" will be found under naialensis, heterolepidotus, s. ihomensis and s. scmivariegatus. Series II. Aberrant individuals with 2 labials entering orbit. 1881. Leptophis Chenonii var. alhovariata Jan (not A. Smith), No. 50, pi. i, fig. 1: (no locality mentioned). 1881. Leptophis Chenonii Jan (not Dumeril & Bihron), No. .'0, pi. 1, figs. 3-4: (no locality mentioned). 1882c. Philothamnns angolen. to 21/4 times its depth. Bocage based the name on an aberrant irregularis in which the tliird labial was excluded from the orbit. From time to time similar individuals have been recorded as occurring alongside normal irregularis throughout much of its range (as here de- tined) but form less than 5 per cent of the population. They are inseparable from the northeastern population which I have named irregularis hattersdyi, as angolensis cannot be applied to them. C. emini (Glinther; 1888a) was based on a snake in which the ventrals lacked lateral keels, such individuals being not at all uncommon. Witte (1933m) recorded both irregularis and emiiii from Elisabethville, Kikondja, Lulenga, Nyonga, Shangugu without it apparently occurring to him that they were synony- mous. The actual credit for synonymizing emini with irregularis goes to Flower (1933) who remarks that in Sudanese snakes the keels "on the ventral shields may be perceptible, just perceptible, or not perceptible. ' ' The question arises as to whether keels are always present in western, or distinct from central, African irregularis. C. vernayi FitzSimons (1932), was synonymized by Bogert (1940) with irregularis in the broad sense in which it was then understood. Sternf eld's (1908b) Togo record of " natalensis" was referred by Angel (1933f) to irregularis, but the data clearly suggests he was dealing with an example of heterolepidotus. I myself (1936h : 94 BULLETIN : MUSEUM OF COMPARATn^B ZOOLOGY 1937c) was mistaken, though with good reason, for considering that heterolepidotus (Giinther) and schvhotzi Sternfeld were synonyms of irregularis (Leach). It may be observed that on the basis of scale-counts there are some slight grounds for separating a northwestern race of irregularis from those in the rest of Africa (for Angola and Congo snakes are essentially like those extending towards the southeast). The actual figures are: Suhcaudals Southeast etc. $ $ 103 (Angola) and 106 (Nyasaland) to 121 (Nyasaland). Northwest $ $ 113 (Liberia) to 127 (Sierra Leone). Southeast etc. 9 5 ? 90 (Angola) and 97 (Nyasaland) to 121 (Angola). Northwest 9 9 100 (Dakar) to 119 (Sierra Leone). Ventrals Southeast etc. $ $ 150 (Angola) and 152 (Nyasaland) to 167 (F. E. Africa). Northwest $ $ 162 (Liberia) to 179 (Portuguese Guinea). Southeast etc. 9 9 153 (Nyasaland), 154 (Angola) to 170 (An- gola; Sudan). Northwest 9 9 162 (Sierra Leone) to 183 (Senegal). liowever, the overlap in average specimens is so extensive that an attempt at separation would inevitably result in deploral)le confusion. Dentition. Maxillary teeth 22 (Liberia) to 28 (Congo, etc.) in six snakes examined by Bogert (1940), who found that they were subequal in some, increasing in size posteriorly in others. Anatomy. Parotids present (Phisalix). Hemipenis. Not bifurcate ; sulcus spermaticus undivided ; basal spines large, merging on the middle third into reticulate calyces which grade into papillae towards the tip. which extends to the seventh or ninth subcaudal (Bogert). Sexual dimorpJiism. None so far as the number of ventrals and subcaudals are concerned. Breeding. At various times I have recorded the following : On September 30, a Misuku Mtns. 9 held 11 eggs measuring ca. 15 X 11 mm. LOVERIDGE : AFRICAN SNAKE GENERA 95 October 9, a Misuku Mtns. $ held 7 eggs measuring ca. 30 x 12 mm. December 21, a Bundibugyo 9 held 5 eggs measuring ca. 22 x 7 mm. December 28, a Chirinda Forest 9 held 8 well-developed eggs (FitzSimons). January 16, Ruwenzori Mtns. 9 held 5 eggs from 28 x 12 to 32 x 11 mm. February 10, a Kisenyi 9 held 8 eggs measuring ca. 25 x 10 mm. Also, on February 1, at Mushongero, Lake Mutanda, a native brought me 193 eggs, allegedly dug from two termitaria. One batch of 8 eggs measured 25 x 14 mm., another of 8 measured 39 X 17 mm., while 3 eggs selected from other batches measured 43 X 18 mm., 30 x 18 mm., and 29 x 18 mm. respectively, their diameter evidently conditioned by the girth of the parent. Each egg contained an embryo nearly ready for hatching, the embryo measuring 203 (143 + 60) mm., and on hatching a few weeks later a $ measured (180 + 80) mm., and a 9 249 (180 + 69) mm. Diet. 1 have recorded a large lizard {Lacerta jacksoni), toads {Bufo r. rcgularis) twice, as also FitzSimons. Mertens has re- ported a tree-frog {Leptopelis sp.?) in one stomach and Laurent recovered sedge-frogs {Hyperolius m. angolensis) from two snakes. 1 found a sedge-frog {Hyperolius schuhotzi) in each of three Idjwi Island irregularis, and an unidentifiable Hyperolius in a Nyasa snake; a frog {Ra/na f. fuscigula) at Lake Mutanda, and two Rana o. oxyrhynchus in Cholo reptiles. Parasites. Hemogregarines were found in Stanleyville snakes by Schwetz; nematodes (Amphicaecum sp. and Ascaroidea sp.) in an Idjwi specimen (Loveridge). Enemies. Two were recovered from the stomachs of vine- snakes {Thelotornis k. kirtlandii) on Cholo Mountain. Temperament. A Kisenyi snake, when suddenly seized, gaped until its jaws were almost in a single plane ; then it struck, the teeth drawing a little blood (Loveridge). Habitat. Occurs from sea-level to 7,000 feet, having been taken in Belgian Ruanda just below the Bihunga Escarpment (Bogert). In the Sudan it has been found in an outhouse and beneath a rock on a mountain top (Corkill). In Uganda, it was captured 96 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY while basking on the bank of a dry watercourse at 6,000 feet, also among vegetation on the bank of the Amaler River at 5,000 feet. In fact it is generally associated with rivers, lakes, and swamps where its principal prey is to be found, though it also occurs in virgin forest (Loveridge). It has been recorded as present in borassus palm "forest" (Cott), in pawpaw tree (Schmidt), in papyrus (Stuhlmann), and the sudd. And it has been suggested that Khartoum records are due to these snakes liaving been transported there with agricultural produce carried on Nile steamers (Flower). Localities. Sudan: Bahr el Jebel; Hellet Nuer; *Kagelu; *Katire ; Lake No ; Metemma ; *Mongalla ; Moro ; Talodi. Ethi- opia: *Mtemma (on Sudan frontier :C.N.H.M. 12719). Uganda: Bihunga Escarpment; *Bukatakata; *Bundibugyo ; *Bussu (an intermediate); *Debasien Mtu. ; Entebbe; Fort Portal; *Gulu, Acholi ; Kabulamuliro ; Kampala ; Katonga River ; Katunguru, Kasinga Channel; Kayonza (Kanjonsa), Kagera River; Kayonza Forest ; Kigezi ; Kishasha Valley ; *Kissolo ; Komo Id. — *Busiro ; Kyagwe Coast ; Lado ^ West Nile District ; Lake Albert ; *Lake Bunyonyi ; Lake Chahafi ; Lake Edward ; Lake George ; *Lake Mutanda — *Mushongero ; Lake Nakwali, Ankole ; *Lira, Lango ; Lugazi, Kyagwe ; *Mabira Forest — *Mubango ; Masaka ; Nkosi Id. ; Rhino Camp, Lado ; *Ruwenzori Mts. — Kilembe, *Mihunga Ridge, Mubuku Valley; Wadelai. Tanganyika Territory: Bu- koba ; Bwanja near Bukoba ; *Kibondo ; *Kihambwe River, Ki- bondo ; *Matengo Mts., *Nyarunga, Kibondo ; *Mwaya ; Rungwe Mtn. ; Tukuyu (as Langenburg) ; *Ugano. Mozambique. Boror (Praso Boror); Charre; Fambani; Quilimane (type of giin- theri). Nyasaland: *Blantyre (type of .s/urawa) ; *Chiradzulu Mtn.; *Cholo Mtn.; Fort Hill; Karonga to Kondowe ; *Limbe; *Misuku Mts.; Mlanje Mtn.; *Mtimbuka; *Nchenachena ; "Nyi- ka Plateau" (1897e) ; *Ruo River; Shire Highlands and Valley. Northern Rhodesia: Kazungulu ; Lealui (Lialui) ; Mumbwa; Up- per Zambezi. Southern Rhodesia: Chirinda Forest; *Chisha- washa ; Mashonaland ; Mazoe ^" ; Salisbury'-*^ ; Selinda Mtn. ; Seluk- we ; Umtali ; Victoria Falls. Bechuanaland Protectorate: Maun (type of vernayi). Southwest Africa: Damaralaud (1887h; 93 Labial arrangement eht'ckod by Walter Rose ; cf. footnote to Salisbury ex- anii'le of JioplogaHtcr. LOVERIDGE : AFRICAN SNAKE GENERA 97 1894a); Otjimbingue (Peters :1867b). Angola (inc. angolensis records): *Bella Vista; Bimbi; *Caconda; Cambondo; Cape- longo ; Cazengo region ; Chingo ; Congulu ; Cuango ; Cubal ; Cu- bango; *Cuma; *Cuuga; Dundo; Duque de Braganca; Loeti River (as neglectus) ; Malange; Moco Mtn. ; Mossamedes; Muita; Mupa; Mutianvo; Ndala (Ndallo) Tando; Pmigo Ndongo; Rio Mbale; San Salvador; Vila da Ponte. Cabinda: Chinchoxo; Landana. Belgian Congo:'^ Aba ; Albertville ; Beni ; Bishakisliaki River; Bobandana; Boma; Bulengo near Lake Mugunga; Bn- runga ; Bm^e-Bale River ; Dika ; Dramba ; Elisabethville ; Far- adje; Ganza; Ishango, Semliki River; Idjwi Id. — *Upper Mulinga River ; Kabenga ; Kabengere ; Kabwe ; Kakyelo ; Kalon- do, Mokoto ; Kalule-Nord River ; Kamalondo Plain ; Kando ; Kankunda River ; Kansenia ; Kantana, Lake Kivu ; Kanzenze ; Kapanga ; Kapiri ; Kasenga ; Kaswabilenga ; Katanda, Lake Ndaraga ; Katobwe ; Kenia ; Kikondja ; Kilwa ; Kilwezi ; Kisala ; Lake Tanganyika; Lukafu; Lukoka River; Lukolela; Lnkon- zohva; Lukuga River; Lululabourg; Lusinga; Mabwe; Mahagi Port; Makaia Ntete ; Makala ; *Mangbetu (Monbuttu: type of cmini) ; Masombwe ; Mati ; *Mayumbe ; Mokabe-Kasari ; Mpala; Mubale-Munte ; Mukandwe ; ]\Iulungu; Mulungwe River; Muna- gana, Lake Rukuru; Mutwanga; Ngesho; Niangara; Nyangwe; Nyonga ; Nyunzu ; Panda ; Pare Upemba ; Pelenge ; Povo Nemlao ; Povo Netonna; Pweto; Rungu; Rwindi; Sake, Lake Kivu; San- doa; Savane; Stanleyville; Tshisliaka; Tshumba ; Upper Tele River; Vista; Yakululu ; Zauibi; Zongo. Belgian Ruanda-Urun- di:^* Below Bihunga Escarpment ; Goma, Lake Kivu ; Kalondo, Lake Ndaraga; *Kisenyi, Lake Kivu; Lulenga : Rwankeri: Shangugu, Lake Kivu; Usumbura. French Congo: Abiras; (labon ; Yakonia. French Equatorial Africa: *F()rt Crampel ; *Fort Sibut. French Cameroon: Cameroon (type of longifrenatus) ; Bipindi; Satsclie (not found on map) ; Tibati; Yaunde. British Cameroon: Kaka region, Bamenda District. Nigeria: Brass, Lower Niger River; *Ibadan; Owerri Province. Togo: Adele (as Bismarckburg) ; Atakpame ; Kete; Moba; Yendi. Ghana: * Ac- cra; *Achimota; *Adjiko, Krobo; Akropong; *Banda Hills; Christian Village; Fantee (type of irregularis). Ivory Coast: 9* To these should be added 40 fresh localities: in the Belgian Congo (8), Ruanda (20) and Urundi (12) recently recorded by Laurent (1956, pp. 161-163). 98 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Abidjan ; Guiglo-Tai. Liberia: *Bolahun ; *Ganta ; *Gbanga ; *Suacoco. Sierra Leone: *Bonthe; *Makeni; *Njala. French Guinea: "Guinea" (type of chenonii) ; Beyla ; Dixine; Ferkesse- dougou ; Kerouane ; Labe, Fouta Djalon ; Nimba Mtn. ; Timbo ; Ziela. Portuguese Guinea: *Bissau ; Bolama ; Caita ; Pitclie ; Ponte. Gambia: *Cape St. Mary; *McCarthy Id. Senegal: Bel Air; Canbarene Road near Dakar; Cap Vert (type of caesius) : Dakar ; Hann ; Kati near Bamakko ; Mbao ; Rufisque ; Sangalkam ; Yoff. French West Africa: Dane near Diebougoii, Upper Volta ; Duong-Huu-Thoi ; Kayo. Range. Senegal, due east to Metemma (if data reliable), Ethiopia on Sudan border, on the White Nile, Sudan, south down Lakes Victoria, Tanganyika and Nyasa to Southern Rhodesia, west to Damaraland (single record), north and west to Senegal. In brief : West and Central Africa south of 15° N. and north of 20° S., but following the Zambezi east to the vicinity of Xyasa- land. Philothamnus natalensis (Smith) 1840. Dendrophis (Philothammus) natalensis A. Smith, 111. Zool. S. Africa, Kept., pi. Ixiv, ligs. 1-3: "Port Natal," i.e. Durban, Natal, South Africa. 1847. Dendropliis natalensis Bianconi, p. 278. 1859. Bianconi, p. 502. 1858c. Ahactulla irregularis var. natalensis Giinther, p. 152. 18()3c. Aliaetulla natalensis Giinther, p. 285. 1890. Miiller, p. (594. 1867a. Leptophis irregularis Steindachner (not Leach), p. 71. 1882c. Philothammus natalensis Bocage, p. 18. 1884a. Rochebrunc, p. 173 (omitted as erroneous). 1887b. Boettger, p. 160. 1887h. Boulenger (part), p. 176 (omit Damaraland). 1951c. Loveridge, pp. 4, 9. 1894a. Chlorophis natalensis Boulenger (part), p. 95 (omit Damaraland). 1898. Werner, 1896-7, p. 144. 1898. Sclater, p. 99. 19081). Boulenger, p. 228. 1908. (iough, p. 24. 1908. Odhner, p. 5. 19101). Boulenger (part), p. 507 (omit Kafue Kiver). LOVERIDGE : AFRICAN SNAKE GENERA 99 1910a. Hewitt, p. 57. 1912. FitzSimons, F. W., p. 87. 1913. Hewitt & Power, p. 162. 1914a. Hewitt, pp. 242, 245, 246. 1929a. Werner, p. 100. 193."). Power, p. 334. 1937a. FitzSimons, V. F., p. 262. 1937e. Hewitt, p. 59, fig. 6. 1950. Fantham & Porter, p. 601. 1950. Rose, p. 271. 1955. Rose, pp. 92, 97. 1903. Cltloropliis angolcnsis Gough (not Boeage), p. 465 (Durban). Further citations of '^ natalensis" will be found under hoplo- (jaster, i. irregularis and heterolepidotus. Common Name. Natal Green-Snake (F. W. FitzSimons). Description. Preoeular 1 ; postoculars 2 ; temporals 2 -f 2, i-arely 1 + 2"' or 2 + 1 ; upper labials 8-9, the fourth and fifth, or fifth and sixth, entering the orbit; lower labials 10-11, the first 5 in contact with the anterior sublinguals. Midbody scale-rows 15; ventrals 150-168^'*^ {£ S 150-165; 9 9 151-168); anal di- vided, rarely entire'*'; subcaudals 114-128 (9 9 114-122; S i 120-128). For characters common to all species, and synopsis of scale- counts, see pp. 49-52, 55. Color. Above, olive green, the interstitial skin black ; usually some scales with a white spot on their outer border. Below, chin and throat white, otherwise greenish white to pale green. Size. Length of i (B.M. 87.12.6.9) 775 (500 + 275) mm.; length of 9 (B.M. coll. Stratham) 1083+ (755 + 328+) mm. Remarks. The type of nataleJisis is not in the British Museum, and the specimen in the Royal Scottish Museum differs from Smith's description l)oth in its fused temporals and greater dimensions (V. FitzSimons). The numerous subcaudals of iiat- alerisis, surpassed only l)y licterolepidotus and semivariegatus ii^ 1 + 2 of Bouleiiger (1804a :0.")) was based on the Kiiigwilliamstown specimen and has been checked. '■"5 Actually 167 and 168 for the two Kin^'williamstown $ $ that Bouleiiger (l8U4a :95) recorded as 169. 07 Entire only in a Natal d" (H. Calloway coll.) in the British Museum. 100 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY {sensu lato), with a trifling overlap by i. irregularis, and possibly the almost constant temporal arrangement of 2 + 2, suggest that »atalcnsis is one of the most highly specialized members of the subgenus Chloroph is. Sexual dimorphism. In 9 9 the subcaudals range from 114 to 122 ; the ventrals overlap. In 5 5 the subcaudals range from 120 to 128. Diet. Sir A. Smith's statement that natalensis reputedly feeds on birds and insects may be discounted until established bj^ con- crete observations. A gecko {Afroedura p. pondolica) is present in the stomach of one Museum of Comparative Zoology specimen. Parasites. To this species Fantham and Porter (1950) attrib- ute the entozoa {Eufrichomastix serpentis), and juvenile cj'sti- cerci that they found in the mucous and serous coats of the stom- ach, or embedded in the peritoneum, of a "Si/o foot" $ and "41/^ foot" 9. If their measurements are correct, however, it would seem more probable that the snakes involved were P. s. semivariegatus. Habitat. Shrubs and trees in the vicinity of Durban, where, being a good climber, the Natal Green-Snake is rarely seen on the ground (A. Smith"). Localities. ^"^ Transvaal: Ilenops Kiver near Pretoria ; Lyden- burg; Middleburg; Pretoria; *Selati, Zoutpansberg District; Vryheid. Natal: C'athkin Peak, Drakensberg Mountains (V. Pitz- Simons coll. :1943); Caversham ; Dargle Road; Durban; *Natal (head in M.C.Z.). Cape Province: Blue Cliff; Clarkesbury ; Duu- brody ; East London; Grahamstown; tiarrismith ; Kimberly ; *Kingwilliamstown ; Mossel Bay; Mqanduli ; *Orange River; Port Elizabeth; *Port St. Johns; Tsomo. Bange^^ : Transvaal, south to Natal and eastern Cape Province. Philothamnus heterolepidottts (Giinther) 1863c. AhaettiUa heterolepidota Giinthev, Ann. Mag. Nat. Hist., (3) 11, p. 286 : Africa. 1872a. Giinther, p. 26. 1866a. Lcptophis heterolepidota Bocage, p. 48. ] S66h. Bocage, p. 69. ''« The two Rhodcsian specimens, i.e. Kafuc Kivor (Boulenger :1910b) and Salis- bury (Boulpn^iT : 10021) ) . are assiinifd to lie hoiilofinstrr ; the Daniaraland snake Iirovert to be an alierrant i. irregularis. LOVERIDGE : AFRICAN SNAKE GENERA 101 1879a. PhilothaDunis hrirrolcpulotns Socage, p. 06. 1882c. Socage, p. 8, fig. 2. 1887a. Socage, p. 185. 1888a. Soettger, p. 60. 1893. Pfeffer, p. 82 {Yiossihly semivariegatus) . 1895a. Socage, p. 88. 1896a. Socage, p. 100 (repeats Pfeffer 's record for Mozambique). 1951c. Loveridge, pp. 5, 10. 1952. Witte, p. 19. 1953. Witte, p. 186, fig. 53. 1954b. Laurent, p. 48. 1954d. Laurent, pp. 292, 305. 1956. Laurent, p. 167, fig. 28, pi. xviii, fig. 2. 1888b. AliaetulJa graeiUima Giinther, Ann. Mag. Nat. Hist., (6) 1. p. 326: Lower Congo, Selgian Congo. 1894a. Chlorophis emini Boulenger (part:not of Giinther), p. 92 (omit type from Monbuttu ) . 19031). Andersson, p. 3. 1908a. Werner, p. 170. 1916a. Loveridge, pp. 78, 83, 84. 1918a. Loveridge, p. 333. 1924b. Loveridge, p. 5. 192911. Loveridge, p. 21. 1933m. Witte (part), p. 88. 1935a. Corkill (part),p. 16. 1951. Monard, p. 150. 1894a. Chlorophis heterolepidotus Boulenger, pp. 95, 358, pi. v, fig. 3. 1896d. Soulenger, p. 631. 1896. Tornier, p. 69. 1897b. Soulenger, p. 278. 1897. Tornier, p. 69 (reprint of 1896). 1897b. Tornier, p. 65. 1898. Soettger, p. 58. 1902a. Werner (part), p. 344 (part irregularis). 1905c. Soulenger, p. 112. 1908b. Stemfeld, pp. 214, 230. 1909a. Sternfeld, p. 14. 1910a. Sternfeld, p. 18 (perhaps omit Pfeffer 's Quilimane record). 1912c. Sternfeld, p. 270. 1915a. Boulenger, p. 205. 1915e. Soulenger, p. 623. 1917. Sternfeld, p. 468. 1918a. Loveridge, p. 333. 102 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1919a. Boulenger, p. 281. 1923. Schmidt, p. 76. 1924b. Loveridge, p. 5. 1926a. Mertens, p. 153. 1929. Hecht, pp. 333, 334. 1929h. Loveridge, p. 23. 1929a. Werner, p. 100. 1931. Monard, p. 104. 1932a. Parker, p. 219. 1933f. Angel, p. 106, figs. 39-39a. 1933m. Witte (part), p. 89. 1934. Pitman, p. 294 (but heterolcpidotus is unknown from Ehodesia). 1934a. Schwetz, p. 380. 1936. Pitman, p. 277, pi. v, fig. 4 ; pi. E, fig. 2. 1937b. Monard, pp. 114, 120. 1938a. Pitman, p. 306. 1938b. Pitman, pp. 39, 90, 91, 306 (pis. as in 1936). 1940b. Monard, p. 17.5 (though possibly irregularis) . 1942e. Loveridge, p. 273. 1950. lonides, p. 101. 1950f. Laurent, p. 128. 1895. Chlorophis ffracillimiis Giinther, p. 528. 1908b. Chlorophis natalensis Sternfeld (not A. Smith), pp. 214, 229. 1909a. Sternfeld, p. 14. 1910d. Chlorophis gracilis Sternfeld, Mitt. Zool. Mus. Berlin, 5, p. 64: Dodo, French Cameroon. 1929. Hecht, p. 332. 1951. Monard, p. 150. 1912c. Chlorophis schuhotzi Sternfeld, Wiss. Ergebn. Deut.-Zent.-Afrika- Exped. 1907-1908, 4. p. 269, fig. 4: Bwanja, near Bukoba, Tan- ganyika Territory. 1915c. Boulenger, p. 622. 1924b. Loveridge, p. 5. 1929. Hecht, p. 332. 1929a. Werner, p. 100. 1916a. Chlorophis irregularis Loveridge (not Leach), p. 84. 1918a. Loveridge, p. 333. 1919d. Chabanaud, p. 567. 1924b. Loveridge, p. 5. 1929h. Loveridge, p. 23. 1935a. Corkill (part), p. 16. 1936J. Loveridge (part), p. 247 (omit Mt. Debasien). LOVERIDGE : AFRICAN SNAKE GENERA 103 1937c. Loveridge (part), p. 273 (omit Fort Sibut). 1941. Witte (part), p. 188 (Kalinga specimen). 1923. Chlorophis bequaerti Schmidt (part), Bull. Am. Mus. Nat. Hist., 49, p. 75, fig. 3: Niangara, Belgian Congo (based on an aberrant heterolepidotus $ with entire anal ; omit para type 9 , which is regarded as a similar aberrant irregularis) . \9Som. Chlorophis ornatus Witte (not Socage), p. 88. l93Sm. Chlorophis hoplogaster Witte (partrnot Giinther), p. 88 (Faradje). Further citations of "heterolepidotus" will be found under i. irregularis. Common Names. Slender Green-Snake (Loveridge) ; Strange- sealed Green-Snake (Pitman) ; calumheremhe (Angola :Boeage) ; chilembe (Cabinda:Bocage) ; lolengne-lengua (Angola :Bocage). Description. Preocular 1, rarely 2"®; postoculars 2, rarely 1^*^°; temporals 1 -f 1, rarely -f 1"^ or 1 + 2; upper labials 7-9, the third and fourth, ^"^ or third, fourth and fifth, normally the fourth, fifth and sixth, entering the orbit; lower labials 9-11,^°- the first four, five, or six in contact with the anterior sublinguals. Midbody scale-rows 15; ventrals 164-193^°' ( S S 164-193; 9 9 169-193) ; anal divided rarely entire^"* ; sub-caudals 109-134 ( 9 9 109-126; 6 3 115-134^''=). For characters common to all species, and synopsis of scale- counts, see pp. 49-52, 55. Color. Above, dark or bronzy green, the interstitial skin black; some scales with a white basal spot. Below, chin and throat white, otherwise greenish white or yellowish to very pale green. 93 Two pre.sent in a Kaimosi 9 (M.C.Z. 29426). 100 On both sides of a Kaimosi d" (M.C.Z. 29423). 101 -I- 1 through fusion with sixth labial on right side of a Tembwe ? (Mus. Congo 4184) whose fourth and fifth labials are also fused on this side, resulting in its determination as "ornatus" (Witte). 102 Actually 12 on left side of a Poll $ (M.C.Z. 44104) resulting in the first seven labials being in contact with anterior sublingual. 103 193 in a Niangara d" (A.M.N.H. 12087). 104 Entire in the holotype d" hequaerti (A.M.N.H. 12080) from Niangara. 105 134 in the type of bequaerti, besides a Lagos d (B.M. 71.4.21.4) and a Katobwe snake (C.N.H.M. 12874). Laurent (1954b) gives 144 for a Dundo J, a figure which far surpasses all species except semivariegatua. 104 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Size. Length of $ (M.C.Z. 48300) 730 (492 + 238) mm., from Kitaya, is exceeded by a $ (Laurent :1956) 745 (475 + 270) mm., from Makungu. 9 (type of heterolepidolus) 662 (450 -|- 212) mm., remeasiired. 9 (type of gracillimus) 722 (480 + 242) mm., remeasured. S (type of hequaerti) 652 (440 + 212) mm. (type of schuhotzi) 565 (380 + 185) mm. (type of f/racilis) 450 (300 + 150) mm. Remarks. The type of Ahaetulla heterolepidota was stated by Giinther (1863e) to have 187 ventrals and 125 snbcaudals; amended by Boulenger (1894a) to 182 ventrals and 117 snb- caudals recounted by Battersby and myself as 182 -\- 4 halves and 119 snbcaudals while I find the former length of 441 + 216 mm. is now 450 + 212 mm. 1 mention this as the British Museum has a "Zanzibar Coast" specimen of almost identical propor- tions (423 + 215 mm.) and scale-counts as given by Giinther for the holotype of heferolepidofa. However, Mr. Battersl)y does not think that a transposition has taken place as the head scales of the Zanzibar snake do not tally with Giinther 's description. A. gracillima Giinther was referred to the synonymy by Pfeffer (1893); C. gracilis Sternfeld by Boulenger (1919a), whose " emini" Avas a composite of heterolepidotus + emini Giinther. which is a synonym of irregularis. It was in this composite sense of ''emini" that I (1936h) synonymized C. schuhotzi Sternfeld with "irregularis." In reality schuhotzi is a synonym of hetero- lepidotus as here understood. To that synonymy I now add the $ type of hequaerti Schmidt which differs from typical heterolepi- dotus only in having an entire anal. The 9 paratype of hequaerti I consider to be an aberrant irregularis with entire anal. Normal examples of both heterolepidotus and irregularis from Niangara were present in the same collection. Sexual dimorphism. Unless there exists some slightly lower average subcaudal count for 9 9 , no other characters for sep- aration of the sexes have been detected. Diet. A sedge-frog {Hyperolius ciniiamomeoventris) removed from a Makungu snake (Laurent :1956), others (Hyperolius ros.^i == pachy derma) and sp. were recovered from the stomachs of three Uganda snakes, also a frog (non det.) in a Tanganyika specimen (Loveridge :1942e) . LOVERIDGE : AFRICAN SNAKE GENERA 105 Habitat. Apparently associated with tropical rain-forest or the gallery forests frinjiingf the equatorial river systems. Localities. Sudan: Barlioi; Khartoum (probably introduced on Nile steamer) ; .lOO miles south of Khartoum; *Torit; White Nile (1882). Uganda: *Bukatakata; Katunguru ; *Lira, Lango ; *Nyakabande ; Rhino Camp, 2°55', West Nile Province. Kenya Colony: *Kaimosi ; *Kakamega ; * Yala River. Tanganyika Ter- ritory: Bwanja near Bukoba (type of schuhotzi) ; *Kibondo; *Kigoma; *Kitaya; Masailand {i\de Sternfeld :1910a) ; a ques- tionable record possibly based on a specimen of semivariegatus) ; *Nyarunga, Kibondo ; Zanzibar coast (i.e. mainland opposite island). ? Mozambique (the sole record is that of Pfeffer (1893), repeated by Sternfeld (1910a), for three snakes from Njama Kette near Quilimane, possibly semiivariegatus) . Angola: Bihe; Bimbi; Caconda; Casangue ; Catengue ; Cuango (Quan- go) ; Cuibula (Quibula) ; Dundo (Dondo) ; Duque de Braganca ; Kutatu (Kutato) ; Muita River; Sao Salvador. Cabinda: Cabin- da (1895a). Belgian Congo:^"" Albertville ; Boma ; Cap Kasem- bondo, Lake Tanganyika ; Duma ; Elisabethville ; Faradje ; Ink- angala ; Kabongo ; Kabwe ; Kabwekanono Lake ; Kadia ; Kando ^"' ; Kapanga; *Katobwe ; Kikondja; Landana; Lofoi; *Lower Congo (type of gracillima) ; Mahagi Port; Masombwe; Mubale-Munte ; Musosa ; *Niangara ; Nyonga ; Pweto ; Tembwe ; Tshikapar ; Sandoa ; Stanleyville ; Zambi. Belgian Ruanda-Urundi:^"" Lake Kivu. French Cameroon: Dodo (type of gracilis) ; Kribi. British Cameroon: *Poli near Caroua. Nigeria: *Lagos. Dahomey: Porto Novo. Togo: Kete. Range. Sudan from the Upper Nile southeast to the mouth of the Rovuma River, Tanganyika Territory (? Mozambique), west to Angola and north to Togo.^"* PHILOTHAMNry SE.MIVARIEGATUS SEMIVARIEGATUS (Smith) 1847. Dmdrnphis (PJiilothamnns) scmivariegata A. Smith, 111. Zool. S. Africa, pis. lix, Ix, Ixiv, figs, la-b; Biishinan Flat, Cape Province 106 To these should he added 13 fresh localities; In the Belgian Congo (6), Ruanda (6) and Urundi (1) recently recorded by Laurent (1936, pp. 167-168). 107 The Kansenia record is transferred to /. irregularis. 108 Monard's (1040b) records from Portugese Guinea — 1300 miles west of Togo — appear to be based on examples of /. irregularis. 106 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY (restricted Bogert:1940) "and the country beyond Kurrichaine, " i.e. Rustenberg, Transvaal. ISfiof. AhactnUa semivaricgata Giinther, p. 285. 18661). Pliilothamnus punctatus Peters, Monatsb. Akad. Wiss. Berlin, p. 889: Zanzibar Coast, i.e. Tanganyika Territory. 1869a. Peters, p. 16, pi. i, figs. 2a-c. 1878a. Peters, p. 206. 1882a. Bocage, p. 289. 1882c. Bocage, p. 14. 1882a. Peters, p. 129, pi. xixA, fig. 1. 1884a. Rochebrune, p. 173 (omitted as unreliable). 1889. Boettger, pp. 289, 294. 1893. Pfeffer, p. 83. 1896a. Bocage, p. 92. 1867a. Lcptophis sp. Bocage, p. 226. 1868a. AhactuUa Eirkii Giinther, Ann. Mag. Nat. Hist., (4) 1. p. 424: Zanzibar. 1869b. Giinther, p. 116. 1869. Dcndrophis melanostigma Jan. Icon. Gen. Ophidiens, No. 32, pi. ii, fig. 3 : Mozambique. 1869b. Philoihomnus semivariegatus Peters, p. 661. 1870b. Peters, p. 114. 1884a. Rochebrune, p. 173 (omitted). 1891a. Boulenger, p. 307. 1891b. Matschie, p. 615. 1893b. Stejneger, p. 728. 1894a. Boulenger, p. 99. 1895a. Bocage, p. 90, pi. xiii, figs. 2a-c 1895c. Bocage, p. 13. 1895. Jeude, p. 229. 1896a. Bocage, pp. 78, 83. 1896b. Bocage, p. 112. 1896d. Boulenger, p. 631. 1896e. Boulenger, p. 216. 1896. Tornier, p. 70. 1897. Bateman, p. 268. 1897b. Boulenger, p. 278. 1897d. Boulenger, p. 375. 1897e. Boulenger, p. 801. 1897g. Boulenger, p. 279. 1897i. Boulenger, p. 20. LOVERIDGE: AFRICAN SNAKE GENERA 107 1897. 1897. 1898. 1898a. 1898. 1898. 1898. 1898. 1899a. 1900b. 1901. 1902a. 1902b. li»02a. 1903b. 1903a. 1907a. 1907J. 1907. 1907a. 1907c. 1908b. 1908. 1908b. 1908. 1908b. 1908c. 1908. 1909d. 1909a. UUOb. 1910a. 1910b. 1910c. 1911c. 1911. 1911b. 1911a. 1912b. 1912. 1912. 1912c. Meek, p. 179. Tornier, p. 65. Boettger, p. 58. Boulenger, p. 720. Ferreira, p. 244. Johnston, p. 361a. Sclater, p. 99. Werner, 1896-7, p. 144. Moequard, p. 219. Boulenger, p. 452. Schenkel, p. 160. Boulenger, p. 446. Boulenger, p. 17. Werner, pp. 339, 344. Andersson, p. 3. Bocage, p. 43. Boulenger, p. 10. Boulenger, p. 486. Lonnberg, p. 15. Roux, p. 76. Roux, p. 735. Boulenger, p. 229. Gough, p. 24. Moequard, p. 5-58. Odhner, p. 5. Sternfeld, pp. 214, 230. Sternfeld, pp. 243, 246. Werner (1907), p. 1873. Boulenger, p. 311. Sternfeld, p. 15. Boulenger, p. 508. Sternfeld, p. 19. Sternfeld, p. 19, fig. 19. Sternfeld, p. 55. Boulenger, p. 16.'i. Lampe, p. 193. Nieden, p. 442. Sternfeld, p. 250. Boulenger, p. 332. FitzSimons, F. W., p. 87. Hobley, p. 49. Sternfeld, p. 270. 108 1913. 19],3. 1913. 1913. 1913a. 1913b. 191oa. 1915i-. 1915(1. 1916. litlCf. 1916a. 1917b. 1917a. 1917. 1918a. 1919a. 1919k. 1920. 1921a. 1921b. 1921. 1922. 1922. 1923e. 1924b. 192na. 1925a. 1927. 1927c. 1928c'. 1928. 1928(1. 1928J. 1929. 1929h. 1929e. 1929a. 1930a. 1930e. 1931. 1931. BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Boettger, pp. 345, 352, 359, 363. Hewitt & Power, p. 162. Klaptocz, p. 285. Lonnberg & Audersson, p. 4. Werner, p. 373, pi. vii, fig. 2. Werner, p. 717. Boulenger, p. 206. Boulenger, p. 623. Boulenger, p. 648. Calabresi, p. 39. Chabanaud, p. 372. Loveridge, p. 84. Chabanaud, p. 11. Phisalix, p. 334. Sternfeld, p. 470. Loveridge, p. 332. Boulenger, p. 282. Boulenger, p. 23. Aders, p. 338. Chabanaud, p. 468. Chabanaud, p. 524. Talk, p. 177. Aylmer, p. 15. Falk, p. 162, photo fig. Loveridge, p. 879. Loveridge, p. 5. Angel, p. 33. Loveridge, p. 72. Calabresi, p. 53. Power, p. 409. Barbour & Loveridge, p. 118. Cott, p. 953. Loveridge, p. 53. Loveridge, p. 74. Flower, p. 204. Loveridge, p. 24. Scortecei, p. 274. Werner, p. 98. Scortecei, p. 212. Scortecei, p. 17. Monard, p. 104. Power, pp. 43, 48. LOVERIDGE : AFRICAN SNAKE GENERA 109 1931c-. Scortecci, p. 206. 19321). Scortecci, p. 4.'). l!t;!31>. Angel, p. 69. 1933f. Angel, p. 110, figs. 40-40a. 1933. Schmidt, p. 13. 1933111. Witte, p. 89. ]934;i. Schwetz, p. 381. 1935a. Coikill, p. 17. 1935. Cott, p. 967. 1935. Power, p. 334. 1937. Andersson, p. 8. 1937:i. FitzSimons, V. F., p. 262. 1937e. Hewitt, p. 60, fig. 5, pi. xviii. 1937b. Monard, pp. 114, 122. 1938. FitzSimons, V. F., p. 156. 1939a. Scortecci, p. 274. 1940a. Scortecci, p. 131. 1945. Leeson, p. 1. 1947. lonides, 1946, p. 70. 1948. lonides, 1947, p. 10. 1949a. Parker, p. 57. 1950. Leeson, pp. 46-48, 66, 133, figs. 20-22, pi. v. 1950. Rose, pp. 272, 314, fig. 9. 1950b. Villiers, p. 82, figs. 100-101. 1951. Monard, pp. 150, 160. 1954. Blair & Graham, p. 286. 1955. Rose, pp. 94, 178, fig. 9.10^ 1881. Philothamnus irref/nlarLi var. Fischer, p. 229, pi. xiv, figs. 5-7. 1884a. Fischer, p. 11. 1882c. Pliilothamnus Smitliii Bocage, Jorn. Sei. Lisboa, 9. p. 12, fig. 5: Bissau, Portuguese Guinea (restricted A.L.) and Angola. 1884b. Sauvage, p. 201 (as Pkilothalmns) . 1886. Dollo, p. 156. 1887a. Bocage, p. 196. 1885. AhactuUa sp. Miiller, p. 683. 1886. Ahaetulla irregularis Parenti & Picaglia (not Leach), p. 74. 1930b. Zavattari, p. 193, 1888. Leptophis punctatus Mocquard, p. 128. 1888b. Ahaetulla Bocagii Giinther, Ann. Mag. Nat. Hist., (6) \. p. 326: Angola. 109 Variously spelled Philothamnus (92, 97), Pholothamnus (95), and Philo- thammus (212). 110 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1895. Giinther, p. 528. 1889. Philothamnus irregularis Pfeffer (not Leach), p. 9. 1902. Ferreira, p. 232. 1893. Philothamnus punctatu^ var. thomensis Pfeffer (not Bocage), p. 84. 1893. Philothamnus punctatus var. sansibaricus Pfeffer, Jahrb. Hamburg. VViss. Anst., 10. p. 83: Zanzibar. 1891:. Ahaetulla punctata Giinther, p. 88. 1895. Philothamnus Kirbii Giinther, p. 528. 1896. Philothamnus variegatus Peracca, p. 2 (lapsus). 1948a. Villiers, p. 203. 1896. Philothamnus dorsalis Tornier (not Bocage), p. 71. 1897. Tornier, p. 65. 1925. Angel, p. 32. 1933m. Witte (part), p. 90 (Faradje). 1896. Philothamnus thomensis Tornier (not Bocage), p. 71. 1897. Tornier, p. 65. 1910a. Sternfeld, p. 19. 19-40. Parker, Moreau & Pakeuham, p. 310 (correct earlier record). 19U8c. Chlorophis irregularis Sternfeld (not Leach), p. 243. 1911. Lepri, 1910, p. 324. 1930a. Scorteeci, p. 212. 1923. Philothamnus nitidus Schmidt (not Giinther), p. 78 (intermediates). 1925a. Angel, p. 32. 1930a. Barbour & Loveridge, p. 772. 1930b. Cldorophis emini Zavattari (not Giinther), p. 193. 1933h. Philothamnus semirariegatus semivariegatus Loveridge, p. 238. 1934c. Mertens, pp. 27, 83, 118, 119. 1934. Pitman, p. 295. 1935b. FitzSimons, V. F., p. 313. 1936h. Loveridge, p. 32. 1936,j. Loveridge, p. 247. 1936. Pitman, p. 42, pi. vi, fig. 1 ; pi. D, fig. 3. 1937c. Loveridge, p. 273. ] 937f . Loveridge, pp. 493, 496, 503. 1937a. Parker, pp. 630. 1937b. Mertens, p. 13. 1937. Uthmoller, p. 110. 1938c. Mertens, p. 439. 1938a. Pitman, pp. 307, 328. 19381). Pitman, pp. 39, 97, 295, 307, 328, i)ls. (as in 1936). 1939b. FitzSimons, V. F.. p. 22. 1 940. Bogert, p. 56. 1941. Morciiu & Piikcnliam, p. 108. LOVERIDGE: AFRICAN SNAKE GENERA 111 1941a. Utlunoller, pp. 26, 42. 1941b. Uthmoller, p. 235. 1941. Witte (part), p. 199 (Bugazia). 19421). Bogert, p. 2. 1942e. Loveridge, p. 274. 19451). Loveridge, p. 2. 1950. lonides, p. 101. 1950f. Laurent, p. 126 (possibly some P. «. dorsalis). 1950a. Villiers, p. 61. 1951a. Loveridge, p. 192. 1951c. Loveridge, p. 11. 1952. Witte, p. 19. 1953e. Loveridge, pp. 262, 318, pi. iv, fig. 1. 1953. Witte, p. 195, fig. 56. 1954. Battersby, p. 247. 1954b. Laurent, p. 48. 1954d. Laurent, pp. 292, 305. 1955a. Merteus, p. 93. 1933h. Philothamnus semivariegatus dorsalis Loveridge (presumably not of Bocage, being 22 intermediates from Mwaya, Lake Nyasa; cf. remarks under s. dorsalis), p. 238. 1937f. Loveridge, p. 496. 1936. Pitman (part), p. 44. 1938a. Pitman, p. 307. 19381). Pitman, pp. 99-100, 307. 1941. Witte, p. 200. 1946b. Philothamnus Mortens, pp. 17, 24 (defense reactions). 1954d. ^^'^' Philothamnus semivariegatus nitidus Laurent (not Giinther), p. 296. 1956. Laurent, p. 180. Further citations of "semivariegatus" will be found under i. hattershyi and s. nitidus. Common Names. Spotted Wood-Snake (Flower) ; sundry other names have been suggested by Pitman (1938b, p. 97). Alakara guinde (Habbe at Bandiagara: Angel); changa (Yao: Lover- idge); ekiimhu (Teita: Loveridge); hasowitu (Pokomo: Lover- idge); kada (Dahomey: Bocage); kalemhwe-lemhwe (Dundo: Laurent) ; kisumera (Makonde at Kitaya: Loveridge) ; kongoani and ukiitiwiti (Amu: Loveridge); luhio (Catumbela: Bocage); 110 But data derived from 1956 reference not incorporated here for reasons given in footnote to P. s. nitidus. 112 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY nilaluwe (Helie: Loveridge) ; namafiva (Fipa: Loveridge) ; nam- ahamha (Konde at Mikindani : Loveridge); nahvinduwindu (Misukii: Loveridge) ; nawirangira (Mawiha: Loveridge) ; nhan- (ja (Gogo: Loveridge); nombo (Benguela: Bocage) ; nyaruhahi or nyaruteti (Toro: Loveridge); nyoka masamba (i.e. garden snake. Nyungwe : Loveridge); ngoe (Sambaa, but supposed to be the young of D. angusticeps: Loveridge) ; salalu (Amba: Loveridge); yarudufu (Nyamwezi: Loveridge). Description. Preocular 1, rarely 2"^; postoculars 2, rarely 1 or 3"^; temporals usually 2 + 2, sometimes 1 + 1, 1 + 2, 2 + 1 or 2 + 3"=^ ; upper labials 8-10, the fourth and fifth ; fourth, fifth and sixth ; fifth and sixth ; or rarely fifth, sixth and seventh entering the orbit ^^*; lower labials 9-11, the first four, five or six in contact with the anterior sublinguals ; midbody scale rows 15, rarely 13"°; ventrals 164-217; anal divided: subcaudals 126- ]62."« m Summary of preocular arrangement on 103 snakes in the M.C.Z. 102 have 1 preocular on each side of the head. 1 has 1 preocular on one side 2 on other (3018S : Kigojro). 112 Summary of postocular arrangement on 105 snakes. 1 has 1 postocular on each side of the head. 103 have 2 postoculars on each side of the head. 1 has 2 postoculars on one side 3 on other (40.t2S : Dchasien ) . 113 Summary of temporal arrangement on 107 snakes. 1 has 1 -f 1 on each side of the head (42109 : Waterberg : fdursalis) . 2 have 1 -j- 1 on one side of the head 1 + 2 on the other. (» have 1 -|- 2 on each side of the head (i have 1 + 2 on one side of the head 2 4- 2 on the other. 2 have 1 -t- 1 or 1 on each side (23259 : Vituri ; & Eritrea). 1 1 + 1 5 have " or " on one side 2 + 2 on the other. 3 have 2 + 1 on each side of the head. V have 2 -f 1 on one side of the head 2 + 2 on the other. 1 has 2 +1 on one side of the head 2 + 3 on the other. 74 have 2 + 2 on each side of the head. 11-1 Summary of upper labial (enterirg orbit) arrangement on 112 snakes 1 has 8 (with :!rd oulv, due to fusion) on each side (M.C.Z. 54825). 2 have 8 (with 4th & 5th) on one side, 9 (with 5th & 6th) on other. 2 have 8 (with 4th & 5th) on one side, 9 (4th, 5th & 6th) on other. 1 has 8 (with 5th only) on one side. 9 (with 5th & 6th) on other. 39 have 9 (with 5th & 6th) on each side. 8 have 9 (with 5th & 6th) on one side, 9 (4th. 5th & 6th) on other. 1 has 9 (with 5th & 6th) on one side, 10 (with 5th & 6th) on other. 48 have 9 (4th, 5th & 6th) on each side. 1 has 9 (4th, 5th & 6th) on one side, 10 (4th, 5th & 6th) on other. 8 have 9 (4th, 5th & 6th) on one side, 10 (5th, 6th & 7th) on other. 1 has 10(5th. 6th & 7th) on one side, 10 (with 6th & 7th) on other. U'l 13 in a Kibwe/.i d (M.C.Z. 4(1530) : als "^ CQ CO t) f-i •a hH cq J t- t- r-i C3 C3 10 in 10 10 in 10 10 \n IC r^ in in 10 in in m in 1-1 CJ CZ W 1— 1 i-H "-1 r-l I— 1 r-( rH T-\ y-^ IC rH r-\ rH j-\ rH rH I-H in CJ «: rH rH rH a CO + « M « •^ ^-i .2 0; ■i 1 s r— 4-) 3 CO 0) •G m 1 =0 1 00 J* 1 i 00 =0 'Si •2 i CO a, CO + 03 1H LOVERIDGE : AFRICAN SNAKE GENERA 133 o + CO CO lO 00 Tj< CO Oi (M to m 00 ti 'M CO OJ I-H I-H CO Ol CO CO CO CO CO (M C] Cl Cl cq 1 lO ■^ Oi 01 lr~ --0 to 00 00 to CO in 00 to rf( S CO CI CO -n CO in •* 'S* -* CO CO i"i CO CO CO y. •-0 + + + + + + + + + + + + + + + ^ o l>-^ ■* 00 o ^ t- t- o 03 in Cl o 00 m 1^ •-0 CO 01 in CO OQ -* co tO in OS to ^ o 00 t^ O] OJ iH OJ CI 0] (M Oi C\] rH rH 01 cq rH Cl n oc Oi 00 «o to ' '-I 00 ^ o (M - cq in 05 o m •Kjg O] CO CO CO ^ (M Cd (M in l>- OS -* t>. t-H (M fH 1— 1 IM (M r-l CO rH rH rH rH rH cq rH 3 e>- e« eo. aj 05 l—H 05 00 ^ -* to CO CO ■* to 00 T« o m 00 tr~ en to CO (O Tt< 00 to 00 o in -*l to in m in in in ch2 i-H 1—1 rH I— 1 rH rH rH ; r-^ rH rH rH rH rH rH rH rH rH IC 00 lO N O in o in in O rH CO •^ in in ot a lO t^ iO CO t^ in CO in ^ ^ to •* -* in m a; I-H I-H 1— ( tH r-l r-t r-i rH rH rH rH rH rH rH rH > ©o. ex o=- lO (>. 5D O O (31 in O to 00 CCl to cq -^ Oi rH •/J lO CO 05 CO m to m -* CO -i< in CO CO •^ rH in rH r-l I-H rH rH r-\ rH rH rH rH rH rH rH rH rH rH ^ 1 in >— 1 ©=• ■4< r-\ t^ O in di (>- OS O t^ OS CO o •^ a CO rH in CO ItO ! ■* ca CJ CO CO CI cq rH tH i-H rH r-i rH rH rH rH rH rH rH rH rH rH to o u o 0) •2 ts •2 to ^^ Cl5 E 1 to s •2 si •2 Co •2 to J- 533 S to •i •2 5 s CiJ rS ■*? s g § S' a rO rO s v 6 tc "^ S •K> ■J2 =c -^ ~ ^" ^ s e e e e e e e •a 1 Oj o ~ c J^ *^ ^ 1* ^' ^." a '■^ a ■ O .• ~ — oc 7* ^"^ M C S. -4-.C ^-> -*- ■>- 0) a J; a; ^ a _, rj Oj Oj iS .-. -^ O -1- •^-' .-CM _ C sxjOJ-M d oj 6X)"'^ a,' to o; o „ 3; M^ OS rH "„'»X.a lO ^S SH ''^ '^ ;_; 3 =d j^ t" n •— ^ 4J a .n ^ -*-'>> kHq^w+^ „ '* ° he ^ ^.M ai! '^ s K o -M a O OO Ol 134 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Recorded distribution of Prosi/mna around the African continent )-4 •— < a > 4) fl 3 ai > ■1) 3 el e) ■*-> ■^-> s B S a a u % s a 5 05 "3 B -3 S 3 3 + s o J= 3 S C3 O a m 3 OS 3 3 3 08 at a 'S m *-w 03 t4 O cS 3 3 a OS a a a 1— t ;a 3 -w CO oj 3 61 Is a OS *s B Si »— « 03 es to- co a 03 u a 3 3 a 03 1 B o; a S a 'c. Sudan * « British Somaliland » Somalia * Uganda Protectorate * Kenya Colony — — — — # * * — — ♦ Tanganyika Territory Zanzibar Island • Mozambique * — — * » — « Nyasaland Northern Rhodesia ♦ Southern Ehodesia * * * South Africa : (Transvaal) » » (Natal & Zululand) * * » (Cape Province) « * « Southwest Africa » » « Angola » » # Belgian Congo • * » » Belgian Ruanda-Urundi » French Congo « » « British Cameroon » Nigeria * * Dahomey « Togoland « Ghana (Gold Coast) * Ivory Coast • Sierra Leone » • Portuguese Guinea « Gambia • Senegal » French West Africa * » LOVERIDGE : AFRICAN SNAKE GENERA 135 1929. Flower, p. 218. 1929a. Werner, p. 141. 1937a. FitzSimons, V. F., p. 262. 1946a. FitzSimons, V. F., p. 355. 1950. Kose, p. 320, fig. 10. 1955. Rose, pp. 96, 178, fig. 10 (reprinted). Further citations of "smulevallii" will be found under .s. hivittata and inelcagris. Common Name. South African Shovel-snout (A.L.) ; Coppery Snake (F. W. FitzSimons). Description. Rostral with sharp horizontal edge ; internasals separated, occasionally in contact by their inner angles ;^^^ pre- frontal single, rarely dividecP^^; frontal large; loreal as long as high or slightly higher ; preocular 1 ; postoculars 2-3 ; temporals rarely 1 -j-2 or 1-|- 3, usually 2 4-2 or 2-\- 3; upper labials 6-8, the third and fourth, rarely fourth and fifth entering the orbit ; lower labials 7-8, the first 3 in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows ; ventrals 130- 169'^* (assumed $ $ 130-155; assumed 9 9 155-169); anal entire ; subcaudals 19-38 (assumed 9 9 19-27 ; definite S S 28-38). Color. Above, rostral and crown yellowish white, or head yel- lowish with a more or less distinct dark brown interocular band and a dark brown crescentic blotch bordering the parietals posteriorly; body pale gray to reddish brown, sometimes each scale edged with darker, the center of each vertebral scale yellow or grayish white, on either side a dorsolateral series of small brown or black spots extend along the back. Below, white or yellowish, uniform or with dusky blotches and a reddish median line along the tail. Size. Length of $ (M.C.Z. 21243) 265 (230 + 35) mm., from Sunnyside; length of 9 (M.C.Z. 21244) 310 (280 + 30) mm., from Meintjeskop, both localities being near Pretoria. Remarks. Sir A. Smith, whose scale counts were often errone- ous, states that the type had 169 ventrals, a figure I should be 133 /ide Boulenger (1894a), possibly by iDclusion of mistakes in the literature. 134 Omitted are the figures of a juvenile recorded by Boulenger (1894a) as having 125 ventrals and 27 subcaudals. l;36 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY inclined to question as it is 5 more than found on any subse- quently recorded specimen. However, FitzSimons (1937a) re- ports that a specimen in poor preservation, but corresponding to the type in lepidosis and measurements, is among the Smith col- lection in the Royal Scottish Museum, Edinburgh. Localities. Transvaal: Assegaibos; Brakkloof; Crocodile-Ko- mati (Comati) River Junction; Harts River; Krabbef ontein ; *Meintjeskop ; *Sunnyside. Natal: Durban. Cape Province: Burghersdorp ; Clanwilliam ; P'ort Richmond ; Graaf Reinet ; Hanover; Kaffirland; Orange River; Tulbagh. Rongc. Union of South Africa, from the Transvaal southwest to Little Xamaqualand where it meets with sundevallii hivittata (including macrospila). Prosy MNA sundevallii bivittata Werner 1867b. Tcmnorhynchus frontalis Peters (part:not Peters 1867b), p. 236, pi. — , figs. 2-2d: Southwest Africa. 1902a. Prosyrnna sundevallii Werner (not Smith), p. 339. 1910b. Sternfeld, p. 20, fig. 20. 1915c. Werner, p. 360. 1929a. Werner, p. 141. 1903. Prnsi/mna sundevallii var. bivittata Werner, Abhand. Konig. Bayer. Akad. Wiss., 22. p. 381 : Africa south of the Orange Eiver, i.e. Little Namaqualand. 1910a. Werner, p. 327. 1915c. Werner, p. 360. 1916g. Chabanaud, p. 489. 1955a. Mertens, p. 95. 1910a. Prosymna siindevaUii var. macrnspiJa Werner, Denkschr. Med.-Xat. Ges. Jena, 16. p. 357: Steinkopf, Little Namaqualand. Common Name. Two-striped Shovel-snout (A.L.). Description. As in the typical form, with prefrontal single, rarely divided"'; temporals 1 + 2, 2 + 2, or 2 + 3"^; ventrals 159-181; subcaudals 22-29. This form agrees with s. sundevallii in having widely separated internasals, but approaches frontalis in its high ventral count. 135 Divided in the adult mentioned by Werner (1910a) ; confirmed by Mertens (1955). i3« 2 + 3 in the holotype of macrospila. LOVERIDGE : AFRICAN SNAKE GENERA 137 Color of var. hivittata. Above, an interocular brown band and a transverse, hourglass-shaped, dark brown blotch on the nape from which there extends backwards a pair of parallel, white- flecked, brown lines three scales in width ; on either flank the three outermost rows of scales are white. Below, white, uniform. Color of var. macrospila. Above, head j^ellowish, a brown band across the nasals and prefrontal ; also brown are the anterior borders of the frontal and supraoculars in addition to a spot under either eye; body yellowish, a vertebral series of angularly shaped, or obliquely elliptical, dark brown spots, a dorsolateral series of similar spots along the back. Below, whitish, uniform. Essexvale ? with 181 A^entrals. Above, dark purplish brown ; rostral lighter ; upper labials cream colored ; frontal with a dull orange (turned brown in alcohol) blotch; parietals with smaller blotches; liack with a vertebral stripe three scales in width but interrupted by numerous X or V-shaped purplish brown mark- ings ; lower flanks have the outermost 11/^ or 2 scale rows cream colored. Below, cream colored, uniform. Based on notes made from this snake in life by D. G. Broadley. Size. Length of 6 (S.M.F. 32542) 245 (221 -f 24) mm., from "Deutsch-Slidwestafrika"; largest 9 (S.R.M. 635) 342+ (315 4- 27+) mm., from Essexvale; length of hivittata type (pre- sumably an adult $ ) 296 (275 + 21) mm.; length of macrospila type (presumably a juvenile 9 ) 156 (141 -|- 15) mm. Remarks. As suggested by its size, macrospila may represent the juvenile coloring for "Werner (1910a) clearly states that he regards it as but another color variety, saying he has both varieties from Steinkopf. In the same paper he mentions three hivittata, the most northerly example coming from Okahandja. Mertens (1955), who rightly recognizes the race, has also seen three specimens of it. The only one I have handled is from Essexvale, the first record from Southern Rhodesia, where it was collected bj^ D. G. Broadley. Localities. Southern Rhodesia: *Essexvale. Little Namaqua- land: South of Orange River (hivittata type) ; Steinkopf (macro- spila type). Southwest Africa: Okahandja; Otjimbingue; Wind- hoek. Range. Southern Rhodesia, west to Little Namaqualand, north to Damaraland, Southwest Africa. 138 BULLETIX : MUSEUM OF COMPARATIVE ZOOLOGY Prosymna lineata (Peters) 1871a. Tcmnorhyiichus Un.eatus Peters, Monatsb. Akad. Wiss. Berlin, p. 568: Matlale, Gazaland, Mozambique. 188ila. Prosymna frontalis Bocage (part:not Peters), p. 288. 1895a. Bocage, p. 99. 1896a. Prosr/mna Snndevallii (sic) Bocage (not Smith), p. 92 1953e. Prosymna lineata Loveridge, p. 26-4. Common Name. Mozaml)ique Shovel-snout (A.L.). Description. Rostral with sharp horizontal edge ; separated l)y a pair of internasals (that are broadly in contact) from the single bandlike prefrontal ; frontal large ; loreal as long as high ; j)reocular 1 ; postoculars 2 ; temporals 1 -(- 2, 2 -}- 2 or 2 -|-3 ; upper labials 5-6, the third and fourth entering the orbit ; lower labials 8, the first 3 in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows ; ventrals 155-169 ( 9 9 155- 169) ; anal entire; subcaudals 18-20 ( 9 9 18-20). Color. Above, head brown with black markings on the pre- frontal ; body dark violet brown, some scales edged with black and others flecked with white and coalescing to form a Jadderlike pattern on the dorsum. Below, yellowish white, uniform. Juvenile 9 from Plumtree. Above, pale gray ; internasals with a black crossbar ; prefrontal with a spectacle-like crossbar as its extremities encircle the orbits ; frontal with a black spot pos- teriorly ; a black, T-shaped, nuchal marking ; dorsum and tail with a series of irregular, light-edged, dark blotches that do not coalesce with the ill-defined line that flanks them. Below, white, uniform. Size. Length of holotype 9 (Berlin Mus.) 282 (259 + 23) mm. In the figures furnished by Peters there was a major (prob- ably printer's) error; I am indebted to Dr. Heinz "Wermuth for those given above. That this was a very large Prosymna was obvious from the head length — 9 (not 11) mm., and body diameter of 9 mm.; length of juv. 9 (S. R. Mus.) 130 (120 + 10) mm., from Plumtree. Remarks. Peters rightly compared his new species to frontalis. with which Bocage erroneously synonymized it. Boulenger over- looked luieata altogether. I obtained the third known example (M.C.Z. 51144) under the peculiar circumstances mentioned LOVERIDGE : AFRICAN SNAKE GENERA 139 below. A fourth was obtained by A. E. Strover, a scholar of Plumtree School, in March, 1954. Still more recently I have seen another 9 taken at Selukwe by A. H. R. Stevenson. Enemies. The tailless, much-chewed remains of a large lineafa were present in the stomach of a topotypic Mozambique Genet {Genetta tigrina mossamhica) that was basking, shortly after sun-up, high in a big baoba!) where I shot it. Localities. Mozambique: Angoche ; *Kasumbadedza near Tete ; Matlale (type). Southern Rhodesia: *Plumtree; *Selukwe. Range. Mozambiciue, west through Southern Rhodesia. Prosymna greigerti Mocquard 19U(Jr. Prosymna Greigerti Mocquard, Bull. Mus. Hist. Nat. (Paris), 12, p. 466 : Lobi region, French Sudan, i.e. French West Africa. 1916g. Chabanaud, p. -438. 1919a. Boulenger, p. 285. 1929a. Werner, p. 142. 1933f. Angel, p. 12."), figs. 47-47a. 1950b. Villiers, p. 87, fig. 112. Description. Rostral with angular horizontal edge, separated by a pair of internasals from the paired prefrontals ; frontal large ; loreal longer than high ; preocular 1 ; postoeular 1 : temporals 1 + 2 ; upper labials 5, the second and third entering the orbit; low^er labials ?, the first 3 in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows ; ventrals 168; anal entire; subcaudals 18. Color. Above, chestnut brown, duller on the neck ; on temporal region, sides of neck, and rest of bodj^ each scale has a white central spot. Below, lips and throat dusky, otherwise whitish, uniform. Size. Length of holotype, presumably a 9 (Paris Mus.), 173 (159.5 + 13.5) mm. Remarks. Since Lt. Greigert collected the type almost lifty years ago, no second example has been taken. The type locality is near the northwest frontier of the Gold Coast and the northeast Ivory Coast. In view of Chabanaud 's recording of an aberrant 140 BULLETIN : MUSEUM OF COMl'ARATIVE ZOOLOGY melcagris with paired interuasals at Dagaua, Senegal, the possi- bility that greigerii (whose paired internasals and paired pre- frontals chiefly separate it from meleagris laurenti) is also an aberrant, should be borne in mind. Prosymna somalica Parker 1930b. Prosy7nna somalica Parker, Ann. Mag. Nat. Hist., (10) 6. p. 605: Hand, 8° X., 47°22' E., at 2,000 feet, British Somaliland. 1932b. Parker, p. 364. 1949a. Parker, p. 64. Common Name. Northern Somaliland Shovel-snout (A.L.). Description. Kostral with angular horizontal edge ; separated by a pair of internasals from the single, very rarely divided (B.M. 1949. 2. 1.91 only) prefrontal; frontal large; loreal longer tlian high; preocular 1, rarely 2 (B.M.1949.2.1.94) ; postocular 1 ; temporals 1 + 2 ; upper labials 6, the third and fourth entering the orbit ; lower labials 8, the first three in contact with the sub- linguals. Midbody scale-rows 15, smooth; ventrals 114-144 ( $ $ 114-126; 9 9 132-144); anal entire; subcaudals 23-38 (9 9 23-29; $ $ 30-38). Color. Above, dark purplish brown, the free edge of each scale margined with lighter. Below, white, usually uniform, the throat sometimes invaded by the darker dorsal coloring. Size. Length of holotype <5 (B.M.1930.5.9.12) 193 (154 -f 39) mm.; length of 9 (Parker: 1932b) 119 (104 -f 15) mm. Remarks. Only a single pair of sublinguals are well developed ; the three following scales do not differ appreciably from the corresponding scales in other species of Prosymna, though in the holotype of somalica they were actually longer than the anterior ])air. While assuming this most northerly representative of the genus is ancestral to its neighbor a. ruspolii to the south, wuth which it has much in common, I agree with Parker as to its specific distinctness. Dentition. Parker (1949a) invites attention to the less special- ized development of the posterior maxillary teeth which, in a 250 mm. somalica, measure only 1.0 mm. as against 1.6 mm. in an amhigua sfuhlmanni of similar size. LOVERIDGE: AFRICAN SNAKE GENERA 141 Habitat. In addition to the type, taken at 2000 feet, Colonel K. II. R. Taylor secured a dozen other examples at altitudes ranging' from 3500 to 4500 feet, in stone-strewn sandy terrain wliose scant veoetation consisted largely of acacia-euphorbia bush interspersed with patches of grass. Localities. British Somaliland: Bohodle; Borama District; Burao; Hand (many positions being designated by latitude and longitude ) . Range. British Somaliland. Prosymna meleagris laurenti subsp. nov. 184;!. Calainaria mrhagriii Reinhardt (part: $ only; omit S), K. Danske Vidcnsk. Selsk. Afhandl., 10. p. 238, pi. i, figs. 4-(i: "Guinea," as old museum specimen, possibly near Gambia? 1849. Prosymna meleagris Gray (part ^^^), p. 80. 18G2. Jan, p. 55 ( ? Sierra Leone). 1894a. Boulenger, p. 249. 1908a. Sternfeld, pp. 409, 425. 1909b. Sternfeld, p. 17. 1915d. Boulenger, p. 649. 1916g. Chabanaud, p. 440 (Frencli Congo and Sudan only). 1919a. Boulenger, p. 285. 1922. Aylmer, p. 15 (but possibly not this race). 1922a. Angel, p. 39. 1929a. Werner, p. 142, fig. 40. 1933f. Angel, p. 126, fig. 48. 1937. Andersson, p. 8. 1938. Angel & Lhote, p. 366. 1950a. Villiers, p. 70. 1950b. Villiers, p. 87, fig. 113. 1951. Monard, p. 151. 1954b. Laurent, p. 57. 19o6c. Loveridge, 1955, p. 45. 1884a. Temnorhynchns su7i(iei^allii Eochebrune (not Smith), p. 149. 1884a. T.emnorhynclius meleagris Eochebrune (not Reinhardt), p. 150. 1884a. Tevinorhynchus frontalis Eochebrune (not Peters), p. 150, 1884a. Tcmnorhynchus ambiguus Eochebrune (not Bocage), p. 151 (all four of these Eochebrune references are ignored). 1918b. Prosymna meleagris eollaris Chabanaud (not Sternfeld), p. 164. 1933m. Prosymna ambigua Witte (part:not Bocage), p. 91 (Kunungu). 137 Inevitably many of the followini; citations are in part only, though pre- ponderately the new form. See also citations under meleagris. I 142 JJULLETIX : MUSEUM OF COMPARATIVE ZOOLOGY Common Name. Sudanese Shovel-snout (A.L.). Holotypc. Museum of Comparative Zoology No. 53383, an adult 6 from Mongalla, Equatoria Province, Sudan, collected by John Owen, Esq., between May and August, 1950. Paratypes. Twenty $ $ and ten $ $ from Equatoria, viz. Museum of Comparative Zoology Nos. 53377-53392 (less 53383), from five localities (Lokwi; Magwe ; Mongalla ; Okaru; Torit) ; also Chicago Natural History iMuseum Nos. -18070-2, 58402-4, 58406-8, 58443, 62233-6, 62238-9, from three localities (Katire; Molongori ; Torit), all collected by John Owen. Diagnosis. Calainaria meleagris was described by Reinhardt on the basis of two specimens (a third, subsequently seen, re- ceives passing mention in a postscript), obviously a £ and 9, though this is not stated. Both were from "Guinea," a name that in 1843 was applied to any of the countries bordering the Gulf of Guinea from Portuguese Guinea to the French Congo. Fortunately Reinhardt states that his two snakes had different origins. One formed part of a collection sent to Copenhagen Museum in 1836 by Herr Chenon, assistant to the "government establishment on the coast." Presumably he refers to Christian- sborg, one-time seat of Danish administration on the coast of Ghana near Accra, for the various species of snakes donated by Chenon are still common in Ghana. The smaller $ , the main basis of Reinhardt 's description, agrees in every Avay with Gbana ni. meleagris as here defined. Reinhardt "s second snake (presumably the larger 5 ) was said to have come from the collection of the "old natural history societv." As its scale-counts fall within the range of the Sudanese race it may have come from the general region of Gambia, alternatively, though less likely at that early date, from the interior of Nigeria or Cameroon. Owing to the description of meleagris being based on both races, the separation of the forms has been obscured until now. Indeed it was not until 1 was able to assemble adequate series from Ghana (22 specimens) and the Sudan (31 examples) that the respective ranges in ventral and sul)caudal counts could be demonstrated. LOVERIDGE : AFRICAN SNAKE GENERA 143 Kao(> No. & Sex Ventrals ("audiils Largest exaiiiiilr m.meleagrls 15 Ghana in only one (C.N.H..M. 58403) of 21i Torit snakes. 141 Six, the third and fourth entering the orbit only on side of a Mongalla ? (M.C.Z. 53384). 1-12 151 in a Torit d (M.C.Z. 53388) ; 184 in a Torit ? (C.N.H.M. 48(i70). 143 19 in a McCarthy Id. snake {fide Andersson :1937) : 21 in several Torit ? ? : 25 in a Lokwi ? (M.C.Z. 53378), and, but not checked, 26 in a Poll specimen I Vienna Mus. ). i« Brown only in a Torit d" (C.N.H..M. 62234). 144 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY upwards to form a collar that almost meets in a point on the median line of the nape ; on the rest of the back each scak^ bears a whitish terminal spot. Below, throat brown in advance of the collar, otherwise under surface white, uniform. Assumino' this Prosy mna from Dagana is referable to the Sudanese form — as would appear from the ventral counts of two Gambia snakes furnished by Andersson — collaris appears to be only a color phase occurring erratically within the ranges of either race of meleagris. There remains a possibility that snakes from the extreme west ma,y represent yet another race of mcha- (jris, something that can be demonstrated only by assembling all the knoAvn material from the region. From Poli, in northwest Cameroon, we have a gravid 9 with heavy black collar which looks rather different from the extensive Sudanese series. In this it would seem to resemble the Yola (Jola) snake mentioned by Sternfeld (1908a) from northeast Nigeria, both localities being near Garoua. Ventral counts of three other Poli snakes in Vienna Museum (but unsexed) indi- cate they are referable to the Sudanese race, though slightly intermediate as might be expected. K^ize. Length of type, the largest 6 (M.C.Z. 58383) 280 (238 4- 42) mm., from Mongalla ; largest 9 (C.N.H.M. 48070) 360 (332 + 28) mm., from Torit. Dentition. Maxillary teeth 6, the anterior 4 minute, the last 2 large; an arrangement strikingly different from that of ango- lensis fide Laurent (1954b). Sexual dimorphism. In addition to the marked difference in the number of ventrals and subcaudals (as may be seen by refer- ence to the Description) the tail is distinctly longer in S $ , a^ may be seen from the following figures: In 13 Sudan 6 $ , tail is included in 11. & B. length 5.6 to 6.3 times; in 9 Sudan 9 9 , tail is included in II. & B. length 10.0 to 11.9 times. Hahifai. Most oi' the Sudan series were hoed up during cultivation, but two of them were found beneath rubbisli in Owen's garden. Localities. Sudan: *Katire ; *Lokwi ; *Magwe ; *Molongori ; *Okaru; *Torit. Belgian Congo: Kunungu. French Congo: (as meleagris; fide Chabanaud). British Cameroon: *Poli near LOVEBIDGE : AFRICAN SNAKE GENERA 145 Garoua. Nigeria: *Yola (Jola). Sierra Leone: {fide Jan:1862). Portuguese Guinea: Mansoa. Gambia: McCarthy Island. Sene- gal: Dagaiia. French West Africa: Dano near Diebon siilc only of a Niircr siiakc 1*7 137 in d^cT from Soinan.va (M.C.Z. 55229) and Lagos (C.X.H.M. 4ni!»l: IBO in ?9 from Achimota (M.C.Z. 5369.S and C.N.H.M. — ). 148 19 in an Achimota 9 (C.N.H.M. — ): ;>« in dd from Achimota and Lagos (C.N.H.M. 41119). LOVERIDGE : AFRICAN SNAKE GENERA 147 than the 9 . It is doubtful if any significance attaches to this variant as it has l)een recorded from as far away as Dagana, Senegal (see Remarks under the Sudanese subspecies). Size. Largest i (M.C.Z. 58698) 257 (220 + 37) mm., from Achimota; largest 9 (M.C.Z. 55224) 279 (255 + 24) mm., from Somanya. Anatomy. No parotids present (Phisalix). Sexual dimorphism. In addition to the marked difference in the number of ventrals and subcaudals, as may be seen b\ reference to the Description, the tail is distinctly longer in (5 5 . Thus : In 12 Ghana c^ $ , tail is included in H. & B. length 5.4 to 6.6 times. In 8 Ghana 9 9 , tail is included in H. & B. length 10.3 to 11.6 times. Hahitai. A Lagos and several of the Achimota snakes were found in roadside gutters intended for carrying off rain water. Most Somanya snakes were dug up by labourers. Localities. Nigeria: Lagos ; Niger. Dahomey: Agouagou. Togo: Adele (as Bismarckburg) ; Mangu ; Misahohe ; Wegbe. Ghana: Accra; *Achimota; *Odumase ; *Pong-Tamale, Northern Terri- tories; *Somanya, Krobo. Also "Guinea" (type locality; prob- ably Christianborg). Possibly the records from Ivory Coast {jide Ghabanaud) and Sierra Leone {-fide Jan) belong here, l)ut no scale counts are available. Range. Southern Nigeria west to Ghana, and probably Ivory Coast possibly even as far as Sierra Leone. Prosvmna frontalis (Peters) ls(J71i. T()U'Hi)fliiincltu.s fro)it(iUs Pctois (part), Monatsb. Akad. Wiss. Berlin, p. 23G, pi — , tigs. 1-ld: Otjinibingue, Southwest Africa. 1894a. Prost/mna frontalis Boiilenger, p. 248. 1898. Selater, p. 99. 1910b. Bouleiiger, p. 508. 1910. LOnnberg, p. 5. 19101i. Sternfeld, p. 20, fig. 21. 1910c-. Sternfeld, p. 55. 1912. FitzSimons, T. W., p. 88. 1916g. Chabanaud, p. 439. 1-18 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY 1929a. Werner, p. 142. 1955a. Mertens, p. 94. 19(12. Prosy mna {Pseudoprosymna) bcrgeri Lindholm, in Lampe, Jahrb. Nassau. Ver. Naturk. (Wiesbaden), 55, p. 57: Eietmond, Gibeon District, Southwest Africa. 1910b. Boulenger, p. 509. 1910b. Sternfeld, p. 20, fig. 21. 1912. FitzSimons, F. W., p. 88. 1914b. Methuen & Hewitt, p. 143. 1915c. Werner, p. 360. 1916g. Chabanaud, p. 439. 1929a. Werner, p. 142. 1938. FitzSimons, V. F., p. 156. Further citations of "frontalis" will be found under s. hivif- tata, lineata and m. laurenti. Common Name. Southwest African Shovel-snout (A.L.). Description. Rostral with angular horizontal edge ; separated by the single bandlike internasal from the single prefrontal ; frontal large ; loreal as long as high ; preocular 1 ; postoculars 2, rarely 1""; temporals 1 -f 2 ; upper labials 6-7, the third and fourth entering the orbit ; low^er labials 8-9, the tirst 3, i-arely 4, in contact with the single pair of sublinguals. Mid body scales smooth, in 15 rows; ventrals 160 ''^'-188 {S S 160 '^''-169; 9 9 180-188) ; anal entire ;subcaudals 34-51 (9 9 34-38; S $ 48-51). Color. Above, head light gray finely dusted with black, a more or less distinct dark crossband along the posterior margin of the prefrontal, or between the eyes, and a dark crescentic blotch on the parietals which may be reduced to an interparietal streak or spot ; except for a black subocular spot, sides of head and neck whitish ; a broad black nuchal band followed by several similar crossbands that become progressively fainter and ill-defined pos- teriorly ; body and tail gray to pale yellowish brown on the sides, each scale edged with pale purplish brown or blackish, their centers often whitish and forming Avavy whitish crossbands. Below, white, yellowish white, or brown, uniform. Size. Length oi & $ (T.M. 17597) 187 (153 + 34) mm.; the h erg eri (toty^Q $ (Wies.Mus.) 160 (131 + 29) mm.; tlie hcrgeri 1 '■' On one siilc only of the liolotyiif. i'>u 1.50 for ii I'sakos ;^ {fidr Werner : lit ITic) reiinires verifyiiij;. LOVERIDGE : AFRICAN SNAKE GENERA 149 eotype 9 (Wies.Mus.) 155 (135 + 20) mm.; the frontalis holo- type S (Berlin Mus.) 300 (244 + 56) mm. ^•''^; largest 9 (S.M.F. 46739) 355 (312 + 43) mm., from Kaiser-Wilhelms-berg. Remarks. The ventrals and subeaudals of the type have been recounted by Dr. H. Wermuth and found to be 162 (not 167) and 48 (not 50) respectively; actually 48 conforms to Peters' illustration. The name fronialis had fallen into disuse for 25 years until revived by Mertens (1955) whose detailed reasons for synony- mizing hergcri should be consulted. Actually, though by a dif- ferent route, I had arrived at the same conclusion in 1950 when the following paragraph was written, though unpublished until now. P. hergeri was based on two snakes in which the posteriorly directed nasal suture was incomplete; such was also the case with a Klipfontein snake taken by FitzSimons (1938). In all other respects — both of lepidosis and color pattern — they con- form so closely to frontalis that one can only conclude the character is a variable one or that Peters' figure is incorrect in this detail. Localities. Little Namaqualand: Klipfontein. Southwest Africa: Churutabis; Gobabis; Kaiser-Wilhelms-berg near Oka- handja; Narudas Sud at 4800 feet; Okahandja; Otjimbingue; Rietmond ; Usakos. Range. Little Namaqualand north to Southwest Africa. Prosymna angolensis Boulenger 1873b. Prosymna frontalis Bocage (not Peters), p. 218. 1882a. Boeage (part: omit Angoche = lineata), p. 288. 1895a. Bocage, p. 98, pi. xi, fig. 2. 1896(1. Boulenger, p. 641. 1915a. Prosymna angolensis Boulenger, Proc. Zool. Soo. London, p. 209: based on Bocage, 1895a, which see for description. No type locality so Huila, 15°5'S., 13°30'E., Angola is now designated. 1916g. Chabanaud, p. 439. 1929a. Werner, p. 142. 1937b. Monard, pp. 114, 122. 151 As remeasured by Dr. Wermuth, Peters' figures were obviously erroneous, they read : "TotaUange 0nil35 ; Schwanz Om057." 150 BULLETIN: MUSEUM OP COMPARATIVE ZOOLOGY 1955a. Mertens, p. 94. 1937b. Prosymna ambigua Mertens (not Booage), p. 13 (Cubal). 1938e. Prosymna amhigua ambigua Mertens (not Bocage), p. 439. For another citation of '^ angolensis" see under a. amhigua. Common Names. Southwest xlnp-ola Shovel-snout (A.L.) ; tjolongo (in Caconda: Bocag:e). Description. Rostral Avith angular horizontal ed<;e ; separated l)y the single bandlike internasal from the single prefrontal : frontal large ; loreal rhomboidal or pentagonal ; preocular 1 ; })Ostoc'ular 1-2 ; temporals 1 -|- 2 ; upper labials 6, the third and fourth entering the orbit ; the first 3 lower labials in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows ; ventrals "145-163" {fide Bocage: probably $ S 145-155; 9 9 155-163); anal entire; subcaudals "17-25" {fide Bocage: prob- ably 9 9 16-19; S $ 25-28). Color. Above, head yellowish brown, with or without black spots of which the most common is a black band across the frontal anteriorly, followed by a pair of black blotches on the supra- oculars and parietals ; a black nuchal spot or collar more or less distinguishable on the paler yellowish-brown specimens, fol- lowed on the back by two longitudinal series of round, black or blackish, spots ; sometimes uniform, or each scale with a light s])ot near its tip and the edges of the scales somewhat darker. Below, including the lips and two or three outermost scale-rows, yellowish white. Size. Largest recorded, presumably a 9 (Bocage :1873b), 360 (331 + 29) mm.: next largest 9 (Mertens :1938e), 327 (305 + 22) mm. from Cubal. Remarks. No type for angolensis was designated by Boulenger, whose basis Avas "frontalis Bocage (1895 :not of Peters)" in Herp. Angola Congo, p. 98, pi. xi, fig'. 2. Bocage lists his ma- terial as coming from seven localities in the high plateau of Benguella and Mossamedes. Mertens (1955, p. 94) errs in giving Duque de Braganca, northwest Angola, as type locality for angolensis, for this was type locality of amhigua Bocage (cf. p. 100). LOVERIDGE : AFRICAN SNAKE GENERA 151 Mertens (1955) suggests the possibility of angolensis being a race of frontalis from which it is chiefly distinguished by its much lower ventral and suhcaudal counts — so much lower, in fact, that the gap is unlikely to be bridged. More probably its relationship lies with lineata Peters of Mozambique, with which it is connected by Southern Rhodesian material. LocaHtics. Southwest Africa: Karakuwisa, Okawango. Angola: Biballa; Caconda ; Cubal; Ebanga; Huila; Maconge (Maconja) ; Mossamedes ; Quibula ; Quindumbo ; Quissange. Range. Southwest Africa north to southwest Angola. Prosyimna ambigua ambigua Bocage 1866a. Prosynina meleagris Bocage (not Reinhardt), p. 47. 1873b. Prosymna amhUiuus Bocage, Jorn. Sci., Lisboa, 4. p. 218: Duque de Braganea, Angola. 1895a. Prosymna ambigua Bocage, p. 99, pi. xi, figs. la-d. 1896d. Boulenger, p. (i-il. 1910. Lonnberg, p. 6. 1915a. Boulenger, p. 208. 1915c. Boulenger, p. 62."). 1929a. Werner, p. 142. 1931. Monard, p. 104. 1937b. Monard (part), pp. 114. 123. 1910. Prosymna meleagris concolor LonnV)erg, Arkiv. ZooL, 7. No. 8, p. 5, fig. 2: Mukimbungu, Lower Congo River, Belgian Congo. 1916g. Chabanaud, p. 440. 1940. Prosymna angolensis Bogert (part:not Boulenger), p. 59 (Cape- longo). 1941a. Prosymma amhii/ua ambigua T'tlniioller (part), p. 43. 1950f. Laurent, p. 129. 1953a.. Laurent, pp. 21, 23. 1953. Witte, p. 206, fig. 65. 1954b. Laurent, p. 52. 1956. Laurent, p. 193. 1952b. Prosymna amhigva bocagei Laurent (part:not Boulenger), p. 200 (for 9 with 154 ventrals, substitute Leopoldville for "Nepoko, Uele"). 1953a. Prosymna ambigua loveridgei Laurent, p. 23 (no description: "un seule specimen, du Mayombe")- 1954b. Laurent, Serv. Cult. Comp. Diani. Angola. Museu Dundo, No. 23, p. 56, fig. 15: Boma, Lower Congo River, Mayombe, Belgian Congo. 152 BULLETIN: MUSEUM OP COMPARATIVE ZOOLOGY Of the foregoing citations several are actually ' ' in part ' ' ; further listings of " amhigua" will be found under m. laurenti, auf/olcufiis, 0. hocagii, a. urundiensis, a. hrevis and a. stuJilmanni. Common Name. Northwest Angola Shovel-snout (A.L.). Description. Rostral with angular horizontal edge^^"; separated l)y the single bandlike internasal from the single prefrontal ; frontal large ; loreal longer than high ; preocular 1 ; postoculars 2, rarely 1''"; temporals 1 + 1, 1 + 2 or 2 + 2'^' ; upper labials 5-8,^'''"' rarely the second and third, or second, third and fourth, normally the third and fourth, entering the orbit ; lower labials 7-8, the first 3 in contact with the single pair of sublinguals. Midbody scales smooth, in 15-17^''" rows; ventrals^" 129-154 {$ i i29-140; 9 9 145-154); anal entire; subcaudals 19-34 ( 5 ? 19-24; $ S 26-34). Color. Above, rostral and upper lip white ; each parietal with or without a whitish spot ; a broad black nuchal collar some- times present; otherwise head and body gray, pale brown or black, the apex of each scale with or without a whitish spot. Below, yellowish white to blackish gray, uniform, or the center of each ventral with a more or less distinct dusky mark, the edges of each scute paler. Size. Largest £ (R.G.M.G. — ), 283 (237 + 46) mm.; largest 9 (R.G.M.G — ), 345 (313 + 32) mm., both from Bokoro (ex. Laurent :1954b). The type of amhigua, evidently a 9 , was given as 225 (113 + 12) mm. The type of loveridgei, also a 9 , was given as 313 (281 + 32) mm.' Remarks. In the synonymizing of )n. concolor from the Lower Congo with a. amhigua from northwest Angola, Laurent (1954b) and I are in complete accord. It seems somewhat illogical, how- ij- The rostral is fused with the internasal in some Bokoro snal had an angnlenxis as a second specimen. LOVERIDGE : AFRICAN SNAKE GENERA 153 ever, to then postulate the presence of two races in the Lower Congo by describing a. loveridgei from there. P. a. loveridgei, as Lanrent points out, differs from all other Congo Prosipinia seen by him in that the horizontal suture, that normally extends backwards from the nostril to the loreal, is, in loveridgei, con- nected by a vertical fissure to the first labial. This has resulted in the splitting off of a small supplementary nasal. On the left side of one (I 3598) of the 50 Liwale a. stuhlmanni 1 examined, there is no horizontal suture whatever, the nostril being in an entire nasal. As long ago as 1923 Schmidt recorded the creation of a supplementary loreal by vertical Assuring of the loreal in one of his Garamba a. bocagii. Since then other in- stances of fissuring have been reported among members of this genus so that it seems reasonable to assume that the only known example of o. loveridgei is nothing more than an aberrant a. amhigua. Apart from its small supplementary shield, the only way in -which a. loveridgei 9 differs from 9 9 oi a. ambigua is in having 24 (as against 19-21) subcaudals. Other more adequately repre- sented species and races of Prosymna display an even wider range than the 6 which would result from extending the range from 19 to 2-t so as to include n. loveridgei. It would appear that I am destined to outlive my namesake which, regretfully, I must consign to the synonymy. Dentition. Maxillary teeth 8, followed by 3 enlarged, lancet- shaped teeth, according to Bogert (1940: dealing with a Cape- longo 9 ), who points out that "the anterior part of the maxilla makes a diagonal suture with a posterior elongation of the premaxilla." apparently an unusual adaptation calculated to provide mechanical support for the snout during burrowing operations. Laurent (1954b), dealing with Congo material, reports the maxillary teeth as 8, the first six small, the last 2 enlarged. In this he sees a dift'erence with ^'angolensis," to which species Bogert had referred the Capelongo snake which I believe to be an a. amhigua. 1,34 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Sexual dimorphism. In addition to the marked difference in the number of ventrals and snbcaudaLs, as may be seen in the Description, 15 (^ 5 (measured by Laurent) have tails included in H. & B. 4.6 to 5.8^'''^ times; 13 9 9 (9 + 1 measured by Laurent) have tails included in H.& B. 8.6 to 10.7 times. Localities. Angola: Capelongo (Bogert :1940) ; Duque de Bra- •lanca (type of amhigua) ; *Missao di Dondi (M.C.Z. 32468) ; Vila da Ponte.^"^ Belgian Congo: Albertville ; Bokoro ; Boma (type of loveridgci) ; Bukena; Congo da Lemba ; Hemptinne St. Benoit, Kasai ; Kalina; Kamina; Kihvezi; Kisantu ; Kunungu ; Leopoldville ; Lomami ; Mabwe ; Mayombe ; Mukimbungu (type of concolor) ; Mushishi; Ndwa ; Thysville ; Tipoyo, Fiji Terri- tory ; Upper Katanga. French Congo: *Brazzaville. Rawje. Northwest Angola, north through Belgian Congo to French Congo. *c>^ Prosyinika ambigua bocagii Boulenger 18971). Prosymna bocagii Boulenger, Anu. Mag. Nat. Hist., (6) 19. p. 278, figs. — : Zoiigo, Ubangi Rapids, Belgian Congo. 190] g. Boulenger, p. 9, pi. iii, fig. 4. 1903b. Bocage, p. 63. ]906c. Moequard, p. 467 (but omit Fassei as syn.). 1910. Lonnberg, p. 6. 1915a. Boulenger, p. 208. 1916g. Ohabanaud, p. 440. 1929a. Werner, p. 142. 1937c'. Loveridge, p. 275. 1908. Aparallactus concolor Werner (not Fischer), (1907), p. 1882 (text). 1916f. Pro.sijmma amhigua Chabanaud (not Boeage), p. 372. 1916g. Chalianaud, p. 439. 1919g. Boulenger, p. 25 {Pro.symma) . 1923. Schmidt, p. 89, figs. 7-8 (but " 9 " is a $). 158 6.5 times In one of Monard's (19.S1) Vila da Ponte snakes, if his measure- ments d'.*.") + •■'.<>) mm. are correct. The assignment of liis two specimens to (unbifjua should be checked for the scale counts (ventrals 129; subcaudals 26). thoush not the locality, agree with a. brevis. The coloration, however, seems to he that of angohtixin. It might be translated as follows: Ahove. gray: pre- frontal with a black. V-shaped marking extending to above the eyes as in front- (ilix. from tlie frctiital across the parietals to the post-parietal scales extends a large nuchal blotch ; bodv reticulated with black, on the dorsum a double row of black, sometimes conriuent, spots tend to be more symmetrical posteriorly (Monard :ltKU). The low number of ventrals rule out the possibility of these Vila da Ponte snakes being referred to either angolensis or frontalis. LOVERIDGE : AFRICAN SNAKE GENERA 155 1933m. Witte (part), p. 91. 1936. Pronymna amhigiia stuhhnanvi Pitman (part mot Pfeffer), p. 126, pi. viii, figs. 1-2; col. pi. H, fig. 2 (Uganda records). 1938a. Pitman (part) p. 310. 1938b. Pitman (part), pp. 39, 120-121, 149, 310, pis. as above. 1952b. ProsymtM amhUjvn hoc^gei Laurent, p. 200. 1954rb. Laurent, p. o4. 1956. Laurent, p. 193. 1956c'. Prosymna avihit/iia ambif^ua Loveridge (not Socage), }). 4.') (9 in reprint ) . Common Name. Northeast Congo Shovel-snout (A.L.). Description. Rostral with angular horizontal edge ; separated by the single band-like internasal from the single prefrontal, which occasionally may enter the orbit ^^^; frontal large; loreal as long as,^"" or longer than, high; preoeular 1, rarely absent ^'^^; postoculars 1, 2 or 3'''- ; temporals 1 + 1,1 + 2 or 2 + 2'''' ; upper labials 5-7, rarely the second and third,^"* normally the third and fourth, entering the orbit ; lower labials 7-8, the first 3, rarely 1, in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows ; ventrals 139-168 ( S S 139-148 ; 9 9 161- 168) ; anal entire; subcauclals 17-32 ( 2 $ 17-21; i S 27-32). Color. Aliove dark brown, head uniform, body seemingly paler owing to the center of each scale being lighter (blue gray), imparting a reticulate appearance whicli extends to the outer- most rows. Below, blackish brown, each ventral and subcaudal shield with a median, transverse, light bar. Size. Largest c^ (A.M.N.H. 12145) 298 (250 + 48) mm., from Garamba; largest 9 (R.C4. 3381) 398 (366 + 32) mm., from Niangara; the holotype 9 (E.G. 44) was 347 (320 + 27) mm. as remeasured by Laurent (1954b). Remarks. See comments under P. a. urundiensis Laurent. 159 On one side of hocuyc type and in a Batangafo 9 (A.N.S.P. 20768). 160 A small supplemeiitarv loreal is split off from nasal in a Garamba d (not ? :A.M.N.H. 12144). i«i In the Lado snake mentioned b.v Werner (1908), if referable to bocagii. 162 Three In a Nopoljo River d" reported by Laurent (1954b). i«3 Both 14-1 and 1 + 2 occur in the Sudan and Uganda : 2 + 2 in an Avakubi d" (Laurent :1954h) . 161 Asymmetrically in a Congo snake (Laurent :19o4b). 156 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Hemipenis. Schmidt (1923), when figuring the extraordinary unforked hemipenis of this species with its single sulcus, remarks that the organ is at least 10 mm. longer than the tail and in consequence must he "telescoped" when withdrawn. He suggests that the tail reduction responsible for this unusual condition has resulted from the adoption of burrowing habits. Sexual dimorphism. In addition to the marked difference in the number of A'entrals and subcaudals, as may be seen in the Description, in 8 known $ $ , tail is included in H. & B. length 5.1 to 6.8 times; in 7 known 9 9, tail is included in H. & B. length 11.0 to 12.8 times. Breeding. On July 2, at ]\Iolongori, a 9 held 6 eggs measuring about 16 to 23 X 7 mm. Localities. Sudan: *Molongori; *Torit. Uganda: *Eastern Province ; Lado = West Nile Province ; Serere, Teso. Belgian Congo:^'""' Avakubi ; Garamba ; Karawi, Uljangi ; Mahagi Port ; Nepoko, Uele ; Niangara ; Upper Ubangi ; Zongo, Ubangi Rapids (type). French Congo: ? Batangafo. Range. Southern Sudan and Uganda, west through northern Belgian Congo to French Equatorial Africa. Prosymna ambigua urundiensis Laurent 1933J. Prosymna amhigua Witte (not Bocage), p. 123. 1953a. Prosymna ambigua urundieyisis Laurent, p. 23 (no description: "un seulc specimen de Xyanza")- ]P541i. Laurent, Serv. Cult. Comp. Diam. Angola. Museu Dundo, No. 23, p. 56: Nyanza on Lake Tanganyika, Belgian Ruanda-Urundi. 1956. Laurent, p. 193. Common Name. Urundi Shovel-snout (A.L.). Description (adapted from Laurent). Preocular 1 ; postoculars 2 ; temporals 1 -f- 2 ; upper labials 6, the third and fourth enter- ing the orbit ; lower labials 8-9, the first 3 in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows ; ventrals 152 (in S ) ; anal entire; subcaudals 32 (in $ ). Size. Length of holotype S (R.G.M.C, 9232) 235 (199 + 36) mm. 165 Omitted are a d and ? (I.R.S.X. 4883) allesodly from Mpese-Iiikisi. Lower Congo, listed in a footnote by Laurent (1954b) who regards data as doubtful. LOVERIDGE : AFRICAN SNAKE GENERA 157 Remarks. That the sexing is correct seems certain from the 32 subcaudals, the same number as present in a P. a. hocagii S (C.N.Il.M. 58405) from as far nortli as Torit in tlie Sudan. The only grounds for the separation of urundie^isis, therefore, rest on its possession of 152 ventrals (139-148 in S S hocagii). A i-ange of 10 in $ ventrals is also found in a. ambigua, a. hrevis, frontalis and ni. laurenti, but is exceeded in half a dozen other Frosymna. If tlie range of a. hocagii is extended to include urundiensis it would give a range of 14 ventrals as opposed to an acceptable range of 16 in $ i of the much better known a. stuhlmanni. Further collecting of Frosymna in Urundi should indicate which is the correct course to follow. Prosymna AMBIGUA BREVis Laureiit 1950f. Prosynoia amhigua ambigua Laurent (not Bocage), p. 129. 1953. Witte (part), p. 206 (Sandoa). 1954b. Proaymna amhigua brcvi>;, Laurent, Serv. Cult. Comp. Diam. Angola, Museu Dundo, No. 23, p. .10, figs. 12-14: Duudo, Angola. Common Name. Northeast Angola Shovel-Snout (A.L.). Descriptio7i (after Laurent). Preocular 1; postoculars 1-2; temporals 1 + 2, fused in one specimen, the second temporal fused with the fifth upper labial in another ; an additional upper labial is asymmetrically present in two snakes ; lower labials 8-9, the first 3 or 4 in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows; ventrals 127-146 ( $ $ 127- 136; 9 9 140-146) ; anal entire; subcaudals 15-30 ( 9 9 15-17; $ $ 25-30). Color. For comments on its variability, see Laurent (1954b). Size. Length of S (R.G.M.C. 7929) 308 (267 + 41) mm.; length of 9 (R.G.M.C. 8276) 351 (325 + 26) mm., both from Sandoa. Remarks. In ventral counts this form overlaps those of a. stuhlmanni, but is apparently distinguishable by the low number of subcaudals in the 9 - In S S , however, the subeaudal ranges overlap so that a specimen of a. siuhlnianni from Abercorn (M.C.Z. 54081) with 139 ventrals and 29 subcaudals might be referred to a. hrevis were it not for the accompanying 9 (M.C.Z. 54082) having 155 ventrals and 20 subcaudals. For further com- ments on the status of this form, see Laurent (1954b). 158 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Sexual dimorphism. The relationship of tail to total length is given by Laurent (1954b) as ranging from .075 to .087 in 9 9, and from .146 to .150 in S S . Localities. Angola: Dundo; Sombo. Belgian Congo: Sandoa, Katanga. Range. Northeast Angola, nortli to southeast Belgian Congo. Prosymna ambigua ruspolii (Boulenger) 1896b. Asthenophis ruspolii Boulenger, Ann. Mus. Civ. Stor. Nat. Geneva, (2) 17. p. 12: Magala, Uniberto Id., Ganale Doria, Somalia. l.S97g. Boulenger, p. 270. 191.1(1. Boulenger, p. Ii49. 1927. Calabresi, p. .14. 1929a. Werner, p. 143. 1930c. Scortecci, p. 18. 1949a. Parker, p. 6.5. 1929c. Prosymna U(/re.'ilis Seorteeci, Atti Soc. Ital. Sci. Xat. (Milano\ 68, p. 272, figs. — : Villaggio Duca degli Abruzzi. Somalin. 1939a. Scortecci, p. 273. 1949a. Parker, p. 63. Common Name. Southern Somaliland Shovel-snout. Description. Rostral with angular horizontal edge; separated by the single bandlike internasal from the single prefrontal; frontal large ; loreal longer than high ; preocular 1 ; postocular 1 ; temporals 1 + 2 ; upper labials 6, the third and fourth entering the orbit; lower labials 8, the first 3 in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows; ventrals 130-154 {S S 130-136; 9 9 143-154); anal entire; subeaudals 23-27 ( 9 9 23-?; c^ (5 32-37). Color. Above, dark reddish to purplish brown ; usually a light spot on the apex of each scale, whose free edge may appear lighter. Below, yellowish gray, somewhat darker anteriorly. Size. Length of S (type of ruspolii) 198 (165 + 33) mm., length of 9 (type of agrestis) 252 (226 + 26) mm., length of largest assumed 9 (Scortecci: 1929a) 295 (262 -f 33) mm. Remarks. As ])ointed out by Parker (1949a, p. 63), there is nothing to distinguish agrestis from ruspolii except the alleged I LOVERIDGE : AFRICAN SNAKE GENERA 159 differences in dentition, and he suggests that the enlarged pos- terior maxillary teetli have been overlooked in the juvenile type of Asthenophis, and the second example of ruspolii, which Scor- tecci called a 9 , is almost certainly a $ . Parker rightly urges a re-examination of the dentition of these two specimens, but, as my attempt to borrow one of the cotypes met with no response, 1 venture to risk censure by synonymizing agrestis with ruspolii. Breeding. One 295 mm. 9 from Belet Amin held 4 eggs, each measuring "275 [presumal)ly 27. 5J x 7 mm." (Scortecci :1939a). Localities. Somalia : Belet Amin ; Jonderma ; Magala ; Villag- gio Duca degli Abruzzi. Range. Somalia. Prosymna ambigua ornatissima Barbour & Loveridge 1928t'. Prosymna ornalis.nma Barbour ^- Loveridge, Mem. Mus. Coinp. Zool., 50. p. 120, col. pi. ii. fig. 2: Nyange, Uluguru Mountains, Tan- ganyika Territor.v. 1928,i. Loveridge, p. 75. 1937f. Loveridge, p. 502. 194La. LTthmoller, p. 40. Conuiion Name. Ornate Shovel-snout (A.L.). Description. Rostral with angular horizontal edge; separated by the single bandlike internasal from the single prefrontal ; frontal large ; loreal longer than high ; preocular 1 ; postoculars 0-1-2; temporals 1 + 2 (tlie anterior one entering the orbit below the postocular on the right side of M.C.Z. 23270 only) ; upper labials 6, the third and fourth entering the orbit; lower labials 7-8, the first 3 in contact with the single pair of sublinguals; midbody scales smooth, in 15 rows; ventrals 127-150 ( $ S 127- 132; $150) ; anal entire; subcaudals 27-40 ( $ 21 ; $ £ 35-40). Color. Above, rostral and rest of head scarlet (whitish in alcohol) except for a black crossband over the prefrontal-frontal suture (frequently reduced to a vertical streak from eye to labials), and an arrow-shaped extension of the black body color- ing reaching to the frontal ; body black with 13 or 14 irregularly transverse, scarlet crossbars which may be interrupted on the vertebral line or broken and alternating. Below, throat pink; rest of undersurface black except where the lateral scarlet blotches impinge on the outer edges of the ventrals. !()() BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY No other member of the genus is so handsomely and apose- niatieally colored {vide plate). Size. Length of paratype S (M.C.Z. 23272) 297 (242 + 55) mm.; length of holotype $ (M.C.Z. 23271) 286 (252 + 34) mm. Breeding. On October 4, the umbilical ventral scutes of a young 143 (117 + 26) mm. male were still unhealed. Habitat. A forest-edge species occurring from 2000 to 2500 feet. For further details see original description. Localities. Tanganyika Territory: Iluguru Mountains — Nyan- ge ; Vituri. Range. Tanganyika Territory. Prosymna ambigua stuhlmanni (Pfeffer) 1891a. Prosymna ambigua Boulenger (not Boeage), p. 306. 1894a. Boulenger, p. 248. 1896a. Boeage, p. 93. 1896. Tornier, p. 71 (reprinted 1897). 1897. Tornier, p. 65. 1902b. Boulenger, p. 17. 1907a. Boulenger, p. 11. 1908b. Bouleng«r, p. 229. 19101). Boulenger (part), p. 509 (omit Angola). 1910a. Sternfeld, p. 21, fig, 23. 1912. FitzSimons, F. W., p. 88. 1923e. Loveridge, p. 880. 1924b. Loveridge, p. 5. 1928c'. Barbour & Loveridge, p. 121. 19331i. Loveridge, p. 244. 1934. Pitman, p. 295. 1893. Ligonirostra StuMmanni Pfeffer, Jahrb. Hamburg Wiss. Anst., 10. p. 78, pi. i, figs. 8-10: Usambara, Tanganyika Territory. 1906b. Prosymna Vas.'ici Moc-qunrd, Bull. Mus. Hist. Xat. (Paris), 12, p. 250 : Mozambique. 1906i-. Mocquard, p. 467. 1916g. Chabanaud, pp. 433, 440, figs. 1-2. 1917e. Chabanaud, p. 225, fig. — (correcting previous drawing). 1909c. Prosymna vartnbili.r, Jalires. Ver. Nat. Wiirttemberg, 65. p. (50: '*East At'ric.-i." 1929a. Werner, p. 164. 1915c. Prnsymna hocagii Boulenp;er (partrnot Boulenger :1897b), p. 625. 1924b. Loveridge, p. ">. 1930,). Pronymnu amhU/iia sttihlmaiini Loveridge, p. 254. 1936. Pitman (part), p. 126, pi. viii, figs. 1-2; eol. pi. H, fig. 2 (omit Uganda records). 1937f. Loveridge, pp. 493, 496. 1938a. Pitman (part), p. 310. 193811. Pitman (part), pp. 39, 1201 21, 149, 310, pis. as above (reprint: omit Uganda records). 1939. 8omeren, p. 155. 1941. Moreau & Pakenham, p. 108. 1941a. Uthmoller, pp. 26, 43. 1941b. Uthmoller, p. 235. 1942e. Loveridge, p. 281. 1950. lonides, p. 101. 1951. Battersby, p. 829. 1951a. Loveridge, p. 192. 1953e. Loveridge, p. 265. 1955e. Loveridge, p. 184. 1937. rro.si/mna amhiciua hncafiii Uthmoller (not Boulenger :1897b), p. 110. Common Name. East African Shovel-snout (A.L.). Description. Rostral with angular horizontal edge; separated by the single bandlike internasal from the single, rarely di- vided,^"'' prefrontal; prefrontal very rarely entering orbit^**'; frontal large ; loreal as long as, or longer than, high ; preocular 1, rarely absent'"- ; postoculars 2, rarely l'''^ or absent''"; temporals 166 Divided in a juvenile ? (U.M.M.Z. 61221) found bottled at Morogoro, but without locality (Loveridge: l!»:^2c). 167 Entering in type of vu-ssci, also in occasional specimens at M.C.Z. io>^ Absent on one side only of an Nyatana i" (M.C.Z. 40.">fi2). and a Lnnilx) -^ (M.C.Z. 1S2IIH). 169 One on 9 sides only of 80 snakes examined. I'd Absent through fusion witli supraocular on right side of a Liwale ? (M.C.Z. .-.i:!ll). 162 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY rarely 1 + 1/'' usually 1 + 2, occasionally 1 + 3'"' or 2 + 2"''; upper labials 5-7, rarely the second and third or second, third and fourth, normally the third and fourth, entering: the orbit^"*; lower labials 7-8, the first 3, rarely 4, in contact with the single pair of sublinguals. Midbody scales smooth, in 15 rows; ventrals 129-155 ($ $ 129-144; 9 $ 144-155); anal entire; subcaudals 19-34 (9 9 19-28; $ $ 29-34). Color. Above, rostral and npper lip j-ellowish white, otherwise head and body gray (young), each scale edged with darker, or plumbeous (adult), uniform or with two dorsolateral rows of Avhite flecks. Below, usually white, uniform or with irregular dusky markings which may form a median line along the tail, rarely brown or black.^"'^ Battersby (1951), after examining 38 East African a. stuhl- )iHJnni (together with 2 Uganda snakes which I refer to a. amhigua), found that 26 had a more or less distinguishable dorsolateral series of white specks on either flank, but no trace of such markings in the more southwesterly material extending into Northern Rhodesia. He decided this speckling was uncor- 1 elated with age, sex or scale-counts. The seeming geographical ditference I believe to be fortuitous. Ske. Length of $ (M.C.Z. 40562) 238 (200 + 38) mm., from Xgatana, largest 9 (M.C.Z. 54082) 323 (293 + 30) mm. fi-oin Abercorn, surpassing even a 9 (B.M. 1902.2.12.91) of 280 (250 + 30) mm., from Mazoe. It would appear that there is an increase in size in the Rhodesias as one approaches the range of the somewhat larger typical race. Ronarks. Mocquard (1906c) erroneously synonymized his vassei with hocagii. Chabanaud (1916g) corrected Mocquard 's miscounts of A'entrals and subcaudals, but himself published a i"i I -|- 1 throujih fusion of the lower tciuiioral with tit'ili iipin'r hil)i;il nii Id't side o fa Mapenya ? (M.C.Z. 40000). 1-' 1 + 3 on 8 sides of 2 Liwale $ 9 (M.C.Z. 51395, 51397). 173 2 -I- 2 on right side of an Ahereorn c' (.M.C.Z. ."40Slt. 174 In the respective ratios of 2-2-7-4. 175 In oue (M.C.Z. 51390) of IS Liwale su;il. 177) Chilorhinophis gerardi tanganyikae Loveridge I933h. Chiloi^hinophis gerardi Loveridge (part), p. 262. 1934. Pitman, p. 298. 1937f. Loveridge, p. 496. 1937. Pitman (part), p. 330 (Nyamkolo only). 19381). Pitman (part), p. 183 (Nyamkolo only). 1947. Witte & Laurent (part), p. 54, figs. 43-45. 1951a. CJiilorhinnphis gerardi tanganyikae Loveridge, Bull. Mus. Comp. Zool., 106, p. 195: Nyamkolo, Lake Tanganyika, Northern Rhodesia. 1953. Witte, p. 263. Common Names. Western Tanganyika Two-headed Snake (English) ; kasimwanamatenga and kalamhanzila (Kirungu). Descripfion. Rostral twice as broad as deep, the portion visible from above as long as one third of its distance from the frontal ; nostril in a single nasal that is not fused with the first labial ; pre- frontals longer than broad, not entering the orbit ; frontal longer ISO Xot :^lo mm. says Dr. J. Eiselt, who kindly sexed the only known specinieii and corrected a number of errors in the original description. LOVERIDGE : AFRICAN SNAKE GENERA 171 than its distance from the rostral, 1^-11/2 times as long as broad, 11/4-2 times as broad as a supraocular, its length included li/o- 1% times in that of a parietal; preocular 1, moderate; eye small, its diameter two-thirds its distance from the mouth; postocular 1, moderate; a moderate to large posterior temporal; upper labials 4, second in contact with a prefrontal, third entering the orbit, fourth largest and in contact with a parietal ; lower labials 5, the first broadlv in contact with its fellow behind the mental, 3 labials in contact with the anterior sublinguals, which are divergent and longer than the posterior. Midbod}^ scales smooth, in 15 rows ; ventrals 308-375 ( S $ 308- 310; 9 375) ; anal divided ;subcaudals 23-26 (2 23; cJ 5 25-26). Color m alcohol. Above, head and neck black, the black de- scending on the sides of the neck suggesting a collar; labials yellow with black spots on rostral, beneath eye, and near gape ; prefrontals, parietals and nape each bearing two small white spots; back yellowish white with a dark brown vertebral stripe flanked by a dark brown dorsolateral stripe on the fifth row and adjacent halves of the fourth and sixth scale-rows, the three stripes extending almost to tij) of tail; the three lower lateral scale-rows uniform white ; posterior third of tail black with lighter flecking. Below, throat and belly immaculate yellowish white, a black crossband on posterior third of tail, whose tip is also black. For color in life see Loveridge (1933h). 8izc. Larger c^ (M.C.Z. 48434) 486 (460 + 26) mm., tail being included in total length oi $ S 17.8-18.6 times ; only known 9 (l.G. 15.824) 569 (545 + 24) mm., its tail being included in total length 23.6 times. Midbody diameters in cJ $ 4.5-5 mm., of 9 about 3.5 mm. Remarks. See Loveridge (1951a, p. 196). Defense. When wriggling along, this snake applies its head to the ground while carrying its tail, which bears a striking resem- blance to the head, upraised as if prepared to defend itself in a rearguard action. (Loveridge: see also Mertens, 1946b, pp. 31,71). Habitat. This slender species inhabits dry savanna both at r,ii.ii and Nyamkolo, where the type was found by men engaged in clearing weeds from the road leading to the London Missionary 172 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Society's station on the bluff overlooking the bay. At the time the countryside was very dry, the rains having ceased a month previously. Localities. Tanganyika Territory: T^jiji; also reported by natives as occurring at Kasanga. Northern Rhodesia: Nyamkolo. Belgian Congo: Lukonzohva, Lake Mweru (Moeru). Range. Tanganyika Territory (in the vicinity of Lake Tan- ganyika) and Northern Rhodesia, northwest to the Belgian Congo (in the vicinity of Lake Mweru). Chilorhinophis gerardi gerardi (Boulenger) 19131j. Apostolepis Gerardi Boulenger, Eevue Zool. Afr., 3, p. 103, fig. — : Kikondja, Lualaba District, Belgian Congo. 1915a. Boulenger, p. 214. 1925. Werner (1924), p. 149. 19281. Witte, p. 9. 1927. Parkcrophis gerardi Barbour & Aniaral (part), p. 25. 1927b. Parker, pp. 81, 82, 84, fig. 1. 1933m. Witte, p. 95. 1933h. Chilorhinophis gerardi 'Loveridge (part :pitations only), p. 262. 1937. Pitman (part), 4. pp. 329, 330. 1938b. Pitman, pp. 182, 183. 1943a. Witte & Laurent, p. 157. 1947g. Laurent, p. 39. 1947. Witte & Laurent (part : omit some citations), p. 54, figs. 43-45. 1950. Battersby, p. 417. 1953. Chilorhinophis gerardi gerardi Witte, p. 261, fig. 90. 1956. Laurent, p. 152. Common Name. Congo Two-headed Snake (English). Description. Rostral broader than deep, the portion visible from above as long as Yi^ to % its distance from the frontal ; nostril in a single nasal that is not fused with the first labial ; prefrontals longer than broad, not entering the orbit; frontal longer than its distance from the rostral, 1% to 2 times as long as broad, IVi to 1% times as broad as a supraocular, its length included 1V4 to 11/2 times in that of a parietal; preocular 1, moderate ; eye small, its diameter equal to its distance from the mouth; postocular 1, moderate; a large (rarely 2) upper tem- poral bordering the parietal ; upper labials 4, second in contact with a prefrontal, third (divided in one specimen) entering the LOVERIDGE : AFRICAN SNAKE GENERA 173 orbit, fourth largest and in contact with a parietal; lower labials 5, first broadly in contact with its fellow behind the mental, 3 labials in contact with the anterior sublinguals, which are diver- gent and longer than the posterior. Midbody scales smooth, in 15 rows; ventrals 263-348 {£ £ 263-294; 9 $ 274-348); anal divided; subcaudals 20-31 (9 9 20-26; S S 27-31). Colo)' in alcohol. Above, head and neck black, the black de- scending on the sides of the neck suggesting a collar; labials speckled with yellow below and behind the eye ; back yellowish with a black vertebral stripe and a dark dorsolateral stripe on the fifth and adjacent halves of the fourth and sixth scale-rows ; posterior third of tail pale blue irregularly blotched with black. Below, throat immaculate white, belly and anterior portion of tail yellowish, the posterior third bluish flecked with black and white. Size. Total length of type 6 (R.G.M.C. 1205) 315 (290 + 25) mm.; largest S (R.G.M.C. 11588) 424 (392 + 32) mm., tail being included in total length of S $ 12.3-15.2 times; largest 9 (R.G.M.C. 748) 513 (486 + 27) mm., tail included in total length of 9 9 17.2-22.2 times. The measurements, and consequently proportions, of the larg- est i furnished by Witte & Laurent (1943a, p. 158; 1947, p. 55) are amended in a letter from de Witte (25.xi.49). Dentition. Maxillary, palatine and pterygoid teeth of a Sinoia snake have been figured by Parker (1927b). Heniipenis. Not bifurcate; basal spines larger but relatively less numerous ; it is possible to distinguish about 4 series of large spines intermingled with spines only half as big ; in the more dilatable part of the organ there are about 40 spines arranged in 7 or 8 chevrons, being most minute and numerous towards the tip, which extends to the fifteenth caudal (Laurent :1956). Localities. Belgian Congo: Elisabethville ; Kankunda, Lua- pula River; Kaswabilenga, Lufiri River; Kikondja, Lualaba Dis- trict; Kisantu, Lower Congo District; Lofoi, Upper Katanga District ; Lukafu, Upper Katanga District ; Lukulu, Tanganyika District ; Mabwe, Lake Upemba ; Mukabe-Kasari, Lualaba Dis- trict ; Tshilunda Village, Lualaba District. Northern Rhodesia: 174 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY (N.M.S.R.— C.I. 43.1.2.: Miss Wilson coll. 1923. Seen A.L.). Southern Rhodesia: Liikosi ; Sinoia Lomagundi District. Range. Southern Belgian Congo, south to Southern Rhodesia. Chilorhinophis butleri Werner l!t()7. Chilorhinophis hutleri Werner, Akad. Anz. Wien, 44. p. 479 (brief notice), and 1908 (for 1907), Sitzb. Akad. Wiss. Wien, 116. y. 1881, pi. iii, figs. 8 ad : Mongalla, Sudan. 1910a?. Sternfeld, p. 33. 1915d. Boulenger, p. 654. 192-11). Loveridge, p. 7. 1925. Werner (1924), p. 148. 1937. Pitman, 4. p. 329, pi. xii, fig. 1 ; eol. pi. M, figs. lib. 1938b. Pitman, p. 181, pi. xii, fig. 1 ; col. pi. M, figs. lib. 1947. Witte & Laurent, p. 56, figs. 46-48. 1950. Battersby (part), pp. 416, 417 (omit synonymy). Common Names. Sudanese Two-headed Snake (English) ; Sudan Black-and-Yellow Burrowing Snake (Pitman). Description. Rostral one and a third times as broad as deep,^^' the portion visible from above about two-fifths^"*" its distance from the frontal ; nostril in a single nasal that is fused with the first labial, so borders the mouth ; prefrontals longer than broad, not entering the orbit; frontal shorter than the parietals; supra- ocular small ; preocular 1 ; postocular 1 or absent ; upper tem- poral large, bordering the parietal; upper labials 4 (if the naso- labial be considered the first), second in contact with a pre- frontal, third entering the orl)it, fourth largest and in contact with a parietal; lower labials ."), the first broadly in contact with its fellow behind the mental, 3 labials in contact with the anterior sublinguals, which are divergent and larger than the posterior ])air.^''" Midbody scales smooth, in 15 rows; ventrals 25G ; anal divided ; svibcaudals 33 pairs. 1*1 Werner's statement to the contrary iiutwitlistandiug, for the rostral of the lidlotype has been measured with vernier callipers by Dr. .T. Kiselt who (letter of 26.x. 54) finds the breadth to be 1.1 mm., the depth .85 mm. 1^- Eiselt's measuremi'nt for the portion of the rostral visibli> from above, is .5 mm., its distance from the frontal 1.25 mm. 1S3 Werner's fi.u'nre of the type, by .Tosef Fleisihmann, is in error. LOVERIDGE : AFRICAN SNAKE GENERA 175 Color. Above, head and neck black, the latter descending on the sides of the net-k suggestino- a collar; labials yellow; occiput un- spotted; back yellowish white with a black vertebral stripe; a black dorsolateral stripe on the fifth scale-row; a black bar across tail. Below, throat and belly immaculate yellowish; on tail the dorsal black cross bar is continued to form a complete ring on the middle third, a few black spots on tip. Size. Total length of holotype S (Vienna Mus.) 818 (283 + 30) mm.,^''* the tail being included in total length 10.4 times. Midbody diameter 4 mm. Remaiks. The type was brought in dead by natives on March 30, ]n05, and named after A. E. Butler, the then Game Warden of the Sudan. Being still the only known specimen, the fore- going description is based on the original one, together with in- formation gleaned from Fleischmann's drawing, besides correc- tions and amendments supplied me by Dr. J. Eiselt after a care- ful re-examination of the type. Fleischmann's misleading figures form the basis of all subsequent illustrations. In assuming that Sternf eld's statement (1910a) "bis 32 cm." represented the measurement of an Amani snake, Pitman (1938b, p. 138) erred, for in round figures this was the length given by Werner for the holotype. Moreover, Sternf eld's description is also based on Werner's, and is not that of the Amani snake he referred to hntleri. This specimen has been missing from the Berlin Museum since I first enquired for it 30 years ago. I sug- gest it may have been a misidentified A pa rail act us tverneri, a common Amani reptile with a color pattei-n not unlike that of Chilorhinophis hutleri. Habitat. Presumably arid upland savanna, possibly in the vicinity of the Nile. Localities. Sudan: Mongalla. Range. Sudan {unknown from Uganda and Tanganyika Terri- tory). Chilorhinophis carpenteri liwalensis Loveridge 1950. CMlorhinopMs carpenteri Battersby (not Parker), p. 416. 1950. lonides, p. 105. i*** Sexed and carefully remeasured for lue by Dr. J. Eiselt (14.ix.o4). 176 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1951a. ChiUyrhinophis carpentcri liwalenMs Loveridge, Bull. Mus. Comp. Zool., 106. p. 19(i: Liwale, Southern Province, Tanfrnnyika Terri- tory. I9r>5e. Loveridge, p. 188. Common Names. Eastern Tanganyika Two-headed Snake (English) ; kitandamha, i.e. the one found among ndandamba beans (Ngindo, but not specific). Description. Rostral nearly twice as broad as deep, the portion visible from above as long as about two times its distance from the frontal ; nostril in a single nasal that is fused with the first labial, so borders the mouth ; prefrontals longer than broad, entering or not entering the orbit; frontal shorter than its dis- tance from the rostral, slightly longer than broad, about twice as broad as a supraocular, much shorter than a parietal ; pre- ocular 1, moderate, minute, or absent ; eye small, its diameter about two-thirds its distance from the mouth ; postocular 1 ; upper temporal large, bordering a parietal ; upper labials 4 (if the naso- labial be considered the first), second in contact with a prefrontal, third entering the orbit, fourth largest and broadly in contact with a parietal ; lower labials 5, the first broadly in contact with its fellow behind the mental, 3, rarely 4, labials in contact with the anterior sublinguals, which are divergent and longer than the posterior. Midbody scales smooth, in 15 rows; ventrals 216-270 {S S 217-238; 9 9 216-270^^^ ; anal divided; subcaudals 18-30 ( 9 9 18-24; i o' 25-30). Color in alcohol. Above, head and neck black, the black de- scending on the sides of the neck suggesting a collar ; labials yel- low ; temporal with or without a white spot ; back yellowish white, a broad brown vertebral stripe on vertebral scale and adjacent halves ; a dark dorsolateral stripe on adjacent halves of the fifth and sixth scale-rows, very rarely (M.C.Z. 50090 only) the lower lateral scales may be faintly edged with brown ; posterior half of tail black, except for some blue-gray mottling about the tip. Below, chin with or without dusky markings, otherwise throat 1S5 288 (verified by A.L.) in a Kihva ? (Brit. Mus.) with IS siihcaudals. unless this represents an undescrihed sul)speeies. Its iuclusiiui would irive a ventral ranpe of T.'i for 9 r. " g. ffernrdi. LOVERIDGE : AFRICAN SNAKE GENERA 177 and belly immaculate yellowish white; proximal portion of tail white, separated by a black bar from a blue-gray area near the tip. In life, according to lonides, the basic color ranges from light yellow to khaki. Size. Largest $ (M.C.Z. 51327) 800 (274 + 26) mm., tail being included in total length of <5 <5 10.5-12.9 times; type 9 (M.C.Z. 50076) 335 (315 + 20) mm., but surpassed by another 9 (M.C.Z. 52832) of 360 (339 + 21) mm.,''*' tail being included in total length of 5 9 14.4-19.6 times. Midbody diameter of 9 type, 3.5 mm. Remarks. See Loveridge (1951a, p. 198). Dentition. Pterygoid teeth on prepared skull, 3 or 4. Diet. \Yorm-lizards {Amphishaena ionidesii) were present in the stomachs of two Li wale snakes. Habits. Chilorhinophis, like adult blind-snakes, usually re- mains underground except following heavy rain. Localities. Tanganyika Territory: Southern Province : Kilwa'*^' ; Liwale ; Masasi District ; Ruponda, Lindi District ; Tunduru Dis- trict. Range. Southeastern Tanganyika Territory. Chilorhinophis carpenteri carpenteri (Parker) 1919. ApostulcpLs gcrardi Carpenter (not Bonlenger), p. 496. 1925. Carpenter, p. 132, pi. vii. 1927. Parlceropliis gerardi Barbour & Amaral (part), p. 25. 1927. Parlcerophis carpenteri Parker, Ann. Mag, Nat. Hist., (9) 20, p. 85, fig. 3: "Anquabe," i.e. Ancuabe, Mozambique. 1937. Chilorhmopliis carpenteri Pitman, 4, pp. 329-331. 19381). Pitman, pp. 182-184 (reprint). 1947. Witte & Laurent, p. 57, figs. 49-50. Common Name. Mozambique Two-headed Snake (English). Description. Rostral much broader than deep, the portion vis- ible from above as long as about ly^ times in its distance from the frontal ; nostril in a single nasal which is fused with the first 186 Also exceeded by a Kilwa ? (Brit. Mus.) measuring 410 (390..5 + ]'.»..")) iiini. which, however, may represent an uiiilescril)ed sulispecies. 187 Unless this Kilwa ? proves to he subspecifically distinct. Kilwa is 125 miles northeast of Liwale. 178 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY labial, so borders the mouth; prefrontals longer than broad, entering the orbit ; frontal shorter than its distance from the rostral, slightly longer than broad, twice as broad as a supra- ocular, much shorter than a parietal; preocular 1, minute; eye small, its diameter equal to its distance from the mouth ; post- ocular 1, moderate; upper temporal large, bordering a parietal; upi^er labials 4 (if the nasolabial be considered the first), second in contact with a prefrontal, third entering the orbit, fourth largest and broadly in contact with a parietal ; lower labials 5, the first broadly in contact with its fellow behind the mental, 3 labials in contact with the anterior sublinguals, which are small and subequal to the posterior. Midbody scales smooth, in 15 rows; ventrals 269 (in only known 9 ) ; anal divided; subcaudals 21 (in 9 ). Color in alcohol. Above, head and neck black, the black de- scending on the sides of the neck suggesting a collar ; labials yellow ; occiput bearing two small white spots ; back yellowish white with a ladder-like, dark brown, vertebral stripe ; a dark dorsolateral stripe on the adjacent halves of the fifth and sixth scale-rows, a narrow brown lateral stripe between the third and fourth scale-rows, lower lateral scales faintly edged with brown ; posterior half of tail black. Below, throat and belly immaculate yellowish white ; a black bar across posterior half of tail sep- arates oft* a light area near the tip. Size. Total length of holotype 9 (B.M. 1918.11.20.1) 280 (264 -|- 16) mm.; tail length included in total length of 9 17.5 times. Midbody diameter 3 mm. Dentition. Pterygoid teeth absent in type {fide Parker). Defense. The defense of this genus was illustrated by the fact that the killer of the type had been induced to strike at the tail, so closely did it resemble the head and act as a directive mark-, writes Hale Carpenter, who found it lying dead in the road through the British militar}^ camp behind Port Amelia, August, 1918. Localities. Mozambique: Ancuabe (Anquabe; Ankwabe), which is 13°02' S., 39° 55' E. Range. Mozambique. LOVERIDGE : AFRICAN SNAKE GENERA 179 BIBLIOGRAPHY ^'*'' A total of 345 works (1803-1956) l)y 126 authors in which have been found references to the species of snakes discussed in the preceding pages. No attempt lias been made to complete citations appearing prior to 1880. Aders, W. M. 1920. In ¥. B. Pearce, Zanzibar: The Island Metropolis of Eastern Africa (London), pp. 326-343. Agassiz, Louis 1848. Nomcnclator Zoologicus: Index Universalis (Solofuri "1846"), pp. 1-393. Amaral, a. do See T. Barbour and A. do Amaral Andersson, L. G. 1901. Bihang till K. Svenska Vet.-Akad. Handl., 27. No. 5, p)i. 1-26, pis. i-ii. 1903b. I7i L. A. Jagerskiold, Results of the Swedish Zoological Expedi- tion to Egypt and the White Nile 1901 (Upsala 1904), 1. No. 4, pp. 1-12 (reprints appeared in 1903). 1916. Goteborgs Kungl. Vetensk.-Vitterh.-Samhalles Handl., (4) 17. No. 5, pp. 1-41. 1937. Arkiv Zool., 29A, No. 16, pp. 1-28, figs. 1-8. See also E. Lonnberg and L. G. Andersson Angel, Fernand 1920a. Bull. Mus. Hist. Nat. (Paris), 26. pp. 197-199, figs. 1-3. 1921a. Bull. Mus. Hist. Nat. (Paris), 27. pp. 42-44, figs. 1-3. 1922a. Bull. Mus. Hist. Nat. (Paris), 28. pp. 39-41. 1923d. In Mission Rohan-Chabot. Angola et Rhodesia (1912-1914) (Paris), 4. pp. 157-169, figs. 1-12, pi. — , figs. 1-2. 1925a. In Voyage de Ch. Alluaud et R. Jeannel en Afrique Orientale (1911-1912). Resultats scientifiques. Vertebrata. (Paris), 2. pp. 1-63, figs. 1-5, pis. i-iii. 1933b. Bull. Mus. Hist. Nat. (Paris), (2) 5. pp. 68-69. 1933f. Les Serpents de 1 'Afrique Oecidentale francaise. (Paris), pp. 1-246, figs. 1-83. 188 Who r<' a date is followed l»v a letter of the alphabet it indicates that durin;: the year cited the author in question published more than one paper on African herpetologv The letter has chronological sijiniticance in a more comprehensive bibliography of African Ilerpetology ( 1SS0-1'J."j8) which it is hoped may be pub- lished in the not too-distant future. 180 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Angel, F., J. Guide, M. Lamotte, and R. Roy 1954. Mem. Inst. Francais Afrique Noire, No. 40, pp. 381-402, fig. 1. Angel, F. and H. Lhote 1938. Bull. Com. Etudes Hist. Sci. Afr. Oecid. Franc, 21, pp. 345-384. Aylmer, G. 1922. Sierra Leone Studies (Freetown;, 5. pp. 7-37. Barbour, T. and A. do Amaral 1927. Bull. Antivonin Inst. Amer., 1, pp. 2.')-27. Barbour, T. and A. Loveridge 1928c. Mem. Mus. Comp. Zool., 50. pp. 87-265, pis. i-iv. 1930a. In Richard Strong, Report of the Harvard-African Expedition upon the African Republic of Liberia and the Belgian Congo. (Cambridge, Mass.), 2, pp. 769-786, pis. i-ii. Bateman, G. C. 1897. The Vivarium (London), pp. 1-424, figs. 1-98, pi. — . Battersby, J. C. 1950. Ann. Mag. Nat. Hist., (12) 3. pp. 413-417, fig. 1951. Ann. Mag. Nat. Hist., (12) 4, pp. 828-829. 1954. Ann. Mag. Nat. Hist., (12) 7. pp. 241-248. Bedrl\ga, J. V. 1892. O. Inst. Rev. Sci. Litt. (Coimbra), 39, pi». 498-507, 042-648, 736-742, 814-820, 901-907. BlANCONI, J. J. 1859. Specimina zoologiea mosambicana (Bononiae), pp. 1-282, 1862. pis. i-xvii. BiBRON, G. See A. M. C. Dumeril and G. Bibron Blair, J. A. and J. Graham 1954. Biochem. Journ., 56, pp. 286-287, figs. 1-3. Blom-Bjorner, Sidney 1946. Journ. E. Africa Uganda Nat. Hist. Soc, 18. (1945), p. 159. LOVERIDGE : AFRICAN SNAKE GENERA 181 BOCAGE, J. V. B. DU 1866a. Joni. Sei. Lisboa, 1. pp. 37-56. 1866b. Joni. Sci. Lisboa, 1. pp. 57-78, pi. i. 1867a. Jorn. Sci. Lisboa, 1. pp. 217-228. 1872. Join. Sci. Lisboa, 4, pp. 72-82, fig. 1873b. Jorn. Sci. Lisboa, 4. pp. 209-227. 1879a. Join. Sci. Lisboa, 7. pp. 85-96. 1882a. Jorn. Sci. Lisboa, 8. pp. 286-290. 1882b. Jorn. Sci. Lisboa, 8. pp. 299-304. 1882c. .Jorn. Sci. Lisboa, 9. pp. 1-19, figs. 1-6. 1886a. Jorn. Sci. Lisboa, 11. pp. 65-70. 1887a. Jorn. Sei. Lisboa, 11. pp. 177-211. 1892a. Jorn. Sci. Lisboa, (2) 2. pp. 179-184. 1893a. Jorn. Sci. Lisboa, (2) 3. pp. 45-46. 1893b. Jorn. Sci. Lisboa, (2) 3. pp. 47-48. 1893d. Jorn. Sci. Lisboa, (2) 3, pp. 141-144. 1895a. Herpetologie cl 'Angola et dn Congo (Lisboa), pp. xx + 203, pis. i-xix. 1895c. Jorn. Sci. Lisboa, (2) 4, pp. 1-15. 1896a. Jorn. Sci. Lisboa, (2)4. pp. 65-104, pis. i-ii. 1896b. Jorn. Sci. Lisboa, (2) 4. pp. 105-114. 1896c. Jorn. Sci. Lisboa, (2) 4. pp. 115-120. 1896e. Jorn. Sei. Lisboa, (2) 4. pp. 176-178. 1897a. Jorn. Sei. Lisboa, (2) 4. pp. 187-206. 1903a. Jorn. Sei. Lisboa, (2) 7. pp. 25-59. 1905. Jorn. Sci. Lisboa, (2) 7, pp. 65-96. BOETTGER, OSKAR 1881b. Abhand. Senckenberg. Naturf. Ges., 12, pp. 393-418, pi. i. 1887b. Ber. Senckenberg. Naturf. Ges., pp. 135-173, pi. v. 1888a. Ber. Senckenberg. Naturf. Ges., pp. 3-108, pis. i-ii. 1889. Ber. Senckenberg. Naturf. Ges., pp 267-316. 1893b. Zool. Anz., 16. pp. 113-119 + 129-132. 1898. Katalog der Reptilien-Sainmlung ini Museum der Senckenberg- ischen Naturforschenden Gesellschaft in Frankfurt am Main. II. Schlangen (Frankfurt a. M.), ix + 160 pp. 1913. In A. Voeltzkow, 1908-1917, Eeise in Ostafrika (Stuttgart), 3, pp. 269-376, pis. xxiii-xxx. BOGERT, C. M. 1940. Bull. Amer. Mus. Nat. Hist., 77. pp. 1-107, figs. 1-18, pi. i. 19421). Amer. Mus. Novit., No. 1178, pp. 1-5. 182 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY BOULENGER, (j. A. 1887h. Zoologist (London), (3) 11, pp. 171-18i;. 1890. Fnuna British India, Rcpt. & Batr., xviii + 541 pp., fi^s. 1-14::. 1891a. Proc. Zool. Soc. London, pp. 305-309. 1892a. In W. L. Distant, A Naturalist in the Transvaal (London), pp. 174-176, fig. — . 1893b. Catalogue of Snakes in the British Museum (Natural History) (London), 1, xiii + 448 pp., figs. 1-26, pis. i-xxviii. 1894a. Catalogue of Snakes in the British Museum (Natural History) (London), 2, xi -f 382 pp., figs. 1-25, pis. i-xx. 1895f. Ann. Mag. Nat. Hist., (6) 16. pp. 32-34. 1895h. Ann. Mag. Nat. Hist., (6) 16, pp. 171-173. 1896a. Ann. Mus. Civ. Stor. Nat. Genova, (2) 16. pp. 545-554. 1896b. Ann. Mus. Civ. Stor. Nat. Genova, (2) 17. pp. 1-14, pi. i. 1896c. Ann. Mus. Civ. Stor. Nat. Genova, (2) 17. pp. 15-23. 1896d. Catalogue of Snakes in the British Museum (Natural History) (London), 3. xiv + 727 pp., figs. 1-37, pis. i-xxv. 1896e. Proe. Zool. Soc. London, pp. 212-217, pis. vii-viii. 1897b. Ann. Mag. Nat. Hist., (6) 19. pp. 276-281, fig. — . 1897d. Ann. Mag. Nat. Hist., (6) 20. pp. 374-376. 1897e. Proc. Zool. Soc. London, pp. 800-803, pi. xlvi. 1897g. Ann. Mus. Civ. Stor. Nat. Genova, (2) 17. pp. 275-280. 1898a. Ann. Mus. Civ. Stor. Nat. Genova, (2) 18, pp. 715-723, pis. ix-x. 1900b. Proc. Zool. Soc. London, pp. 433-456, figs. 1-2, pis. xxvii-xxxii. 1901g. Ann. Mus. Congo, Zool. (1) 2. pp. 1-14, pis. i-v. 1902b. Proc. Zool. Soc. London, 2. pp. 13-18, pis. ii-iv. 1902d. In H. H. Johnston, The Uganda Protectorate (London), 1. pp. 445-447. 1905c. Ann. Mag. Nat. Hist., (7) 16. pp. 105-115, pi. iv. 1905f. Mem. Real Soc. Esp. Hist. Nat. (Madrid), 1. pp. 183-186. 1905h. Proc. Zool. Soc. London, pp. 248-255. 19061. Ann. Mus. Civ. Stor. Nat. Genova, (3) 2, pp. 196-216, figs. 1-9. 1907a. Mem. Proc. Lit. Philos. Soc. Manchester, 51. pp. 1-12. 1907J. Proc. Zool. Soc. London, pp. 478-487, figs. 140-141, pis. xxi-xxii. 1908b. Ann. Natal Mus., 1, pp. 219-235, figs. 1-3, pis. xxxv-xxxvi. 1908c. Ann. Mus. Civ. Stor. Nat. Genova, (3) 4. pp. 5-7, fig. — . 1909a. Ann. Mus. Civ. Stor. Nat. Genova, (3) 4. p. 193. 1909b. Ann. Mus. Civ. Stor. Nat. Genova, (3) 4. pp. 302-307. 1909d. Ann. Mus. Civ. Stor. Nat. Genova, (3) 4. pp. 310-311, fig. — . 1909f. Proc. Zool. Soc. London, pp. 944-952, figs. 295-299. 1909g. Trans. Zool. Soc. London, 19. pp. 237-247, pis. viii-ix. 1910b. Ann. S. African Mus., 5, pp. 455-538. LOVERIDGE : AFRICAN SNAKE GENERA 183 1911c. Aim. Mus. Civ. Stor. Nat. Gonova, (3) 5, pp. 161-169. 1912b. Ann. Mus. Civ. Stor. Nat. Genova, (?-) 5, pp. 329-.332. 1912c. In P. A. Talbot, In the Rliadow of the Bush (London & New York), p. 470. 1913b. Eevue Zool. Afr., 3, pp. 103-105, figs. — . 1915a. Proc. Zool. Soc. London, pp. 193-223, figs. 1-2. 1915c. Proc. Zool. Soc. London, pp. 611-640, figs. 1-3. 1915d. Proc. Zool. Soc. London, pp. 641-658. 1919a. Proc. Zool. Soc. London (Feb. 1920), pp. 267-298, figs. 1-3. 1919g. Revue Zool. Afr., 7, pp. 1-29. Brien, Paul 1939. Ann. Soc. Roy. Zool. Belgique, 69, pp. 119-137, map. BucHHOLZ, R. and W. K. H. Peters 1875a. Monatsb. Akad. Wiss. Berlin, pp. 196-212, pis. i-iii. BUTTIKOFER, JOHANNES 1890. Reisebilder aus Liberia. 2, Die Bewohner Liberia's Thierwelt (Leiden), pp. 435-447 and 478, pis. xix-xxxii. Calabresi, Enrica 1916. Monitore Zool. Ital. (Firenze), 27. pp. 33-45, pi. ii. 1918. Monitore Zool. Ital. (Firenze), 29, pp. 122-124. 1925a. Atti Soc. Ital. Sci. Nat. (Milano), 64. pp. 100-109. 1925b. Atti Soc. Ital. Sci. Nat. (Milano), 64, pp. 119-125, fig. — . 1927. Atti Soc. Ital. Sci. Nat. (Milano), 66. pp. 14-60, pi. i. Cansdale, G. S. 1948d. Nigerian Field, 13, pp. 43-50, photos. Carpenter, G. D. H. 1919. Journ. E. Africa Uganda Nat. Hist. Soc, 15. p. 496. 1925. A Naturalist in East Africa (Oxford), pp. 1-187, figs. — , map, pis. i-viii. Chabanaud, Paul 1916f. Bull. Mus. Hist. Nat. (Paris), 22. pp. 362-382, figs. 1-23. 1916g. Bull. Mus. Hist. Nat. (Paris), 22. pp. 434-440, figs. 1-2. 1917b. Bull. Mus. Hist. Nat. (Paris), 23. pp. 7-14, figs. 1-3. 1917e. Bull. Mus. Hist. Nat. (Paris), 23, pp. 219-225, figs. 1-9. 1918b. Bull. Mus. Hist. Nat. (Paris), 24, pp. 160-166. 1919d. Bull. Mus. Hist. Nat. (Paris), 25. pp. 567-568. 1921a. Bull. Com. Etudes Hist. Sei. Occ. Franc. (Paris), pp. 445-472, map, pis. i-iv. 1921d. Bull. Mus. Hist. Nat. (Paris), 27. pp. 519-525. 184 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Chubb, E. C. 1909a. Proc. Zool. Soc. London, pp. o90o97. 1909b. Ehodesia Mus. Bulawayo, 8th Ann. Report, pp. 34-36. Cope, E. D. 1860. Proc. Acad. Nat. Sci. Philadelphia, pp. 241-266, 553-.^66. 1863. Amer. Journ. Sci. Arts, (2) 35. pp. 45.5-458. CORKILL, X. L. 1935a. Sudan Govt. Mus. (N.H.) Publ. No. 3 (Khartoum), pp. 1-40, map. COTT, H. B. 1928. Proc. Zool. Soe. London, pp. 923-961, pis. i-iv. 1935. Proc. Zool. Soc. London, pp. 963-975, pi. i. CUNHA, J. G. DE B. E. 1935. Mem. Estudos Mus. Zool. Univ. Coinibra, (1) No. 83, pp. 1-16. 1937. Compte Eendus XII Congres Internat. Zool., Lisbonne, 1935, 3, pp. 1777-1786. Daudin, F. M. 1803c. Histoire naturclle, generale et particuliere, des Reptiles (Paris), 7, pp. 1-436, pis. Ixxxi-xcii. Dekeyser, p. L. and A. Villiers 1954. Bull. Inst. Franc. Afrique Noire, 16 (A), pp. 957-970. DoLLO, Louis 1886. Bull. Mus. Roy. Hist. Nat. Belgique, 4, pp. 1.51-160, figs. — . Dumeril, a. H. a. 1860. Arch Mus. Hist. Nat. (Paris), 10. pp. 137-240, pis. xiii-xix (Sometimes cited as 1859 or 1861, but given as 1860 in Roy. Soc. Catalogue). Dumeril, A. M. C. and G. Bibron 1854a. Erpctologie generale ou Histoire naturelle complete des Reptiles (Paris), 7a. xvi -\- 780 pp., pis. Ixiii-lxx. Falk, Kurt 1921. Wochensch. Aquar.-Terrarienkunde (Braunschweig), 18, p. 177. 1922. Wochensch. Aquar.-Terrarienkunde (Braunschweig), 19, pp. 160- 164, figs. — . LOVERIDGE : AFRICAN SNAKE GENERA 185 Fantham, H. B. and A. Porter 1950. Pio(>. Zool. Soc. London, 120. pp. 599-646, pis. i-iii. Febreira, J. B. 1898. .Torn. Sei. Lisboa, (2)5. pp. 240-246. 1900. Jorn. Sci. Lisboa, (2) 6. pp. 49-53. 1902. Jorn. Sci. Lisboa, (2) 6. pp. 231-233. 1903a. Jorn. Sci. Lisboa, (2) 7, pp. 9-16. 1904. Jorn. Sci. Lisboa, (2) 7, pp. 111-117. 1905. Jorn. Sei. Lisboa, (2)7. pp. 159-171, pi. — . Fischer, J. G. 1881. Abhand. naturwiss. "Verein Bremen, 7. pp. 225-238, pis. xiv-xvii. 1884a. Jahrb. Hamburg. Wiss. Anst., 1. pp. 33-39, pi. iii. FiTZINGER, L. J. F. J. DE 1826. Xeue Classification der Eeptilieu . . . Wien, viii -f 66 pp., pi. i. 1843. Systema Eeptilium . . . Fasciculus primus . . . (Vindobonae), pp. 1-106. 1861. Sitzungsb. K. Akad. Wiss. Wien, 42. pp. 383-416. FiTZSlMONS, F. W. 1912. The Snakes of South Africa; their Venom and the Treatment of Snake Bite (Cape Town, ed. 2; ed. 3 appeared in 1919) xvi + 547, figs. 1-162 + A-Z, col. frontis. FiTZSlMONS, V. F. 1932. Ann. Transvaal AIus., 15. pp. 519-550. 1935b. Ann. Transvaal Mus., 16. pp. 295-397, figs. 1-30, pis. x-xi. 1937a. Ann. Transvaal Mus., 17. pp. 259-274, pi. x. 1938. Ann. Transvaal Mus., 19. pp. 153-209, map, pis. ii-iv. 1939b. Ann. Transvaal Mus., 20. pp. 17-46. 1946a. Ann. Transvaal Mus., 20. pp. 351-377, pis. xiv-xvii. 1946b. Ann. Transvaal Mus., 20. pp. 379-393. 1948a. Ann. Transvaal Mus., 21. pp. 73-80, figs. 1-3, pi. i. Flower, S. S. 1900. Proc. Zool. Soc. London, pp. 967-968. 1925b. Proc. Zool. Soc. London, pp. 911-981. 1929. In List of the Vertebrated Animals exhibited in the Gardens of the Zoological Society of London, 1828-1927. (London), 3. pp. 1-272. 1933. Proc. Zool. Soc. London, pp. 745-851, map. 186 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Gendre, E. 1909. Coniptes Rendus Sof. Linn. Bordeaux, 63, pp. cv-cvi. GouGH, L. H. 1903. Zool. Jahrb. Syst., 17, pp. 457-468. 1908. Ann. Transvaal Mus., 1. pp. 17-45. Graham, J. See J. A. Blair and J. Graham Gray, J. E. 1849. Catalogue of the Specimens of Snakes in the Collection of the British Museum (London), xvi + 125 pp. Greef, S. R. 1885. Sitzungsb. Ges. Bedford. Naturw. Marburg, for 1884, pp. 41-80. GuiBE, Jean. See F. Angel and J. Guibe GuNTHER, Albert 1858c. Catalogue of the Colubrine Snakes in the Collection of the British Museum (London), xvi + 281 pp. 1860a. Proc. Zool. Soc. London, pp. 427-430, figs. — . 1862b. Ann. Mag. Nat. Hist., (3) 9, pp. 52-59, 124-132. pis. ix-x. 1863. Proc. Zool. Soc. London, pp. 16-17, fig. — . 1863c. Ann. Mag. Nat. Hist., (3) 11, pp. 283-287. 1863d. Ann. Mag. Nat. Hist., (3) 12, pp. 348-365, pis. v-vi. 1866a. Ann. Mag. Nat. Hist., (3) 18, pp. 24-31, pis. vi-vii. 1868a. Ann. Mag. Nat. Hist., (4) 1, pp. 413-429. 1869b. Zool. Record, 6, pp. 105-122. 1872a. Ann. Mag. Nat. Hist., (4) 9, pp. 13-37, pis. iii-vii. 1888a. Proc. Zool. Soc. London, pp. 50-51. 1888b. Ann. Mag. Nat. Hist. (6) 1. pp. 332-335, pis. xviii-xix. 1893. Proc. Zool. Soc. London for 1892, pp. 555-558, pis. xxxiii-xxxv. 1894a. Proc. Zool. Soc. London for 1893, pp. 616-628, pis. liii-lvii. 1894b. Proc. Zool. Soc. London for 1893, pp. 628-632, pi. Iviii. 1894c. Proc. Zool. Soc. London, pp. 84-88, pi. viii. 1895. Ann. Mag. Nat. Hist., (6) 15. pp. 523-529, pi. xxi. 1896. Ann. Mag. Nat. Hist., (6) 17, pp. 261-285. Hallowell, Edward 1857. Proc. Acad. Nat. Sci. Philadelphia, pp. 48-72. HeCHT, GiJNTHER 1929. Zool. Anz., 81, pp. 329-335. LOVERIDGE : AFRICAN SNAKE GENERA 187 Hewitt, John 1910a. Ann. Transvaal Mus., 2. pp. 50-71. 1910b. Ann. Transvaal Mus., 2. p. 73. 1912. Rec. Albany Mus., 2. pp. 264-285. 1914a. S. Afr. Journ. Sci., 10. pp. 238-253, maps. 1937e. In A Guide to the Vertebrate Fauna of the Eastern Cape Prov- ince, S. Africa (Grahamstown), pp. 1-118 and GIT, pis. i-xxviii. See also P. A. Mothuen and J. Hewitt. Hevi^itt, J. and J. H. Power 1913. Trans. Roy. Soc. S. Africa, 3. pp. 147-176. HOBLEY, C. W. 1912. Journ. E. Afr. Uganda Nat. Hist. Soc, No. 5, pp. 43-56, figs. — . lONIDES, C. J. P. 1947. Journ. E. Africa Nat. Hist. Soc. for 1946, 19, p. 70. 1948. In M. S. Moore, Ann. Rep. Game Pres. Dept. Tanganyika Terr, for 1947, pp. 10-11. 1950. Tanganyika Notes and Records, No. 29, pp. 98-108, figs. 1-8. 19-T2a. In G. H. Swynnerton, Ann. Rep. Game Dept. Tanganyika Terr, for 1950, pp. 9-10. Jan, Georg 1862. Arch. Zool. Anat. Fisiol. (Modena), 2, pp. 1-76, pis. v, vii, xvii, xviii. 1863a. Arch. Zool. Anat. Fisiol. (Modena), 2. pp. 213-330. 1866. Iconographie generate dcs Ophidiens (Milano), No. 16, pis. i-vi. 1869. Loc. cit., No. 32, pis. i-vi. 1876. Loc. dt., No. 48, pis. i-vi. 1881. Loc. Git., No. 50, pis. i-vi. Jeude, T. W. van Lidth de 1895. Notes Leyden Mus., 16. pp. 227-230. Johnston, H. H. 1897. British Central Africa (London), xx + 544 pp., 220 illus., 6 maps. 1898. Second edition of 1897 with some changes in pagination. Kingdon, J. 1948. Bull. E. Africa Nat. Hist. Soc, No. 7, p. 10. KiiAPTOCZ, Adalbert 1913. Zool. Jahrb. Syst., 34. pp. 279-290. 188 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY LacEPEDE, B. G. E. DE 1789. Histoire naturelle des Quadrupedes oviparcs et des Serpents (Paris), 2. pp. 1-527, pis. i-xxii. Lamotte, Maxime. See F. Angel and M. Lamotte Lampe, Eduard 1902. Jahvb. Nassau. Ver. Natuik. (Wiesbaden), 55, pp. 1-66. 1911. .Jnhrb. Nassau. Ver. Naturk. (Wiesbaden), 64, pp. 137-236. Laurent, Raymond 1947e. BulL Mus. Roy. Hist. Nat. Belgique, 23. No. 16, i)p. 1-12. 1947s. Bull. Cercle Zool. Congo, 18. pp. 39-40. 1950b. Revue Zool. Bot. Af r., 43. pp. 349-352. 1950f. Serv. Cult. Comp. Diam. Angola, Museu Dondo No. 10, pp. 7-17. 1952b. Revue Zool. Bot. Afr., 45. pp. 198-203. 1953a. Bull. Cercle Zool. Congolais, 21. pp. 21-29. 19541). Serv. Cult. Comp. Diam. Angola, Museu Dondo No. 23, pp. 35-84, figs. 1-23. 1954d. Bull. Soe. Zool. France, 79. pp. 290-310. 1956. Ann. Mus. Roy. Congo Beige, Sci. Zool., 48. pp. 1-390, figs. 1-50, pis. i-xxi. Leach, W. E. 1819. In T. E. Bowdicli, Mission from Cape Coast Castle to Ashantec . . . (London), Appendix IV, p. 494. Leeson, Frank 1945. Key to Gold Coast Snakes having Colubride-Type Head-Shields (Accra), pp. 1-6. 1950. Identification of Snakes of the Gold Coast (London), x + 142 pp., tigs. 1-65, pis. i-xxxiii. Lepri, Giuseppe 1911. Boll. Soc. Zool. Ital. Roma for 1910, (2) 11. pp. 317-328. Lhote, H. See F. Angel and H. Lhote Lichtenstein, M. H. C. 1823. Verzeichniss der Dubletten des Zoologischen Museums der . . . Universitilt . . . (Berlin), x -f 118, pi. — . LiNDHOLM, W. A. 1902. In E. Lampe, Jahrb. Nassau. Ver. Naturk. (Wiesbaden), 55, pp. 1-66. LOVERIDGE : AFRICAN SNAKE GENERA 189 LONNBERG, EiNAR 1907, In Y. Sjostedt, Wissenschaftliche Ergebnisse der Schwedischen Zoologischcii Expedition nach deni Kiliinaiidjaro, dcni Meru imd den umgebenden Massaisteppen 1905-1906 (Uppsala), No. 4, pp. 1-28, pi. i. 1910. Arkiv Zool., 7. No. 8, pp. 1-8. 1911. K. Svenska. Yetensk.-Akad. Hand!. (Stockholm), 47, No. G, pp. 1-42, pis. i-ii. LoNNBERG, E. and L. G. Andersson 1913. Arkiv Zool., 8. No. 20, pp. 1-6. liOVEiRiDGE, Arthur 1916a. Jouni. E. Africa Uganda Nat. Hist. Soc, 5. pp. 77-87. 1916b. Journ. E. Africa Uganda Nat. Hist. Soc, 5. pp. 109-124. 1917b. Journ. E. Africa Uganda Nat. Hist. Soc, 6. pp. 170-185. 1918a. Joiun. E. Africa Uganda Nat. Hist. Soc, No. 13, pp. 315-338. 1923b. Journ. Tanganyika European Civil Serv. Assoc, pp. 10-28. 1923e. Proc. Zool. Soc. London, pp. 871-897. 1924b. Journ. E. Africa Uganda Nat. Hist. Soc, Spec Suppl. No. 3, pp. 1-16. 192.')a. Proc. Zool. Soc London, pp. 71-74. 1928d. Proc U. S. Nat. Mus., 73. Art. 17, pp. 1-69, pis. i-iv. 1928g. Bull. Antivenin Inst. Amer., 2, pp. 32-41, figs. 5-6. 1928J. Bull. Antivenin Inst. Amer., 2. pp. 72-76. 1929h. Bull. U. S. Nat. Mus., No. 151, pp. 1-135, pi. 1. 1933h. Bull. Mus. Conip. Zool., 74. pp. 197-416, pis. i-iii. 1935c. Bull. Mus. Comp. Zool., 79. pp. 1-19. 1936h. Field Mus. Nat. Hist., Zool. Ser., 22. pp. 1-111. 1936J. Bull. Mus. Comp. Zool., 79. pp. 209-337, pis. i-ix. 1937c. Proc. Acad. Nat. Sci. Philadelphia, 89. pp. 265-296. 1937f. Bull. Mus. Comp. Zool., 79. pp. 481-541, pis. i-iv. 1938d. Proc New England Zool. Club, 17. pp. 49-74. 1941c Proc. U. S. Nat. Mus., 91. pp. 113-140. 1942e. Bull. Mus. Comp. Zool., 91. pp. 237-373, pis. i-vi. 1942g. Fauna (Philadelphia), 4. pp. 118-121, figs. 19451). Notulae Naturae Acad. Nat. Sci. Philadelphia, No. 151, pp. 1-4. 1946a. Journ. E. Africa Nat. Hist. Soc, 18. pp. 97-115, figs. 1-4. 1951a. Bull. Mus. Comp. Zool., 106. pp. 177-204, fig. — . 1951c Bull. Inst. Roy. Soc Nat. Belgique, 27. pp. 1-12. 1951d. Bull. Zool. Nomencl., 6, pp. 88-S9. 1953e. Bull. Mus. Comp. Zool., 110. pp. 141-322, pis. i-v. 1953h. Quart. Journ. Florida Acad. Sci., 18. pp. 139-1.10, photo. 190 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 1955e. Journ. E. Africa Nat. Hist. Soc, 22. pp. 168-198, fig. — . 1956c. Sudan Notes Eecords for 1955, 36, pp. 37-56, map. 1957a. Tanganyika Notes Eecords, No. 43, pji. 1-19. See also T. Barbour and A. Loveridge Matschie, Paul 1891a. Zool. Jahrb. Syst., 5. pp. 605-611. 1891b. Zool. Jahrb. Syst., 5. pp. 612-618. 1892. Sitzungsb. Ges. Naturf. Freunde Berlin, pp. 101-110. 1893c. Mitt. Fors. Gel. deutschen Sehutzgeb., 6, pp. 207-215. Meek, S. E. 1897. Field Columbian Mus., Zool. Ser., 1. pp. 161-184, fig. — 1910. Field Mus. Nat. Hist., Zool. Ser., 7. pp. 403-414. Mertens, Egbert 1926a. Senckenbergiana, 8, pp. 137-155. 1934c. Zoologica (Stuttgart), 32. No. 84, pp. 1-210, pis. i-vi. 1937b. Abhand. Senekenberg. Naturf. Ges., No. 435, pp. 1-23. 1938b. Abhand. Senekenberg. Naturf. Ges., No. 442, pp. 1-52, pis. i-x. 1938e. Senckenbergiana, 20. pp. 425-442, figs. 1-6. 1940a. Zool. Anz., 131, pp. 239-250, fig. — . 1941. Zool. Anz., 135. pp. 275-281. 1946b. Abhand. Senekenberg. Naturf. Ges., No. 471, pp. 1-88, pis. i-viii. 1955a. Abhand. Senekenberg. Naturf. Ges., No. 490, pp. 1-172, map, pis. i-xxiv. Methuen, p. a. and J. Hewitt 1914b. Ann. Transvaal Mus., 4, pp. 118-145, figs. 14-16, pi. xir. MocQUARD, Francois lS87b. Bull. Soc. Philom. Paris, (7) 11. pp. 62-92, pis. i-ii. 1888. Mem. Soc. Philom. Paris, Cent., pp. 109-134, pis. xi-xii. 1896b. Compte-Eendu Soc. Philom. Paris, No. 19, pp. 44-45. 1896c. Bull. Mus. Hist. Nat, (Paris), 2. pp. 59-60. 1897b. Bull. Soc. Philom. Paris, (8) 9. pp. 5-20. 1899a. Bull. Mus. Hist. Nat. (Paris), 5. pp. 218-219. 1902b. Bull. Mus. Hist. Nat. (Paris), 8. pp. 404-417. 1906b. Bull. Mus. Hist. Nat. (Paris), 12, pp. 247-253. 1906c. Bull. Mus. Hist. Nat. (Paris), 12, pp. 464-467. 19081). In E. Foa, R/'sultats scientifiques des Voyages en Afriquo d'Edouard Foa (Paris), pp. 557-558, map. LOVERIDGE: AFRICAN SNAKE GENERA 191 MoNARD, Albert 1931. Bull. Soc. Neuchatel. Sci. Nat., 55. pp. 89-111, figs. 1-5. 1937b. Arqu. Museu Boeage (Lisboa), 8. pp. 1-154, figs. 1-3. 1940b. Arqu. Museu Bocage (Lisboa), 11. pp. 147-180, figs. 1-7, pis. i-ii. 1951. Mem. Inst. Frau^. Afrique Noire, No. 1, pp. 123-185, figs. — . MOREAU, R. E. and R. H. W. Pakenham 1941. Proc. Zool. Soc. London for 1940, 110 (A), pp. 97-128, fig. 1. 1942. Proe. Zool. Soc. London, 112, (A), pp. 61-63. Morrison, F. D. See G. L. Orton and F. D. Morrison MULLER, FRIEIDERICH 1882a. Verhandl. Naturf. Ges. Basel, 7. pp. 120-165, pis. ix-xi. 1882b. Verhandl. Naturf. Ges. Basel, 7. pp. 166-174. 1884. Verhandl. Naturf. Ges. Basel, 7. pp. 274-299. 1885. Verhandl. Naturf. Ges. Basel, 7. pp. 668-717, pis. Lx-xi. 1890. Verhandl. Naturf. Ges. Basel, 8. pp. 685-705, pi. x. MilLLER, LORENZ 1910. Bayer. Akad. Wiss., Kl. 2, 24. pp. 543-626, pi. — . Nieden, Fritz 1911b. Sitzungsb. Ges. Naturf. Freunde Berlin for 1910, pp. 441-452. See also R. Sternfeld and F. Nieden Odhner, Teodor 1908. Arkiv Zool., 4, No. 18, pp. 1-7. Orton, G. L. and F. D. Morrison 1946. Copeia, pp. 15-17. Pakenham, R. H. W. 1947. Ann. Mag. Nat. Hist., (11) 14. pp. 134-140. See also R. E. Moreau and R. H. W. Pakenham See also H. W. Parker, R. E. Moreau and H. W. Pakenham Parhnti, Paolo and Luigi Picaglia 1886. Atti. Soc. Nat. Modena, (3) 5, pp. 26-96. Parker, H. W. 1927b. Ann. Mag. Nat. Hist., (9) 20. pp. 81-86, figs. 1-3. 1930b. Ann. Mag. Nat. Hist., (10) 8. pp. 603-606. 1932a. Journ. Linn. Soc. London, Zool., 38. pp 213-229, figs. 1-10. 192 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY 1932b. ProL-. Zool. Soe. London, pp. 335-367, figs. 1-3. 1936c. Novit. Zool. (Tring), 40. pp. 115-146. pp. 1-2. 1937a. Ann. Mag. Nat. Hist., (10) 20. pp. 629-632. 1949a. Zool. Verhand. Bijksnnis. JVatur. Hist. Leiden, pp. 1-116, figs. 1-11, map. Parker, H. W., R. E. Moreau and E. H. W. Pakenham 1940. Ann. Mag. Nat. Hist., (11) 5. pp. 309-314. Pbrracca, M. G. 1896. Boll. MuR. Zool. Anat. (;oiii[». Univ. Torino, 11. No. 25.j, pp. 1-4, figs. — . 1909. In Luigi Amedeo, Due d'Abruzzi, II Ruwenzori. Parte scientifica (Milano), 1. pp. 16.5-180. 1910. Boll. Mus. Zool. Anat. Conip. Univ. Torino, 25. No. 624, pp. 1-6. 1912. Ann. Museo Zool. Univ. Napoli, 3. No. 25, pp. 1-8. Peters, W. K. H. 1854b. Monatsb. Akad. Wiss. Berlin, pp. 614-628. 1855. Arch. Naturg., 21. Abt. 1, pp. 43-58 (reprint of 19541)). 1861a. Monatsb. Akad. Wiss. Berlin, pp. 358-360. 1866b. Monatsb. Akad. Wiss. Berlin, pp. 884-892. 1867b. Monatsb. Akad. Wiss. Berlin, pp. 234-237, pi. — . 1869a. In C. C. von der Dec-ken, Ecisen in Ost-Afrika (Leipzig), 3. Abt. 1, pp. 11-18, pis. i-ii. 1869b. Ofver. K. Vet.-Akad. Forh., pp. 657-662. 1870b. Monatsb. Akad. Wiss. Berlin, pp. 110-115, pi. — . 1871a. Monatsb. Akad. Wiss. Berlin, pp. 566-581. 1875a. Monatsb. Akad. Wiss. Berlin, pp. 196-212, pis. i-iii. 1876a. Monatsb. Akad. Wiss. Berlin, pp. 117-123, pi. i. 1877e. Monatsb. Akad. Wiss. Berlin, pp. 611-621, pi. — . 1878a. Monatsb. Akad. Wiss. Berlin, pp. 194-209, pis. i-ii. 1881d. Sitzungsb. Ges. Naturf. Freunde Berlin, pp. 147-150. 1882a. NaturAvissenschaftliohe Reise naeh Zklossanibiiiue anf Befehl sei- ner Majestiit des Konigs Friedrieli Wilhelm IV in den Jahren 1842 bis 1848 ausgefuhrt. Zoologie, 3, .Aniphil.ieii. (Berlin), xv -f 191 pp., pis. i-xxx. See also R. Buchholz and W. K. H. Peters I'KEFFER, GEORG 1889. Jahrb. Hanilmrg. Wiss. Anst., 6, pi). 1-38. 1893. .lahrl.. Hamburg. Wiss. Anst., 10. pp. 71-105, pis. iii. LOVERIDGE : AFRICAN SNAKE GENERA 193 Phisalix, Marie 1917a. Bull. Mus. Hist. Nat. (Paris), 23. pp. 331-336. Pitman, C. R. S. 19129. Pioc. Zool. Soc. London, pp. 133-146. 193-4. Report ou a Faimal Survey of Northern Rhodesia (Livingstone), pp. 292-312, index i-xxxii, maps. 1936. Uganda Journal (Kampala), 3. pp. 211-229, 259-281, pis. iii-v; 4. pp. 41-61, 126-150, pis. vi-x, col. pis. A-G. 1937. Uganda Journal (Kampala), 4. pp. 319-349, pi. xiii, col. pis. L-M. 1938a. Uganda Journal (Kampala), 5, pp. 160-224, pis. xv-xviii, col. pis. R-W, diagrams I-II. 19381). A Guide to the Snakes of Uganda (Kampala), xxi + 362, diagrams I-II, maps, pis. i-xxviii, col. pis. A-Q (re]>rint of preceding articles in book form). Porter, A. Sec H. B. PANTnAJt and A. Porter Power, J. H. 1927c. Trans. Roy. Soc. S. Africa, 14, jip. 405-422, fig. 1, map, pis. xviii- xxii. 1931. Trans. Roy. Soc. S. Africa, 20. pp. 39-49, figs. 1-2. 1935. Proc. Zool. Soc. London, pp. 333-346, figs. 1-10, pi. i. See also J. Hewitt and J. H. Power I'liATo, Alberto del 1893. Le Raccolte zoologiche fatte nel Congo dal Cav. Giuseppe Corona (privately printed i)am])hlet. Premiata Casa Editrice L. Battel. Parma), pp. 1-14. Reichenow, Anton 1874. Arch. Naturg., 40, Abt. 1, pp. 287-298, pi. Lx. Reinhardt, J. T. 1843. K. Danske Vidensk. Selsk. Afhandl., 10, pp. 233-277, pis. i-iii. Rochebrune, a. T. de 1884a. Faune de la Senegamltie. Reptiles. (Paris), pp. 1-221, pis. i-xx (a totally unreliable work). Rose, Walter 1929. Veld and Vlei (Cape Townj, xxiii + 240, figs. ^, photos 1-125. 1950. The Reptiles and Amphiljians of Southern Africa (Cape Town), xxviii + 378 pp., frontis. and photos 1-213. 195.1. Snakes — Mainly South African. (Cape Town), xvi -|- 213 pp., frontis. and 89 -|- 34 figs. li)4 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Roux, Jean 1907a. Revue Suisse Zool. (Geneve), 15. pp. 75-86. 1907c. Zool. Jahrb. Syst., 25. pp. 403-444, pis. xiv-xv. 1910. Revue Suisse Zool. (Geneve), 18, pp. 95-103. Roy, R. See F. Angel, . . . and R. Roy Sauvage, H. E. 1884b. Bull. Soc. Zool. France, 9. pp. 199-208, pi. vi. SCHENKEL,, EHRENFRIED 1901. Verhandl. Naturf . Ges. Basel, 13. pp. 142-199. ScHLEGEL, Hermann 1837. Essai sur la Physionomie des Serpens. (Leide), 1. xxviii + 251 + xvi pp.; 2. xvi + 60fi pp.; Atlas with maps 1-3; pis. i-xxi. Schmidt, K. P. 1923. Bull. Amer. Mus. Nat. Hist., 49, 1-148 pp., figs. 1-15, maps 1-19, pis. i-xxii. 1933. Ann. Carnegie Mus., 22, pp. 1-15, figs. 1-2, pis. i-ii. Schouteden, Henri 1933. Revue Zool. Bot. Afr., 23, pp. 233-238. SCHWETZ, J. 1934a. Revue Zool. Bot. Afr., 25. pp. 379-384. 1934b. Revue Zool. Bot. Afr., 26. Suppl., pp. 23-25. SCLATER, W. L. 1898. Ann. S. African Mus., 1, pp. 95-111, pi. v. ScoRTEcci, Giuseppe 1929c. Atti Soc. Ital. Sci. Nat. (Milano), 68. pp. 245-278, figs. — , pi. xii. 1930a. Atti Soc. Ital. Sci. Nat. (Milano), 69. pp. 193-217, figs. — . 1930c. Boll. Mus. Zool. Anat. Univ. Torino, (3) 41. No. 10, pp. 1-2(5, figs. — . 1931c. Atti Soc. Ital. Sci. Nat. (Milano), 70. pp. 203-215. li)32b. Atti Soc. Ital. Sci. Nat. (Milano), 71. pp. 39-49, figs. — . 1939a. Ann. Mus. Civ. Stor. Genova, 63. pp. 263-291. 1940a. Jn Missione biologica nel paesc dei Borana raccolte zoologicbe. (Reale Accad. d 'Italia, Roma), 2. pp. 125-150 (repaged 5-30 in separate). 1943. In Missione biologica Sagan-Omo, Zool. (Roma), pp. 267-305, figs. 1-5. LOVEBIDGE : AFRICAN SNAKE GENERA 195 Sjostedt, Yngve 1897. Bihang till K. Sveiiska Vet.-Akad. Ilandl., 23, Part 4, No. 2, pp. 1-36, pis. i-iii. Smith, Sib A. 1838- Illustrations of the Zoology of South Africa, Reptiles. (Lon- 1849. don), pis. i-lxxviii, App., pp. 1-28. Smith, M. A. 1928. Ann. Mag. Nat. Hist., (10) 1, pp. 494-497. SOMEREN, V. G. L. VAN 1939. Journ. E. Afr. Uganda Nat. Hist. Soc, 14, pp. 155-160. Steindachner, Franz 1867a. Reise der osterreichisehen Fregatte Novara um die Erde in den Jahren 1857, 1858, 1859, unter den Befehlen des Conunodore B. von Wullerstorf-Urbair. Zool., 1, 1-98 pp., pis. i-iii. Stejneger, Leonhard 1893b. Proc. U. S. Nat. Mus., 16, pp. 711-741. Sternfeld, Robert 1908. Sitzb. Ges. Naturf. Freunde Berlin, pp. 92-95. 1908a. Mitt. Zool. Mus. Berlin, 3, pp. 397-432. 1908b. Mitt. Zool. Mus. Berlin, 4. pp. 207-236. 1908c. Mitt. Zool. Mus. Berlin, 4, pp. 237-247. 1909a. In Die Fauna der deutschen Kolonicn. (Berlin), Ser. 2, ]it. 1, pp. iv + 29, figs. 1-42, map. 1909b. hi Die Fauna der deutschen Kolonien. (Berlin), Ser. 1, pt. 1, pp. iv + 28, figs. 1-34, map. 1910a. In Die Fauna der deutschen Kolonien. (Berlin), Ser. 3, pt. 2, pp. iv _|_ 47^ figs. l-ri4, mail. (X. B. 1910d actually antedates 1910a). 19101). In Die Fauna der deutschen Kolonien. (Berlin), Ser. 4, pt. 1, pp. iv + 45, figs. 1-50, map. 1910c. Mitt. Zool. Mus. Berlin, 5, pp. 51-56. 1910d. Mitt. Zool. Mus. Berlin, 5, pp. 61-65. 1911a. Sitzb. Ges. Naturf. Freunde Berlin, pp. 245-2.31. 1911b. Mitt. Zool. Mus. Berlin, 5. pp. 393-411. 1912c. Ill Wiss. Ergeb. der deutschen Zentral-Afrika-Expedition 1907- (1913) 190S. (Leipzig, 1913, but separate 1912), 4, pp. 197-279, figs. 1-4, pis. vi-ix. 1917. In Wiss. Ergeb. der. zweiter deutschen Zentral-Afrika-Expedi- tion 1910-1911. (Leipzig, 1917), 1, pp. 407-510, pis. xxii-xxiv. 196 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY Sternfbld, Robert, and Fritz Nieden 1911. :\ritt. Zool. :\Iiis. Berlin, 5. pp. 383-385. Themido, a. a. 1941. Mem. Estudos :\Ius. Zool. Uuiv. Coimbra, (1), No. 119, pp. 1-32. Theobald, William 1868. Journ. Asiatic Soc. Bengal, Extra No., pp. 1-88, App. 1-111, 4 pis. — . 1876. Descriptive Catalogue of the Reptiles of British India (Cal- cutta), X 4- 238 pp., Syn. i-xxxviii, App. i-xiii, 5 pis. (6 in M. C. Z. copy). TORXIER, GUSTAV 1896. Die Kriechthiere Deutsch-Ost-Afrikas, Beitriige zur Systeniatik also und Descendenzlehre. (Berlin, reprinted 1897) xiii -f 164 pp., 1897. figs. 1-11, pis. i-v. 1897b. Arch. Naturg., 63, Abt. 1, pp. 63-66. 1898. In C. W. Werther, Die mittleren Ilocliliinder des nordliclien Deutseh-Ost-Afrlka. (Berlin), pp. 281-304, figs. 1-15. Traill, T. S. 1843. In II. Schlegel, Essay on the Physiognomy of Serpents (London), vii -f- 254 pp., pis. i-iii. Uthmoller, Wolfgang 1937. Temminckia (Leiden), 2, pp. 97-134, maps. 1938. Zool. Anz., 124. pp. 41-48. 1941a. Arch. Xaturg., (2) 10, Abt. 1. pp. 1-70, figs. 1-20, map, pis. i-ii. 19411). Zool. Anz., 135, pj). 225-242. Vieira, a. X. Y. 1886. O. Inst. Rev. Sci. Lift. (Coimbra), (2), 34, p]). 234-241 (herpe- tology on pp. 237-238). Villiers, Andre 1948a. Bol. Cult. Guine Portuguesa, No. 9, pp. 201-204. 1950a. La Collection de Serpents de I'l. F. A. N., Catalogues VI, Inst. Frang. Afrique noire. (Dakar), 1-155 pp., figs. 1-12. 19501j. Bull. Inst. Fram^. Afrique noire, 12, pp. 984-998. 1951. Etud. Dahomeennes, No. 5, pp. 17-43, 2 pis. 1951a. Bull. Inst. Franc,-. Afrique noire, 13, pp. 813-836, figs. 1-3. 1952b. Bull. Inst. Fran.'. Afrique noire, 14, pp. 881-898. 1954b. Bull. Inst. Franc. Afrique noire, 16 (A), pp. 1234-1237, figs. 1-2. See also P. L. Dekeyser and A. Villiers LOVERIDGE : AFRICAN" SNAKE GENERA 197 Werner, Franz lS95b. Verb. Zool.Bot. Ges. Wien, 45. pp. 190-194, pi. v. ]897h. Verb. Zool.-Bot. Ges. Wien, 47, pp. 395-408. 1898. Jahresli. Alihand. Natunv. Ver. Magdeburg, 1896-1897, pp. 139- 148. 1898a. Verb. Zool.-Bot. Ges. Wien, 48. pp. 191-213, figs. — , pi. ii. 1899a. Verb. Zool.-Bot. Ges. Wien, 49. pp. 132-157. 1901h. Verb. Zool.-Bot. Ges. Wien, 51. pp. 634-639. 1902a. Verb. Zool.-Bot. Ges. Wien, 52. pp. 332-348. 1903. Ahband. Konig. Bayer. Akad. Wiss., 22. pp. 381-384, pi. — . 1907. Akad. Anz. Wien, 44. p 479. 1908. Sitzb. Akad. Wiss. Wien, 116. Abt. 1, pp. 1823-1926, pis. (1907) i-iv. 1908a. Tbird Eep. Wellcome Trop. Res. Lab. Kbartoum, pp. 169-172, fig. 42, pis. xvii-xx. 1909c'. Jahresb. Ver. Nat. Wurttemberg, 65. pp. 55-63. 1909d. Jabrb. Hamburg. Wiss. Anst., 26. pp. 205-247, figs. 1-14. (1908) 1910a. DcnksL'hr. Med.-Nat. Ges. Jena, 16. pp. 279-370, figs. 1-15. 1913a. In A. Brebm, Tierleben. (Leipzig), cd. 4, 5. xvi + 598 pp., figs. — , maps, pis. — . 1913b. Denkscbr. Akad. Wiss. Wien, 88. pp. 714-719, pi. i. 1915c. In W. Micbaelsen, Beitrage zur Kenntnis der Land-und Siiss- wasserfauna Deutsch-Siidwestafrikas. III. (Hamburg), 325-376 pp., pi. vii. 1923c. Ann. Naturhist. Hofmus. (Wien), 36. pp. 160-166. 1924b. Sitzb. Akad. Wiss. Wien, 133. Abt. 1, pp. 29-56, figs. 1-9. 1925. Arcb Naturg., 90. Abt. A, pp. 108-166, figs. 1-18. (1924) 1925. Sitzb. Akad. Wiss. Wien, 134. Abt. 1, pp. 45-66, figs. 1-4. 1929a. Zool. Jabrb. Syst., 57. pp. 1-481, figs. 1-48. WiTTE, G. F. DE 1927d. Revue Zool. Afr., 15. pp. 320-332. 1928i. Revue Zool. Bot. Afr., 16. Suppl., p. 9. 1933J. Revue Zool. Bot. Afr., 24. pp. 111-123, figs. — . 1933m. Ann. Mus. Congo beige, Zool., (1)3. pp. 55-98, pis. i-iv. 1941. Exploration du Pare National Albert. Mission G. F. de Witte (1933-1935). Batraciens et Reptiles. Inst. Pares Nat. Congo Beige, Fase. 33, xviii + 261 pp., figs. 1-50, map, pis. i-xlvi. 1952. Exploration Hydrobiologique du Lac Tanganyika (1946-1947). Inst. Royal Sci. Nat. Belgique, 3. Faac. 3, 1-22 pp., pi. i. 198 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY 1953. Exploration du Pare National de rUpcmba. Mission G. F. de Witte en collaboration avec W. Adam, A. Janssens, L. van Meel et R. Verheyen (1946-1949). Reptiles. Inst. Pares Nat. Congo Beige, Fasc. 6, 1-322 pp., figs, i-iii, pis. i-xli, map. 195.5. Bull. Seances Acad. Roy. Sci. Colonials (N.S.) 1, pp. 203-225. Witte, G. F. de and R. Laurent 1943a. Revue Zool. Bot. Afr., 37, pp. 157-189. 1947. Mem. Mus. Royal Hist. Nat. Belgique, (2) fasc. 29, pp. 1-134, figs. 1-132. Worthington, E. B. 1929. A Report on the Pishing Survey of Lakes Albert and Kioga. (Cambridge, Eng.), 1-136 pp. (herpetology on pp. 122-124). Zavattari, Edoardo 1930b. In S. E. E. de Bono, Missione scientifiea per 1 'Eritrea. (Parma), 171-205 pp., figs. 1-19. Bulletin of the Museum of Comparative Zoology AT HAEVAED COLLEGE Vol. 119, No. 2 FOUR NEW RAJIDS FROM THE GULF OF MEXICO By Henry B. Bigelow and "William C. Schroeder Museum of Comparative Zoology and Woods Hole Oopanosrraphic Institution CAMBRIDGE, MASS., U.S.A. printed for the museum .July, 1958 Publications Issued by or in Connection WITH THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE Bulletin (octavo) 1863 — The current volume is Xo]. 119. Breviora (octavo) 1952 — No. 89 is current. Mejioirs (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 op the Nev^ 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 Comporcrtive Zoology AT HARVAED COLLEGE Vol. 119, No. 2 FOUR NEW RAJIDS FROM THE GULF OF MEXICO By Henry B. Bigelow and William C. Schroeder Museum of Comparative Zoology and Woods Hole Oceanographic Institution CAMBRIDGE, MASS., U.S.A. PRINTED FOR THE MUSEUM July, 1958 No. 2 — Four New Rajids from the Gulf of Mexico^ By Henry B. Bigelow and William C. Schroeder Museum of Comparative Zoology and Woods Hole Oceanographie Institution Recent otter trawling operations of the U. S. Fish and Wildlife Service vessel "Oregon" in the Gulf of Mexico have yielded four new species of batoids, three of them falling in the genus Raja and one in Cruriraja. These, added to about a dozen new species of elasmol)ranehs described from the Gulf during the past seven years, and records of others previously unknown from the western Atlantic, emphasize the richness of the Gulf fauna. We are indebted to Stewart Springer and Harvey R. Bullis, Jr. of the U. S. Fish and Wildlife Service for the opportunity of de- scribing these species. Drawings are by Eugene N. Fischer. Raja oregoni sp. nov. Figures 1, 2 Study Material. Mature male, 1440 mm in total length, holo- type, U. S. Nat. Mus. No. 156710, and an immature male 1069 mm in total length, paratype, M. C. Z. No. 39617, both from the northern part of the Gulf of Mexico, the holotype from Lat. 29°10'N, Long. 88°13'W, in 205 fathoms, "Oregon" station 1247, the paratype from Lat. 28°32'N, Long. 86°20'W, in 260 fathoms, "Oregon" station 1277. Distinctive Characters. The combination of features distinctive for oregoni among hard-nosed skates of the genus Raja are : anterior contour of disc deeply concave, with narrowly projecting snout, of the shape pictured in Figure 1 ; upper surface of pectorals without conspicuous markings of any sort ; disc, rear- ward of the spiracles, without an}- large thorns, its midbelt bare even of prickles; tail with three rows of large close-set thorns extending to base of first dorsal fin ; lower surface with pore openings in head region marked conspicuously with black dots. 1 Contribution No. 932 from Tlie Woods Hole Oceanographie Institution. 202 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Comparison with previously known species. It is only with species sharing the diagnostic features listed in the preceding section that comparison seems called for, a limitation Avelcome Figure 1. Eaja oregoni, dorsal view of type, mature male 1440 mm long; end of tail showing dorsal fins and caudal fin about x 0.2; tail thorns from mid row about x 1. because about 110 recognizably distinct species of the genus Raja have been named already from one part of the oceans or another (Bigelow and Schroeder 1953, p. 147). BIGELOW AND SCHROEDER : NEW RAJIDS 203 Among the skates known from the North Atlantic (including the Gulf of Mexico) oregoni most nearly resembles laevis Mitchill 1817, the common barn-door skate of the northeastern American Figure 2. Eaja oregoni, ventral view of specimen shown in Figure 1. coast and its eastern Atlantic representative, hatis Linnaeus 1758. But the thorns on the tail of oregoni are much larger than on laevis or on hatis, closer set, and with those on the lateral rows 204 BFLLETIX : MUSEUM OF COMPARATIVE ZOOLOGY along the rear half of the tail directed rearward in very charac- teristic shape (Fig. 1). Other differences are that the snout is noticeably longer in oregoni (length anterior to orbits about 4 times as great as the distance between orbits) than in either laevis or hafis (2.4-3 times) in males of about equal size; that the rostral cartilage is narrower and that the black pore markings of the ventral surface are restricted to the area anterior to the abdominal region whereas in laevis and in hatis they are present on the rearward part of the disc as well. The only other western North Atalntic skates that oregoni resemljles at all closely in general appearance, combined with thorn pattern, are E. olseni Bigelow and Schroeder 1951 and R. teevoni Bigelow and Schroeder 1951 but there is no danger of confusing it with either of these. Conspicuous respects in which it differs from olseni are in a shorter interspace between first and second dorsal fins, with only 1 thorn (3-6 thorns in olseni) ; in a larger number (31 to 48) of thorns in the mid-row on tail (only 13-26 in olseni) with the lateral rows closely and regularly spaced (irregularly and wddely spaced in olseni). Other differences of perhaps equal significance taxonomically, though less noticeable, are that in oregoni the expanded outer- posterior margin of the nostrils is smooth edged (fringed in olseni, Bigelow and Schroeder 1951, fig. 2B; 1953, fig. 54 a B), that the upper surface of the snout, with the outer-anterior edge of the pectorals, is prickly (smooth in olseni), that the nuchal region is naked in oregoni but is armed with a strong thorn in large specimens of olseni^ (not, however, on the juveniles). On the other hand, no trace is to be seen on the upper surface of oregoni of the white dots marking the pores that are a conspicu- ous feature of olseni, nor has the latter any dark dots on the under surface as is the case in oregoni. Oregoni resembles teevani closely in anterior contour of its disc but differs in having 3 rows of prominent thorns on the tail (one row on teevani), by having a tail which narrows posteriorly as is usual among skates (on teevani the posterior half of tail is as wide or wider than the anterior), and by having a space, occupied by a thorn, between its dorsals (dorsals are confluent on teevani), and teevani lacks dark pores on the under surface. ('(I 1 Sppciiiipiis at hand inohide a nearly mature male olseni 510 mm long recently llected by "Oregon" in the Gulf of Mexico at Station 1514. BIGELOW AND SCHROEDEB : NEW RAJIDS 205 Oregoni is set apart, by the combination of characters classed above as "diagnostic," from all hard-nosed skates yet known Figure 3. Baja flavirostris, immature male 374 mm long, "William Scoresby" station 79, from Lat. 51°01'30"S, Long. 64°59'30"W, in British Museum (Natural History); pelvic fins about x 0.4; mouth, nostrils and nasal curtain about x 0.8. 206 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY from the east coast of South America, or from the Magellanic region, with the exception of R. flavirostris Philippi 1892 the recorded range of which extends from Argentine and Patagonian waters to Chile. And it differs sharply from half grown and larger flavirostris^ in a conspicuously longer tail (46-50 per cent of total length contrasted with 41.8-42.5 per cent in the half- grown flavirostris with which we have compared it ; in lacking the large nuchal thorn that is characteristic of flavirostris ; also in the greater size, more regular spacing, and shape of the thorns in the lateral rows on the tail (cf. Fig. 1 with Fig. 3 of flavirostris) . The only skates known from European or tropical west African Avaters, and from southern Africa, with w^hich oregoni shares an outer anterior contour of the shape pictured in Figure 1, com- bined with a disc that is wholly bare of thorns posterior to the spiracles, but a tail that is armed with 3 regular rows (one median and two laterals) of close-set thorns, are : Raja hatis Linnaeus 1758, R. alba Lacepede 1803 (including R. marginata Lacepede 1803, partly grown specimens and some adults), and R. miraletus Linnaeus 1758 (some specimens). And it is marked off from all of these by characters that cannot reasonably be credited to variation (whether individual or regional) or ex- plained away on the basis either of sex or of the stage of growth of the particular specimens that have served as the bases for published accounts. Thus it differs from hatis in the same respects in which it differs from laevis of the North American coast (p. 00), and this also applies to the skate that has been reported as hatis from South Africa (Thompson 1914, p. 156; Barnard 1925, p. 70; Smith 1949, p. 66) though available infor- mation regarding the latter is scanty. And while it resembles alba of European waters and the skate that has been reported under that name in southern African waters from Walfish Bay on the west around the Cape to Natal on the east, first bj' von Bonde and Swart (1924, p. 5) both as alba and as stabuliforis 1 We are indebted to the British Museum for the loan of three flavirostris, a male 374 mm long, a female 353 mm, and a juvenile male of 180 mm. from the series reported upon and pictured by Norman 1937, p. 13, flg. 4. BIGELOW AND SCHROEDER : NEW RAJIDS 207 Garman 1913, next as marginata (Barnard 1925, pp. 63, 65), and more recently as alha by Smith (1949, p. 66), the skin of the tail of oregoni is naked between the thorns (described as "strongly spinulose" for alba by Clark, 1926, p. 48), as well as over the inner-posterior parts of the disc as a whole while it is more or less prickly on adults of alha. Further, the mid-row of thorns on its tail does not extend forward on to the disc as it does in alba, when adult, and the black pore markings characteristic of the lower surface of oregoni (Fig. 4) are not present on alba. Again, oregoni resembles B. miraletus in its pointed snout, smoothness of posterior part of disc and presence of 3 rows of thorns on the tail. But its outer anterior contour is much more deeply concave than that of miraletus; it lacks the 2 nuchal thorns characteristic of the latter; the number of thorns in each of the rows along is tail is much greater, there being about 50 in the median row on the male pictured in Figure 1, contrasted with 14 pictured by Rey for an adult male of miraletus 402 mm long, and 25 recorded by him for a female, not counting those between the dorsal fins (Rey, 1928, p. 573, pi. 13, fig. 2). Also, the inter- space between the two dorsals is noticeably shorter in oregoni than in miraletus; and while black pore marks are present on the lower surface of oregoni, they are lacking on miraletus. And oregoni is much the larger growing member of the pair, it being unlikely that miraletus grows longer than about 500 mm, nor does oregoni show any signs of the large, light-blue centered eye spots, one of which marks the upper surface of each pectoral of miraletus. It is interesting, in passing, that in the presence of the black pore markings on its lower surface oregoni parallels not only hatis, but also oxyrhynchus Linnaeus 1758 (including macrorhyn- chus Bonaparte 1832-41) among European species. But there is no question of specific unity in this case, the snout of oregoni being less narrowly pointed than that of oxyrhynchus with the outer anterior contour of the disc much less deeply concave ; the 3 rows of large tail thorns of oregoni, so conspicuous a feature of the adult male, being represented on the adult male oxyrhyn- chus by a single median row of about 8 to 12 thorns; the two dorsal fins separated by an interspace with a thorn in oregoni 208 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY being confluent at the base, or nearly so, in oxyrhynchus; and the black pore marks on the lower surface, which are confined Figure 4. Baja orcgoni, ventral view of head to show dark nuu-ous pores, about X 0.1. Lower left, section of tail from 8th to 16th thorns in mid row, from first dorsal fin, about x 0.-5, same specimen as in Figure 1. Lower right, Raja laevis, section of tail from 3rd to 9th thorns in mid row, from first dorsal fin, about x 0.5, a male 118;" mm long from off Nantucket, Massa- chusetts. BIGELOW AND SCHROEDER : NEW RAJIDS 209 mostly to the head region in oregoni being distributed over the abdominal region as well in oxyrhyncJius. In its concave outer anterior outlines, with narrow snout, and in the black dots marking the lower surface of its head, oregoni most nearly resembles B. rhina Jordan and Gilbert 1880 among skates of the west coast of North America. But it differs from rhina in lacking the nuchal spine or spines that characterize rhiyia, as well as in the orderly arrangement of the thorns on its tail which on rhina are usually arranged irregularly with the rows interrupted. Furthermore, the mid zone of disc from the scapular region on to the tail is smooth on oregoni but has a band of small thorns with stellate bases on rhina. Similarly, the narrowness of its "nose" combined with its lack of thorns anywhere on the disc rearward from the spiracles, contrasted with the presence of 3 rows of thorns along the tail, marks off oregoni among hard-nosed skates known from the Pacitic coast of Central and South America, except flavirostris which was originally described from Chile. Respects in which it differs from flavirostris are summarized above. Among the dozen or so narrow-nosed species of Raja that have been reported, with adequate published accounts, from the Indo- West Pacific region, three only are described, with a fourth pictured, as with the openings of the pores black-marked on the lower surface, and oregoni cannot be identified with any one of these. Thus oregoni lacks a nuchal thorn which is present and conspicuous in australis Macleay 1884 from Australia and Tas- mania and the 3 rows of thorns on the tail of oregoni are longer and more regular than in australis. While oregoni lacks a nuchal thorn and the tail thorns are arranged in regular rows, tohae Tanaka 1916 from Japan has 2 thorns on the nuchal region and its 3 rows of tail thorns are irregular. Oregoni, lacking thorns along the midline of disc, differs from porosa Giinther 1874, from the north Chinese coast, which has 3 or 4 conspicuous thorns in that area, and the 3 rows of thorns on the tail of oregoni distinguish it from gudgcri Whitley 1940, from Western Aus- tralia, which has but a single row as well as a much longer snout. Neither can oregoni be referred to any of the other Indo-West Pacific species that it resembles at all closely in the anterior con- tour of its disc, for no known member of this group parallels 210 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY oregoni in the presence of three regular rows of large close-set thorns extending from the base of the tail rearward as far as the first dorsal fin, but combined with a total lack both of thorns and of prickles on the mid belt of the disc posterior to the spiracles. While it agrees with ynacracauda Ishiyama 1955 from Japan in shape of snout and disc and also in presence of dark pores below/ it differs from macracauda in having 3 rows of thorns on the tail (only 1 row in macracauda). Description of holotype. Proportional dimensions in per cent of total length r Disc. — Extreme breadth 69.8; length 53.2. Snout length. — In front of orbits 17.4; in front of mouth 18.5. Orbits. — Horizontal diameter 2.8 ; distance between 4.5. Spiracles. — Length 2.1 ; distance between 6.4. Mouth. — Breadth 8.0. Exposed nostrils. — Distance between inner ends 8.3. Gill openings. — Length, 1st 1.4; 3rd 1.8; 5th 1.2; distance between inner ends, 1st 13.8 ; 5th 8.1. First dorsal fin. — Vertical height 1.5; length of base 3.6. Second dorsal fin. — Vertical height (damaged); length of base 3.8. Pelvics. — Anterior margin 10.9. Distance. — From tip of snout to center of cloaca 50.0 ; from center of cloaca to 1st dorsal 37.2 ; to tip of tail 50.0 ; from rear end of 2nd dorsal base to tip of tail 4.2. Interspace between. — 1st and 2nd dorsals 1.3. Disc nearly 1.4 times as broad as long, maximum angle in front of spiracles about 72° ; anterior margins sinuous, being rather strongly concave from tip of snout to opposite orbits where they are slightly convex, hence again concave, this second con- cavity culminating about three-fifths the distance between tip of snout and outer corners of disc which are sharply rounded; posterior margins about straight near outer corner, then gently rounded, as is inner margin. Axis of greatest breadth about 72 per cent back from tip of snout to axils of pectorals. Tail with a 1 Two young specimens examined by us, through the kindness of Dr. Ishiyama. - The measurements from which the proportional dimensions have been cal- culated were taken on a horizontal line between perpendiculars at given points (see Bigelow and Schroeder 1953, fig. 1 and p. 4). BIGELOW AND SCHROEDER : NEW RAJIDS 211 narrow lateral fold, low down on each side, beginning almost imperceptibly beyond level of tips of pelvics by a distance equal to about twice the interorbital width, widening posteriorly and continuing almost to extreme tip of tail being much the widest opposite dorsals and caudal membrane ; length of tail from center of cloaca to origin of first dorsal fin 0.75 times as great, and to its tip equal to distance from center of cloaca to tip of snout. A row of 7 or 8 small thorns along inner margin of each orbit, ending about opposite middle of spiracle. End of snout covered with coarse prickles ; smaller prickles over rostral process, on translucent areas in front of orbits and in space between orbits. A band of small thorns and coarse prickles extends along edge of disc beginning rearward from tip of snout by a distance about equal to that l)etween spiracles, and ending a little in advance of alar thorns, being coarsest in the area opposite and a little posterior to the spiracles. No thorns in the scapular region. A median row of about 48 thorns beginning on tail opposite axils of pectorals and extending to first dorsal fin, of unequal sizes, some with points directed obliquely rear- ward, others Avitli points broken or worn ; a row of thorns low down on each side of the median row, beginning a little in advance of tips of pelvics and ending opposite second dorsal fin, 31 on the left and 34 on the right, those with points still intact directed rearward parallel, or nearly so, to axis of tail ; 1 thorn between dorsals. Alar thorns prominent, in two to four rather regular rows, 43 on left side of disc, 38 on right, rest of body, pelvics and tail smooth, including dorsal fins and skin over eyes. A short row of 9 or 10 mucous pores in nuchal region on each side, parallel to midline, at a distance from it equal to about two-thirds length of orbit, the pores not ringed with dark as conspicuously as on garricki. Lower surface with prickles in front of and aside the mouth, densest toward end of snout and along margins of disc where they extend rearward about one-half the distance toward outer angle of pectoral ; a few minute prickles scattered on disc between mouth and fifth pair of gill openings; rest of lower surface naked. 212 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY Snout in front of orbits 6 times as long as orbit, its length in front of mouth 2.2 times as great as distance between exposed nostrils. Distance between orbits about 1.6 times as great as length of orbit. Orbit 1.3 times as long as spiracle. Nasal curtain fringed; expanded posterior (outer) margin of nostrils smooth, without fringes. Upper and lower jaAvs strongly arched centrally. Teeth |^, more nearly in longitudinal rows than in quincunx, with circular or ovate base, those in upper jaw in median sector of mouth with a prominent narrow upright cusp, the adjacent teeth pointing very slightly toward corner of mouth where the cusps become low triangular ; lower teeth similar. Distance between first gill openings 1.7 times as great as between exposed nostrils ; between fifth openings 0.97 times ; first gill openings 1.2 times as long as fifth and about 0.2 as long as breadth of mouth. First and second dorsals similar in size and perhaps in shape (second dorsal damaged). Interspace between dorsals 0.36 times as long as base of first dorsal. Second dorsal confluent with caudal membrane. Caudal membrane base equal to first dorsal base. Pelvics deeply concave, strongly scalloped along anterior side of excavation, moderately so rearward ; an- terior margin 0.74 times as long as distance from its own origin to rear tip of pelvic ; anterior lobe slender, including four radial cartilages besides the first stout one ; posterior lobe moderately convex, with a narrowly rounded tip, extending 0.36 times the distance from axil of pectorals tow^ard first dorsal. Claspers reach- ing beyond tips of pelvics by a distance equal to seven-tenths the distance from front of orbits toward tip of snout. Color. Upper surface brownish, without spots or other mark- ings. Lower surface pale bluish-gray on disc, pelvics and clasp- ers ; tail pale brown ; scattered black pores anterior to abdominal region, most numerous around mouth and in front of level of mouth to near tip of snout. The other known specimen of this species, an immature male 1069 mm long, differs in the following respects. The disc is about 1.3 times as broad as long; maximum angle in front of spiracles 70° ; axis of greatest breadth 70 per cent back from iSp of snout toward axils of pectorals. The tail is relatively shorter, the dis- tance from center of cloaca to its tip being 0.85 times the distance from center of cloaca to tip of snout. Because of the relatively BIGELOW AND SCHBOEDER : NEW RAJIDS 213 shorter tail on the immature male, the snout is longer relative to total leng-tli (21.5 per cent to eye and 23.2 per cent to mouth opening, compared with only 17.4 and 18.5 per cent, respectively, for the mature male), the disc proportionately wider, and longer (76.5 per cent and 57.2 per cent, compared with 69.8 and 53.2 ])er cent, respectively, for the larger male). There are 5 small thorns along inner margin of each orbit and 1 or 2 immediately in front of orbit ; fewer prickles on end of snout ; and the anterior edge of the disc is smooth. The median row of thorns on tail numbers 31, the side rows 24 on the left and 22 on the right, these latter ending opposite first dorsal fin and there is no thorn be- tween the dorsals. On the lower surface the prickles along the margins of disc extend rearward about three-fifths the distance toward outer angle of pectoral. The snout in front of orbits is 7 times as long as orbit, its length in front of mouth 2.5 times as great as distance between exposed nostrils. The jaws are less strongly arched and there are ^ teeth, with low triangular cusps, only those near outer corners being slightly oblique. The second dorsal, which is undamaged on this specimen, is similar to the first in shape but with a slightly shorter base and the interspace between the dorsals is relatively shorter, being about 0.14 times as long as the base of first dorsal. The anterior margin of the pelvic is almost as long as the distance from its own origin to rear tip of pelvic and the posterior lobe extends 0.31 times the distance from the axils of pectorals toward first dorsal. The elaspers fail to reach the tips of the pelvics by a distance equal to length of fifth gill opening. Known only from the offing of Cape San Bias, Florida, in 205-260 fathoms. Named in recognition of the fishery explora- tions of U. S. Fish and Wildlife Service vessel "Oregon" in the Gulf of Mexico and the Caribbean Sea. Raja garricki sp. nov. Figures 5, 6, 7 Study Material. Mature male, 975 mm in total length, holo- type, U. S. Nat. Mus. No. 156711, and another male 1019 mm in total length, paratype, M. C. Z. No. 39616, both from the northern part of the Gulf of Mexico, Lat. 28°32'N, Long. 86°20'W, in 260 fathoms, "Oregon" station 1277. 214 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Figure 5. Raja garridki, dorsal view of type, mature male 975 mm long; end of tail showing dorsal fins and caudal fin, about x 0.5. BIGELOW AND SCHROEDER : NEW RAJIDS 215 Figure 6. Raja garricki, ventral view of specimen shovm in Figure 5 ; right hand nostril and nasal curtain, about x 1 ; upper and lower teeth from center and from near outer part of jaws, about x 2.4. 216 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Distinctive Characters. Characters in combination which dis- tinguish garricki from other species of the genus Raja are : margins of disc anterior to outer angles of pectorals sinuous (at least in the males, the female has not been seen) ; snout pointed (Fig. 5) ; disc with a single row of large thorns along the midbelt extending from the nuchal region to the first dorsal fin, with an interrupted row on either side on the tail ; a single large scapular thorn on each side of the midrow ; lower surface with dark mucous pores anterior to the abdominal region (Fig. 7). _ _ Comparison with previously known species. The only western Atlantic skates with a disc somewhat the shape of garricki and with dark pores on the under surface are laevis, oregoni, flavi- rostris, platana Giinther 1880 and agassizi Miiller and Henle 1841. But garricki differs from all of these (excepting some specimens of agassizi) by having a continuous row of thorns along the mid-line of disc and tail, extending from the nuchal region to the dorsals whereas the midrow of thorns on the above mentioned species does not extend much, if any, forward beyond the axils of the pectorals. And garricki may be distinguished from agassizi by the position of its first dorsal which originates much nearer to tip of tail than to tips of pelvics (nearer pelvics than tip of tail on agassizi) and by a dorsal fin interspace much shorter than base of either first or second dorsal. Also, the dark pores on the under surface of garricki are much more con- spicuous and more numerous than on agassizi. In the eastern Atlantic, including southern Africa, only batis and oxyrhynchus have a narrow snout together with dark pores below, and both of these lack the thorns along the midline of the disc that are present on garricki. Garricki falls with rhina and flavirostris of the eastern Pacific in general shape of disc and in the presence of dark pores below. But garricki differs in spination from rhina by having scapular thorns and a single row of midline thorns on the disc but no addi- tional small thorns in the midzone. Flavirostris has one nuchal thorn but no other midline thorns on disc. Garricki somewhat resembles lempricri Richardson 1844-48, porosa Giinther 1874, and austraUs Macleay 1884 in the Indo- Australian region in having a narrow snout and dark pores BIGELOW AND SCHROEDER : NEW RAJIDS 217 (Munro 1956, p. 15) on the lower surface but its row of thorns along the midzone of disc sets it apart from these species which lack thorns in this area and from gudgeri AVhitley 1940 which has but a very short row along the middle of the back as well as a very long and narrow snout resembling that of oxyrhynchiis. Garricki closely resembles macracauda Ishij-ama 1955 from Japan in shape of disc and in having dark pores below but, except for 2 nuchals, macracauda lacks thorns along the midline of disc and has no scapular thorns. Description of holotype. Proportional dimensions in per cent of total length : Disc. — Extreme breadth 78.3; length 59.0. Snout length. — In front of orbits 18.3 ; in front of mouth 19.0. Orbits. — Horizontal diameter 3.5 ; distance between 5.4. Spiracles. — Length 3.1 ; distance between 6.9. Mouth. — Breadth 8.5. Exposed nostrils. — Distance between inner ends 9.1. Gill openings — Length, 1st 1.5; 3rd 1.8; 5th 1.5; distance between inner ends, 1st 15.7 ; 5th 9.4. First dorsal fin. — Vertical height 2.8 ; length of base 4.0. Second dorsal fin. — Vertical height 2.6 ; length of base 3.7. Pelvics. — Anterior margin 10.6. Distance. — From tip of snout to center of cloaca 55.4 ; from center of cloaca to 1st dorsal 30.7, to tip of tail 44.6 ; from rear end of 2nd dorsal base to tip of tail 5.0. Interspace between. — 1st and 2nd dorsals 1.2. Disc about 1.3 times as broad as long, maximum angle in front of spiracles about 78° ; anterior margins weakly concave just pos- terior to tip of snout, thence slightly convex followed by a stronger concavity culminating about midwaj^ between tip of snout and outer corners which are sharply rounded ; posterior luargiixs at first about straight from outer corner, then gently rounded, as is inner margin. Axis of greatest breadth about 62 per cent of distance back from tip of snout toward axils of pectorals. Tail with a narrow lateral fold, low down on each side, beginning abruptly in advance of tips of pelvics by a dis- tance ecjual to about three-fourths the interorbital width and con- tinuing almost to extreme tip of tail, widest opposite dorsals 218 BULLETIN : MUSEUM OF COMPARATRTI ZOOLOGY and caudal membrane; length of tail from center of cloaca to origin of first dorsal fin 0.56 times as great, and to its tip 0.80 times as great, as distance from center of cloaca to tip of snout. A patch of 11 small thorns immediately in front of left orbit and 7 in front of right followed by a row of 7 or 8 larger thorns along inner margins of orbits ending slightly posterior to rear margins of spiracles. End of snout densely covered with small thorns and prickles extending rearward over rostral process in diminishing numbers; a band of small thorns and prickles extends along edge of disc from tip of snout to about opposite foremost alar thorns, narrow near end of snout, then widening and the prickles coarsening half way from snout to opposite orbits, again smaller rearward. A single stout thorn on each side of median row in scapular region; a median row of prom- inent thorns from nuchal region to first dorsal fin, of which 4 are in the nuchal-scapular area, the second of these being very small, followed by 32 thorns of which 7 are anterior to pectoral axil and 25 on tail, these latter somewhat closer together and more uneven in size ; a row of 3 large thorns aside the median row on left side of the mid-tail region and 5 on the right side, unevenly spaced ; 2 prominent thorns between dorsal fins. All the thorns on body and tail that have retained their sharp points are directed obliquely rearward. Alar thorns prominent, in two to three irregular rows, 26 on left side of disc, 30 on right ; rest of body, pelvics and tail smooth as is space between orbits and skin over eyes ; first and second dorsals with a few scattered prickles. A short longitudinal row of 11 dark pigmented mucous pores on the right side of first nuchal thorn and of 8 on the left side, at a distance from the midline about two-thirds as great as diameter of orbit. The presence of mucous pores, in a conspicuous pattern in this region, has thus far been found only on Raja fyllae Liitken 1887, laevis and in the new species oregoni and clarkii among skates known from the western North Atlantic. (See Bigelow and Schroeder, 1954, p. 57.) Lower surface wnth prickles in front of and aside the mouth, most dense toward end of snout and along margins of disc where they extend rearward about three-fifths the distance toward outer BIGELOW AND SCHROEDER : NEW RAJIDS 219 angle of pectoral; a few exceedingly small widely scattered prickles over parts of the scapular and abdominal regions and on pelvics; rest of lower surface naked. Snout in front of orbits 5.2 times as long as orbit, its length in front of mouth 2.1 times as great as distance between ex- posed nostrils. Distance between orbits about 1.6 as great as length of orbit. Orbit 1.1 as long as spiracle. Nasal curtain fringed; expanded posterior (outer) margins of nostrils smooth, without fringes. Upper and lower jaws strongly arched cen- trally. Teeth || more nearly in quincunx arrangement than in longitudinal rows, with circular or ovate base, those in upper jaw in median sector of mouth with a prominent upright cusp, the adjacent teeth with cusps pointing toward corner of mouth, those near corners of mouth with flattened crowns ; lower teeth similar but median teeth with cusps at slight angle, which in- creases outwardly. Distance between first gill openings 1.7 times as great as between exposed nostrils; between fifth openings 1.0 times; first gill openings 1.1 times as long as fifth and about 0.2 as long as breadth of mouth. First and second dorsals similar in size and shape. Interspace between dorsals 0.33 as long as base of first dorsal. Caudal membrane base equal to that of first dorsal. Pelvics deeply concave, strongly scalloped along anterior side of excavation, moderately so rearward ; anterior margin 0.58 as long as distance from its own origin to rear tip of pelvic ; anterior lobe slender, including four radial cartilages beside the first stout one ; posterior lobe moderately convex, with a narrowlj' rounded tip, extending about one-half the distance from axil of pectorals toward first dorsal. Claspers reaching beyond tips of pelvics by a distance equal to seven-tenths the distance from front of orbits toward tip of snout. Rostral cartilage firm, extending nearly to tip of snout. An- terior pectoral rays reaching about three-sevenths the distance from front of orbits toward tip of snout. Color. Upper surface plain brownish without spots or other markings. Lower surface pale bluish dusky on disc, pelvics and claspers, tail tending more to pale brownish. Scattered black 220 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY pores anterior to abdominal region, most numerous around mouth and in front of level of mouth to near tip of snout. (Fig. 7.) The other known specimen of this species, a male 1019 mm long, differs as follows : maximum angle of disc in front of spiracles 74°, axis of greatest breadth about 67 percent of dis- tance back from tip of snout toward axils of pectorals ; length of tail from center of cloaca to origin of first dorsal fin is 0.53 times as great and to its tip 0.78 times as great as from center of cloaca to tip of snout. There is a patch of 8 small thorns along anterior edge of each orbit ; fewer prickles over rostral process which is Figure 7. Baja garricJci, ventral view of head to show dark mucous pores, about X 0.1. naked toward orbits ; 5 thorns in the mid row in the nuchal- scapular area, followed by 41 thorns to first dorsal, of which 8 are anterior to pectoral axils and 33 are on tail ; a row of 5 unevenly spaced large thorns on left side of tail aside the median row, 2 on right ; 31 alar thorns on left and 33 on right side of disc. The mucous pores in nuchal region are in a series of 9 on each side. On the lower surface the prickles along the anterior margin of disc extend less than halfway toward outer angle of pectoral. BIGELOW AND SCHROEDER : NEW RAJIDS 221 Known only from the offing of Cape San Bias, Florida, in 260 fathoms. Named for Dr. J. A. F. Garrick of Victoria University College, Wellington, in recognition of his work on elasmobranchs of New Zealand. Raja clarkii sp. nov. Figures 8, 9 Study Material. Immature male, 665 mm in total length, holo- type, U. S. Nat. Mus. No. 156712, another male 580 mm long, and a female 747 mm long, paratypes, M. C. Z. No. 39618, all from the northern part of the Gulf of Mexico, Lat. 28°32'N, Long. 86°20'W, in 260 fathoms, "Oregon" station 1277. Diagnostic Features. The presence of 1 to 3 pairs of prominent white roundish or barlike markings on the upper surface of disc in combination with a broad snout and a band of formidable and very sharp thorns which extend along the lower surface from the tip of snout almost to the extreme outer margin of pectorals sets off clarkii from all other know^n members of the Rajidae. Description of holotype. Proportional dimensions in per cent of total length : Disc. — Extreme breadth 68.7 ; length 55.0. Snout length. — In front of orbits 14.2 ; in front of mouth 15.3. Orbits. — Horizontal diameter 3.2; distance between 3.6. Spiracles. — Length 3.3; distance between 6.0. Mouth. — Breadth 8.7. Exposed nostrils. — Distance between inner ends 4.5. Gill openings. — Length, 1st 1.2; 3rd 1.6; 5th 1.3; distance between inner ends, 1st 14.9 ; 5th 8.9. First dorsal fin. — Vertical height 1.7 ; length of base 4.3. Second dorsal fin. — Vertical height 1.7 ; length of base 4.3. Pelvics. — Anterior margin 11.7. Distance. — From tip of snout to center of cloaca 53.4 ; from center of cloaca to 1st dorsal 34.6 ; to tip of tail 46.6 ; from rear end of 2nd dorsal to tip of tail 3.2. Interspace between. — 1st and 2nd dorsals 0.6. Disc about 1.25 times as broad as long; outline of snout as in Figure 8 ; anterior margin rather strongly convex in front of orbits, thence concave with outer corners broadly rounded; pos- 222 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY terior and inner margins gently rounded. Axis of greatest breadth about 70 per cent of distance back from tip of snout ;*_ .* o.- ■-' i ■i-j<^m ^Simmm'-. ■■'■■■ ^/'"i^TY:!iii:'^ V-X- .i^v* ■■ ^-' ;*^.- -^' 1 . . s-v.^c-*'j..?Ny .,x,9- 1^%v;-i Figure 8. ^aja clarkii, dorsal view of type, immature male 665 mm long ; end of tail showing dorsal fins and caudal fin, about x 0.5. to axils of pectorals. Tail with a narrow lateral fold, low down on each side, beginning just beyond tips of pelvies and continu- BIGELOW AND SCHBOEDER : NEW RAJIDS 223 ing almost to extreme tip of tail, widening posteriorly, widest opposite dorsals ; length of tail from center of cloaca to origin of /.-'.■;..''iii'V- w^-mwum-- Figure 9. Ha^a cJarhii, ventral view of specimen shown in Figure 8 ; right- hand nostril and nasal curtain, about x 1.1. first dorsal fin 0.65 times as great and to its tip 0.87 times as great as distance from center of cloaca to tip of snout. 224 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY A small thorn in advance of each orbit by a distance of about Yg orbit's diameter; a larger thorn immediately next to inner anterior edge of orbit, with one at posterior end of orbit and another smaller one inward toward median line, this last opposite rear end of spiracle. Snout and disc in front of orbits and space between orbits with small thorns, many with stellate bases, and with prickles; two stouter thorns over rostral process, located 27 and 47 per cent, respectively, in the distance from tip of snout toward orbits ; a band of small thorns extends along edge of disc from near tip of snout to about opposite scapular region. A triangular patch of three thorns on each side of median row in scapular region ; a median row of 34 thorns from nuchal region to first dorsal, of which 4 are in the nuchal-scapular area, followed by 7 anterior to pectoral axils and by 23 on tail, all of these of about equal size and equally spaced ; a row of thorns on each side of the median row beginning about opposite axils of pelvics and continuing to opposite first dorsal, many with sharp hooked points directed rearward ; 1 thorn be- tween dorsals. Alar thorns not yet exposed ; rest of disc with scattered prickles, very sparse in some areas and mostly minute ; also small thorns and prickles on tail in addition to the three prominent rows, mostly low down, from near axils of pelvics nearly to tip of tail ; pelvics smooth ; dorsals with an occasional prickle. A longitudinal row of 12 dark ringed mucous pores on right side and 13 on left outward from first nuchal thorn by a distance about one-third the diameter of orbit. Lower surface with prickles along rostral process out to tip of snout with a few scattered over translucent area in front of nostrils ; also a band of strong, very sharp thorns all of about the same size, curving inward or obliquely rearward, extending from near end of snout almost to outer angle of pectoral, in a band about 0.4 times as wide as distance between orbits, these charac- ters in combination being a striking feature of this species; rest of lower surface smooth. Snout in front of orbits 4.4 times as long as orbit, its length in front of mouth 3.4 times as great as distance between exposed nostrils. Distance between orbits about 1.1 as great as length of orbit. Orbit about the same length as spiracle. Nasal curtain fringed; expanded posterior (outer) margins of nostrils with a BIGELOW AND SCHROEDER : NEW RAJIDS 225 few fringes. Upper and lower jaws moderately arched. Teeth -|^, in part in longitudinal rows and in part more or less in quincunx, with rounded base and triangular cusp, inclined toward corners of mouth. Distance between first gill openings 3.3 times as great as dis- tance between exposed nostrils ; between fifth openings 2.0 times ; first gill openings about as long as fifth and about 0.14 as long as breadth of mouth. First and second dorsals similar in size and shape. Interspace between dorsals 0.14 as long as base of first dorsal. Caudal membrane with base a little shorter than that of first dorsal base. Pelvics deeply concave, stronglj' scalloped along anterior side of excavation, moderately so rearward ; anterior margin 0.78 as long as distance from its own origin to rear tip of pelvic ; anterior lobe slender, including four radial cartilages besides the first stout one ; posterior lobe moderately convex, with a narrowly rounded tip, extending a little less than half the distance from axils of pectorals toward first dorsal. Claspers reaching beyond tips of pelvics by a distance equal to less than half diameter of orbit. Rostral cartilage firm, extending nearly to tip of snout. Anterior pectoral rays reaching anterior edge of disc but falling short of end of rostral cartilage which continues into snout protuberance. Color. Upper surface pale brown with darker punctulations scattered over disc, pelvics and more or less of tail. Three l)airs of prominent white markings on disc, of which one pair is opposite spiracles, bar-like, somewhat curved, and a little longer than interorbital space ; a second pair about half as long and oi^posite scapular region and a little closer together than first pair ; a third pair about size of second pair, in line with first pair and opposite eighth thorn from nape in median series. Lower surface white with a wide irregular grayish band along posterior and inner margins, extending from a little in advance of outer angles to axils of pectorals and along rear margin of pelvics; also a pair of grayish spots opposite cloaca and a few small ones here and there on disc, with some gray markings on tail and on claspers. The other two known specimens of this species, a male 580 mm long and a female of 747 mm agree very closely with the holotype 226 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY in proportional dimensions and in the formidable band of sharp thorns on the under surface along the anterior margin of the disc. The chief differences are as follows. They have a disc 1.2 times as wide as long, and 65.3 and 65.0 per cent, respectively, of the total length of the specimens, as compared with 1.25 times as wide and 68.7 per cent as long as the holotype, while the length of tail from center of cloaca to origin of dorsal fin is 0.69 times as great and to its tip 0.90 times as great as distance from center of cloaca to tip of snout on both specimens as compared with 0.65 and 0.87 times, respectively, on the holotype. The two thorns over the rostral process on the 580 mm male are located 34 and 58 per cent in the distance from tip of snout to orbits while the 747 mm female has three such thorns located 23, 44, and 60 per cent in this distance. There are 40 thorns in the median row on the 5bO mm male of which 28 are posterior to the pectoral axil, and there are 38 thorns on the 747 mm female of which 27 are posterior. The mucous pores number 9 on one side and 11 on the other on the smallest skate and 14 on each side of the largest which also has more numerous prickles on lower surface over the translucent area in front of nostrils than on the other two specimens. The tooth count of the 580 mm male is f}^, and of the 747 mm female ^^^ . The upper surface of the female has the three pairs of prom- inent white markings of about the same relative size and location as on the holotype but below there are no gray spots in the cloaeal region while there is an oval gray spot on each side of the rostral process near its tip. The small male has only one pair of white spots above, opposite the spiracles, roundish, and about half the size of the orbit. Known only from the northern part of the Gulf of Mexico, Lat. 28°32'N,'Long. 86°20'W, in 260 fathoms. Named for Robert S. Clark in recognition of his revision of European skates and rays. Cruriraja rugosa sp. nov. Figures 10, 11 Study Material. Immature male, 367 mm in total length, holo- type, U. S. Nat. Mus. No. 156713 from the northeastern part of the BIGELOW AND SCHROEDER : NEW RAJIDS 227 Gulf of Mexico, in 200-300 fathoms, trawled by the "Oregon," station number not known. Distinctive Characters. Cruriraja rugosa differs from all other known members of its genus in that the under side of its tail is covered with minute prickles, this area being smooth in all other species. Comparison with previously known species. From other west- ern Atlantic species it differs as follows : from atlantis Bigelow and Schroeder 1948, by the short space between its dorsals, equal to about one-fourth the length of first dorsal base (about 2^2 times the dorsal base on atlantis) ; from poeyi Bigelow and Schroeder 1948, by its prickles on disc, the presence of two nuchal thorns and of a single row of thorns on the midline of tail (disc without prickles, no nuchal thorns, midline row of thorns on tail divides into two rows posteriorly on poeyi). It differs from the two species of Cruriraja known from South Africa, as follows : from parmomaculata (von Boude and Swart 1924, p. 9, pi. 21, fig. 2) in having a wide space lacking thorns along the midline of disc between the 2 nuchal thorns and axils of pectorals, by lacking scapular thorns, and by its plain coloration {parmor)%acu- lata, described from a young specimen 181 mm in total length, has a continuous row of thorns from the nuchal region to the first dorsal, 1 scapular thorn on each side and 14-18 blackish brown spots of varying size on the disc). Rugosa with its widely interrupted row of thorns along the midline of back, its lack of thorns in the scapular region and the rounded rear margins of its dorsal fins may be readily distinguished from durhanensis (von Bonde and Swart 1924, p. 11, pi. 22, fig. 1) which has an uninterrupted row of thorns along the midline of back from nape of first dorsal on a male 232 mm long, 2 scapular thorns on each shoulder and dorsal fins which are pointed posteriorly ; and on a female 311 mm long the midrow of thorns ends about halfway along tail, leaving a wide thornless space l)efore first dorsal. Description of holotype. Proportional dimensions in per cent of total length : Disc. — Extreme breadth 59.4; length 46.0. Snout length. — In front of orbits 12.0 ; in front of mouth 15.0. Orbits. — Horizontal diameter 4.5 ; distance between 2.8. 228 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY if I . • '^ Figure 10. Cruriruja rut/osa, type, immature male 367 mm long; pelvic fins about x 0.5; section of tail about x 2; first 2 thorns in front of dorsal fin about x 3. BIGELOW AND SCHROEDER : NEW RAJIDS 229 Spiracles. — Length 2.2 ; distance between 6.5. Mouth. — Breadth 6.3. Exposed nostrils. — Distance between inner ends 5.5. Gill openings. — Length, Lst 1.5; 3rd 1.5; 5th 1.2. First dorsal fin. — Vertical height 2.7 ; length of base 4.5. Second dorsal fin. — Vertical height 2.6 ; length of base 3.3. Pelvics. — Length of limb 15.0. Distance. — From tip of snout to center of cloaca 40.6 ; from center of cloaca to 1st dorsal 48.0; to tip of tail 59.4; from rear end of 2nd dorsal base to tip of tail 2.5. Interspace between. — 1st and 2nd dorsals 1.1. Disc about 1.3 times as broad as long, maximum angle in front of spiracles about 87° ; anterior margins from snout to outer corners of pectorals slightly sinuous, the corners abruptly rounded ; posterior and inner margins gently rounded. Axis of greatest breadth about 70 per cent of distance back from tip of snout to axils of pectorals. Tail with a lateral fold, low down on each side beginning about 55 per cent of the distance from straight. Teeth ||, arranged in quincunx, with ovate base and axils of pelvics toward tip of tail, very narrow anteriorly but widening considerably on approaching tip of tail; length of tail from center of cloaca to origin of first dorsal 1.2 times as great and to its tip 1.45 times as great as distance from center of cloaca to tip of snout. Inner margin of left orbit with 6 thorns, 2 of which are an- terior, 1 midway and 3 posterior ; right orbit with 3 anterior and 3 posterior thorns, the last two in each ease being opposite the spiracles. A staggered row of 4 thorns on anterior part of rostral process followed by a pair of rostral thorns a little more than half the distance from tip of snout toward a line connecting anterior margins of orbits ; a band of sharp, backward pointing thorns along margins of disc, beginning opposite anterior edge of orbits and ending about opposite scapular region, the thorns decreasing in size and the band narrowing posteriorly. Two prominent nuchal thorns followed by a naked area and then by a midrow of 47 thorns beginning about 11/2 eye's diameter in advance of axils of pectorals and continuing uninterrupted to first dorsal fin, becoming staggered posteriorly, but in a single row ; an additional row of smaller thorns each side of the midrow, 230 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY low down on tail, beginning about opposite tips of pelvics and extending nearly to first dorsal where it merges with a band of prickles ; all the tail thorns pointing strongly rearward ; a single thorn between dorsals. Upper surface of disc (including skin over eyes) covered with minute prickles except along posterior margins and on extreme tip of snout ; pelvics naked ; tail prickly, everywhere except along the midzone ; dorsals densely prickly. Lower surface of disc entirely smooth, but tail covered with minute prickles from about opposite tips of pelvics to opposite origin of second dorsal. Snout in front of orbits 2.8 times as long as orbit, its length in front of mouth about 2.7 times as great as distance between exposed nostrils. Distance between orbits 0.65 times as great as Figure 11. Cruriraja rugosa, ventral view of head, about x 0.5; right- hand nostril and nasal curtain, about x 2. length of orbit. Orbit twice as long as spiracle. Nasal curtain with a very short, blunt fringe, expanded outer margins of nostrils smooth, Avithout fringes. Upper and lower jaws nearly short triangular cusp. Distance between first gill openings twice as great as distance between exposed nostrils; between fifth openings about 1.2 times; first to fourth gill openings about 1.3 times as long as fifth and about 0.25 as long as breadth of mouth. First and second dorsals similar in shape the first slightly the larger. Interspace between dorsals 0.25 times as long as base of first dorsal. Caudal fin very small. Anterior division of pelvics long and slender reaching BIGELOW AND SCIIROEDER : NEW RAJIDS 231 beyond tips of posterior lobe when pulled back ; posterior lobe with 12 rays, quadrate, the rear margin gently rounded and faintly scalloped, the tips reaching but little beyond rear margin of disc. Claspers failing to reach tips of pelvics by a distance equal to half the diameter of eye. Rostral cartilage firm, narrow, extending nearly to tip of snout. Anterior pectoral rays reaching 22 per cent the distance from front of orbits toward tip of snout. Color. Upper surface of disc brownish with a bluish tinge, without spots or other markings ; tail pale brownish ; dorsal brownish with a pale area basally. Below pale bluish on disc, the tail pale brown. Known only from a single specimen, an immature male 367 mm long, trawled in 200-300 fathoms in the northeastern part of the Gulf of Mexico. EET^EEENCES Barnard, K. H. 192.5. A monograph of the fishes of South Africa. Part I. Ann. S. African Mus., vol. 21, part 1, pp. 1-418, 18 figs. BiGELOW, H. B. AND W. C. SCHROEDER 1948. New genera and species of batoid fishes. Jour. Mar. Res., vol. 7, no. 3, pp. 543-566, 9 figs. 1951. Three new skates and a new chimaerid fish from the Gulf of Mexico. Jour. Washington Acad. Sci., vol. 41, no. 12, pp. 383-392, 4 figs. 1953. Fishes of the western North Atlantic. Mem. Sears Found. Mar. Res., no. 1, part 2, x + 588 pp., 127 figs. 1954. Deep water elasmobranchs and chimaeroids from the northwest- ern Atlantic Slope. Bull. Mus. Comp. Zool., vol. 112, no. 2, pp. 37-87. BoNAPARTK, Charles Lucihn 1832- Iconografia della fauna Italica . . ., vol. 3, Pesci, 78 pis. (not 1841. 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. 232 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Clark, R. S. 1926. Rays and skates — Sei. Invest. Fisher. Board Scotland, 1926, no. 1, 66 pp., atlas, 36 pis. CiARMAN, Samuel 1913. The Plasiostoinia (sharks, skates and rays). Mem. Mus. Conip. Zool., vol. 36, xiii -j- 515 pp., and atlas, 75 pis. CJllNTHER, ALBBaiT 1874. Third notice of a collection of fishes recently obtained by Mr. Swinhoe in China, Ann. Mag. Nat. Hist., ser. 4, vol. 13, pp. 154- 159. 1880. Report on the shore fishes — " Challenger " Rept. Zool., vol. 1, part 6, 82 pp., 32 pis. ISHIYAMA, REIZO 1955. Studies on the rays and skates l)olonging to the family Rajidae, found in Japan and adjacent regions. Jour. Shimonoseki College of Fisheries, vol. 4, no. 1, pp. 43-51. Jordan, D. S. and C. H. Gilbert 1880. Description of a new species of ray, Haia rhvna, from the coast of California. Proc. U. S. Nat. Mus., vol. 3, pp. 251-253. Lacepede, B. G. E. 1803. Histoire Naturelle Poissons, vol. 11, 399 pp., 3 pis. Linnaeus, C. 17.')8. Systenia Naturae, 10th ed., Holmiae, vol. 1, 824 pp. Lutken, C. F. 1887. Korte Bidrag til Nordisk iclithyographi. 4. Vidensk. Medd. naturh. Foren. Copenhagen, pp. 1-4, pi. 1. Macleay, William 1884. Some results of tra^vl fishing outside Port Jackson. I'roi-. Linn. Soc. N. S. Wales, vol. 8, pp. 457-462. Mitchill, S. L. 1817. Memoir on ichthyology — Amer. Monthly Mag. Crit. Rev., vol. 2, pp. 321-328. Muller, Johannes and F. G. J. Henle 1841. Systematische Beschreibung der Plagiostomen. Berlin, x.xii + 200 pp., 60 pis. BIGELOW AND SCIIROEUER : NEW RA.JIDR 233 MuNRO, Ian S. R. 1956. Handbook of Australian fishes. In: Fisheries Newsletter, vol. 15, no. 9, 32 pp. Canberra. Norman, J. R. 1937. Coast fishes. Part II. The Patagonian region. Discovery Re- ports, vol. 16, pp. 1-150, pis. 1-5. Philippi, R. a. 1892. Algunos pecos de Chile. An. Mus. Nac. Chile, sec. 1, Zool., 16 pp., 6 pis. Rey, Luis Lozano 1928. Fauna Iberiea. Feces, vol. 1, 690 pp., 20 pis. Inst. Nac. Ciencias, Madrid. Richardson, John 1844. Zoology . . . H. M. S. "Erebus" and "Terror", part 7, Fishes. 1848. 139 pp., 60 pis. London. Smith, J. L. B. 11)49. The sea fishes of southern Africa, 550 pp., 1232 figs., 103 pis. Central News Agency, S. Africa. Tanaka, S. 1916. A new species of Japanese fish. (In Japanese.) Dobwts. Zool., Tokyo, vol. 28, pp. 313-314. Thompson, W. W. 1914. Catalogue of fishes of the Cape Province. Mar. Biol. Report, Prov. Cape of Good Hope, no. 2, pp. 132-167. von Bonde, C. and D. B. Swart 1924. The Platosomia (skates and rays) collected by the S. S. "Pickle". Rep. 3, Fish. Mar. Biol. Surv., S. Africa (1922), Spec. Rept. 5, 22 pp., pis. 20-23. Whitley, G. P. 1940. The fishes of Australia. Part I. Royal Zool. Soc. New South Wales, 280 pp. Bulletin of the Museum of Comparcrtive Zoology AT HARVARD COLLEGE Vol. 119, No. 3 THE GENERAL HISTOLOGY AND TOPOGRAPHIC MICROANATOMY OF AUSTRALORBIS GLABRATIIS By Chia-Tung Pan The Department of Tropieal Public Health, Harvard School of Public Health, Boston, Massachusetts With Eighteen Plates CAMBRIDGE, MASS., U.S.A. PRINTED FOR THE MUSEUM July, 1958 Publications Issued by or in Connection WITH THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE Bulletin (octavo) 1863 — The current volume is Vol. 119. Breviora (octavo) 1952 — No. 89 is current. Memoirs (quarto) 1864-1938 — Publication was terminated with Vol. 55. JoiiNSONiA (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 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. 119, No. 3 THE GENERAL HISTOLOGY AND TOPOGRAPHIC MICROANATOMY OF AUSTRALORBIS GLABRATUS By Chia-Tung Pan The Department of Tropical Public Health, Harvard School of Public Health, Boston, Massachusetts With Eighteen Plates CAMBRIDGE, MASS., U.S.A. PRINTED FOR THE MUSEUM July, 1958 No. 3 — The General Histology and Topographic Microanatomy of Australorhis glabj'atus^ By Chia-Tung Pan^ TABLE OF CONTENTS PAGE INTRODUCTION 238 MATERIAL AND METHODS 239 DESCRIPTION 241 Topographic Microanatomy 241 Histology 241 I. Epithelium 242 Epithelium covering the exposed body surface ; epithelium covering the area usually protected by the shell; epithelium covering the mantle cavity. II. Connective Tissue 244 Fibroblasts; pigment cells; vesicular cells; mucous cells ; collagenous-like fibers ; delicate fibrils on the individual muscle fibers. III. Muscular Tissue 248 Granular muscle or heart muscle ; intermediate granular type of muscle ; smooth muscle. IV. Nervous System and Sensory Organs 251 Central ganglia ; peripheral ganglion cells and peripheral ganglion ; nerves ; statocyst ; eyes ; osphradium ; tentacles ; sensory cells at the mar- gin of the lips and at the bases of the tentacles. 1 This Investigation was supported (in part) by a research grant (E-513-C) from the National Institute of Allergy and Infectious Diseases, National Insti- tutes of Health, Public Health Service. 2 Grateful appreciation is expressed to Dr. Thomas H. Weller, Dr. Donald L. Augustine and Dr. Eli Chernin of this Department for their generous advice and encouragement in preparing this manuscript. The author wishes to thank Dr. Edward H. Michelson for his generous supply of Australorbis glabratus used in this study and for his advice in malacolojrical information. Acknowledgment is also made to Mr. Parker A. Glass, Department of Legal Medicine, Harvard Medical School, for taking the photomicrographs. 238 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY V. Circulatory System 259 Heart ; arteries ; veins ; blood sinuses ; amoebo- cytes or wandering phagocytes; hemopoietic tissues. VI. Respiratory System 263 VII. Renal Organ 264 Tubular portion of the kidney; saccular portion of the kidney. VIII. Alimentary System 266 Lips and oral cavity ; buccal mass ; radula ; radu- lar sac ; radular carrier ; esophagus ; stomach ; prointestine ; midintestine ; postintestine ; cecum ; ibuccal gland ; salivary gland ; liver. IX. Reproductive System 280 Common genital (hermaphroditic) organs: ovo- testis; collecting canal; hermaphroditic duct or ovisperm duct. Male genitalia : sperm duct or vas efferens ; pro- static gland ; vas deferens ; penial complex. Female genitalia: albumen gland; carrefour; oviduct ; nidamental gland ; uterus ; vagina ; spermatheca or seminal receptacle. SUMMARY AND CONCLUSION 292 BIBLIOGRAPHY 295 INTRODUCTION In a series of investigations on the potentialities of biological control of schistosomiasis, our efforts thus far have been concen- trated on the accumulation of fundamental biological data on the snail vector, Australorhis glahratus. It was early realized that a knowledge of the normal histology of this snail was essential in order to permit a recognition of specific pathological altera- tions which might be induced by microorganisms. The present study was undertaken to satisfy this objective. The gross anatomy of A. glahratus has been described by Baker (1945), and Paraense and Deslandes (1955). However, information on the histology of this snail is sparse (Faust and PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 239 Hoffman, 1934; von Brand and Files, 1947; Marcuzzi, 1950; Paraense and Ueslandes, 1955) and inadequate as a basis for studying pathological changes. Despite the fact that the para- sitic trematodes of man are known to utilize at least one fresh- water molluscan intermediate host, we have not discovered a comprehensive study of the histology of a single important molluscan host. The paucity of knowledge of the normal histology of fresh-water snails has been recognized recently by workers in this field, and some papers dealing with the digestive and genital tracts of a few' fresh-water snails (Holm, 1946; Carriker and Bilstad, 1946 ; Abdel-Malek, 1954 a, b) have appeared. The excellent monograph by Baecker (1932) on the micro- morphology of Helix and other pulmonates, proved to be excep- tionally helpful in the present studies. MATERIAL AND METHODS Laboratory-raised Australorhis glabratus of Puerto Rican ori- gin were used. Our colony was started from stocks sent by Dr. Redginal I. Hewitt of the Lederle Laboratories, New York, and Dr. Donald A\ Moore then at New York University. Although snails of various sizes were used, the descriptions given here are based mainly on specimens measuring between 9 and 20 mm. in size. Snails were removed from the shell by crushing gently be- tween two glass slides. The shell fragments were then sep- arated with fine forceps under a dissecting microscope. With practice, intact snails could usually be obtained readily in this manner. The shell-free snails were immediately fixed either in a coiled or stretched position. A small amount of fixative was injected into the mantle cavity, employing a 27-gauge needle attached to a tuberculin syringe to insure rapid fixation of internal organs. Air in the mantle cavity was always removed by gentle manipulation under a dissecting microscope. Zenker's or Maximow's fixative gave the best results for cyto- logical study but hardened the genital tract and certain genital organs considerably. Formic acid Bouin's or Benin's mixture had less tendency to harden the tissue, but the cytological struc- tures were less well preserved. A solution of 5 per cent formic 240 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY acid in Zenker's stock was tried with results equal to those with Zenker's. Carnoy's and Newcomer's mixtures and 10 per cent neutral formalin were used for special purposes. Fixation was considered complete after 4 to 48 hours with aqueous mixtures, and after 3 to 6 hours with alcoholic solutions. Tissues were imbedded in paraffin (M. P. 56° C— 58° C) at 60° C and sectioned between 6 and 13 microns. Since it was necessary to remove all of the sand grains in the digestive tract for com- plete serial sections, the method of Carriker and Bilstad (1946) was used to obtain sand-free snails. Lillie-Mayer's, Delafield's, Mallory's and Heidenhain's hema- toxylins were used for nuclear staining, and eosin Y or phloxine B were employed as counter stains. Maximow's hematoxylin- azure Il-eosin in thin sections gave excellent results for cytologi- cal studies. Special staining techniques used for the identifica- tion of various tissue components or structures were: Mallory's and Gomori's trichrome stains for connective tissue elements; Wolbach's Giemsa stain variant; Feulgen reaction for deoxyri- bonucleic acid ; aldehyde f uchsin and acid orcein for elastic tissue ; Hotchkiss' (1948) periodic acid-Schiff (PAS) for carbohydrate ; thionin (at pH 4.0 to 5.0) and toluidine blue for metachromatic substance; Bielshowski-Glees' and Bodian's silver impregnation; Korson's technique (1951) for nucleic acids; alcian blue 8GS (Steedman, 1950) for acid mucopolysaccharides; and Millon's reaction (Pearse, 1953; Lillie, 1954). In the course of this histological study, two types of micro- organisms — an acid-fast bacillus and a yeast-like organism — were encountered. Both organisms were always found associ- ated with histopathologic changes (Pan, 1956). Since these histopathologic changes frequently aided in identifying the nor- mal histologic structures of A. glahratus, reference will be made to the above-mentioned two microorganisms in the later sections. The terminology used to designate various organs is based on Baker's monograph (Baker, 1945) on planorbid snails. The measurements of snails are given in shell diameter and thickness. PAN : HISTOLOGY OP AUSTRALORBIS QLABRATUS 241 DESCRIPTION TOPOGRAPHIC MICROANATOMY Since a knowledge of the topography of various organ systems is necessary for the description of the histology of Australorbis glahratus, a brief account of its topographic microanatomy is given. One longitudinal and four cross-sections were selected to show the relations of various organs. Although the sections were selected on an arbitrary basis, it is believed they provide an adequate perspective of the microanatomy of the organ systems. The photomicrographs of these sections are self-explanatory, and, therefore, only important landmarks are indicated. Figure 1 is an approximate median longitudinal section of a stretched snail measuring 15.0 x 5.1 mm. Figures 2 to 5 were taken under higher magnification and represent four areas of Figure 1. Figure 6 is a cross-section through the ganglion ring, and Figure 7 is through the spermatheca; these were from the same snail which measured 12.8 x 4.0 mm. Figure 8 is a cross- section through the stomach of a snail (10.7 x 4.1 mm.), and Figure 9 is through the overlapping area of the liver and ovo- testis of a snail measuring 20.0 x 6.5 mm. The important landmarks which should be noted in Figures 1 to 9 are to be found in the appropriate captions. HISTOLOGY Baecker (1932) divided the various systems of land pulmo- nates into two major categories, tissue {Gewebe) and organ {Organe), and made several subdivisions for each category. This classification is logical but has considerable overlapping. The following classification of the various tissues and organ systems of Australorbis glabratus was adopted in the present study : I. Epithelium II. Connective Tissue III. Muscular Tissue IV. Nervous System and Sensory Organs V. Circulatory System VI. Respiratory System VII. Renal Organ VIII. Alimentary System IX. Reproductive System 242 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY I. Epithelium Anatomically the epidermal tissue can be divided into three major zones: (a) that exposed to the outer world, including the head-foot organ and the mantle collar, (b) that always protected by the shell, including the mantle, coluraellar muscle, liver and ovotestis, and (c) the mantle cavity surface (respiratory sur- face). Baecker (1932) demonstrated that the epidermal tissue in land pulmonates consists only of a simple epithelium over the body surface. This is also the case in A. glabratus except in one small area at the dorsal rim of the mantle collar where there is pseudostratified columnar epithelium (Figs. 1 e, 11). The epidermal sheet rests on a basement membrane which is supported by a layer of connective tissue containing various amounts of smooth muscle fibers. The coelomic cavity is lined incompletely with flat cells. Epithelium covering the exposed body surface. The surface of the foot or the sole is covered with a layer of tall columnar epithelial cells (ca. 22.5 microns in 18.8 x 6.5 mm. snail) pos- sessing oval to elongated vesicular nuclei, relatively rich in chromatin. The free surface of these cells is heavily invested with long cilia. A row of basal bodies, each of which Ls con- nected with a cilium, lies immediately beneath the plasma mem- brane. The basement membrane can be seen clearly in sections cut at right angles and stained with Gomori's, Mallory's tri- chrome, or PAS. The epidermal cells on the foot surface appear to rest on the basement membrane with their short processes embedded therein. The cytoplasm of these cells is granular and lightly basophilic. The nucleus is located between the middle and basal third of each cell (Fig. 10). The cilia become sparse and dis- appear over the side wall of the head-foot organ where the tall, columnar, simple epithelium of the foot surface is replaced by low columnar-to-cuboidal simple epithelium. The epidermal cells of the mantle collar are also of this type except in a small area at the dorsal rim where there is pseudostratified columnar epithelium (Fig. 11). The cytoplasm of these cells is granular, moderately basophilic, and contains yellowish-brown pigment. This pseudo- stratified epithelium has a glandular appearance like that of the prointestine and may prove to have functions connected w'ith shell regeneration. PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 243 Epithelium covering the area usually protected hy the shell. This area includes the mantle surface, most of the eolumellar muscle, the liver and ovotestis (Fig. 1). The epithelial cells are characterized by having a dense deposit of brownish-l)lack pig- ment in the cytoplasm, which usually obscures cellular struc- tures. This pigment, named melain by Simroth in 1903 (Baecker, 1932), is not seen in the other two zones. Since this pigment stains green with Giemsa or thionin and reduces ammine silver, it is probably related to melanin of higher animals. The epi- thelial cells covering the mantle are low columnar to cuboidal with parabasal, round nuclei (Fig. 12). These cells are trans- formed into flat or squamous epithelial cells in the region of the liver, ovotestis and eolumellar muscle. The basement membrane and supporting connective tissues are delicate. The nuclei of these cells contain fewer chromatin granules than do those of cells covering the head-foot area. Epithelium covering the mantle cavity. Three types of epi- thelium can be recognized in the mantle cavity. These are flat cells, cuboidal cells without cilia, and columnar cells which may be ciliated. The flat cells cover the surface of the wall of the coelomic cavity. The rest of the mantle cavity surface, except the surface of the three ridges (rectal, dorsal and kidney ridges), is covered with a sheet of cuboidal epithelial cells with round, parabasal nuclei (Fig. 13). The rectal, dorsal and kidney ridges are covered with a sheet of columnar cells that are very tall at the summits of the ridges and are provided with dense cilia in this area. These tall cells measure 30 microns in height in a snail of 12.8 x 4.0 mm. Unlike those of the foot surface, they rest on the basement membrane without anchoring short proc- esses. The nuclei are oval, rich in chromatin and located in the middle third of the cytoplasm. The granular cytoplasm has three zones, the surface and the basal zones staining acidophilic, and the middle zone basophilic. Cilia arise from basal bodies located beneath the plasma membrane. The nuclei of the cuboidal cells are spherical and located near the basement membrane. The granular cytoplasm of the flat and of the cuboidal cells stains lightly basophilic throughout and these cells lack cilia. These two types of cells also have a smooth basal surface where 244 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY it is in contact with the basement membrane. Mucus-secreting cells (goblet cells) are scattered among the columnar and cu- boidal cells and secrete mucus through the interstices between epithelial cells. Pigment does not occur in the epithelial cells in the mantle cavity. II. Connective Tissue The connective tissue of A. glahratus occupies the region be- tween the organs and tissues and forms a membranous covering or sack enclosing the organs in the body cavity. The connective tissue contains various cellular and fibrous components and may be classified as to type depending on the relative proportion of the individual components. Baecker (1932) recognized four cellular components {Fibroblast en, Blasenzellen, Kornchenzellen, and Pigment zellen) and two fibrillar components {praekollagene Fibrillen and Gitterfasern) in the connective tissue proper of land pulmonates. In A. glabratus the cellular components con- sist of: (a) fibroblasts, (b) pigment cells, (c) vesicular cells, and (d) mucous cells, and the fibrous components consist of collag- enous-like fibers and delicate fibrils on the muscle fibers {Gitter- fasern of Baecker). Two distinctly different types of connective tissue can be recog- nized. These are loose "vascular" connective tissue and dense connective tissue. Between these extremes, intermediate forms occur. The loose "vascular" connective tissue (Fig. 14) is characterized by an open network of slender fibroblasts which, in section, appear as numerous oval or irregularly round perfora- tions, the " Zirkulationsliicken" of Kisker (1923). These per- forations, or circulation spaces, hold the hemolymph of the snail, which stains a homogeneous pink with eosin Y (Fig. 14 a). Few additional cellular and fibrillar elements are present in this type of connective tissue. So-called "concretions" occur in the mesh- works. These are crystalline-like structures of irregular shape and size (Fig. 14 c) and stain light blue with hematoxylin or light pink with PAS. The loose "vascular" connective tissue is found characteristically in the dorsal wall of the coelomic cavity, in the pseudobranch, the liver and the ovotestis (Figs. 6, 7, 9). PAN : HISTOLOGY OP AUSTRALORBIS GLABRATUS 245 The dense connective tissue contains numerous fibroblasts, pig- ment cells and some amoeboeytes as well as large amounts of collagenous-like fibers. There is no fibroblastic meshwork, but the cellular and fibrillar elements are embedded in the ground substance to form a compact tissue mass. Smooth muscle fibers are also abundant in this type of connective tissue. Since the ground substance, fibroblasts and collagenous-like fibers all stain pinkish blue in hematoxylin-eosin preparations, the ground sub- stance may obscure the cytoplasm of the fibroblasts and also the collagenous-like fibers. " Zirkulationsliicken" are present, but they are small in size and number and are inconspicuous. The dense connective tissue is seen characteristically in the foot proper and in the core of the tentacles (Fig. 15) . Staining of this tissue with aldehyde fuchsin and acid orcein yielded negative results; therefore, as in land pulmonates (Baecker, 1932), ^l. glahratus is without true elastic fibers. Fibroblasts. The fibroblasts of A. glabratus appear to have the ability to transform into a variety of cell types and to be involved in the repair of damaged tissues. In section they are fusiform or spindle-shaped but may have several branching processes (Fig. 16). These processes attach to or end on nearby fibroblasts and form the meshwork of the loose "vascular" con- nective tissue. The elongated or oval nucleus of the fibroblast has a delicate nuclear membrane, one or two nucleoli and moder- ately rich, fine chromatin. In stained material, the cytoplasm usually appears to be scanty and can be visualized only at the poles of the nucleus. It stains an almost homogeneous, light pinkish blue in hematoxylin-eosin preparations. Various amounts of brown to brown-black pigment are fre- quently seen in the fibroblasts; and since intermediate forms may be seen, especially in pathological tissue (Fig. 17), it is considered that pigment cells may be derived from fibroblasts. The possible transformation of fibroblasts into amoeboeytes will be described later. Pigment cells. These cells approximate 15 x 21 microns (Fig. 17). They are very irregular in shape, being round, oval or elongated, and may bear processes, as described for this type of cell in land pulmonates by Baecker (1932). The nucleus is also irregularly shaped, may be round, oval or lentiform and is 246 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY frequently eccentric with relatively rich chromatin. The cyto- plasm contains acidophilic granules which may be obscured or completely replaced by brown to brown-black coarse pigment granules. These pigment granules react to various stains sim- ilarlj' to the pigment of the epithelial cells. While pigment cells are distributed throughout the connective tissue, they are es- pecially abundant in the rectal ridge and the renal ridge, and in pathological tissues. They may derive from fibroblasts as de- scribed in the preceding section. Their presence in large numbers was frequently connected with certain microbial infections (Pan, 1956). Marcuzzi (1950) described the pigment cells in A. glabra- tus as being excretory in function. Stein and Mackin (1955) reported that increased numbers of pigment cells, in oysters, were associated with certain infections. Vesicular cells. Baecker (1932) described and discussed vesicu- lar cells (Blasenzellen) in land pulmonates. Carriker and Bil- stad (1946) found them in Lymnaea stagnalis appressa. These cells are irregular in shape, being round, ovoid, elongate and spindle-form, and measure up to 70 x 35 microns (Figs. 18, 52). The nuclei are relatively small, quite regular in shape, usually round or oval, and contain few chromatin granules. Binucleated cells are seen frequently. The nuclei are often eccentric and may be attached to the cell membrane. The cell membrane usually is discrete and stains well with eosin, aniline blue or fast green. The characteristic homogeneously-staining cytoplasm may con- tain a fibrillar iietAvork which stains less intensely than the plasma membrane, and reacts strongly with PAS suggesting the presence of carbohydrate material. In hematoxylin-eosin prep- arations the cytoplasmic mass may appear condensed around the nucleus, thus creating a space in the cytoplasm. The vesicular cells are distributed in many parts of the body ; they form small groups in the radular carrier and the dorsal wall of the buccal cavity (Figs. 49, 52). The connective tissue of the liver and ovotestis also contains many of these cells. Schaeffer (Baecker, 1932) postulated that the vesicular cells serve primarily as elastic supporting structures due to the fluid contents enclosed in the thick cell membrane. Since they occur in large numbers in the radular carrier and in the dorsal wall of the buccal cavity, they appear to have this supportive function. PAN: HISTOLOGY OF AUSTRALORBIS GLABRATUS 247 However, the large amount of PAS-positive material also suggests other functions, such as the storage of carbohydrates. This latter function may be especially important in the connective tissue of the liver and ovotestis where it is probably connected with diges- tion and gametogenesis respectively, as suggested by Faust (1920). Mucous cells. Scattered, well-developed mucous cells occur in small numbers throughout the connective tissue proper in A. glahratus. They are also found concentrated in large numbers in two glands: (a) a foot gland and (b) a buccal gland. The buccal gland will be described in the section on the alimentary system. Unlike the land pulmonate (Baecker, 1932), the foot gland of A. gldbratus does not have an excretory duct. Each mucous cell comprises a secreting unit and secretes mucus through a gradually tapering process (Fig. 19). The cell is usually tear- drop or retort-shaped but may be pleomorphic. In size these cells vary considerably but usually measure about 23 microns at the widest diameter. The relatively round or oval nucleus is extremely rich in chromatin and possesses an eccentric, large nucleolus. When the cells are tilled with secretion materials, the nuclei tend to lose their normal structure and become pyknotic. As has been described by Baecker (1932) and Car- riker and Bilstad (1946) in other species, these cells contain various cytoplasmic structures depending on the stage of secre- tory activity. In hematoxylin-eosin preparations, the cytoplasm of actively secreting cells is filled with large basophilic granules which usually obscure the nucleus. These basophilic granules are strongly PAS-positive but do not show metachromasia with thionin or toluidine blue 0. They stain with orange G in Mallory s trichrome stain. After the secretory materials are released, the cytoplasm loses the basophilic substance, becomes finely reticular, and shows only weak reaction to PAS stain. In the foot gland, several of the mucous cells form subgroups, and the secreting processes bundle together and extend toward the epithelial sheet. The muscle fibers of the foot run through and between the subgroups. These muscles probably serve to force the secreted mucus through the intercellular spaces of the epi- thelial cells. 348 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Collagenous-like fibers. The connective tissue fibers of land pulmonates resemble the collagenous fibers of higher animals (Baecker, 1932) and are described as collagenous-like fibers from Lymnaea stagnalis by Carriker and Bilstad (1946). In A. gla- bratus these fibers are more delicate than the muscle fibers, and are differentiated poorly in hematoxylin-eosin preparations tak- ing a light pinkish-blue hue. They are colored blue by Mallory's triple stain and green by Gomori's trichrome, but they are not impregnated by Bielschowski-Glees' or Bodian's silver stain. They are more abundant in dense connective tissue than else- where. Fibers seen in the basement membrane of epithelial sheets are more delicate than the collagenous-like fibers in the connective tissue proper. The staining reaction of both types is similar. Delicate fibrils on the individual muscle fibers. Baecker (1932) described delicate fibrils {" Gitterfasern'') lying on the surface of sarcolemma of land pulmonates; these can be stained with aniline blue or by silver impregnation. The perimycium of A. glabratus contains similar fibers which encircle the muscle fibers. Although they stain with aniline blue in Mallory's tri- chrome, silver impregnation fails to stain them. They are ex- tremely delicate and recognizable with difficulty unless special stains are employed. From the staining characteristics, these fibrils are probably modified collagenous-like fibers. III. Muscular Tissue It has been pointed out by Olson (1942) that although the muscle fibers of the molluscs have been much studied, there is yet no real agreement as to their structure. Baecker (1932) summarized the work of previous investigators and stated that two types of muscle fibers, smooth and striated, are present in land pulmonates, and that the latter is a very primitive type. Marcuzzi (1950) described both smooth and striated muscle fibers in A. glabratus. "We observed three different types of muscle fibers in our preparations, but we have not been able to observe fibers which possess periodic striations comparable to those of higher animals. Baecker (1932) pointed out that PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 249 the muscle of land piilmonates does not react to stains exactly as that of the higher animals, and in this respect the muscle fibers of molluscs are regarded by him as somewhat related to collagenous-like fibers of connective tissue. This is also true in A. glahratus, in which muscle fibers, especially smooth fibers, may occasionally stain as do collagenous-like fibers. Thus, some of the smooth fibers or portions of smooth fibers may stain with fast green in Gomori's trichrome or with aniline blue in Mallory's triple stain instead of with Chromatrope 2 R or Orange G re- spectively. The three types of muscle fibers observed in A. glahratus are (a) granular muscle or heart muscle, (b) intermediate granular muscle, and (c) smooth muscle. Granular muscle or heart muscle. Marcuzzi (1950) reported that the heart muscle of A. glahratus is of the obliquely striated type of Plenk (1924). We could not demonstrate striations (cross, or oblique) in the muscle fibers of this organ by any of the ordinary cytological techniques (iron hematoxylin, phos- photungstic acid hematoxylin, toluidine blue 0, hemalum, phlox- ine B and eosin Y). In cross-section the fibers appear round, oval, angular or irregular in shape. The fiber is enveloped by a delicate, somewhat refractile membrane (plasmalemma or sar- colemma) which stains lightly basophilic. The fibers measure ±12 microns in thickness. No myofibrils could be recognized, but the fiber is filled with relatively coarse acidophilic granules. These granules are distributed quite evenly within the fibers, but may at times be concentrated in the peripheral zone (Fig. 20) . Occasionally these granules are seen to be connected by very delicate fibrils. In longitudinal sections (Fig. 20) the fiber is spindle-shaped with a thickened mid-portion at the site of the nucleus. The coarse acidophilic granules also are more or less concentrated in the peripheral zone. Myofibrils were not recognized in the longitudinal positions. The nucleus is single, round to oval in shape and has but little chromatin. There is no clear area around the nucleus, as occurs in vertebrates. The heart muscle fibers of A. glahratus anastomose freely to each other as in vertebrates, but there are no intercalated discs. Connective tissue is scarce, and the fibroblasts attach here and there on the muscle fibers. 250 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY The fibroblasts sometimes contain a small amount of brown granular pigment. There is no true endocardium, but the outer surface of the heart is protected by a well-developed epicardium (Figs. 33, 35). This type of granular muscle fiber is seen only in the heart of A. glahratus. Infrequently the granules in some fibers may be so arranged that striation is suggested. Marcuzzi (1950) ap- parently observed this condition. The heart muscle of A. glahra- tus forms only primary bundles. Intermediate gramdar type of muscle. This type of fiber is apparently an intermediate form between the granular muscle (heart muscle) and smooth muscle, and is confined to the buccal mass. It may measure up to 10 microns in thickness. In cross- section the muscle fibers appear more irregular in shape than in heart muscle and present round, oval, triangular and quad- rangular outlines. There are two clearly defined zones : a rather heavily stained peripheral zone and a lightly stained central zone (Fig. 21). The refractile sarcolemma is thinner than that of the heart muscle. Delicate myofibrils are evenly packed in the peri- pheral zone and stain intensely acidophilic. The central zone contains a number of coarse granules which are stained less acidophilic than the myofibrils. In longitudinal sections (Fig. 21) the fibers are spindle-shaped, having a slightly thickened middle portion. Myofibrils run paral- lel to the long axis of the fiber in the more heavily stained per- ipheral zone. The inner portion stains lightly acidophilic and contains evenly scattered, coarse granules which occasionally may be linked with fine threads. The granules are, as in the case of heart muscle, arranged at times so as to show an irregular, cross- striation-like appearance. This striated appearance is not con- stant and when it does appear in a preparation only a few such fibers show it. Each fiber contains one centrally-located nucleus at the thickened middle portion of the fiber. The nucleus is oval, poor in chromatin granules and usually possesses a nucleolus. This type of muscle fiber has been described for many molluscs by various authors, especially by Olson (1942) who studied it in the radular retractor of Busy con sp. :. PAN : HISTOLOGY OF AUSTEALORBIS GLABEATUS 251 The muscle of the buccal mass forms primary and secondary bundles. Small amounts of endomj^sium bind individual fibers to form primary bundles which are in turn bound by perimysium to form secondary bundles. These interstitial tissues are col- lagenous in nature and can be stained with PAS, Mallory's or Gomori's trichrome stains. Fibroblasts may attach on the sur- face of sarcolemma and may contain brown pigment granules. Smooth muscle. This type of muscle cell is comparable to that of vertebrates and constitutes the major musculature of A. glahratus. It is found in the columellar muscle, foot muscle, ali- mentary canal, genital tracts, tentacles, and other areas. Its fibers are the smallest of the three muscle types, measure less than 9 microns in thickness, and show considerable morphologic variation. In cross-section the fibers appear round to ovoid or angular in shape. In longitudinal sections they are fusiform with a thickened mid-portion containing the nucleus. The fibers stain uniformly acidophilic and do not contain granules except rarely about the nucleus. Myofibrils are abundant and can be recog- nized readily running along the long axis of the fibers in properly fixed specimens (Fig. 22). They are distributed evenly in the fibers and are packed together tightly. A sarcolemma cannot be recognized as such, but a very thin, refractile outline sur- rounds the fiber. The single nucleus is elongate or oval. It lies along the long axis of the fiber and is somewhat eccentric in position. There is usually a nucleolus ; the chromatin granules are small and few. Dark brown pigment may be present around the nucleus, espe- cially at its poles. Most of the fibers in the foot run singly in various directions through the dense connective tissue. In the columellar muscle the fibers form primary and secondary bundles which run along the long body axis. The endomysium and perimysium are more abundant here than in the buccal mass musculature. IV. Nervous System and Sensory Organs The nervous system of A. glahratus consists of central ganglia, peripheral ganglion, individual ganglion cells, and nerves. The sensory organs observed in this study are : statocysts, eyes. 252 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY osphradium, tentacles, and five groups of small sensorj'- cells, one each at the base of each tentacle and the margin of each lip respectively. Central ganglia. According to Baker (1911 and 1945) and Baecker (1932) the central ganglion ring of pulraonate snails is characteristically composed of 11 separate ganglia which are linked by communicating branches of commissures to form a ganglion ring around the esophagus immediately behind its union with the buccal mass. There are two cerebral ganglia above the esophagus, and below the esophagus there are two buccal ganglia, two pedal ganglia, two pleural ganglia, two visceral ganglia and a single abdominal ganglion (Baker 1945). The arrangement of the ganglia in A. glahratus is essentially the same as in other pulmonate snails. The paired visceral ganglia are not of equal size, the left ganglion being somewhat larger than the right one. All of the parallel ganglia and their commissures are covered by a common connective tissue sheath, the epineurium. Histologically, these ganglia are similar in arrangement and composition (Figs. 2, 6, 23). The epineurium is a relatively thick connective tissue layer which is rich in cellular and fibrous elements. The collagenous-like fibers are relatively coarse, loosely packed, and, for the most part, run parallel to each other along the long axis of the ganglion. A few cross fibers are also present. The cellular elements are similar to the fibroblasts of the connective tissue. Each ganglion is directly covered by a sheath, one cell thick, the perineurium. It stains more intensely than the epineurium with connective tissue stains such as Mallory's or Gomori's triehrome. Arteries and blood spaces are present in the epineurium. Vesicular cells seen in the epineurium of land pulmonates by Baecker (1932) were not observed in the ganglia of A. glahratus. Ganglion cells (Figs. 6, 23, 24) are located on the periphery of the ganglia and many are in contact with the perineurium; however, areas where the nerve root and commissure leave a ganglion are devoid of ganglion cells. The center of the ganglion contains neurofibrils sent out by the ganglion cells and is thus filled with neurofibrils which run in various directions in bundles. PAN : HISTOLOGY OP AUSTRALORBIS GLABRATUS 253 A few fibroblast-like slender cells occur among these neuro- fibrils in the ganglion and larger nerves. These cells under patho- logical stimuli may become hyperplastic and phagocytic and apparently are equivalent to the glia cells of the vertebrate cen- tral nervous system. According to Nabia (Baecker, 1932) two types of ganglion cells are present in the gastropod nervous system ; one, the so- called ordinary ganglion cell with a relatively large amount of cytoplasm, and the second, the chromatinic ganglion cell, with sparse cytoplasm and of small size. In A. glahratus the chroma- tinic ganglion cells form two prominent groups in the dorso- posterior corner of each cerebral ganglion. The nuclei of these cells, measuring 7x4 microns, are slightly larger than those of fibroblasts, and the cytoplasm is seen as a small rim around the nucleus. The so-called ordinary ganglion cells may measure 59.5 X 52.5 microns, with nuclei of 42.0 x 24.5 microns. The nucleus has a definite nuclear membrane, contains a large num- ber of coarse chromatin granules held in a linin network, and one or two nucleoli (Fig. 24). The nucleolus is located eccen- trically and may be oval. The cytoplasm contains a large amount of basophilic material in the form of coarse granules or small plaques (especially in thionin-stained preparations) which re- semble Nissl's bodies in the vertebrates. Although silver im- pregnation (Bielschowski-Glees' and Bodian's method) was not successful in impregnating neurofibrils, many of the Bouin-fixed preparations, when overstained with hemalum, contained fibrillar networks in the cytoplasm of the larger ganglion cells. Neuroglia cells ma}' surround the larger ganglion cells, but none of them was observed inside the cytoplasm of ganglion cells (i.e., the " Trophospongien" of Holmgren, 1905). The structure of the central ganglia, as descri])ed above, is somewhat similar to the central nervous system of higher animals in that the neurons are located in the peripheral zone (gray substance) and the neurofibrils are concentrated in the central zone (white substance). Baecker (1932) also pointed out this correlation in the central nervous system of land pulmonates. However, mye- linated nerve fibers were not demonstrated in A. glahratus; neither were they seen in land pulmonates by Baecker (1932). 254 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY This author also described neuroglia cells in the ganglia of land j)ulmonates and observed that the cells were concentrated in the peripheral zone of the ganglia. In .1. glahrntus, neuroglia cells are similar morphologically to fibroblasts and are scattered evenl}' in the ganglia in small numbers (Fig. 23). Hyperplasia and transformation of neuroglia into phagocytes were observed in a few snails which were infected with an unidentified yeast- like organism. In some instances partial or complete replace- ment of ganglion cells and neurofibrils of affected ganglia by neuroglia was also observed (Fig. 25). Peripheral ganglion cells and peripheral ganglion. Peripheral ganglion cells are scattered singly or in small numbers in various organs and tissues. These cells are morphologically similar to those in the central ganglia. There appears to be only one peripheral ganglion present in A. glabratus. This ganglion is associated with the osphradium (Fig. 29) and will be described later. Nerves. The structure of the nerve in A. glabra f us resembles that of the Helix nerve described by Baecker (1932). The bun- dles of neurofibrils are covered by the connective tissue sheath (peri- and epineurium) of the ganglion in the root area. The perineurium is lost shortly' below the ganglion. The epineurium is continuous with that of the ganglion and is relatively thick in the larger nerves. In small peripheral nerves the epineurium becomes one cell thick and has only a few collagenous-like fibers. Most of the fibers run along the long axis, but circular fibers are also present. These stain light blue with Azan triple stain, green with Gomori's trichrome, and pinkish-blue with hematoxylin- eosin. The neurofibrils stain bluish-red and dark red with Mallory's and Gomori's trichrome respectively. Neuroglia cells are present among the neurofibrils and are more abundant in larger nerves. They are usually spindle-shaped and run with or obliquely to the long axis of the nerve (Fig. 26). No myelin- ated neurofibrils were observed in A. glahratus. The nerve trunks and large nerves of A. glabra! us can be recognized without dif- ficulty in hematoxylin-eosin preparations by their staining characteristics and structures. In cross-section of the larger nerves a connective tissue substance may form incomplete septa PAN : HISTOLOGY OF AUSTRALORBIS QLABRATUS 255 at the periphery. Since no myelinated neurofibrils were demon- strated in A. glabratus, the nerve of this snail belongs to the "incomplete type" of Schultze (Baecker, 1932). The axis cylinders, or neurofibrils, in individual nerves, are embedded in a ground substance which takes up fast green or aniline blue. Statocysf. The statocyst, or the balancing organ of the pul- monates, has been described in detail for the genus Helix by Baecker (1932). The statocyst in A. glabratus appears to have structures similar to those of Helix. In Australorhis it is a paired organ located in the latero-posterior corner of each pedal ganglion at the root of the commissure to the pleural ganglion. It is an oval-shaped sac measuring approximately 94.5 x 66.5 microns (20 x 6.4 mm. snail )and lies embedded in the epineu- rium of the pedal ganglion (Fig. 23). Unlike Helix, the main nerve to the statocyst of A. glahratiis originates from the pedal ganglion, but it also receives a smaller nerve from the pleural ganglion. The wall of the statocyst consists of (1) an outer layer, (2) a middle layer and (3) an inner or epithelial layer. The outer layer is a connective tissue sheath, rich in circular collagenous-like fibers, and is fused into the epineurium of the pedal ganglion. The middle layer is a very thin homogeneous capsule which can be recognized only with a connective tissue stain. It is considered by Baecker (1932) to be the basement membrane of the epithelial layer in Helix. The inner or epithelial layer is composed of two types of epithelial cells which inter- mingle and form a membrane one-cell thick. One type of cell is small, flat and fibroblast-like with chromatin-rich nuclei but with little cytoplasm and indistinct boundaries. The other type includes cells which are half-moon in shape with the convex side on the basement membrane. They usually measure 25 x 12.5 microns and are the "giant cells" of Baecker (1932). The nucleus is small, oval, and is somewhat peripherally situated. On the basal side of the nucleus, the basophilic cytoplasm may contain large vacuoles supported by fibrillar structures. Neuro- fibrils may be traced into the cytoplasm of these larger cells but are not present in the smaller or first type. The free surface of the "giant cells" is covered with long, but sparse, cilia. The lumen of the statocyst sac contains ovoidal bodies, the statoliths. They measure 4x3 microns and stain lightly basophilic (Fig. 24). The peripheral zone appears denser than the central zone. 256 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Eyes. The eyes of pulmonates are well developed and are com- parable in structure to those of vertebrates. Smith (1906) made a detailed study of the eyes of pulmonates, and the more recent studies are summarized b}" Baecker (1932). According to the former author, land pulmonates have the following eye struc- tures : optic capsule, cornea, retina, lens, \'itreous humor, optic nerve and accessory retina. The eyes of A. glabraius are located latero-posteriorly to the base of each tentacle and are embedded in the dorsal wall of the head. All of the above-mentioned eye structures for pulmonates, except the accessory retina, were recognized in A. glabraius (Fig. 27). In addition, a few delicate smooth muscle fibers were observed to be attached to the optic capsule. The optic capsule consists of a connective tissue sheath wdiich is one cell thick on the surface of the cornea, but becomes two to three cells thick near the optic nerve where it fuses with the epineurium or connective tissue sheath of the latter. The cornea and retina together form a closed sac which constitutes a wall, one cell thick, enclosing the lens and the vitreous humor. Be- tween the cornea and the epithelium of the head there is a small space which is filled with tissue fluid and contains a few wander- ing cells or amoebocytes. The cornea is of non-pigmented, squamous epithelium and is in close contact with the lens on the inner surface and with the optic capsule on the outer surface. It covers approximately one-third of the surface of the eyeball. The retina (Fig. 27) makes up the remaining portion of the sac. As shown for Planorhis trivolvis by Smith (1906), there are three difl'erentiated regions in the layer of simple epithelium. The outer zone is in contact with the optic capsule and com- prises the non-pigmented but nucleated portions of the retinal cells. The middle region consists of the constricted portions of the sensory cells and the pigmented portions of the pigmented or supportive cells. The sensory cells in this region are usually obscured by the heavily pigmented, thick cytoplasm of the pigment cells. The inner zone is made up of the rods of sensory cells and is in contact with the vitreous humor. Both types of retinal cells (the sensory and the pigment) are attached to the optic capsule by radiculae (Smith, 1906). The pigment cells are PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 257 more slender and basophilic than the sensory cells at the periph- eral zone. The nuclei of the pigment cells generally are nearer the capsule and are smaller than are the nuclei of the sensory cells. The rods constitute those portions of the sensory cells which extend beyond the middle pigment zone, and are light receptors according to Smith (1906). In the P.F.F.-fixed and heraatoxylin- phloxine B-stained preparations, the rods appear as club-shaped objects surrounded by a radially striated thick mantle. The lens is spheroidal, occupies most of the ocular sac, and is strongly acidophilic. The peripheral zone of the lens is homo- geneous, but the central zone is porous, perhaps an artifact due to fixation and staining. The vitreous humor is very small in amount and stains lightly pink. The optic nerve originates from the cerebral ganglion and has the structure of a medium-sized nerve. Osphradiiim. Baecker (1932) did not describe the osphradium, or the so-called olfactory organ, as occurring in land pulmonates. The osphradium in A. glahratus is located where the mantle collar joins the neck of the snail between the median line and the pneumostome siphon (Fig. If). It is a somewhat elongated pear- shaped sac, about 300 x 120 microns in a snail measuring 15 x 5.1 mm. The opening of the osphradium is obscure in living specimens. In section, the lumen is lined with a layer of tall columnar epithelial cells covered with long, dense cilia (Figs. 28, 29). Two types of columnar epithelial cells can be identified. One has a basal oval nucleus which is rich in chromatin, and the cytoplasm is filled with basophilic granules. The second type is a slender cell compressed between cells of the former type and has a central elongated nucleus. The cytoplasm is scanty and cannot be clearly recognized. The epithelium of the osphradium is replaced by the cuboidal cells of the neck and mantle surfaces at the opening to the exterior. The epithelium rests directly on a layer of smooth muscle fibers (mostly circular and some longi- tudinal) that contain a few fibroblasts (Fig. 28). The lower portion of the sac is surrounded by a peripheral ganglion (Fig. 29). The structure of this peripheral ganglion is similar to that of the central ganglia. Neurofibrils from these ganglion cells 258 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY appear to penetrate through the muscle layer and end on the epithelial cells. The ganglion attached to the osphradium receives a thick branch nerve from the left visceral ganglion. Tentacles. There is but one pair of tentacles in A. glabratus. The tentacle is the most abundantly innervated and the most flexible structure of the snail. Thus, it is a very delicate tactile sense organ. A thick nerve trunk arises from the cerebral ganglion and, after reaching the root of the tentacle, passes through the central core of the tentacle to the very tip. Anatom- ically, the tentacle is a gradually tapering cylinder. On the cross-section it has a round contour except at the base where there is a leaf -like enlargement on the lateral side (Fig. 30). The core of the tentacle near the root consists of dense connective tissue which becomes less dense toward the tip. A central artery runs up the core to the tip where it empties into the peripheral blood sinuses. The connective tissue of the core sends out radial strands of fibroblasts to the epithelial sheet (Fig. 31). The nerve trunk is embedded in this core of connective tissue and also sends out many branches to the epithelium along the radial connective tissue bridges ; the bridges also contain delicate muscle fibers. Longitudinal muscle fibers and pigment cells are abun- dant in the central connective tissue core. The hemolymph fills the interstices between the central core and the peripheral epi- thelium. The simple short columnar-covering epithelium bears dense cilia. The basement membrane is not distinct toward the tip. Beneath the basement membrane there are a few delicate circular muscle fibers intermingled with fibroblasts. Postulating from the histological structure, the extension of the tentacle is probably accomplished by the filling of the blood sinuses with hemolymph through the central artery. The retraction of the tentacle is apparently accomplished b}^ the contraction of the longitudinal and circular muscle fibers with resultant emptying of the hemolymph from the sinuses. Sensory cells at the margin of the lips and at the bases of the tentacles. There are five specialized groups of sensory cells, one associated with each leaf-like enlargement at the base of the tentacle and one at the margin of each lip. These cells are subdivided into many clumps by fibroblasts and are in close con- tact with the basement membrane (Fig. 32). They resemble the PAN : HISTOLOGY OF AUSTRALORBIS GLABEATUS 259 ehromatinic ganglion cells of the central ganglia. These sensorj^ cells are short, fusiform in shape, contain chromatin-rich ovoid nuclei and have a verj- basophilic, vacuolated cytoplasm. One end of each fusiform cell fuses into the bundle of neurofibrils sent out from the branches of the nerve innervating the tentacle. The other end of the cell is directed peripherally and tapers oif into a filament. Many of these cells are in direct contact with the epithelial cells of their respective areas. Although the exact nature of these cells is not clear, they appear to function as peripheral sensory cells. V. Circulatory System According to Baker (1945) the circulatory system of planorbid snails consists of a heart, arterial system, venous system, and blood sinus system. Histological description of these organs in land pulmonates is given by Baecker (1932). Our histological studies of the circulatory system of A. gla- bratus fully agree with Baker's anatomical observations. How- ever, the "loose vascular" connective tissue should be men- tioned as an integral part of the circulatory system. Heart. The heart consists of two chambers : a caudal pear- shaped, muscular ventricle, and a cranial pear-shaped, thin- Avalled atrium. These two chambers are joined at their wide bases by a constriction where there is a pair of muscular valves (Fig. 33). The valves are directed into the ventricle and are thin muscular sheets covered by cells resembling those of the epi- cardium. The junction between the ventricle and aorta is also provided with a thin muscular valve which is directed into the aorta (Fig. 34). The muscle of the heart is of the granular type already described. The atrium has a very thin muscular wall and is more dis- tensible than the ventricle. In sectioned material the lumen of the atrium is usually several times larger than that of the ventricle. The thick muscular wall of the ventricle is formed by a three- dimensional mesh of branching and anastomosing muscle fibers, slightly more densely woven adjacent to the epicardium. The wall of the heart (both chambers) consists principally of longi- tudinal and circular muscle fibers. Some of the branching fibers 260 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY form trabeculae that cross the lumen to the opposite wall. Fibro- blast -like, fusiform cells occur among the muscle fibers and are especially prominent in certain pathological conditions. In Helix, Baecker (1932) regarded these fusiform cells as peri- mysium and not as endocardium. There appears to be no true endocardium in A. glabraius. The outer surface of the heart is covered with a continuous layer of cells which comprise the epicardium (Figs. 33, 35). These cells are round to short, columnar in shape, with distinct boundaries when the heart is contracted, but become flat and with- out clear boundaries when it is distended. In the contracted position of the heart, the nuclei of the epicardial cells are round to oval, subbasal in position and poor in chromatin granules. The cytoplasm is finely or coarsely vacuolated and contains baso- philic filamentous material (Fig. 35). The structure of the epi- cardium appears to be adapted for great distention. The heart is enclosed in the pericardial sac which is connected to the lumen of the saccular portion of the kidney by a reno- pericardial canal (see section on kidney) (Fig. 46). The peri- cardial sac is bordered by a portion of the saccular kidney, part of the mantle and by membranous tissue which is an extension of the mantle into the mantle cavity. The internal surface of the sac is lined by a sheet of flat cells which contain a moderate amount of dark brown pigment. Arteries. In large arteries, including the aortae, the wall is composed of three layers (Figs. 36, 37). The lumen is lined with a layer of fibroblast-like cells which are very thin and barely recognizable under most conditions. Baecker (1932) stated that these lining cells as well as the connective tissue cells of the heart, are not truly endothelial in nature. From our observations in Australorhis, these cells may undergo hyperplasia and hyper- trophy under pathologic stimuli ; they then may become detached and transformed into phagocytes (Figs. 36, 40). Therefore, we believe these lining cells bear endothelial properties. Beneath the lining cells there is a laj^er of smooth muscle fibers, most of which are circular, but longitudinal fibers are also present. The muscular layer is covered with a varying amount of connective tissue. In small arteries the muscle and connective tissue layers may be absent, but the lining cells are well defined. PAN : HISTOLOGY OF AUSTBALORBIS GLABRATUS 261 Veins. The veins have no definite wall and are tissue spaces which are lined incompletely with fibroblast-like cells (Fig. 38) ; thus the veins cannot be ditt'erentiated from the surrounding tissues. The veins connect freely with the blood sinus system. Blood sinuses. The blood sinuses are tissue spaces which are interlaced abundantly by fibroblasts, some of which form a trabecula-like support. The most conspicuous blood sinuses are those in the mantle (Fig. 39). The blood or hemolymph appears to be forced into the "loose vascular" connective tissue via arteries, and after bathing the organs and tissues is collected in the blood sinuses. It is finally returned to the atrium of the heart via the pulmonary and renal veins. Both of these veins pass along the lateral sides of the kidney and unite near the blind end of the saccular portion of the kidney before entering the atrium of the heart. The hemolymph in the "loose vascular" connective tissue is apparently squeezed forward by the smooth muscle fibers present in that tissue, as described by Baecker (1932). Our description of the vascular system of A. glahratus essentially agrees with that given by Baecker for land pulmonates. However, hyperplasia and trans- formation of the lining cells into amoebocytes are probably noted for the first time in the pulmonate snails. Amoebocytes or wandering phagocytes. In Helix, Baecker (1932) described nucleated blood corpuscles which he regarded as equivalent to the leucocytes in higher animals; he called these nucleated blood corpuscles "itwioc&oc^^^n" and these were said to have phagocytic functions. In A. glahratus, only nucleated blood corpuscles occur and these cells closely resemble the amoebocytes described by Baecker in Helir (Fig. -40). Several kinds of amoebocytes have been described in molluscs by various workers (George and Ferguson, 1950; Wagge, 1955). Although there may be different types in A. glahratus we did not attempt their differentiation, and the nucleated cellular components of the hemolymph are all included as amoebocytes. They occur in small numbers in the circulatory system as well as in connective tissue. Because of the semi-open circulatory system of A. glahratus, the amoebocytes are normally fairly evenly distributed in the con- nective tissue, but in certain pathological states a large number may localize in diseased tissue areas (Fig. 41). In one type of 262 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY inflammatory reaction observed, a large number of micro- organisms were present in the cytoplasm which then became rounded and enlarged from two to three times its normal size (Fig. 42). In sections, the amoebocyte of A. glabratus measures approximately 9 x 12 microns, but there is considerable variation in size and shape. However, many are round or oval with occa- sional lobose pseudopodia. The nucleus is vesicular, and round, oval or lentiform. It contains a moderate amount of coarse granu- lar chromatin and is usually eccentric in pcsition (Fig. 40). The cytoplasm is lightly basophilic and granular and usually shows coarse vacuolation. Hemopoietic tissues. According to Baecker (1932), the origin of the amoebocyte in land pulmonates is unknown. However, recent workers have observed that the fibroblasts and epithelial cells of the mantle in Helix sp. transform into amoebocytes under certain conditions (Haughton, 1934; Crawford and Barer, 1951; Wagge, 1951, 1955). No mention of specially differentiated hemo- poietic tissues was made by these workers. Haughton (1934) indicated that the blood vessels in the invertebrates were the place of origin of the amoebocytes. Our observations suggest that possible normal sites of production for amoebocytes are the blood sinuses and the wall of the saccular portion of the kidney which forms part of the pericardial sac. The fibroblasts which form the trabecula-like supports of the blood sinuses in the mantle frequently round up, but remain attached to the wall of the sinus l;y a cytoplasmic process. The typical elongated nucleus of the fibroblast is lost in this transition form and becomes vesicular as in the mature amoebocyte. The wall of the saccular portion of kidney bordering the pericardial cavity is composed of primi- tive tissue probably of mesenchymatous origin (Fig. 43). The cellular components are round or oval in shape but with irregular outlines and with processes which join with those of the neighbor- ing cells to form a cellular reticulum. They are fairly closely packed and imbedded in a ground substance which contains some collagenous-like fibers. The cytoplasm is lightly basophilic, may contain several vacuoles, and in general closely resembles that of the mature amoebocyte. The nuclei are vesicular and also re- semble those of amoebocytes. Mitotic figures are frequently seen in this tissue. The cellular reticulum has numerous blood spaces PAN: HISTOLOGY OF AUSTRALORBIS GLABRATUS 263 which are connected with the blood sinuses of the saccular portion of the kidney. The appearance of this tissue resembles closely the medulla of lymph nodes in the vertebrates. Extreme hyper- plasia in this tissue was noted in certain patholooical conditions and the cellular components were observed to contain many microorganisms. Amoebocytes free in the blood spaces were regularl}^ seen in this tissue. From its histological structure as well as its behavior in pathological states, this tissue may be regarded as hemopoietic tissue or lymphoid tissue, and it is de- scribed for the first time in this study. In pathologic conditions, the fibroblasts in various parts of the body of A. glabratus, especially the rectal ridge and the kidney ridge, were also observed to participate in production of amoe- bocytes. Under such circumstances transition forms are fre- (juently noted in the connective tissue. Although the amoebocyte has been described by various authors as being concerned with digestion and wath the repair of damaged shells (Yonge, 1946; Wagge, 1951; George, 1952; Wagge and Mittler, 1953), we found little activity of amoebocytes in A. glabratus Avhich may be correlated with these functions. We observed few amoebocytes around the digestive tract of this snail. The main activity of the amoebocytes in A. glabratus, according to our observations, appears to be phagocytosis of foreign bodies, especially microorganisms. VI. Respiratory System The respiratory system of A. glabratus consists of a pneumo- stome equipped with a pneumostome siphon (pseudob ranch), and the mantle cavity. The pneumostome is surrounded by a portion of mantle collar and the neck. It is loosely divided into two openings by the pseudobranch which protrudes out of the mantle cavity. The pneumostome siphon occupies the right opening and is actually a curled, small flap of tissue arising from the neck and possessing a moderately loose vascular connective tissue, as stroma. The epithelial cells of the mantle collar and the siphon are cuboidal to short columnar in type. In the stroma of these structures, pigment cells and smooth muscle fibers are present in moderate amounts. 264 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY The wall of the mantle cavity is provided with three ridges which protrude into the lumen and apparently serve to increase the respiratory surface. These are the kidney ridge, rectal ridge, and tlie dorsal ridge which is situated between the kidney ridge and the rectal ridge on the left side. The first two ridges con- tain two veins each and the dorsal ridge contains one vein. In each case, the veins run along the long axis of the ridge. Al- though the epithelium in the mantle cavity serves as respiratory epithelium, no special differentiation of these cells from the other, non-respiratory epithelia could be recognized. The cilia of the tall columnar epithelial cells on the summits of the three ridges apparently serve to guide the flow of water which enters the mantle cavity. The three types of epithelial cells of the mantle cavity surface have been described earlier. Since the sup- porting connective tissue underlying the basement membrane of the area between the ridges is very thin and the epithelium is therefore in close contact Avith the blood sinus system, the cuboidal epithelial covering of this area probably has an active respiratory function. Baecker (1932) described this form of respiratory cell in land pulmonates. but the tall ciliated columnar epithelial cells on the summit of the ridges were not men- tioned. A few delicate smooth muscle fibers are present beneath the basement membrane. In addition to the veins in the ridges, the mantle is supplied with a bloood sinus system (Fig. 38) which communicates freely with the venous system and is also in inti- mate contact with the respiratory epithelium. The three ridges freciuently have been observed to be invaded by microorganisms with resulting hyperplasia of fibroblasts and amoebocytes. VII. Renal Organ Baker (1945) described the kidney of the Planorbidae as con- sisting of a small upper saccular portion, and elongated lower tubular portion, and a short ureter curving nearly 160° to the left before it opened into the mantle cavity at the pneumostome. Baecker (1932) considered the tubular portion of the kidney in land pulmonates as a secondary ureter, while Abdel-Malek (1952) held that both portions of the kidney in Planorbidae are similar PAN : HISTOLOGY OF AUSTBALORBIS GLABRATUS 265 histologically but different anatomically. In A. glabratus the two portions of the kidney are histologically and anatomically distinct. Tuhular portion of the kidney. The characteristic epithelial cells of the tubular portion of the kidney are low columnar to cuboidal and rest on a thin basement membrane which is sup- ported by a sheet of fibroblasts and a few smooth muscle fibers (Fig. 44). These epithelial cells thus border on the venous and sinus systems and are, therefore, in intimate contact with a rich supply of blood. The epithelial sheet of this portion of the kidney shows an irregular wavy appearance in cross section. The cells are somewhat variable in appearance. In well-fixed material the nuclei are usually round or oval, very rich in chromatin granules and vesicular in type (Fig. 44). They are usually located near the peripheral zone. The cytoplasm is vertically striated from the superior surface to the basement membrane. At times a few small vacuoles may be present in the cytoplasm near the lumen. Again, the vacuoles may be numerous, coalesce to form larger vacuoles, and occupy about two-thirds of the cell. Thus, the nuclei may be forced toward the basement membrane and the acidophilic striations then become delicate, displaced, and con- fined to the basal third. It is possible that these striations are intracellular canaliculi. The vacuoles contain variable amounts of PAS-positive materials. The ureter is lined with the same type of epithelial cells as the tubular portion except at the area about the opening where the cells show a transition to the epithelium of the mantle cavity. Saccular portion of the kidney. The wall of the saccular por- tion forms many prominent folds which run with the long axis of the kidney. The folds may reach the opposite wall. The epi- thelial cells of the wall of the saccular portion are columnar and are taller than those of the tubular portion but are arranged in the same manner. The folds consist of two epithelial sheets with a blood space between them (Fig. 45). The nuclei are round to oval and are basal in position. In each epithelial cell there is usually a large vacuole which frequently contains a crystalline concretion. The vacuole usually pushes most of the cytoplasm to the basal third of the cell. Unlike the cytoplasm in the tubular portion, the cytoplasm of the cells in the saccular 266 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY portion contains coarse acidophilic granules and filamentous material. Little PAS-positive material is present in the vacuoles. The crystalline concretions were thought by Baecker (1932) to be the urine-substance in land pulmonates. They do not take ordinary stains. They are lightly refractile, round bodies with a central core and appear yellowish-brown in hematoxylin-eosin preparations (Fig. 45). In fresh specimens the presence of numerous concretions gives a yellowish-orange color to the saccular portion of the kidney. The concretions are very weakly PAS-positive. The lumen of the saccular portion is connected with the pericardial sac by the "renopericardial canal." This canal has been described in Helix by Niisslin and in Avion and Limax by von Rolle (Baecker, 1932). The canal in Australorhis is lined bj' a sheet of acidophilic cuboidal epithelium at the opening to the pericardial sac. These gradually transform to tall columnar cells toward the renal opening. The free surface of these cells is heavily covered with long cilia (Fig. 46). In specimens prepared in alcoholic fixatives, such as Newcomer's or Carnoy's, and stained M'ith thionin, a few mucous cells may be observed scattered among the epithelial cells. The mesenchymatous tissue in the wall of the saccular portion of the kidney, which forms part of the wall of the pericardial sac, has been described in the section dealing with amoebocytes. VIII. Alimentary System The histology of the alimentary system in pulmonates has been described in detail by Baecker (1932) and von Haffner (1923) for Helix sp. and by Carriker and Bilstad (1946) for Lymnaea siagnalis appressa. Baker (1945) described the gross anatomy of the alimentary system of Australorhis glahratus and Marcuzzi (1950) studied its histology. In general, this system in A. gla- hratus is comparable to that of Helix or Lymnaea, but with cer- tain variations. Morphologically and functionally, the alimentary system of A. glahratus can be divided into two parts: the digestive tract and the glandular organs. The digestive tract includes the PAN : HISTOLOGY OP AUSTRALORBIS GLABRATUS 267 buccal mass, the esophagus, the stomach and the intestine. The glandular organs comprise three glands : the buccal, the salivary, and the so-called ''liver." The general structure of the digestive tract is the same through- out, with variations as to size and shape. The lumen is lined with a sheet of simple columnar epithelium which rests on a basement membrane. The basement membrane is supported by two layers of smooth muscle fibers, an inner longi- tudinal and an outer circular layer. The thickness of each layer varies considerably. A sheath of connective tissue envelopes the outer muscle layer in the region where the digestive tract lies free in the body cavity. Lips and oral cavity. The oral cavity is bordered externally by three lips, two upper and one lower, located in front of the buccal mass. In cross section, three lips form a "T "-shaped space, which with the buccal mass comprises the oral cavity. The marginal epithelium is composed of ciliated, tall columnar cells like those on the foot surface. These cells are gradually replaced toward the buccal mass cavity by cuticular cells. The cuticular layer which appears bluish-gray and homogeneous in hema- toxylin and eosin preparation thickens toward the buccal mass and is transformed into three wedge-shaped horny jaws at the margin of the buccal mass (Fig. 47). The horny jaws stain orange to orange-pink in H-E preparations and appear striped, each stripe arising from individual epithelial cells. Beneath the epithelial sheet is a distinct basement membrane resting on thick dense connective tissue. Delicate smooth muscle fibers run in various directions in the connective tissue. In it are also em- bedded many mucous cells as well as a group of sensory cells (Fig. 32). A small amount of brown pigment may be observed rarely in the epithelial cells. Buccal mass. The buccal mass, called the "pharynx" in Helix by Baecker (1932), functions primarily as a scraping and swal- lowing organ (Fig. 2). In addition to the complicated muscular layers it contains three distinct structures: i) the chitinous radular ribbon, ii) the radular sac, and iii) the radular carrier. The histology of the muscle of this organ has been described in an early paragraph (see Muscular Tissue). The epithelial 268 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY sheet which covers the oral cavity consists of tall columnar cnticu- lar cells except on the floor where interrupted by the radular carrier. The cnticular cells contain centrally located oval nuclei which are rich in coarse chromatin granules (Fig. 48). Manj^ acidophilic striations are present in the slightly basophilic cytoplasm, and these run from the basement membrane to the cuticular laj^er. They contain larger amounts of coarse granules of brown pigment than do the epithelial cells of the lips. The pigment granules are usually located in the zone between the cuticular layer and the nuclei. The cuticular layer appears bluish- gray in H-E preparation, is strongly positive with PAS-stain, and decreases gradually in thickness toward the esophagus where it is replaced by cilia. A group of vesicular cells which resemble those of the radular carrier is embedded in the dorso-anterior wall of the buccal mass and probably serves as an elastic cushion (Fig. 49). These cells will be described in a later paragraph. Radula. The radula, a structure of taxonomic value, is a ser- rated chitinous ribbon originating in the radular sac. It passes forward on the floor of the buccal mass toward the lower lip and terminates near the horny jaw. The radular ribbon is sup- ported by the odontophoral cartilage (or radular carrier) which acts as an elastic cushion to control its movement along with the radular protractor and retractor muscles. The radula con- sists of two parts, the serrated chitinous dentins, and a thin basal plate. The histology of the radula of Anstralorhis is essentially similar to that of Helix (Baecker, 1932). The dentins are acidophilic and are embedded in the thin, homogeneous basal plate which is lightly basophilic. A thin layer of connective tissue binds the radular ribbon to the dorsal surface of the radular carrier. Rachdar sac. The radular sac is located at the ventro-posterior portion of the buccal mass near the junction of the buccal mass and esophagus. In cross section, it is roughly horseshoe-shaped with the convex side facing ventrally (Fig. 50). It is partially embedded in the musculature of the buccal mass and, together with the latter, is covered by a common connective tissue sheath. The radular ribbon is on the periphery and follows the outline of the horseshoe, being held between two sheets of epithelial cells. Both sheets appear to be an extension and modification of the PAN : HISTOLOGY OF AUSTRALORBIS QLABRATUS 269 epithelial sheets of the buccal cavity. The epithelial cells be- tween the radular ribbon and the outer connective tissue sheath consist of a layer of columnar cells (Fig. 51) which, in Lymnaea stagnalis, were called the "subradular epithelium" by Carriker and Bilstad (1946). These cells contain central, ovoid nuclei and basophilic cytoplasm and apparently give rise to the basal plate of the radular ribbon. The boundary between the subradular epithelium and the basal plate is largely obscured. Acidophilic striations occur in the cytoplasm in the zone between the nuclei and the basal cell membrane. No basement membrane is present. The other epithelial sheet (the supraradular epithelium) consists of cuboidal or low columnar cells, with more or less central, round nuclei (Fig. 51). These cells also contain rich basophilic gran- ules and apparently give rise to the serrated dentins. Carriker and Bilstad (1946) described syncytium formation of these cells in Lymnaea, but the cell boundaries of the supraradular epi- thelium in Australorhis are clearly apparent. The core or center of the radular sac is filled with a special type of supportive tissue, the collostyle (Fig. 50). Baecker (1932) described this particular tissue in Helix as "Gallertge- wehe" and Carriker and Bilstad (1946) refer to it in Lymnaea as a gelatinous-like supportive tissue. The cells of the collostyle are irregular in shape and size, being round, ovoid, elongated or spindle-shaped and measuring 10 x 4 microns to 64 x 24 microns. They are packed together like epithelial cells. The nuclei are relatively small, round or oval, and poor in chromatin. The cytoplasm is lightly acidophilic and homogeneous but may con- tain filamentous material at times. Little PAS-positive material was demonstrated in these cells, but they stain weakly with alcian blue 8 GS. Muscle fibers of the radular retractor join the collo- style at the ''opening" of the horseshoe (Fig. 50). Radular carrier. The radular carrier is frequently referred to as the odontophoral cartilage and is composed of three elements bound together by connective tissue to form the shape of a boat ; the floor of the boat is on the dorsal side to support the radular ribbon. The three pieces of tissue also form an enclosed sac which is connected to the circulatory system by a large blood vessel (Fig. 53). Since the radular carrier appears to have a pumping action, 270 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY and since an accessory heart has been reported in other mol- luscs (Michelson, 1956), the relationship of the lumen of the radular carrier with the circulatory system in A. glabratus sug- gests the possibility that it serves as an accessory heart in addi- tion to possessing a supportive function. Histologically, the radular carrier is composed of vesicular cells and muscle fibers. The muscle fibers are the same type as those in the wall of the buccal mass, and run vertically to the surface of the radular carrier. The vesicular cells are polygonal, measure 70 x 30 microns and usually have relatively small nuclei (7 microns) which are eccentric and poor in chromatin (Figs. 18, 52). The cytoplasm is enclosed in a conspicuous cell membrane and con- tains a network of fine acidophilic filaments and some amorphous material. Both the fibrillar network and tiie cloudy material are strongly stained with PAS technic. Binucleated cells are some- times seen. The vesicular cells in the antero-dorsal wall of the buccal mass are smaller than those in the radular carrier, but otherwise are similar. Both the inner and outer surfaces of the radular carrier are covered with thin connective tissue sheaths (Fig. 52). The nature of the vesicular cells and their functions have been discussed by Baecker (1932) and are thought to be primitive cartilage cells (the prototypes of chondrocj^tes) and to serve as an elastic cushion for the radular ribbon. Esophagus. The esophagus originates at the dorso-posterior wall of the buccal mass and runs posteriorly, parallel to and along the right side of the genital tracts. It joins the crop in the vicinity of the albumen gland. The ciliated simple epithelium is arranged in several longitudinal folds (Fig. 54). These folds clearly differentiate the esophagus from the postintestine where the folds are circular (Fig. 60). The epithelium consists of tall, uniform columnar cells (33 microns in a snail of 14.8 x 5 mm.) throughout the entire length of the esophagus (Fig. 55). They contain basal oval nuclei which are very rich in chromatin. The cytoplasm is filled with basophilic granules and may contain several vacuoles. The external surface is covered with long dense cilia (6 microns). Mucous cells or goblet cells are usually not found among the epithelial cells but do occur beneath the basement membrane in the muscle layers. The goblet cells meas- ure 10 microns at the largest diameter, are tear-drop in shape and PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 271 contain relatively large nuclei (6 microns), rich in chromatin. In hematoxylin-eosin preparations these cells are difficult to recognize, but they stand out conspicuously in PAS-stained prep- arations. The goblet cells are morphologically distinct from buccal gland cells in that the former are smaller in size, are embedded singly among the muscle fibers and the cytoplasm is not basophilic. They secrete PAS-positive materials through the intercellular spaces of the epithelial lining. Beneath the thick basement membrane are two layers of smooth muscle fibers ; an inner longitudinal, and an outer circular layer. The muscle layers of the esophagus are thicker than those of the intestine. The longitudinal layer is thicker than the circular layer toward the buccal mass, but this characteristic is reversed near the crop. Stomach. Baker (1945) recognized three anatomically distinct parts in the stomach of planorbids : 1) the crop, 2) the gizzard and 3) the pylorus. Histologically, the crop and pylorus are similar. The simple epithelium of the two parts is composed of columnar cells (25 microns in an 18 x 5.5 mm. snail) and many goblet cells (Fig. 56). The chromatin-rieh nuclei are oval and subbasal in position. The cytology of the epithelial cells does not differ materially from that of the esophagus except in size ; the region also has many more goblet cells. Between the thin basement membrane and the outermost connective tissue sheath, there is a type of supportive tissue which does not resemble any other connective or supportive tissue observed in this snail (Pig. 56). It may attain 700 microns in thickness (18 x 5.5 mm. snail), but varies depending on the degree of distention of the organ. It resembles poorly differentiated mesenchymatous tissue with scarce cellular ele- ments but is rich in collagenous-like fibers and ground substance. Two types of cellular elements are present in addition to smooth muscle fibers and amoebocytes. One of the cellular elements re- sembles a fibroblast and is more or less concentrated near the basement membrane. The other consists of irregularly shaped cells which vary considerably in size (4-18 microns). The nuclei of the latter cells are round or oval, usually eccentric in position and generally very poor in chromatic material. In hematoxylin- eosin preparations, the cytoplasm may appear vacuolated and 272 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY contain a few lightly acidophilic granules and brown pigment. The cell is, for the most part, obscure and difficult to recognize without special staining technics. The cell membrane stains green in Gomori's trichrome, appearing to enclose a vacuolated cytoplasm containing a nucleus. The vacuoles are filled with PAS-positive material in the form of granules or droplets. The ground substance is lightly basophilic, homogeneous and stains lightly with PAS technic. Collagenous-like fibers are lightly refractile and poorly stained in hematoxylin-eosin preparations but take a deep blue in Azan stain. The smooth muscle fibers are not layered as in the intestine or esophagus, but run in various directions embedded in the ground substance of the supportive tissue. The gizzard has the thickest muscular wall of the entire di- gestive tract and contains at least 10 layers at its middle portion. The inner surface is covered with simple columnar epithelial cells which measure up to 25 microns (18 x 5.5 mm. snail) and contain subbasal, oval nuclei. Except for the cuticular layer on the lumen surface of the epithelial cells of the gizzard these cells resemble those of the crop and esophagus. However, no mucus- secreting goblet cells were observed in the walls of the gizzard. Circular, longitudinal, oblique and radial muscle layers alternate in the wall, and weave a conspicuous pattern characteristic of the gizzard (Fig. 57). Cuticular epithelial cells have been ob- served in Lymnaea stagnalis (Carriker and Bilstad, 1946) and in Helix (Baecker, 1932) in the stomach. Many sand grains are usually present in the gizzard. Thus, the gizzard appears to serve primarily as a mechanical grinder of ingested foods. Intestine. Baecker (1932) subdivided the intestine of Helix into Duodenum, Blindsack and Enddarm. We prefer, however, to subdivide the intestine of A. glahratus after the scheme used by Carriker and Bilstad (1946) for Lymnaea stagnalis, i.e., prointestine, midintestine, postintestine and cecum. Prointestine. The prointestine makes a circular loop around the stomach after leaving the pylorus and is gradually trans- formed into the midintestine at the right side of the stomach. The epithelial sheet is most conspicuous in this portion of the intestine, occupies more than three-quarters of the intestinal wall, and displays an active secretory function. The epithelial PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 273 cells form a pseudostratified epithelium which may reach 88 microns in thickness in a snail of 18 x 5.5 mm. The nuclei are oval, are rich in chromatin, and lie at several levels (Fig. 57). The cytoplasm of some of these epithelial cells is filled with numerous basophilic granules, but in the majority it appears alveolar, the spaces being filled with a cloudy substance. Since this substance is strongly PAS-positive and is constantly dis- charged into the lumen of the intestine, it is likely that the prointestine is a secretory zone and corresponds to the glandular portion of the intestine of A. glabratus described by Marcuzzi (1950). The free surface of the epithelial cells is covered with dense short cilia (Fig. 58). The basement membrane is thin and difficult to discern. The muscular layers are also very thin and are composed almost entirely of circular fibers. The con- nective tissue sheath is relatively thick and consists of two to three layers of fibroblasts which may contain a few brown to brownish-black pigment granules. Midintestine. The midiutestine, beyond the prointestine, runs posteriorly along the dorsal side among the liver lobules, makes a "U" turn in the anterior third of the liver and continues anteriorly on the ventral side to become the postintestine near the pericardial sac. The intestinal wall is thinner in this portion than in any other area of the intestine including the esophagus. The epithelium is composed of the same type of ciliated columnar cells (18 microns in 18 x 5.5 mm. snail) as in the pylorus; how- ever, there are more goblet cells (Fig. 59). There is no folding of this simple epithelium. The inner longitudinal muscle layer, almost absent in the wall of the prointestine, becomes more prominent in this portion. Postintestine. The postintestine originates in the vicinity of the pericardial sac and is distinguished from the latter by circular folds which are formed by the epithelial sheet. It then enters the loose vascular connective tissue at the base of the central ridge (rectal ridge) and passes forward to the anus at the left rim of the pneumostome. The circular folds are con- spicuous in longitudinal section and readily differentiate the postintestine from other portions of the digestive tract (Fig. 60). The columnar epithelial cells are taller than those of the midintestine and measure 42 microns in a snail of 18 x 5.5 mm. 274 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY (Fig. 61). The epithelial cells are similar in morphology to those found in the esophagus and the raidintestine, but the mucus- seereting goblet cells are fewer than in the midintestine. The muscle layers shift their relative position as compared to the midintestine, so that the inner layer becomes circular and the outer layer longitudinal. Toward the anus (Fig. 62), the circular muscle fibers increase in number and form the anal sphincter. The basement membrane around the anus thickens considerably and contains much fibrillar material; some of the fibrils appear to run into the basal zone of the epithelial cells. Cecum. The blind sac or cecum opens into the prointestine at its junction with the pylorus and shares a common opening with the hepatic duct. There are two low, opposing, longitudinal folds which, as in Lymnaea stagnalis (Carriker and Bilstad, 1946), delimit the incurrent and the excurrent tubules within the cecum. In addition to these folds the epithelial cells also form many fold-like circular elevations similar to those of the hepatic duct. Those cells covering the wall of the incurrent tubule, which is on the side of the hepatic duct, resemble the epithelial cells of the hepatic duct and possess a few goblet cells which secrete PAS-positive material. Those cells covering the excurrent tubule resemble the epithelial cells of the prointestine and ac- tively secrete PAS-positive material (Fig. 63). Muscle fibers are scarce, but outer circular and inner longitudinal fibers are present. No basement membrane was observed. A thin connec- tive tissue sheath covers the outer surface of the blind sac. Buccal gland. Carriker and Bilstad (1946) studying L^/»niflt'a stagnalis, claimed to have described the buccal gland for the first time. In Helix pomatia, this gland was called Nalepa's gland by Pacaut and Vigier (1906) ; however, since the cytology of the secretory cells could not be difi^erentiated from that of the secre- tory cells of the salivary gland proper, Meisenheimer (1912) in- cluded it as part of the salivary gland. This group of gland cells in Australorbis is histologically distinct from the secretory cells of the salivary gland and should be identified as the buccal gland. The major portion of this gland is located on the dorsal wall of the buccal mass, but extends to its ventral wall, as Avell as to a small portion of the esophagus. In fresh specimens the gland gives a characteristic yellow color to this area. The buccal gland PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 275 is composed of numerous mucus-secreting cells which form many small islands embedded among the muscular bundles of the buccal mass wall (Fig. 64). It is a ductless gland; the contractions of the surrounding muscle bundles apparently force the secretions through the intercellular spaces into the oral cavity. The in- dividual secretory cells are usually pleomorphic but frequently they are flask-sliaped. Large cells measure 15 microns at the base (in a snail of 18 x 5.5 mm.) when they are filled with PAS-positive material. The nuclei are round or slightly oval with chromatin practically filling the entire nucleus. The cyto- plasm is strong!}' basophilic and also shows a positive reaction to PAS. However, it does not show metachromasia with toluidine blue or thionin. The buccal gland cells are similar to mucous gland cells in the foot and cannot be differentiated from the latter. Salivary gland. The salivary gland of A. glahratus is a paired tubular organ, the distal ends of which are joined, thus forming a loop on the dorsal wall of the esophagus at its middle third. Each gland has a short duct which opens to the buccal cavity on the lateral aspect of the esophagus. The gland does not have an acinar structure as it does in Lymnaea stagnalis (Carriker and Bilstad, 1946), but the tubular wall bulges slightly to form haustra-like folds which resemble those of the vertebrate colon. No basement membrane was demonstrated in the epithelial sheet with the staining methods used. The epithelium rests on a thin connective tissue sheath which contains some dark brown pig- ment. Sub-epithelial smooth muscle fibers were described in Helix by Baecker (1932), but we did not observe them in A. glahraius. The ducts are short and lined wdth cuboidal, non- secretory epithelial cells that have long cilia on the surface. The nuclei are usually round, rich in chromatin granules and situated centrally. The cytoplasm is filled with acidophilic granules and may also contain a few vacuoles. The secretory cells gradually appear among the duct epithelial cells lying be- tween the non-secretory cells in the basal zone. These secretory cells gradually increase in size and number toward the gland proper; they compress the duct epithelial cells and become the principal cell (Fig. 65). The flattened duct epithelial cells of the gland proper are very slender and are hardly recognizable 276 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY among the secretory cells except for their acidophilic, thread-like cytoplasm with a centrally located basophilic nuclear outline. The epithelial cells of the duct retain the long cilia and probably function as supportive structures for the secretory cells as well as propelling secreted materials (saliva) to the buccal cavity by the movement of the cilia. Pacaut and Vigier (1906), studying Helix, and Carriker and Bilstad (1946) studying Lymnaea sfagnaJis, differentiated several secretory stages in the secretory cells, but it was difficult to differentiate comparable stages in Australorhis. The actively secreting cells are tall columnar and measure 50 microns in height with round nuclei of 15 microns in an 18 x 6.0 mm. snail (Fig. 66). The nucleus is extremely rich in coarse chromatin, usually contains a large nucleolus and re- sembles the nucleus of a ganglion cell. Since cell function and regeneration are not synchronous, cells of various sizes are usually seen in a given area. The nuclei also are not of equal size and are located on two or three levels, i.e., basal, central or subperipheral (Fig. 66). The cytoplasm is filled with a fine, basophilic reticular net which holds large amounts of secretory materials in droplets or coarse granules. These materials some- times show light acidophilia, sometimes remain unstained in hematoxylin-eosin preparations, but are usually stained by PAS which frequently reacts strongly and masks the basophilic reticu- lar net. Active secretion of the glandular cells apparently occurs by rupture of the cell membrane and thus the cell contents are discharged into the lumen of the gland (Fig. 66). Since de- generated nuclei of glandular cells frequently occur among the cell debris in the lumen, the secretory cells probably disinte- grate completely after secretion (holocrine). Young secretory cells differentiate from fibroblasts in the connective tissue sheath. The cytoplasm of the differentiating fibroblast first thickens and becomes pyramidal, with the base of the cell resting on the con- nective tissue sheath. These young cells gradually grow larger and assume the morphology of columnar secretory cells with round, chromatin-rich nuclei ; the cytoplasm accumulates a baso- philic reticular net as well as PAS-positive material during the process. PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 277 Liver. The liver is a massive digestive gland and occupies the caudal two-thirds of the snail body together with the ovotestis ; the latter is approximately one-fifth of the size of the liver. The liver is a compound tubular gland, consisting of a main hepatic duct with one short dorsal branch and numerous secretory lobules (Figs. 1, 5, 68). The main duct runs anteriorly along the ventral side of the lobules, passes into the cecum and thence via a common opening into the prointestine at its junction with the pylorus. The short, dorsal branch unites wdth the main duct shortly before its point of union with the cecum (Fig. 5). The simple columnar epithelium of the hepatic duct is raised in numerous fold-like, circular elevations which, in cross section, are dome-shaped, finger-shaped or fungoid in appearance (Fig. 67). These elevations decrease posteriorly in number and height and the epithelial sheet finally becomes smooth. The elevations are caused by different sizes of epithelial cells. Those in the elevated areas measure 25 microns in height and elsew^here meas- ure 14 microns (in an 18 x 5.5 mm. snail). The epithelium of the raised areas near the opening of the duct into the cecum has a pseudostratified structure, but this is lost as the epitlielium passes posteriorly. The free surface of the cells is covered with cilia. The cilia of each cell arise from a point near the nucleus, radiate to the surface, and penetrate through the basal bodies before becoming free in the lumen. This type of structure for ciliary apparatus has been observed in Helix by Baecker (1932). Mucus-secreting goblet cells are also present among the ciliated columnar cells. Except for the basal bodies of the cilia, the cytology of the epithelial cells of the hepatic duct is similar to that of the epithelium of the pylorus. Mitotic figures are occa- sionally seen in the epithelium of the hepatic duct. No basement membrane is recognized, and the epithelium is covered with a thin, connective tissue sheath which contains a few collagenous- like fibers. A few^ delicate smooth muscle fibers, mostly longi- tudinal, are seen between the connective tissue sheath and the epithelial sheet near the opening. The lobules of the liver are also covered with a thin, connective tissue sheath which is con- tinuous wdth that of the hepatic duct. No basement membrane was observed in the lobules. Each lobule is embedded in and separated from its neighbor by a loose vascular connective tissue 278 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY which is continuous with that of the ovotestis. Numerous blood spaces, with varying numbers of vesicular cells and pigment cells, are present in the interstices of the connective tissue. The vesicular cells are similar to those described for the radular carrier and are perhaps storage-cells for polysaccharides. Von Brand and Files (1947) described storage of glycogen in the liver and ovotestis of Australorhis glahratus but did not specify where. Faust (1920), Baecker (1932), and Carriker and Bilstad (1946) agree that only two types of cells (lime cells and digestive cells) occur in the liver lobules of the snails studied. In Austra- lorhis, mucus-secreting goblet cells are present in addition to the digestive cells and lime cells. The goblet cells are sparsely scattered among the other two types of cells and cannot be dif- ferentiated from the young digestive cells except when PAS stain is used. They take a deep stain with leucofuchsin and are morphologically similar to those present in the other parts of the digestive tract. The lime cells are more abundant than the goblet cells and are usually pyramidal or rhomboidal in shape with their bases lying on the connective tissue sheath (Fig. 70). The larger cells measure 40 microns across (20 x 6.4 mm. snail) with relatively large, round nuclei (measuring 12 microns) which are extremely rich in chromatin granules and usually contain a large nucleolus. The cytoplasm is tilled with basophilic granules and fibrillar reticulum; it may also include vacuoles of various sizes, but not inclusion bodies or the yellow excretory bodies of the type noted in the cytoplasm of the digestive cells (Fig. 68). The lime cells are embedded between digestive cells and usually do not reach the lumen surface except after break- down of the latter in the secretory process. It has been observed that after the surface cell membrane ruptures, the nuclei may lose chromatin granules and separate from the cytoplasm. There- fore, it would appear that the secretory processes of the lime cells are holocrine in nature. The digestive cells constitute the principal glandular epi- thelial cells and show considerable polymorphism. Krijgsman (1925) described four stages in the digestive cells of Helix poniatia and classified their secretory function as being apocrine in nature. Similarly four physiologic stages could be recognized PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 279 in the functioning of digestive cells of Austral orhis. The diges- tive cells appear to break down completely during secretion (holocrine), and new cells arise or are differentiated from cells of the connective tissue sheath. The fibroblasts are first trans- formed into columnar cells which resemble the epithelial cells of the hepatic duct except that there are no cilia. The nuclei become round, are located subbasally and are rich in chromatin granules (Pig. 69). The cytoplasm is filled with basophilic granules and a reticular net. No inclusion bodies appear at this stage, and the lumen surface of the lolmle is smooth (Fig. 69). In the second stage, the cytoplasm begins to accumulate secretory materials in the peripheral zone which is filled with coarse acidophilic granules (Fig. 59). Owing to the uneven accumula- tion of these acidophilic granules, the lumen surface of the lobule becomes irregular and cells become pleomorphic. In the third stage, vacuoles of various sizes appear in the cytoplasm peripheral to the nuclei which are now forced to a basal position (Figs. 68, 70). Yellow globular inclusion bodies, or so-called excretory bodies, are frequently found in the vacuoles, at first very small and several in number but later becoming larger in size and fewer in number. They are not stained in H-E prepara- tions, assume their own yellow color, are lightly positive to PAS, and may also be stained with fast green {gruene Granida of Krijgsman, 1925 and 1929). The digestive cells reach their peak of growth at the third stage and measure 70 microns in height (20 x 6.4 mm. snail). At the fourth stage the cell mem- brane on the lumen surface ruptures, and the cell discharges its contents. The nuclei lose their rich chromatin granules, float free in the empty cells and finally disappear. The cytoplasm near the nucleus retains its basophilia until nearly replaced by vacuoles just prior to cell dissolution. The lumen of the lobules at the beginning of secretion is filled with cellular debris which transforms into an amorphous cloudy mass. This cloudy mass is lightly acidophilic and also stains lightly with PAS. Although the liver is directly connected with the intestinal tract, bacteria and other microorganisms are seldom seen in the hepatic lumen. Fedele (Baecker, 1932) studying " Opisthobranchium, " ascribed secretory, phagocytic, absorptive and excretory functions to the 280 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY liver cells. We were able to obtain evidence only of secretory and possibly excretory functions in the cells of liver lobules in A. glahratus. IX. Reproductive System Since pulnionate snails may be identified by the characters of the reproductive system, the anatomy and histology of the repro- ductive organs have been studied heretofore in greater detail than those of the other systems (Baker, 1945; Hubendick, 1955). The anatomical descriptions of the reproductive organs of A. glahratus by Baker (1945) and Paraense and Deslandes (1955) are essentially similar except that the latter authors recognized the common collecting- canal of the ovotestis as histologically different from the diverticula and hermaphroditic duct. The genital system of A. glahratus can be subdivided into a) the common genital org^ans (hermaphroditic organs), b) male genitalia and c) female genitalia. Common genital (hermaphroditic) organs. The connnon geni- tal organs consist of a highly-branched ovotestis with its collect- ing canal and a hermaphroditic duct. The acini of the ovotestis are histologically and functionally distinct from the collecting canal. In spite of the morphological differences. Baker (1945) apparently regarded this collecting canal as part of the hermaph- roditic duct. The cephalic or anterior portion of the collecting canal is expanded to form a pouch-like structure which abruptly narrows at its end to join the seminal vesicle or the origin of the hermaphroditic duct by a thin, "S "-shaped tube (Fig. 75). The hermaphroditic duct conveys mature male and female sexual cells into the vas efferens (sperm duct) and carrefour respectively. Ovotestis. The walls of the acini of the ovotestis are composed of thin connective tissue, two to three cells thick, with abundant collagenous-like fibers. This wall, "Ancel's layer" (Ancel, 1902), contains cells which, in section, are morphologically simi- lar to fibroblasts. Although Abdel-Malek (1954 a, b) in Helisoma and Biomphalaria and Merton (1930) in PlanorUs described germinal epithelium with a basement membrane in the ovotestis, we observed no such specially differentiated layer in Australorhis. PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 281 It appears, however, that the germinal cells differentiate from the innermost cells of Ancel's layer by a thickening and trans- formation of the cytoplasm and nucleus. In section, the female germinal cells or ova are usually located at the apices of the acini, and the male germinal cells are arranged along the side walls (Fig. 71). The very early stages (spermatogonia) of the male germinal cells are frequently located near the atrium of an acinus. The maturing stages of the male cells generally line the wall of the acinus from the atrium toward the apex (Fig. 71), and the maturing spermatozoa attaching to the basal or Sertoli cells are close to the area where the ova are developing. The developing young male germinal cells are attached to Sertoli cells by cytoplasmic stalks, which in section frequently are not apparent, and thus some of the germinal cells appear to be free in the lumen (Fig. 90). The youngest male germinal cells are tear-drop in shape and measure 6 microns at the base. The chromatin-rich nuclei till more than two-thirds of the cytoplasm which is also extremely rich in basophilic material. The sperma- togonia separate by mitotic division to form spermatocytes which enlarge considerably before undergoing another division. The cytoplasm of the spermatocytes gradually loses its basophilia and becomes slightly acidophilic. Another mitotic division takes place (Fig. 72) resulting in the formation of spermatids. The spermatids are about the size of spermatogonia and half the size of the spermatocytes. The small round nuclei of the sperma- tids are at first compact, but before the transformation into spermatozoa is complete the chromatin material becomes con- centrated and crescent-shaped (Fig. 89). The nucleus finally becomes helicoid, and the cytoplasm gives rise to an acidophilic tail portion or flagellum by elongation (Figs. 89, 90). Although meiosis must take place in the process of spermatogenesis, no special study was made of the reduction divisions in A. glahratus. The mature spermatozoon (Fig. 96) has a slightly flattened cephalic portion, shaped like a corkscrew, and a long flagellum with two loosely-wound spiral coils encircling the axis cylinder. The Sertoli cells also differentiate from Ancel's layer at an early stage of spermatogenesis, but they remain inconspicuous until the spermatogonia reach the spermatid stage. The Sertoli cells vary in size and shape considerably, appearing as half-moon 282 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY or columnar-shaped. They contain one or occasionally two oval nuclei which are rich in chromatin, and contain a large nucleolus. A wedge-shaped constriction may occasionally be seen in the nucleus of Sertoli cells (Fig. 89). The cytoplasm is filled with a fine reticulum which may be acidophilic or basophilic. Vacuoles in varj'ing number may be present in the Sertoli cells. The female germinal cells are nearly always located at the apex of each acinus in small numbers, and also differentiate from the cells in Ancel's layer (Fig. 73). The mature ovum measures 98 microns in diameter and contains a round, slightly eccentric nucleus measuring 36 microns. The nucleolus is round, eccentric, and has a basophilic, half moon-shaped paranucleolus along the margin. The cytoplasm stains purple or bluish-purple in hema- toxylin-eosin preparations, is coarsely granular, and may also contain a few small vacuoles. PAS-positive material may be present at times, as reported by von Brand and Files (1947). The nurse cells, which are equivalent to the Sertolis of the male germinal cells, adhere to the ovum to form an enveloping sac. A follicular cavity may occur between the nurse cells and ovum. Abdel-Malek (1954 a) described the presence of a follicu- lar cavity in Helisoma trivolvis. Degenerating female germinal cells are frequently seen among the maturing ova. Collecting canal. The collecting canal of the ovotestis is lined with a sheet of cuboidal cells which may have the appearance of transitional epithelium w'hen the canal is not distended. No basement membrane is present, and the epithelial sheet is directly covered with a connective tissue sheath rich in collagenous-like fibers. A few delicate, smooth muscle fibers are present beneath the epithelium. The collecting canal is capable of great disten- tion to accommodate the large numbers of germinal cells (mostly spermatozoa) produced at certain stages of reproduction. When the canal is distended, the epithelial cells become almost flat with the compressed nuclei occupying a central position. The epi- thelial sheet forms several longitudinal folds when the canal is not distended. The epithelial cells are then compressed and appear wedge-shaped with dome-like surfaces forming irregular wavy outlines in the lumen, and the oval nuclei are irregularly located at several levels, thus giving the epithelial sheet a transi- tional appearance (Fig. 74). Short cilia may be seen on these PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 283 cells. PAS-positive material can be demonstrated in the cyto- plasm peripheral to the nucleus. Tlie acidophilic cytoplasm is usually finely granular, containing a delicate reticulum. Hermaphroditic duct or ovisperm duct. The hermaphroditic duct in Australorhis Ls histologically distinct from the collecting canal of the ovotestis and is not capable of great distention (Fig. 77). The duct joins the collecting canal via a small-calibre, "S "-shaped tubule (Fig. 75). Beyond this junction the hermaph- roditic duct enlarges and possesses many diverticula which are usually filled with mature spermatozoa. This portion of the duct is generally called the seminal vesicle (Baker, 1945) (Fig. 76). The lumen of the duct, as well as the diverticula, is lined with simple cuboidal epithelium with central, round nuclei rich in chromatin. The epithelial cells in various diverticula may vary slightly in size. Contrary to the report of Paraense and Deslandes (1955), we have found the free surface of the epithelial cells to be covered with short, dense cilia. The cytoplasm is filled with lightly basophilic granules and may also contain vacuoles. Abdel-Malek (1954 a) described a syncytial form of the diverticu- lar epithelium in Helisoma, but this character was not observed in Australorhis. The cytoplasmic boundary is particularly clear in preparations stained with Azan or Gomori's trichrome. Cilia are present throughout the entire duct. Beyond the seminal vesicle, the duct gradually narrows and becomes very small before joining the carrefour. Smooth muscle fibers are present beneath the epithelial sheet. A thin connective tissue sheath covers the outer surface. The duct joins the carrefour tangentially so that part of its wall is elevated from the surface of the carre- four (Fig. 97). Male geAiitalia. Sperm duct or vas efferens. The male geni- talia consist of the prostatic gland and genital tract. The genital tract is subdivided into sperm duct (vas efferens), vas deferens, and penial complex (Baker, 1945; Paraense and Deslandes, 1955). The prostatic gland is formed by evaginations of the sperm duct and possesses the same histological structure as the duct. Although the sperm duct in Australorhis is very important in the reproductive process of this hermaphroditic snail, the pre- cise origin of the duct has not been described. This is probably due to the extremely small calibre of the duct at its origin, thus 284 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY making the exact location of its departure from the hermaphro- ditic duct difficult to find by gross dissection. Abdel-Malek (1954 a) showed that the sperm duct of Helisoma arises from the hermaphroditic duct just before the latter empties into the carrefour. The sperm duct in Australorbis originates from the hermaphroditic duct at its opening into the carrefour ; thus both ducts in reality have a common opening into the carrefour. For a distance of approximately 200 microns from its origin, the sperm duct is about 33 microns across and is histologically dis- tinct from the rest of the duct. Tlie epithelium of this portion consists of low cuboidal, non-secretory cells covered with long, dense cilia (Fig. 78 a). The cj^toplasm is acidophilic, and the nucleus is compact. The epithelial sheet is invested with many smooth muscle fibers, most of which are circular and constitute the inner layer. The muscle layers are covered with connective tissue which is common to the carrefour, the end portion of the hermaphroditic duct and the first portion of the oviduct. Con- tinuing toward the penial complex, the epithelial cells of the sperm duct gradually increase in height to become columnar, and secretory cells appear among the non-secretory (Fig. 78 b). The secretory cells finally predominate and compress the non-secre- tory cells which become small and obscure (Fig. 81). Both types of cells alternate in the epithelial sheet. The secretory cells measure 52 microns (20.0 x 6.4 mm. snail), and have round to oval, chromatin-rich nuclei (7 microns) located in the basal zone (Fig. 81). The cytoplasm in actively secreting cells contains a basophilic reticulum, the meshes of which are filled with secretory droplets. These droplets appear first as fine granules but later coalesce and become droplets or globules. In hematoxylin-eosin preparations, the secretory granules stain blue or blue-purple, and the droplets stain pale blue or are refractory to staining. In PAS technic, the secretory granules stain weakly, and the droplets stain strongly. The non-secretory cells are as tall as the secretory and are ciliated. The cytoplasm is usually so com- pressed between the secretory cells that it appears as an acido- philic thread. The central nuclei are then so small and compact that only a blue oval outline can be distinguished. The epithelial sheet is invested with a few delicate circular, smooth muscle PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 285 fibers but is without a basement membrane. A thin connective tissue sheath containing some dark brown pigment covers the outer surface. Prostatic gland. The prostatic gland is histologically similar to the glandular portion of the sperm duct and represents evaginations of the latter. The vas deferetis. The sperm duct tapers abruptly in the vicin- ity of the duct of the seminal receptacle and is transformed into the non-secretory vas deferens (Fig. 82). The vas deferens con- tinues toward the praeputium, enters the tissue of the neck near the male genital opening and makes a "U" turn on the dorsal surface of the praeputium; then lies free in the coelomic cavity and joins the verge. Histologically, the portion of the vas defer- ens between the sperm duct and the "U" turn (proximal leg) is different from the portion between the "U" turn and the verge (distal leg). The columnar epithelial cells of the sperm duct are replaced abruptly by densely ciliated, non-secretory, cuboidal cells which rest on a thick basement membrane. The nuclei of the cuboidal cells are round, central in position and rich in chromatin granules. Circular muscle fibers which are sparse in the sperm duct become abundant. Longitudinal fibers do not appear in the outer layer until the proximal leg reaches the "U" turn. The proximal leg is narrower (66 microns at its origin and 33 microns near the "U" turn in a snail of 20 x 6.4 mm.), is usually oval in cross section, and contains few longitudinal muscle fibers (Fig. 83). The distal leg is thicker (uniformly 122 microns in a snail of 20 x 6.4 mm.), round in cross section, and the outer longitudinal muscle layer is as thick as the inner cir- cular layer (Fig. 84). The muscle fibers of the distal leg, es- pecially the longitudinal fibers, contain a large amount of PAS- positive material around the nuclei, and then appear vacuolated in cross section in hematoxylin-eosin preparations (Fig. 84). A thin connective tissue sheath covers the outer surface of the vas deferens. Penial complex. The penial complex is composed of the verge, the vergic sac and the praeputium (Baker, 1945). The inner surface of the verge is lined with a layer of low cuboidal cells containing ovoid, central nuclei and bearing dense, long cilia. 286 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY The epithelial cells rest on a basement membrane which is almost as thick as the epithelial sheet. Beneath the basement membrane are two layers of muscle fibers, an inner circular and an outer longitudinal. A thin connective tissue sheath covers the outer surface (Fig. 85 a). The vergic sac is lined with a sheet of nonciliated cuboidal cells. The basement membrane is invested with a layer of inner circular and outer longitudinal smooth muscle fibers. The outer surface of the vergic sac is covered with a relatively thick con- nective tissue sheath containing rich dark brown pigment. The sac is frequently seen in an invaginated position in the fixed specimen (Fig. 85 b). The praeputium opens on the neck behind the base of the left tentacle and is characterized by a highly-developed muscular wall, especially in the region of the two opposing longitudinal ridges called pilasters (Fig. 86). The pilasters extend into the lumen which appears "H "-shaped in cross section. The simple epithelium consists of columnar cells with dense, long cilia. The nuclei are very rich in chromatin, ovoid in shape and subbasal in position. A thick basement membrane supports the epithelial sheet. Two layers of circular muscle fibers are present, one in contact with the basement membrane and the other in contact with the connective tissue sheath covering the outer surface. Be- tween these two layers, there is a zone of connective tissue W'hieh is richly invested with both longitudinal and radial muscle fibers. The longitudinal fibers are separated into bundles by radial fibers and connective tissue (Fig. 86). The longitudinal bundles are especially well-developed in the pilasters. No mucous cells are present in the epithelial sheet, but they occur in the con- nective tissue, apparently discharging their secretions into the lumen via the intercellular spaces of the epithelial cells. Abun- dant small pigment cells are found in the connective tissue sheath and among the muscle fibers. The region between the praeputium and the vergic sac is called the diaphragm (Baker, 1945). This is a muscular ring with papilla-like protrusions of epithelial folds into the lumen (Fig. 87). The epithelial cells are low columnar, bearing a cuticular covering on the surface and have round chromatin-rich PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 287 nuclei in the subbasal zone. An inner layer of circular and an outer layer of longitudinal muscle fibers lie beneath the basement membrane. Female genitalia. The female genitalia include the albumen gland, carrefour, oviduct, nidamental gland, uterus, spermatheca (seminal receptacle) and vagina (Baker, 1945). The carrefour, oviduct, nidamental gland, uterus and vagina comprise a con- tinuous duct which carries the mature ova to the exterior. The microscopic anatomy of the female genitalia follows a general scheme like that previously described in Lymnaea stagnalis by Holm (1946). The simple epithelial sheet contains two types of cells, a massive secretory and a slender supportive cell. These two cells alternate quite regularly to form the duct lining. There is usually no basement membrane beneath the epithelium. Deli- cate smooth muscle fibers are seen l^etween the epithelium and the outer connective tissue sheath. Alhumen gland. This is a compound tubular gland situated on the dorsal side of the stomach. Two histologically distinct structures are present : a main excretory duct and a large number of highly branched secretory tubules. The main excretory duct opens into one pole of the carrefour which histologically is an enlarged end portion of the duct (Fig. 79). The simple epi- thelium of the duct consists of low columnar cells covered with long dense cilia. The cytoplasm contains acidophilic striations running vertically to the surface and a central, oval nucleus. The lumen is frequently filled with a cloudy, lightly basophilic secretion of the gland. No basement membrane was observed beneath the epithelium which is invested by inner longitudinal and outer circular smooth muscle layers. The connective tissue sheath covering the outer surface is continuous with that of the gland. The epithelium of the secretory tubules contains two types of cells, a massive secretory and a slender supportive cell. The secretory cells vary considerably in size and structure depending on the phase of secretory activity. In a snail of 20 x 6.4 mm., the}' measure 12 x 7 microns as secretory globules begin to accumulate. The nuclei measure 5 microns, are round, central in position, extremely rich in chromatin, and contain a large nucleolus. The cells gradually enlarge and measure 30 x 22.5 288 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY microns just before the globules are discharged. The secretory materials appear first as fine granules embedded in an ampho- philic (purple-red in H-E preparation), cytoplasmic reticulum (Fig. 88). They coalesce and form one or more large globules which are secreted into the lumen through the ruptured cyto- plasmic membrane. The scantj^ cytoplasm -which is forced to the basal region at this stage is highly basophilic and cannot be differentiated clearly from the darkly stained (pyknotic) nucleus. The globules stain a light blue in H-E preparations, blue in Azan, bluish-green in Gomori's trichrome, are strongly positive with PAS, but do not stain with thionin or toluidine blue 0. The secretory materials in the lumen also react in a similar manner with these stains. Von Brand and Files (1947) reported that the galactogen was responsible for the PAS reaction in this gland. Millon's reaction was also applied with a moderately positive result. The supportive cells are usually compressed tightly be- tween the massive secretory cells and show poorly defined cytol- ogy. The acidophilic cytoplasm appears as a thin thread con- taining an elongated blue nucleus in the center. The nuclei of the supportive cells may be displaced either toward the basal or the peripheral zone. In the latter case, they were described in land pulmonates by Cavalie and Beylot (1902) as "Cellules centrotuhuleuses," but in Australorhis are actually cells of a supporting nature, as described by Baecker (1932) . The secretory tubules are covered by a thin connective tissue sheath. The tubules are separated by connective tissue containing small pig- ment cells and abundant blood spaces. Carre four. Baker (1945) called the bean-shaped swelling at the end of the albumen gland duct the "carrefour" (Fig. 97). It had earlier been described and considered to be the site of fertilization of the ova in hermaphroditic land pulmonates (Meisenheimer, 1907 and 1912). The hermaphroditic duct opens tangentially into the carrefour opposite the opening of the duct of the albumen gland. Macroscopically, it appears partially embedded in the wall of the carrefour. The oviduct opens on the side wall of the carrefour between the sperm duct and the duct of the albumen gland. Except that the epithelial sheet con- tains some mucous cells and is infolded, the structure of the carre- four cannot be distinguished from that of the duct of the albumen PAN : HISTOLOGY OP AUSTRALORBIS GLABRATUS 289 gland and therefore, the carrefour should be regarded as the enlarged end portion of this duct (Figs. 79, 80). From its micro- anatomical and histological structures, it appears that the func- tion of the carrefour in Aiistralorhis is more likely a device for separating the male and female germinal cells into their respec- tive genital tracts than a fertilization site as reported by Meisen- heimer (1907) for Helix pomatia, and in Helisoma and Biom- phalaria by Abdel-Malek (1954 a, b). Since this structure has not been described in dioecious snails, and since the sperm duct is extremely small in calibre at its opening to the carrefour and shares a common opening with the hermaphroditic duct, the spermatozoa are probably delivered to the sperm duct directly, while the ova, too large to enter the sperm duct, are passed to the carrefour to be delivered to the oviduct. Oviduct. The oviduct originates from the carrefour, runs anteriorly in the coelomic cavity along the sperm duct and expands into the nidamental gland as a pouch-like swelling in the vicinity of the posterior end of the prostatic gland (Baker, 1945) (Fig. 97 e). The epithelial cells of the oviduct are dif- ferentiated from those of the carrefour by their glandular struc- ture, basophilic staining, and the absence of dense cilia. The lumen of the duct is folded irregularly. The secretory cells are massive and columnar measuring 87 microns in height (15 x 5.1 mm. snail) and contain relatively small compact nuclei in the basal zone (Figs. 79, 80, 91). The cytoplasm is filled with a basophilic, fine reticulum, containing amorphous secretory ma- terial. The secretory material stains strongly with alcian blue and PAS, and shows metachromasia with thionin (beta meta- chromasia). These reactions indicate the presence of acid muco- polysaccharides. The non-secretory cells were called sustentacu- lar cells by Paraense and Deslandes (1955), and are tightly compressed between the secretory cells. They are poorly differ- entiated and usually appear as light acidophilic threads contain- ing thin, central, ellipsoidal nuclei. Contrary to Paraense and Deslandes (1955), we found the sustentacular cells to be sparsely covered with long cilia on the surface. Like those of the sperm duct, the cells are apparently supportive structures for the ovi- duct. No basement membrane was observed and the epithelial 290 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY sheet is invested directly with a few delicate smooth muscle fibers. A thin connective tissue sheath covers the outer surface. Nidamental gland. The anterior end of the oviduct is enlarged considerably and is twisted into a half circle to form a pouch- like structure called the "nidamental gland" by Baker (1945) (Fig. 98). Its lumen is almost completely occluded by irregular epithelial folds. The secretory and sustentacular cells are the same size as those in the oviduct ; however, the staining reactions of the secretory cells are strikingly different (Fig. 91). Over a small area facing the spermatheca the epithelial sheet contains only non-secretory, low columnar, acidophilic cells (about one- fourth of the height of secretory cells) possessing abundant cilia. These cells are also sustentacular in nature and are not modified by the pressure of the secretory cells. The slender sustentacular cells seen in the oviduct are also present among the secretory cells of the nidamental gland. The majority of the secretory cells are filled with acidophilic secretory globules which press the small, compact oval nuclei to the basal zone. These globules are strongly stained with PAS and show purple metachromasia with thionin (beta metachromasia) but are not stained by alcian blue, thus indicating a different chemical nature from the secretory sub- stance in the oviduct. A small number of the secretory cells stain lightly basophilic in H-E preparations and lack secretory glob- ules. Their cytoplasm is filled with cloudy material containing fine granules, which stain positively with PAS but show much weaker metachromasia with thionin than do the acidopliilic cells. The cloudy material does not stain with alcian blue. The smooth muscle fibers are minimal in the wall covered by the secretory cells but are more abundant in the wall covered only by the low columnar sustentacular cells. Utenis. The uterus is a short segment of the female genital tract that connects the nidamental gland to the vagina. It is not differentiated macroscopically from the nidamental gland except for its wider lumen. Histologically, differences are apparent in the epithelial sheet and in the muscle layer (Fig. 92). The secretory cells in the uterus resemble the basophilic secretory cells of the nidamental gland. At the area of transformation from the nidamental gland, the secretory cells are approximately the same size as those in the latter, but gradually become smaller PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 291 toward the vagina. The cytoplasm is reticular and contains baso- philic secretory material. This material is PAS-positive but is negative with alcian blue and shows no metachromasia with thio- nin. The compact nuclei are oval and are located between the mid- and basal zones. The non-secretory sustentacular cells are more abundant and less tightly pressed between the secretory cells than those in the nidamental gland. These cells become very numerous and finally replace the secretory cells completely to transform into vaginal epithelium (Fig. 93). Cilia are abun- dant owing to the increasing number of sustentacular cells. The non-secretory area of the epithelium in the nidamental gland is continuous with the uterus and ends at the junction of the latter with the vagina. No basement membrane is present beneath the epithelium. The muscle fibers become more numerous toward the vagina and form inner longitudinal and outer circular layers. The outer connective tissue sheath contains small pigment and mucous cells. Vagina. The vagina is the only segment in the female genital tract wherein the epithelial sheet rests on a basement membrane. The irregular epithelial folds of the uterus are rearranged into several longitudinal folds. The epithelial sheet is composed of non-secretory, heavily ciliated, columnar cells, among which occur a few mucus-secreting goblet cells. The cytoplasm is acidophilic, containing oval nuclei in the basal zone. The inner longitudinal muscle fibers become less abundant, while the outer circular fibers become very well developed toward the female genital opening. This aperture is located between the male genital opening and the anus. The lumen usually appears occluded by the long cilia and the epithelial folds. The outer connective tissue sheath is fused with the surrounding tissue of the neck and contains abundant small mucous cells (Pig. 95). Spennatheca or seminal receptacle. The spermatheca is a pear- shaped sac which opens into the vagina at its middle and pos- terior thirds. The short duct of the spermatheca is similar his- tologically to the vagina. The heavily ciliated, simple columnar epithelium of the duct forms several longitudinal folds and rests on a relatively thick basement membrane (Fig. 94). The inner longitudinal and outer circular muscle layers beneath the base- ment membrane are well developed. The anterior half of the 292 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY duct is embedded in conueetive tissue in common with tlie vagina and a portion of the proximal leg of the vas deferens. The cilia of the epithelial cells are long and dense near the opening but become shorter and more sparse toward the sac epithelium and finally disappear on the latter. The longitudinal muscle fibers of the duct run into the spermatheea. The epi- thelium of the sac is composed of columnar cells which decrease in size from the duct to the end of the sac. Many folds are pres- ent when the sac is empty. The columnar cells become cuboidal when the sac is distended. The chromatin-rich nuclei are round and located in the basal zone. The cytoplasm is filled with a fine basophilic reticulum and may occasionally contain several small vacuoles (Fig. 85). The cytoplasm superficial to the nucleus stains with PAS but does not take up alcian blue or show metachromasia with thionin. The epithelial cells apparently possess secretory functions. The spermatozoa in the spermatheea may occasionally lose their structure completely and appear degenerate. -o"- SUMMARY AND CONCLUSION A histological study of Australorhis glahratus has been carried out as a necessary requisite for subsequent histopathological investigations on this snail. The results are presented as a sys- tematic description of nine organ systems or tissues. It is believed that this study represents one of the first of such descriptions of a fresh-water snail, although similar descriptions exist for the land pulmonates. The epidermal tissue consists of simple epithelium except for a small area of the mantle collar where pseudostratified epithelium is present. The epithelia of the foot surface, the tentacles and a small portion of the mantle cavity are heavily ciliated. Two types of connective tissue are recognized: a) the dense connective tissue containing abundant fibroblasts, pigment cells, collagenous-like fibers and ground substance but lacking con- spicuous circulation spaces; b) the loose "vascular" connective tissue characterized by the presence of numerous circulation spaces formed by a meshwork of fibroblasts. The fibroblasts ap- pear to be an important cellular element owing to their ability PAN : HISTOLOGY OP AUSTRALORBIS GLABRATUS 293 to transform into amoebocytes under certain stimuli and to repair damaged tissue. The muscular tissue contains three types of muscle fibers: a) the granular muscle fibers, confined to the heart; b) the inter- mediate granular muscle fibers found only in the buccal mass ; c) the smooth muscle fibers composing the rest of the muscular tissue of the other organs. The structure of the heart muscle in A. ylahratus differs from that of the land pulmonates which possess obliquely striated fibers. Eleven ganglia forming a ring around the esophagus comprise the central nervous system. These ganglia are connected to one another by commissures. The ganglion cells are arranged periph- erally, and the nerve fibers centrallj^, in each ganglion. The statocyst, which is imbedded in the epineurium at the dorso- posterior corner of each pedal ganglion, is lined with two types of epithelial cells and contains a number of ovoid statoliths. The eyes are well developed and structurally are composed of the optic capsule, cornea, retina, lens, vitreous humor and optic nerve. The osphradium is a pear-shaped sac enveloped at its base by the osphradial ganglion and is located at the junction between the mantle collar and the neck on the left side of the median line. This organ has not been described in land pul- monates and is believed to be a sense organ present only in aquatic snails. The tentacle has a dense connective tissue core containing a central artery, and has peripheral blood sinuses in the region between the core and the epithelial sheet. There are five groups of conspicuous peripheral sensory cells, one at the base of each tentacle and one at the margin of each lip. The heart is provided with two valves, the atrioventricular and aortic valves. No endocardium was observed, but the epi- cardium is well defined. The only type of blood cell, the amoebocyte, is probably formed under normal conditions in lymphoid tissue located in the wall of the kidney, and in the blood sinuses. The lymphoid tissue has not been reported (in the pulmonate) prior to this study and in appearance resembles the medulla of a lymph node in the vertebrates. The respiratory surface is covered with four types of epi- thelial cells : flat, cuboidal, columnar without cilia, and ciliated columnar. The last type of epithelial cells has not been reported in the land pulmonates. 294 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Unlike Helisoma sp., the renal organ of A. glabratus consists of two distinct parts, an anterior tubular and a posterior saccular portion. The lumen of the kidney is connected with the peri- cardial sac by the renopericardial canal which is lined with heavily ciliated columnar cells. The alimentary system consists of a digestive tract and three digestive glands. The digestive tract is covered for the most part with a simple, ciliated epithelium containing a number of goblet cells and is provided with two layers of smooth muscle fibers. The buccal gland comprises a group of mucous cells which can not be distinguished morphologically from the cells of the foot mucous gland. The salivary gland is composed of a pair of simple tubular organs with haustral-like folds in the gland proper and is morphologically different from that in Lyninaea sp. or in the land pulmonates where a compound tubular structure has been reported. The holocrine, simple glandular epithelium secretes PAS-positive material, which is delivered to the buccal cavity via the ducts. The liver is a compound tubular gland which delivers its secretion to the prointestine via the main hepatic duct. The simple, glandular epithelium has no basement membrane and contains two types of cells, the digestive and lime cells. The hepatic ducts are provided partly with circular, fold-like elevations which are formed by epithelial cells of vary- ing heights. In contrast to the findings in Helisoma sp., Biomphalaria sp. and the land pulmonates, the acini of the ovotestis of A. glabratus are lined with cells which can not be differentiated morphologi- cally from fibroblasts and do not possess a basement membrane. The common collecting canal of the ovotestis is provided with a transitional epithelium and is capable of great distention. The mature germinal cells in the common collecting canal are de- livered via the hermaphroditic duct to the carrefour where they are separated into the respective genital tracts. The carrefour, which has been found onlj- in the hermaphroditic snails, is his- tologically an enlarged end-portion of the main duct of the albumen gland in A. glabratus. The vas efferens and prostatic gland are histologically similar and are the glandular portions of the male genitalia. The vas deferens, verge, vergic sac and praeputium are muscular structures and are considered devices PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 295 for copulation. The albumen gland, oviduct, nidamental gland and uterus are the glandular portions of the female genitalia. The vagina is a muscular tube and serves primarily as a copu- latory organ. BIBLIOGRAPHY Abdel-Malek, E. T. 1952. Morphology, bionomics and host-parasite relations of Planor- bidae (Mollusca: Pulmonata). Ph.D. Thesis, University of Michigan, 226 pp. 1954a. Morphological studies on the family Planorbidae (Mollusca: Pulmonata) I. Genital organs of Helisoma trivolvis (Say) (sul)- family Helisomatinae F. C. Baker, 1945). Trans. Amer. Micros. Soc, 73: 103-124. 1954b. Morphological studies on the family Planorbidae (Mollusca: Pulmonata) II. The genital organs of Biomphalaria hoissyi (subfamily Planorbinae, H. A. Pilsbry, 1934). Trans. Amer. Micros. Soc, 73: 285-296. Ancel, p. 1902. Histogenese et structure de la glande hermaphrodite d'Helix pomatia (Linn.). Arch. Biol., Paris, 19: 389-652. Baecker, E. 1932. Die Mikromorphologie von Helix pomatia und einigen anderen Stylommatophoren. Zeitschr. Gesam. Anat., Ergeb. Anat. Ent- wick., 29: 449-585. Baker, F. C. 1911. The Limnaeidae of North and Middle America, recent and fossil. Spec. Pub. Chicago Acad. Sci., No. 3, pp. i-xvi, 1-539. 1945. The molluscan family Planorbidae. Univ. Illinois Press, Urbana, Illinois, 530 pp. Carriker, M. E., and N. M. Bilstad 1946. Histology of the alimentary system of the snail Lymnaea stag- nalis appressa Say. Trans. Amer. Micros. Soc, 65: 250-275. CAVALii;, M. M., AND Beylot 1902. Sur la glande albuminipare de I'escargot. C. E. Soc. Biol. Paris, 54: 297-298. 296 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY Crawford, G. N. C, and K. Barer 1951. The action of formaldehyde on living cells as studied by phase- contrast microscopy. Quart. Jour. Micros. Sci., 92: 403-452. Faust, E. C. 1920. Pathological changes in the gastropod liver produced by fluke infection. Johns Hopkins Hosp. Bull., 31: 79-84. Faust, E. C, and W. A. Hoffman 1934. Studies on schistosomiasis mansoni in Puerto Rico III. Biologi- cal studies. 1. The extra-mammalian phases of the life cycle. Puerto Rico Jour. Publ. Health and Trop. Med., 10: 1-47. George, W. C. 1952. The digestion and absorption of fat in lamellibranchs. Biol. Bull., 102: 118-127. George, W. C, and J. H. Ferguson 1950. The blood of gastropod molluscs. Jour. Morphol., 86: 315-327. Haffner, K. 1923. Ueber den Darmkanal von Helix pomatia L. Zeitschr. wiss. Zool., 121: 126-169. Haughton, I. 1934. Amoebocytes and allied cells in Invertebrata. Jour. Roy. Micros. Soc, 54: 246-262. Holm, L. W. 1946. Histological and functional studies on the genital tract of Lymnaea stagnalis appressa. Trans. Amer. Micros. Soc, 65: 45-68. Holmgren, E. 1900. Weitere Mitteilungen ueber die " Saf tkanaelchen " der Nerven- zellen. Anat. Anz., 18: 290-296. HOTCHKISS, R. D. 1948. A microchemical reaction resulting in the staining of polysac- charide structures in fixed tissue preparations. Arch. Biochem., 16: 131-141. HUBENDICK, B. 1955. Phylogeny in the Planorbidae. Trans. Zool. Soc. London, 28: 453-542. PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 297 KiSKER, L. G. 1923. Ueber Anordnung und Bau der mterstitiellen Bindesubstanzeii von Helix pomatia L. Zeitschr. wiss. Zool., 121: 64-125. KORSON, R. 1951. A differential stain for nucleic acids. Stain Tech., 26: 265-270. Krijgsman, B. J. 1925. Arbeitsrhythnius der Verdauungsdruesen bei Helix pomatia I. Teil: Die natuerlichen Bedingungen. Zeitschr. vergl. Physiol., 2: 264-296. 1929. Arbeitsrhythnius der Verdauungsdruesen bei Helix pomatia II. Teil: Secretion, Resorption und Phagocytose. Zeitschr. vergl. Physiol., 8: 187-280. LiLLIE, R. D. 19.')4. Histopathologic technie and practical histochemistry. Blackiston Co., Inc., New York, 501 pp. Marcuzzi, G. 1950. Notas sobre la anatomia y la histologia del caracol Australorbis glabratus (Moll. Gastropoda). Arch. Venezolanos Patol. Trop. y Parasit. Med., 2: 1-74. Meisenheimer, J. 1907. Biologie, Morphologie und Physiologic des Begattungsvorganges und der Eiablage von Helix pomatia. Zool. Jb., Abt. System., 25: 461-502. 1912. Die Weinbergschnecke Helix pomatia L. W. Klinkhardt, Leipzig, 140 pp. Merton, H. 1930. Die Wanderungen der Geschlechtszellen in der Zwitterdruese von Planorbis. Zeitschr. Zellf. und Mikr. Anat., 10: 527-551. Miohelson, E. H. 1956. Studies on the biology of the Genus Ceratodes (Mollusca: Pilidae). Ph.D. Thesis, Harvard University, 171 pp. Olson, M. 1942. Histology of the radula protractor muscles of Busycon canali- culatum Linn. Biol. Bull., 82: 190-194. 298 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY Pacaut, M. M., and p. Vigier 1906. Les glandes de Nalepa (glandes salivaires anterieures) de I'eseargot. C. R. Soc. Biol. Paris, 60: 230-232. Pan, C. 1956. Studies on the biological control of schistosome-bearing snails : a preliminary report on pathogenic microorganisms found in Australorhis glabratus. Jour. Parasit., 42 (Sect. 2) : 33. Paraense, W. L., and N. Deslandes 1955. Observations on the morphology of Australorhis glabratus. Memo. Inst. Oswaldo Cruz, 53: 87-103. Pearse, a. G. E. 1953. Histochemistry. Theoretical and applied. Little, Brown and Co., Boston, 530 pp. Plenk, H. 1924. Die Muskelfasern der Schnecken. Zugleich eine kritische Studie Ueber die sogennante Schraegstreifung. Zeitschr. wiss. Zool., 122: 1-78. Smith, G. 1906. The eyes of certain pulmonate gastropods, with special reference to the neurofibrillae in Limax maximus. Bull. Mus. Comp. Zool. Harvard Coll., 48: 233-283. Steedman, H. F. 1950. Alcian blue 8 GS: A new stain for mucin. Quart. Jour. Micros. Sci., 91: 477-479. Stein, J. E., and J. G. Mackin 1955. A study of the nature of pigment cells of oysters and the rela- tion of their numbers to the fungus disease caused by Dermo- cystidium marinum. Te.xas Jour. Sci., 7: 422-429. VON Brand, T., and V. S. Files 1947. Chemical and histological observations on the influence of Schistosoma mansoni infection on Australorhis glabratus. Jour. Parasit., 33: 476-482. PAN : HISTOLOGY OF AUSTRALORBIS GLABRATUS 299 Wagge, L. E. 1951. The activity of amoebocytes and of alkaline phosphatases during the regeneration of the shell in the snail, Relix aspcrsa. Quart. Jour. Micros. Sei., 92: 307-321. 1955. Amoebocytes. Internat. Kev. Cytol., 4: 31-78. Wagge, L. E., and T. Mittler 1953. Shell regeneration in some British molluscs. Nature, 171: 528- 529. YONGE, C. M. 1946. Digestion of animals by lamellibranchs. Nature, 157: 729. EXPLANATION OF PLATES Magnifications of photomicrographs and drawings are indi- cated in tlie appropriate legends for the figures. Fixatives and staining technics used for each figure are also given. Small letters indicating various detail structures are independent for each figure. Drawings were made Avith the aid of a camera lucida. All figures were reduced to % on reproduction. The abbreviations used in the legends are as follows : Zenker : formic acid — Zenker's fixative; Newcomer: Newcomer's fixative; PFF: formic acid — Bouin's fixative; Methyl: absolute methyl alcohol ; H-E : hematoxylin-eosin ; H-Phlox : hematoxylin-phloxine B ; H-Az II-E : Hexatoxylin-azure Il-eosin ; PAS : periodic acid — Schiff reaction; Gomori : Gomori's trichrome ; Azan : Mallory's trichrome; Fresh: fresh, unfixed and unstained specimen. PLATES Plate 1 Fig. 1. Median, longitudinal section of a stretched snail (PFF, H-Phlox, Gx). Various positions or directions of snail body are indicated by arrows a to d; a, anterior; b, posterior or caudal; c, dorsal; d, ventral; e, location of the pseudostratified epithelium; f, location of the osphradium. Fig. 2. Enlarged head region of Figure 1 (36x) ; a, lower lip; b, radular carrier; c, radular sac; d, praeputiuin ; e, central ganglion ring; f, foot gland. Fig. 3. Enlarged coluniellar muscle region of Figure 1 (36x) ; a, colu- mellar muscle; b, salivary gand; c, vagina; d, uterus; e, nidamental gland; f, oviduct; g, sperm duct; li, tuliular i)ortion of the kidney; i, mantle cavity. PLATE 1 Plate 2 Fig. 4. Enlarged stomat-h region of Figure 1 (36x) ; a, ovidiu't; b, sperm duet; c, saccular portion of the kidney; d, albumen gland; e, prointestine; f, gizzard; g, muscular wall of the gizzard. Fig. 5. Enlarged anterior hepatic region of Figure 1 (36x) ; a, main hepatic duct; b, dorsal branch of the hepatic duct; c, hepatic lobules; d, midintestine. Fig. 6. Cross section through the central ganglion ring (Zenker, H-E, 36x) ; a, columellar muscle; b, ganglion ring; c, radular sac; d, esophagus; e, salivary gland; f, praeputium ; g, vagina; h, postintestine; i, rectal ridge; j, dorsal ridge; k, tubular portion of the kidney. A^'V 5 PLATE 2 Plate 3 Fig. 7. Cross section through the spermatheca (Zenker, Gienisa, 36x) ; a, columellar muscle; b, salivary gland; c, esophagus; d, spermatheca; e, visceral ganglion; f, vagina; g, postintestine; h, kidney ridge; i, tubular portion of the kidney. Fig. 8. Cross section through the stomach (PFF, H-Phlox, 36x) ; a, gizzard; b, muscular wall of the gizzard; c, crop; d, prointestine; e, ventricle of the heart; f, pericardial sac; g, saccular portion of the kidney; h, albumen gland; i, descending crus of the midintestine ; j, ascending crus of the midintestine. Fig. 9. Cross section through the overlapping area of the liver and ovo- testis (PFF, H-Az II-E, 42x) ; a, hepatic lobules; b, common collecting canal of the ovotestis; c, acini; d, loose vascular connective tissue. Fig. 10. Simple epithelium of the foot surface (Zenker, H-E, 1275x). °}-^^^j PLATE 3 rinte 4 Fig. 11. Pseudostratified epithelium of the mantle collar (Zenker, H-E, oOOx) ; see Figure 1 e. Fig. 12. Pigment epithelium of the mantle (Zenker, H-E, 1275x). Fig. 13. Respiratory epithelium in tlie mantle cavity (Zenker, H-E, 1-275X). Fig. 14. Loose vascular connective tissue (Zenker, H-E, oOO.x) ; a, Zirkit- lationslik'ken of Kisker or hlood sjiuce; b, smooth muscle fiber; c, crystalline concretion ; d, fibroblast. Fig. 15. Dense connective tissue (Zenker, H-E, 585x) ; a, blood space; b, smooth muscle fiber; c, mucous cells; d, fibroblasts. Fig. 1(5. Fibroblasts (Zenker, H-E, 2200x). Fig. 17. Pigment cells (Zenker, H-E, I600x). PLATE 4 Plate 5 Fig. 18. Vesicular cells (Zenker, H-E, 920x). Fig. 19. Mucous cells in the foot gland (Zenker, II-E, 950x). Fig. 20. Granular muscle (Zenker, H-E, 920x) : a, fibrol)lasts. Fig. 21. Jnterniediate granular muscle (Zenker, H-E, 1275x). Fig. 22. Smooth muscle in the gizzard (Zenker, HE, 920x). Fig. 23. Pedal ganglion and statocyst (Zenker, H-E, 292x) ; a, ganglion; 1), statocyst; c, commissure; d, epineurium. PLATE 5 Plate 6 Pig. 124. (nmglion colls .-ukI statocyst containing statoliths (Zcnki'i', II-E, 750x) ; ;i, ganglion cell with a proniinont process; h, neurofibrils; c, giant cell of the statocyst; d, statoliths; e, epineurium. Fig. 25. Pathology of a ganglion affected by a yeast-like organism (Zenker, 11-E, oUOx) ; note hyperplasia of and nodnle formation by neuroglia cells, and damaged ganglion cells. Fig. -<;. Nerve (Xe\vconu:'r, Il-Az II-E, o85x) ; a, ei)ineuriuni ; b, neuroglia cells. Fig. '27. Eye (Zenker, II-E, 325x ) ; a, cornea; b, retina with its pigment layer; c, lens; d, optic nerve. Pig. 1''■■: " o f . . , c 38 PLATE 8 Plate 9 Fig. 40. Aiiioebocytes in the lumen of an artery (PFF, H-Az II-E, 160()x). Fig. il. Xodule formation bv anioeboeytes and fibroblasts and infiltration liy amoeliocytes (PFF, H-Az II-E, .58nx) ; note that many of the anioeboeytes i-ontain yeast-like organisms whidi appear as dark dots. Fig. 42. Amoel)OC-ytes from the lumen of a heart (PFF, H-Az II-E, L'2()0x); note several yeast-like organisms in the cytoplasm. Fig. -±3. Hemopoietic tissue in the wall of the saccular portion of the kidney (Zenker, H-E, T.oOx) ; a, reticular cells; b, blood spaces; c, amoe l)0cytes ; d, kidney cells. Fig. 44. Epithelium of the tubular portion of the kidney (Zenker, H-E, oS.ix) ; note the acidophilic striations in the cytoplasm. Fig. 45. Epithelium of the saccular portion of the kidney (PFF, H-Az II-E, o8;jx) ; a, crystalline concretions (urine suljstanee of Baecker) ; b, i)l()od space; c, kidney lumen. ^%t:- N '1 4! V } PLATE 9 Plate 10 Fig. 46. Renoperieardial canal (Zenker, Giemsa, 292x ) ; a, kidney lumen; b, lumen of tlie pericardial sac; c, kidney epithelium; d, renoperieardial canal. Fig. -47. Horny jaw in the upper lip (Zenker, H-E, 250x). Fig. 48. Cuticular epithelium in the cavity of the buccal mass (Zenker, H-E, .585x) ; a, cuticular layer. Fig. 49. Vesicular cushion in the dorsal wall of the Iniccal mass (Zenker, H-E, 250x) ; a, vesicular cells; b, intermediate granular muscle; c, lilood space; d, cuticular epithelium of the oral cavity. Fig. .lO. Radular sac (Zenker, H-E, 212x) ; a, collostyle ; b, radula ; c, supraradular epithelium; d, subradular epithelium; e, musculature of the buccal mass. Fig. ill. Epithelia of the radular sac (Zenker, H-E, oS.lx) ; a, supra- radular epithelium ; b, subradular epithelium ; c, radula ; d, collostyle. 48 Ik/' b t 46 ■""s^L^ji- r^^N^ ... 49 " .««* PLATE 10 Plate U Fig. 52. Eadular carrier (Zenker, Gomori, oOOx ) ; a. vesticular cells; 1), intermediate granular muscle fibers; c, epithelia. Fig. 53. Artery connecting the lumen of the radular carrier (Zenker, Gomori, 500x) ; a, artery; h, radular carrier; c, lumen of the radular carrier. Fig. 54. Cross section of the esophagus (Zenker, H-E, 250.x) ; a, epithe- lium; b, longitudinal muscle laj-er; e, circular muscle layer. Fig. 55. Epithelium of the esophagus (Zenker, H-E, 920x). Fig. 56. Epithelium and its supportive tissue of the crop (Xewcomer, Gomori, G-lOx). Fig. 57. Musculature in the wall of the gizzard (Newcomer, Gomori, lOOx) ; a, cuticular epithelium. PLATE 11 Plate 12 Fig. 58. Pseudostratifiod epithelium of the iirointestine (Newcomer, Gomori, 750x). Fig. 59. Epithelium of the midintestine and regenerating epithelium of the liver lobule (2nd stage) (Newcomer, Gomori, 750x) ; a, midintestine; b, liver lobule; c, vacuole. Fig. 60. Longitudinal section of the postintestine (Newcomer, Gomori, lOO.x) ; note the circular epithelial folds. Fig. 6]. Epithelium of the postintestine (Newcomer, Gomori, 500.\). Fig. 62. Anus (Zenker, H-E, 190x). Fig. 63. Longitudinal section of the cecum (Newcomer, H-E, lOOx) ; a, epithelium resembling that of prointestine; b, epithelium resembling that of hepatic duct; c, liver lobules. PLATE 12 Plate 13 Fig. (i4. Buceal gland (Newcomer, PAS, 500x) ; a, island of the gland cells; b, secretory material being discharged toward the epithelial sheet; c, cuticular epithelium of the oral cavity. Fig. 65. Salivary gland, approximately 14 from the opening (Zenker, H-E, oOOx) ; note two types of epithelial cells. Fig. 66. Epithelium of the salivary gland proper in active secretion (Zenker, H-E, oOOx) ; note the loss of the cell membrane on the lumen; a, saliva; b, coelomic cavity. Fig. 67. Epithelium of the main hepatic duct (Newcomer, Gomori, 720x). Fig. 68. Hepatic lolmles with epithelium at the 3rd stage (Zenker, H-E, 292x) ; note the irregular lumen surface of the epithelial sheet and globular inclusion bodies of the digestive cells; a, inclusion globule. Fig. 69. Regenerating epithelium (1st stage) of the hepatic lobule (New- comer, Gomori, 7.j0x) ; note the smooth epithelial surface and the round nuclei. >. -^ .-.i,^:^^ '£ ' •.'*^> r PLATE 13 Plate 14 Fig. 70. Hepatic epithelium at the 3rd stage (Zenker, H-E, 750x) ; a, digestive cells; b, lime cells; c, inclusion globules. Fig. 71. Spermatogenesis (PFF, H-Az II-E, 292x); a, ovum; 1), male germinal cells; c, mature spermatozoa. Fig. 72. Mitosis in the spermatogenesis (PFF, H-Az II-E, 920x). Fig. 73. Mature ovum (Zenker, H-E, oOOx) ; a, nurse cells; b, male germinal cells in the ad.jacent acinus. Fig. 74. Transitional epithelium of the common collecting canal of the ovote.stis (Zenker, H-E, 920x). Fig. 7~). Connecting "S' '-shaped tubule between the common collecting canal and the hermaphroditic duct (Newcomer, Gomori, 425x) ; a, lumen of the collecting canal; b, connecting tul)ule; c, seminal vesicle (part of the hermaphroditic duct). ^ • >! ^^m^^ IT ^72 f 74 9 4' f\ ji' * *•"• • PLATE 14 Plate 15 Fig. 76. Seminal vesicle (Zenker, H-E, 250x) ; note the lun;en filled with numerous mature spermatozoa ; a, blood space. Fig. 77. Hermaphroditic duct (Zenker, H-E, oOOx). Fig. 78. Origin of the sperm duct (PFF, H-E, -130x ) ; a, nonsecretory portion of the sperm duct; b, transition from nonsecretory to secretoiy portion of the sperm duct; c, carrefour; d, oviduct; e, albumen gland; f, artery. Fig. 7il. Duct of the albumen gland (PFF, H-E, 325x) ; a, duct; b, carrefour. Fig. 8(t. Epithelium of the carrefour (PFF, H-E, 320x)3; a, carrefour; b, oviduct. Fig. 81. Secretory portion of the sperm duct (Zenker, H-E, 370x). # - Ik. ■I -* . a' ,Kf •^;. .79 b .'' '^"■'-J' r 1^ f _-y^' ♦,r-«^ ■_•, « <*•..«- •^-M »'^.' '8li PLATE 15 Plate 1(5 Fig. 82. Transition from sperm duct to vas deferens (Zenker, H-E, 190x) ; a, sperm duet; b, vas deferens. Fig. 83. Proximal leg of the vas deferens (Zenker, H-E, oSox). Fig. 84. Distal leg of the vas deferens (Zenker, H-E, oOOx). Fig. 85. A'erge and vergic sac (Zenker, H-E, 190x); a, verge; 1>, vergic sac; c, invaginated portion of the vergic sac; d, spermatheca. Fig. 86. Cross section of the praeputium (Zenker, H-Phlox, 120x ). Fig. 87. Diaphragm of the penial complex (Zenker, H-E, 190x ) ; a, diaphragm; ]>, lumen of the praeputium. Fig. 88. Albumen gland (Zenker, H-E, 250x). *^3. ■-- . ,- 83 vc ■ 1^^ ^^^^ .*^ 85 f'^ '*^17|' PLATE 16 Plate 17 Fig. 89. Transformation of the spermatids into mature spermatozoa (PFF, H-Az II-E, 585x) ; a, mature spermatozoa; b, transforming sperma- tids ; c, Sertoli cell. Fig. 90. Transformation of the spermatids into mature spermatozoa (Zenker, H-E, 500x) ; a, transforming spermatids; b, spermatogonia. Fig. 91. Oviduct and nidamental gland (PFF, H-Phlox, 190x) ; a, ovi- duct ; b, nidamental gland. Fig. 92. Nidamental gland and uterus (PFF, H-Phlox, 180x) ; a, nida- mental gland ; b, uterus. Fig. 93. Transition from the uterus to the vagina (Zenker, H-E, 295x) ; a, uterus; b, vagina. Fig. 94. Duet of the spermatheca (Zenker, H-E, 58.5x) ; a, mucous cells. Fig. 95. Vagina (Zenker, H-E, 190x). F^~ ' " *^B \' > ^-'-.^a t^ • ' A -•^ S^ '•'- ^: Igr i '*'-^j^ 5W .- fi^. '^ ^ P^S: O ■"■-,