-J VVi O IT o 2 -J 2 an LIBRARIES SMITHSONIAN INSTITUTION NOIlOiliSNI NVINOSHIIINS S31 z r; z r- z TvASOvTX 2 .rr. O o D3 :o > ;o ^7 - m pc^ ^ rn ^ rn TIOn" NOliniliSNI NVINOSHilWS S3 I d V« 8 n“tl B RAR I ES^SMITHSONIAN^INS' C/) Z w z .»,• c/5 2 a LI B RAR I ES^SMITHSONIAN INSTITUTION NOlinillSNl NVIN0SHims^S3 1 cn c/3 — CO nON^NOliniliSNl'^NVlNOSHimS S3IBVBan*^LIBRARI ES^ SMITHSONIAN'^INS’ z 2 ^ ^ ^ I => m t W H- y w XjvASA^ rn - .W' pi \,jvAsv^ m E CO " ± CO r: co an LIBRARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVINOSHillNS S3 I Z % Z V CO Z CO - ^ - 7^^ s s E z ~i /yMfWiA z .v\ z ~ / 2 ''W > >■ 2 x^^vos^ > C/3 Z As 2 CO Z ION WOIXnillSNI_NVINOSHimS S3IBVaan_LIBRARIES SMITHSONIAN INS1 j? LlI v<^?rn77>«v ^ Z o z 5 ^ o -I ^ z — -J z ian LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNl NVINOSHIIIAIS S3 1 r- z r- ^ ^ Z r- z o Z o o m ^ ^ noN NoiinmsNrNViNosHims S3 1 avdan“u brar i es Smithsonian insi CO z ^ z X £n. dC^ ^ > • Z c/) Z CO *Z cO lan libraries Smithsonian institution NoiiniiiSNi nvinoshiiims S3i X CO X CO X CO x-v i I 2^ o - o o 2 ^ -J 2 ^ j 2 libraries SMITHSONIAN INSTITUTION NOlirullSNI NVINOSHimS SBIBVlia “ 2: r- . z . ^ W 'I ^ '^NOliniliSN! NVINOSHimS S3iaVaan LIBRARIES SMITHSONIAN INSTITUTI — (f) "Z » V . 2 .V- S S ^- / ^NoiinmsNiI^NViNOSHiivMS^sa I avaan^'i-iBRARi es^smithsonian^instituti Z 00 PO > ^ _ I W' z - “libraries SMITHSONIAN INSTITUTION NOlifUliSNI NVINOSHilWS SaidVda 5 5 2 - J 2 _j 2 LIBRARIES SMITHSONIAN INSTITUTION NOIiniliSNI NVINOSHXIWS S3tUVH^ Z r- 2: r~ . 2 C/5 =: C/3 ± c/> NOIiniliSNI NVINOSHXIWS S3lHVaan LIBRARIES SMITHSONIAN INSTITUTI w z ^ ^ w 2 w 2 -■ ^ ^ ? Xtvi5V4rx v< > 2 > (/y 2 (/} * ^ C/3 ■* LIBRARIES SMITHSONIAN INSTITUTION NOIiniliSNI NVINOSHXIWS S3IdVdE CO 2 (O ~ _ CO CONTRIBUTIONS IN SCIENCE DECEMBER 3 1, 1970 5 ? S O'? Lies TABLE OF CONTENTS and AUTHOR INDEX 1970 Nos. 176-205 LOS M I ANGELES COUNTY MUSEUM FE8 1 81971 4'3RAR1£5. Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 2 CONTRIBUTIONS IN SCIENCE TABLE OF CONTENTS No. 176. No. 177. No. 178. No. 179. No. 180. No. 181. No. 182. No. 183. No. 184. No. 185. No. 186. No. 187. No. 188. No. 189. The status and habits of Grauer's broadbill in Uganda (Aves: Eurylaemidae), by Herbert Friedmann. 4 pp., January 30, 1970. A new species of Neadmete (Neogastropoda) from the Pliocene of California, by Jack D. Mount. 4 pp., 1 fig. February 25, 1970. Description of a new species of net -winged midge from the Great Basin, with a key to the North American species of the genus Dioptopsis (Diptera: Blephariceridae), by Charles L. Hogue, 10 pp. , 13 figs. February 25, 1970. Geographic variation and systematic status of Mexican lyre snakes of the Trimorphodon tau group (Colubridae), by Roy W. McDiarmid and Norman J. Scott, Jr. 43 pp. , 7 figs. February 25, 1970. Studies on North American bees of the genus Hylaeus. 5. The subgenera Hylaeus, S. Str. and Paraprosopis (Hymenoptera: Colletidae), by Roy R. Snelling. 59 pp. , 12 figs, February 27, 1970. Type specimens of fossil invertebrata in the Los Angeles County Museum of Natural History, exclusive of Paleoentomology, by Edward C. Wilson and Donald E. Bing. 20 pp, February 27, 1970. Aquatic adaptations of the water mouse, Rheomys underwoodi, by Andrew Starrett and George F. Fisler. 14 pp, April 17, 1970, Observations and records of Myotis (Pizonyx) vivesi Menegaux (Chiroptera: Vespertilionidae), by Donald R. Patten and Lloyd T, Findley, 9 pp. April 17, 1970. Conocardium langenheimi Sp. N. (Mollusca: Bivalvia) in the Lower Permian Series of the McCloud limestone, northern California, by Edward C. Wilson. 14 pp, , 34 figs, April 17, 1970, Five new epitoniid gastropods from the west coast of the Americas, by Helen DuShane. 6 pp. , 5 ,figs. April 17, 1970. A generic revision and catalog of the western hemisphere Glenurini with the description of a new genus and species from Brazil (Neuroptera: Myrmeleontidae), by L, A. Stange. 28 pp, , 8 figs. April 17, 1970. Further studies on audible vocalizations of the Amazon freshwater dolphin, Inia geoffrensis, by Melba C. Caldwell and David K. Caldwell. 5 pp. , 2 figs. May 4, 1970. Four new terebrid gastropods from the eastern Pacific jslands, by Twila Bratcher and R. D. Burch. 6 pp. , 8 figs. May 4, 1970. Reinstatement of the turrid genus Bellaspira Conrad, 1868 (Mollusca: Gastropoda) with a review of the known species, by James H. McLean and Leroy H. PoOTman. 11 pp. , 16 figs. May 4, 1970, TABLE OF CONTENTS 3 No. 190. No. 191. No. 192. No. 193. No. 194. No. 195. No. 196. No. 197. No. 198. No. 199. No. 200. No. 201. No. 202. No. 203. No. 204. No. 205. A new species of fossil Kelletia (Molliisca: Gastropoda) from the Lomita Marl, Late Cenozoic of San Pedro, California, by Leo G, Hertlein. 8 pp. , 3 figs. May 20, 1970. Studies on the frogs of the genus Leptodactylus (Amphibia: Leptodactylidae). VI. Biosystem atics of the Melanonotus group, by W. Ronald Heyer. 48 pp. , 11 figs. May 20, 1970. System atics of the lizards of the gekkonid genus Phy llod actylus of mainland South America, by James R, Dixon and Raymond B. Huey. 78 pp. , 14 figs. June 26, 1970. Hadrosaurian dinosaur bills —morphology and function, by William J. Morris. 14 pp. , 5 figs. July 1, 1970. Another new night lizard (Xantusia) from Durango, Mexico, by Robert G, Webb. 10 pp. , 2 figs. July 1, 1970. The birds of the Kalinzu Forest, southwestern Ankole, Uganda, by Herbert Friedmann and John G. Williams. 27 pp, July 1, 1970. Two new species of catfishes of the genera Nannorhamdia and Imparales (family Pimelodidae) from Central America, by William A, Bussing. 11 pp. , 2 figs, July 1, 1970. The social wasps of Lower California, Mexico (Hymenoptera: Vespidae), by Roy R. Snelling. 20 pp, , 4 figs. July 28, 1970, Additions to the known avifauna of the Bugoma, Kibale, and Impenetrable Forests, west Uganda, by Herbert Friedmann and John G. Williams, 20 pp, August 31, 1970. Fish remains, mostly otoliths and teeth, from the Palos Verdes Sand (Late Pleistocene) of California, by John E. Fitch, 41 pp, , 6 figs. November 13, 1970, A new species of Batrachoseps (slender salamander) from the desert of southern California, by Arden H. Brame, Jr. 11pp., 5 figs. November 13, 1970, A new salamander (genus Oedipina) of the uniformis group from western Panama, by Arden H. Brame, Jr. and William E. Duellman, 8pp,, 3 figs. November 13, 1970. The ethology of the wasp, Pseudomasaris edwardsii (Cresson), and a description of its immature forms (Hymenoptera: Vespoidea, Masaridae), by Philip F. Torchio, 32 pp., 33 figs, November 24, 1970. A review of the extinct avian genus, Mancalla, by Hildegarde Howard. 12 pp,, 1 fig. November 24, 1970, A new moth fly of the genus Psychoda from crabholes on the Kenya coast (Diptera: Psychodidae), by Charles L. Hogue, 5 pp, , 7 figs, November 24, 1970, Further information on the breeding biology of the honey guides, by Herbert Friedmann, 5 pp, November 24, 1970, 4 CONTRIBUTIONS IN SCIENCE Bing, Donald E. Brame, Arden H, , Jr Bratcher, Twila Burch, R, D. Bussing, William A. Caldwell, David K. Caldwell, Melba C. Dixon, James R. Duellman, William ] DuShane, Helen Findley, Lloyd T, Fisler, George F. Fitch, John E. Friedmann, Herbert Hertlein, Leo G. Heyer, W. Ronald Hogue, Charles L. Howard, Hildegarde Huey, Raymond B. McDiarmid, Roy W. McLean, James H. Morris, William J, Mount, Jack D. Patten, Donald R, Poorman, Leroy H. Scott, Norman J., Jr Snelling, Roy R. Stange, L. A, Starrett, Andrew Torchio, Philip F. Webb, Robert G, Williams, John G. Wilson, Edward C. AUTHOR INDEX No. 181 Nos. 200, 201 No. 188 No. 188 No. 196 No. 187 No. 187 No. 192 No. 201 No. 185 No. 183 No. 182 No. 199 Nos. 176, 195, 198, 205 No. 190 No. 191 Nos. 178, 203 No. 203 No. 192 No. 179 No. 189 No. 193 No. 177 No. 183 No. 189 No. 179 Nos. 180, 197 No. 186 No. 182 No. 202 No. 194 Nos. 195, 198 Nos. 181, 184 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE January 30, 1970 1. 7 J THE STATUS AND HABITS OF GRAUER’S BROADBILL IN UGANDA (Aves: Eurylaemidae) By Herbert Friedmann Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor THE STATUS AND HABITS OF GRAUER’S BROADBILL IN UGANDA (Aves: Eurylaemidae) By Herbert Friedmann^ Abstract: Five examples of the very rare broadbill, Pseudo- calyptomena graueri, from the Impenetrable Forest, southwestern Uganda, prove to be the same as topotypical east Congo birds, and not to have consistently narrower bills as was earlier sus- pected. The field notes accompanying the specimens cause a complete revision of our knowledge of this bird. It is primarily a denizen of the forest undergrowth, not of the tree-tops, and it feeds largely on seeds, buds and flowers, plus such insect larvae and beetles that may occur with them, and is not primarily an aerial feeder on flying insects as has been assumed in the litera- ture. In the course of a second collecting survey of the vertebrate fauna of the Impenetrable Forest, Kigezi, extreme southwestern Uganda, under the spon- sorship of National Science Foundation grant GB 7787, to the Los Angeles County Museum of Natural History Foundation, Robert Glen and Andrew Williams obtained five specimens of Grauer’s green broadbill, Pseudocalyp- tomena graueri, in May and June, 1969. These were particularly significant because two years earlier A. L. Archer had collected one male in that area, which had a very narrow, laterally compressed bill, only 8.5 mm broad at the gape, and 5.2 mm wide at the proximal (basal) end of the nares, compared with 10.6 to 1 1.5 mm at the gape and 7.3 to 7.6 mm at the basal edge of the nostrils in eastern Congo examples. In discussing this specimen Friedmann and Williams (1968: 22) stated that additional specimens would be of inter- est to determine whether or not the Impenetrable Forest birds might prove to be a racially distinct narrow billed population. The present five examples show that this is not the case, that the first one was merely unusual in the narrowness of its bill. The measurements in millimeters of the five 1969 examples, now in The Los Angeles County Museum of Natural History, are as follows: 2 males— width of bill at gape 13.2, 13.6; width of bill at basal edge of nostril 6.9, 7.1; 3 females— width of bill at gape 13.4, to 13.6; width of bill at basal edge of nostrils 7 to 7.5. As may be noted from these figures, kindly measured for me by J. R. Northern, the present series have slightly broader, not narrower, bills than the eastern Congo topotypical examples measured earlier. No Congo material was studied in connection with these five specimens, but it is obvious that the variational range in bill width in the Kigezi birds completely overlaps that of topotypical graueri, and that the two populations are not separable. ^Director, Los Angeles County Museum of Natural History, Los Angeles, Cali- fornia 90007 1 Contributions in Science No. 180 9 That the two are not different is actually easier to understand than was the impression of a local difference suggested by Archer’s single specimen in 1967. The extreme limits of the entire, and relatively narrow, area of high- land forest known to be inhabited by this broadbill are barely more than 150 miles apart, an area that seems to be too small and ecologically too continu- ously uniform to allow for geographic differentiation. It is true that all the eastern Congo records are from the mountains to the west of the deep gorge of the Ruzizi river, but this valley does not constitute a barrier between those highlands and the mountainous forests of Kigezi. The eastern Congolese high- lands extend north from the northern end of the Ruzizi river along the west shore of Lake Kivu to the Rwanda border and to that of extreme southwestern Uganda, where they are coterminous with the so-called Impenetrable Forest. Furthermore there is no reason for thinking that the broadbill is either very local or sporadic within this small total area, even though the actual specimen locality records are still few in number. Definite records from the Kivu dis- trict, eastern Congo, are the type locality about 80 kilometers west of the Ruzizi river, northwest of Lake Tanganyika; Miki, Muusi and Luvumba. In southwestern Uganda, the bird is known only from the Bwindi and Ruhizha areas of the Impenetrable Forest, Kigezi Province. Knowledge of Grauer’s broadbill as a bird, and not merely as a museum taxon, is drastically altered by the field observations by Glen and Williams that accompanied the specimens. Archer, in his field notes on the first (1967) Kigezi specimen, wrote that he shot the bird from a perch about 60 feet up in a tall tree. Rockefeller and Murphy ( 1933: 28) similarly found this species moving about in the upper branches of fairly tall trees. Chapin (1953: 23) noted that the green broadbill was usually found 25 to 75 feet up in forest trees near native plantations, and that in their behavior the birds were quite flycatcher-like, making short but rapid dashes after insects, which they appar- ently caught on the wing. Chapin’s statements were based entirely on the notes of Rockefeller and Murphy as at that time he had had no personal experience with the bird in life. Therefore the observations of Rockefeller and Murphy are taken directly from their paper. The first individual they saw was in a “. . . large tree with abundant juicy berries, a feeding place for many other species of birds”; it was not active and seemed not to be feeding. “When shot, it was catching insects about twenty feet above the ground in a vine-draped forest tree . . .” Another one was seen later “. . . slumped down close to the end of one of the uppermost boughs, then with a hop and a flit of the wings it caught an insect over its head, realighting on the perch it had just left . . .” These two observers wrote that the “normal level” for Grauer’s broadbill was from 20 to 75 feet up in trees. The picture one gets from the reports of Glen and Williams is quite dif- ferent. They found the species to be a denizen, not of the higher branches. 1970 The Status and Habits of Grauer’s Broadbill 3 but of the upper portions of the undergrowth, about eight feet from the ground. When an individual of the species was seen high up in a tree it had just flown there from the nearby lower vegetation. Glen considered its general behavior and actions very much like those of a waxbill rather than a flycatcher. The species looks extremely small in the field, and the short tail gives the bird a crombec-like (Sylvietta) appearance in flight, especially when seen from above. The birds were found, not always singly, but at times in loose groups of 2 or 3 individuals, searching about for food among the branches of a euphor- biaceous tree, Neoboutonia sp., with apparent lack of concern about the ap- proach of the observers. The stomach contents of the five present specimens reveal not the slight- est indication of any aerial feeding habits. The stomach of one of them con- tained a single small beetle, some small seeds, and small flower buds; the sec- ond contained only flower buds; the third had the remains of small pulpy fruits; the fourth only small white flowers; and the fifth contained one small snail, one beetle, several insect larvae and a white seed. These stomach contents are in general agreement with the fact that Rockefeller and Murphy saw one of these broadbills in a berry-laden tree, even though they witnessed no feeding. In their paper they failed to report on the contents of the stomachs of any of their specimens. Inasmuch as the only other African broadbill genus, Smithornis, is known to indulge in a specialized courtship flight behavior, one wonders if the state- ment referred to above, of Pseudocalyptomena making short rapid flights might relate to a similar courtship behavior and not merely to a feeding pattern. On one occasion Glen and Williams watched this species making short flights through the upper undergrowth of the forest and noted that the flight was of a slow and gliding nature. Even on short flights the birds rarely flapped their wings after starting but seemed to glide by preference. In Smithornis rufolateralis Chapin {loc. cit: 18) was convinced that the noises accompanying the flights was produced in the wings. He noted that the primaries had unusually stiff shafts and that the plane of their webs was twisted slightly on their distal portions. “. . . four or five of the outer primaries (but not the outermost) have unusually narrow webs toward the base, and it seems possible that air may pass between them during this special flight . . .” He noted {loc. cit: 22-23) that in Pseudocalyptomena the same remiges were fairly similar, but with “. . . just a little of the downward twist of the inner web toward the tip . . .” Whether this slight difference could be responsible for the absence of the flight sound in the green broadbill is not at all clear, but as our collectors made no entry in their notebooks of any such sound, it would seem that these flights were unaccompanied by any special noises. The collectors did report two types of call notes from Brauer’s broadbill: on one occasion they heard one of these birds give a soft cree-cree repeated three times; and at another 4 Contributions in Science No. 180 time they heard a one-syllabled high-pitched prrrp at intervals of about thirty seconds. The new field notes are also of interest in that they tend to allocate Pseudocalyptornena ecologically more closely with Smithornis. The latter genus is known to nest and to feed fairly low down in the forest vegetation, the nests so far reported having been close to the ground or not more than eight feet from it. The earlier observations that conveyed the impression that the green broadbill was entirely a tree-top bird and an aerial feeder on insects must be emended in the light of new knowledge. Literature Cited Chapin, J. P. 1953. The birds of the Belgian Congo. III. Amer. Mus. Nat. Hist. Bull. 75A: 1-821. Friedmann, H., and J. G. Williams. 1968. Notable records of rare or little known birds from western Uganda. Rev. Zool. Bot. Afr., 77, no. 1-2: 1 1-36. Rockefeller, J. S., and C. B. G. Murphy. 1933. The rediscovery of Pseudocalyp- tomena. Auk, 50: 23-29. Accepted for publication December 16, 1969 LOS ANGELES COUNTY MUSEUM ! UMBER 177 CONTRIBUTIONS iSf. IN SCIENCE February 25, 1970 A NEW SPECIES OF NEADMETE (NEOGASTROPODA) FROM THE PLIOCENE OF CALIFORNIA By Jack D. Mount Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Eos Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor A NEW SPECIES OF NEADMETE (NEOGASTROPODA) FROM THE PLIOCENE OF CALIFORNIA By Jack D. Mount^ Abstract: Neadniete schwartzi, a new species of gastropod belonging to the family Cancellariidae, is described from the Upper Pliocene Fernando Formation near Newport Bay, Orange County, California. Introduction During February, 1967, earth moving equipment on the east bluff of upper Newport Bay, Orange County, California, uncovered a highly fossili- ferous portion of the lower part of the Fernando Formation. This locality re- mained exposed until August of the same year when further construction destroyed it. Between these months extensive collecting at the exposure by me and my associates yielded an interesting fauna of over 200 species of larger invertebrates characterized by Nuculana fossa (Baird, 1863), Patinopecten healeyi (Arnold, 1906), Pecten hemphilU Dali, 1879, Glycymeris grewingki Dali, 1909, Exiloidea rectirostris hertleini Bentson, 1940, Neptunea tabiilata colmaensis (Martin, 1914), Boreotrophon raymondi (Moody, 1916) and My a tnuicata Linne, 1758. This assemblage suggests a Late Pliocene age based on a two-fold division of the Pliocene of California (Vedder, 1960). While studying the material several new species of mollusks were noted. One of these belongs to the genus Neadniete Habe, 1961, which was only recently recognized in the west American fauna (Kanakoff and McLean, 1966). This new species is described herein. References to the Los Angeles County Museum of Natural History, Section of Invertebrate Paleontology are hereafter abbreviated as LACMIP. Acknowledgments I am grateful for the assistance of the following persons. Dr. Robert F. Meade, California State College at Los Angeles, and Dr. James H. McLean, Los Angeles County Museum of Natural History, gave helpful advice and critically reviewed the manuscript. Mr. Jack Schwartz, Mr. Boris Savic and Mr. and Mrs. F. L. Grouard aided in collecting the type material. Photographs are by Mr. Louis W. Kundrath and Mr. James W. Shuttleworth of Kundrath Studio of Photography, Baldwin Park, California. iDepartment of Geology, California State College, Los Angeles, 90032. 1 2 Contributions in Science No. 177 Figure 1. Neadmete sduvartzi, new species. Holotype, LACMIP 1779. Height 35.6, diam. 16.9 mm. 1970 A New Species of NEADMETE 3 Systematic Description Family Cancellariidae Gray, 1853 Genus Neadmete Habe, 1961 Neadmete schwartzi new species Figure 1 Diagnosis: Shell relatively large; whorls 8, evenly convex, weakly tabu- late; 16 spiral ribs on body whorl; outer lip reflected. Description: Shell relatively large for genus, long, slender, fusiform; spire elevated. Whorls 8, moderately inflated, evenly convex, weakly tabulate on early whorls becoming obsolete on penultimate and body whorls. Nucleus simple, smooth, whorls 2. Axial sculpture of approximately 16 low, rounded ribs, becoming obsolete on last four whorls. Spiral sculpture of low, narrow, rounded ribs, 3 on postnuclear whorls and 16 on body whorl; with single smaller rib intercalated between major ribs on the apical whorls and the first nine ribs of the body whorl; interspaces slightly wider than ribs. Sutures moderately impressed. Aperture subeliptical, approximately one half the height of the shell; outer lip smooth, thin, slightly reflected, with 15 internal spiral lines; canal short and shallow. Columella long, straight, with two folds and eight plicae formed by extension of primary spiral ribbing. Type Material: Holotype, LACMIP 1779. Four paratypes: LACMIP 1780; California Academy of Sciences (CAS), Department of Geology type collection 13203; University of California at Los Angeles (UCLA), Depart- ment of Geology type collection 47313; and collection of Jack D. Mount (JDM) 106. Measurements for the type specimens are listed in Table 1. Type Locality: LACMIP locality 471; 33° 38' 21" N, 117° 53' 02" W; in a 6 foot thick lens of sandy cobble conglomerate which occurs approximately 425 feet stratigraphically above the base of the formation. At the present time it is located under the residence at 2161 Vista Entrada, Newport Beach, California. Age and Formation: Late Pliocene, Fernando Formation. Discussion: Neadmete schwartzi is nearest to the Recent N. circum- cincta (Dali, 1873:59) but differs from that species principally in the larger size, more evenly convex, less tabulate whorls, more spiral ribs and in the presence of the reflected outer lip. N. circumcincta is recorded from Alaska, with a specimen in the Los Angeles County Museum of Natural History from Hope Island, off the northern end of Vancouver Island, British Columbia. N. sutherlandi Kanakoff and McLean, (1966:4) from the Lomita Marl (Plio- Pleistocene) of Los Angeles County, California, is the only other fossil species of Neadmete and it is readily distinguished from N. schwartzi by its larger size, more pronounced tabulation, and smaller number of spiral ribs. Etymology: This new species is named for Mr. Jack Schwartz, who 4 Contributions in Science No. 177 collected the holotype, in recognition of his contributions to the mineralogy of California. Table 1. Measurements (in mm) of the type specimens of Neadmete schwartzi, new species. Specimen no. Height Diameter Height of Aperture LACMIP 1779 35.6 16.9 16.8 LACMIP 1780 28.2 14.8 13.3 CAS 13203 35.3 17.9 16.5 UCLA 47313 33.7 17.1 15.7 JDM 106 31.3 15.7 15.1 Literature Cited Dale, W.H. 1873. Preliminary descriptions of new species of Mollusca from the coast of Alaska, with notes on some rare forms. Proc. Calif. Acad. Sci., 5: 57-62, pi. 2. Kanakoff, G. P., and J. H. McLean. 1966. Recognition of the cancellariid genus Neadmete Habe, 1961, in the west American fauna, with description of a new species from the Lomita Marl of Los Angeles County, California. Los Angeles County Mus., Cont. in Sci., 116: 1-6. Vedder, J. G. 1960. Previously unreported Pliocene Mollusca from the southeastern Los Angeles basin, in Short papers in the geological sciences. U. S. Geol. Survey Prof. Paper 400-B, B326-B328. Accepted for publication October 31, 1969. 'it .B ! I y ..^f . ■■ i LOS ANGELES COUNTY i“c?7.7j> MUSEUM ^ CONTRIBUTIONS IN SCIENCE UMBER 178 February 25, 1970 DESCRIPTION OF A NEW SPECIES OF NET-WINGED MIDGE FROM THE GREAT BASIN, WITH A KEY TO THE NORTH AMERICAN SPECIES OF THE GENUS DIOPTOPSIS (DIPTERA: BLEPHARICERIDAE) By Charles L. Hogue Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor DESCRIPTION OF A NEW SPECIES OF NET-WINGED MIDGE FROM THE GREAT BASIN, WITH A KEY TO THE NORTH AMERICAN SPECIES OF THE GENUS DIOPTOPSIS (DIPTERA: BLEPHARICERIDAE) By Charles L. Hogue^ Abstract: The adult, pupa and larva of a new species of Dioptopsis are described and figured from material collected in Great Basin ranges in northeastern California and Nevada. The adult is peculiar among its cogeners in lacking mandibles and having infuscate wings. It is further characterized along with the immature stages in a key presented for all the North American member of the genus. The amandibulate condition (with associated modifications of other head structures) of this species and others is discussed and a new term, colocephalous, is introduced to designate it. Thanks to the exceptional field work of John F. Emmel and Oakley Shields over the past three years, material has accumulated of a previously undiscovered species of Dioptopsis in the Great Basin. For this contribution to my studies on the Blephariceridae of North America, I wish to express my appreciation to these spirited collectors. I wish also to thank J. A. Powell and A. Stone and their respective institutions. University of California, Berkeley — California Insect Survey Collection [CIS] and Agricultural Research Serv- ice, U.S. Department of Agriculture — U.S. National Museum [USNM], for the loan of additional and comparative material. The majority of the specimens recorded are in the Los Angeles County Museum of Natural History [LACM]. Dioptopsis dismalea, new species (Figs 1-13) Dioptopsis n. sp. Hogue, in press. Bull. Calif. Insect Survey. FEMALE (based on allotype; Figs. 6-8) Size. — A medium sized blepharicerid. Measurements (lengths in mm): Body: 3.6 Wing: 5.7 egs: fore mid hind femur 2.5 2.5 3.4 tibia 2.5 2.3 3.1 tarsus 1 1.2 1.0 1.4 2 0.3 0.3 0.4 3 0.25 0.25 0.25 4 0.2 0.2 0.2 5 0.4 0.4 0.4 ^Senior Curator of Entomology, Los Angeles County Museum of Natural History, Los Angeles, California 90007. 1 2 Contributions in Science No. 178 Figures 1-8. Dioptopsis dismalea, n. sp., Adult. 1. Wing, 2. Terminal antennal seg- ments, male. 3. Head, male. 4. Phallosome complex, male genitalia. 5. Male genitalia. 6. Terminal antennal segments, female. 7. Head, female. 8. Female genitalia. 1970 New Species of Net-Winged Midge 3 Figures 9-13. Dioptopsis dismalea, n. sp. 9. Larva. 10. Details of typical median abdominal segment. 11. Pupa, lateral view of anterior end. 12. Pupa (female). 13. Terminal end of male pupa. 4 Contributions in Science No. 178 Coloration. — General integument melanic, well sclerotized. Head, in- cluding antennae, all black. Mesoscutum brownish laterad, large antero- median rectangular area black, pruinose mesad, shiny black laterad; mesono- tal sutures, supraanterior pronotal areas, supraalar, scutoscutellar notches and postscutellum, all shiny black. Scutellum gray pollinose. Pleuron generally brown, lower sternopleuron and metepimeron contrasting black. Haltere even gray. Legs gray-black, darker at joints and distal half of tarsal segments 5. Wing membrane tinged sooty-black. Abdomen evenly black, slightly pruinose. Head. — Figure 7. Eyes widely separated; upper eye division small, 7-9 rows of ommatidia along maximum vertical arc; not well differentiated from lower division, no unfacetted strip, ommatidia of both divisions equal in size. Antenna short, 14 segmented, shape and length of flagellar segments subequal except basal which is approximately 2.0 length of others, ultimate segment slightly smaller than penultimate (Fig. 6) . Mouth parts short, mandibles entirely lacking, hypopharynx truncate, with weak marginal teeth. Palpus 4-segmented, terminal segment (4 and 5) swollen ventromesad, subtriangular in shape. Thorax. — Scutellum with dense lateral setal patches, connected by a single row of setae across caudal perifery; a few supraalar and prescutellar bristles, anterior pronotum bare; other sclerites nude. Legs with tibial spurs 0-0-2, spurs of hind tibia unequal. Wing shape, venation and trichiation as figured (Fig. 1). Genitalia. — Figure 8. Oviscapt deeply cleft mid-posteriorly. Spermathe- cae moderate in size, one smaller than other two and elongate-ovoid, others elongate-ovoid, slightly constricted mesally. MALE (based on holotype; Figs. 1-5) General. — Slightly larger than female. Measurements (lengths in mm) : Body: 3.6 Wing: 6.0 Legs: fore mid hind femur 2.9 3.3 4.3 tibia 3.1 2.8 4.1 tarsus 1 1.8 1.4 1.8 2 0.6 0.5 0.6 3 0.4 0.4 0.4 4 0.3 0.3 0.3 5 0.4 0.4 0.5 Coloration. — Generally darker than female. Essentially sooty-black all over; abdominal pleural membranes contrasting white. Wing membrane tinged sooty-black as in female. Head. — Figure 3. Practically indistinguishable from that of female. Eyes widely separated. Hypopharynx truncate, marginal teeth absent. Tip of labrum 1970 New Species of Net-Winged Midge 5 entire. Antenna short; 14-segmented, relative size and shape of segments as in female. Thorax. — Chaetotaxy as in female, a few bristles on anterior pronotum. Legs with tibial spurs 0-0-2, hind pair very unequal. Wing as in female. Genitalia. — Figures 4 and 5. PUPA (Figures 11-13) Size. — Small to medium. Measurements (overall length in mm) : 5 range 4. 9-5. 6, mean (n = 4) 5.1. $ range 5. 2-6.0, mean (n = 10) 5.5. Coloration. — Dorsally all black, no pattern. Structure. — Figures 1 1-13. LARVA (final instar; Figures 9 and 10) Variation: — Fortunately material has recently come to hand from a third locality, in the Jarbidge Mountains, to indicate that this species is discordantly variable in respect to several features of the male genitalia and one pupal character. Formerly, specimens whose differences suggested the existence of two geographical subspecies were only available from the Warner Mountains of California and Ruby Mountains of Nevada. In the genitalia of the male the following characters are notably variate. ( 1 ) Angle of projection of the 9th tergite lobes. The lobes tend to lie straight and parallel, thus angled at about 90° from the posterior margin of the seg- ment, in the Ruby Mountain populations and variously curved and con- vergent in the other two populations. (2) Extent of bristles on the sides and venter of the basistyle. These bristles are mostly lateral but a few stray mesoventrad, especially in the Nevada populations. (3) Shape of tip of the paramere. Usually the apex of this organ is rounded and slightly inturned with the canal running nearly to the end. In some specimens from the Jarbidge Mountains, the apex is extended beyond the canals to form a spatulate process. (4) Shape of dorsal lobe of outer dististyle. In dorsal aspect this lobe appears club shaped and straight in the Jarbidge Mountains specimens but tend to be capitate and incurved in males from the Ruby and Warner Mountains. (5) Length of ventral projection of the inner lobe of the dististyle. This process variously exists as a short point hardly longer than the body of the main lobe, to a long fingerlike projection curving up and beyond the latter. A dimorphism occurs in the extent of dorsal integumentary stippling on the pupa. In the California populations, these structures terminate abruptly short of the lateral margins leaving a clear zone around most of the abdominal periphery. Nevada pupae show the stippling strongly up to, and even carrying ventrad of, the lateral edge of the abdominal segments. Range: The species is now known only from extreme western and eastern mountains in the Great Basin and Range geological system. It is certain to be found throughout the intervening ranges where streams occur at elevations of over 7000 feet. 6 Contributions in Science No. 178 Diagnosis'. The pigmented wing of this species is unique among Dioptop- sis, the wing membrane of all other species being hyaline. Other characteristics to distinguish adults and the immatures are given in the keys below (see Discussion). Type Locality and Material : The types of this species come from streams draining Mount Bidwell through a large meadow in the Warner Mountains of extreme northeastern California (Modoc County) called Dismal Swamp ( 119° 10' W, 41° 59' N) . Specific collections are as follows: HOLOTYPE: ^ , (genitalia on slide No. CLH 6809-3) and ALLOTYPE, $, (genitalia on slide No. CLH 6809-6) : Dismal Swamp, 7200 feet, 15-16 July 1968 (J. Emmel and O. Shields) [LACM]. PARATYPES: 5 $ $ , 6 $ $ : same data as holotype (one pair com- pletely dissected, on slides) [LACM, USNM]. ^ : creek just west of Dismal Creek, 7300 feet, 15 July 1968 (Emmel and Shields) [LACM]. ADDITIONAL SPECIMENS: CALIFORNIA, Modoc County. 65 pupae and pupal skins: creek just west of Dismal Creek, 7300 feet, 15 July 1968 (Emmel and Shields) [LACM]. 103 pupae: Dismal Creek, 7300 feet, 15 July 1968 ( Emmel and Shields) [LACM]. 3 pupae: stream west slope Mount Bidwell, 1 mi. S Dismal Swamp, 8000 feet, 16 July 1968 (Emmel and Shields) [LACM]. 66 pupae, 75 larvae: Dismal Swamp, 8000 feet, 15 July 1948, 1950 (W. W. Wirth) [USNM, CIS]. NEVADA, Elko County. 27 pupae and pupal skins, 81 larvae: stream connecting Island Lake and Lamoille Creek, Ruby Mountains, 8800-10,000 feet, 8 August 1967, 29 June 1968 (Emmel and Shields) [LACM]. 16 S S ,15 pupae: Bonanza Gulch, 7800-8000 feet, Jarbidge Mountains, 23 July 1969 (Emmel and Shields) [LACM]. 18 pupae: creek west of Angel Lake, 8600 feet. East Humboldt Range, 26 July 1969 (Emmel and Shields) [LACM]. Biological Notes : A]1 loca]ities where D. dismalea has been taken lie between 7300 to 10,000 feet elevation. The habitats are accordingly subalpine and alpine but vary to a considerable extent in several features important to blepharicerid ecology. According to Emmel (personal communication) , “The Dismal Swamp is a large wet meadow (not a swamp, really) with willow thickets around the edges. There are scattered patches of thick Pine forest on the surrounding slopes. The “swamp” is not in a basin or valley as such, but is on the edge of a high plateau which slopes toward the east, just east of the crest of the Warner Mts. Several very small streams flow into the Dismal Swamp from the slopes above it on the west and south sides. Probably the largest of these streams are two flowing from the slopes of Mt. Bidwell. The streams flow through open aspen forest just above the edge of the large wet meadow; it is in these areas of open aspen forest where blepharocerid material was most 1970 New Species of Net-Winged Midge 7 common. Larvae or pupae were not found more than several hundred yards away from the Swamp (although the streams [terminated] not too far above the areas where pupae stopped appearing), nor in the Swamp itself. Larvae and pupae were found on stones ranging in size from 1" diameter up to 6" diameter (which were probably the largest rocks). The stream rocks were generally very dark in coloration, although not usually as dark as the pupae on them. A few pupae were also found in a small stream high on the western slope of Mt. Bidwell where there was no forest; small willow thickets surrounded this stream. In all areas where blepharocerid material was collected, the stream habitat here was in contrast to other blepharocerid stream habitats in which we have collected, in that these streams were mere trickles (although this was a very dry year) rather than rushing torrents of abundant water as is usually the case. The observation that larvae and pupae were on relatively small rocks (the only ones available) is also of interest. The habitat of this species is noticeably different from that of . . . the Ruby Mountains locality . . . [which] . . . looks very much like a typical stream in the high altitudes of the eastern slope of the Sierra Nevada. Lamoille Canyon has been well glaciated; many streams flow down its steep slopes in the early summer. The stream in which larvae and pupae were found flows down an open slope into Lamoille Creek; blepharocerid material was found in the swiftest water, on dark rocks (probably granite). The whole area looks not unlike the higher altitudes of the Sierra Nevada, and it is well watered. Aspens are more abundant than conifers at this altitude; trees occur in scattered patches, never as extensive forest.” The site in the stream above Angel Lake in the East Humboldt Range is like the preceding, a nearly vertical narrow cascade down a rough granite face of a steep glaciated valley. There is scarcely any vegetation in the immediate vicinity of the stream. On the other hand at Bonanza Gulch, in the Jarbidge Mountains, the gently tumbling stream in the bottom of a broad ravine flows through a jumble of log and debris jams and is heavily overgrown with tall clump grass and willows. Discussion : The new species clearly shows a complete lack of mandibles. This is a normal condition in only one other North American blepharicerid, Blepharicera ostensackeni Kellogg, but occurs consistently in many exotic species and in populations (subspecies ?) of Dioptopsis aylmeri (Garrett) in the eastern drainage of the Sierra Nevada (Hogue, in press) . Associated with the amandibulate condition is a complete rearrangement in head proportions including a much shortened labrum and other mouthparts, weak hypopharyn- geal teeth, widely separated eyes and reduced upper eye divisions. This type of head may be designated, “colocephalus” (Gr. Kolos, stunted, incomplete) . The genetic and phylogenetic significance of colocephaly which is ana- malous in some species and normal in others, deserves special study. I have expressed (Hogue, in press), as has Edwards (1929:35-36), the opinion that such a total reorganization of head construction could come about by pleio- 8 Contributions in Science No. 178 trophic effects of simple mutations. Thus what would appear to be a profound set of differences upon which new species or other taxa might be based actually may easily and frequently arise in individuals or populations (even becoming dominant in one or more geographic areas as has presumably occurred with eastern Sierran D. aylmeri) of a single species. A trimorphic or even quadrimorphic condition could exist wherein a species may have males and females alternately with or without mandibles and with associated modi- fied head structure. Such a hypothesis was long ago advanced by Muller but was refuted by other dipterists (Alexander, 1958:814) on the grounds that examples of dimorphic females actually belonged to different species. I believe the phenomenon needs further study to determine its extent, cause, and adaptive significance. With the discovery of D. dismalea, the genus Dioptopsis in North Amer- ica now consists of six species: 1. aylmeri (Garrett, 1923) . Can. Ent. 55:244 {cheaini Garrett, 1925, to be placed in synonomy in forthcoming work by author) . Widespread in west- ern Canada and the United States. 2. sequoiarum (Alexander, 1952) . Bull. Brook. Entomol. Soc. 47:91. Cali- fornia, western Sierra Nevada. 3. dismalea Hogue, 1969. Great Basin Ranges. 4. arizonica Alexander, 1958. Bull. Brook. Entomol. Soc. 53:50-51. Arizona, Sierra Ancha Mountains. 5. alpina Hogue, 1966. Los Angeles Co. Mus. Nat. Hist., Contrib. Sci., 120: 1-5. Northern and southern California. 6. markii (Garrett, 1925) . Seventy New Diptera, p. 5. Recorded only from northern California, Washington and Alberta. The first three species are very close relatives and may be united into an Aylmeri group. The primary basis of the grouping is a common larval type. In spite of the colocephalous heads of adult dismalea giving it a markedly different character from aylmeri (with the exception of certain populations as men- tioned above) and sequoiarum, the species are indistinguishable in the larval stage. Arizonica and alpina together form a second natural grouping on the basis of features I have outlined elsewhere (Hogue, 1966:5). Markii stands alone with very different larval, pupal and adult characteristics. The following keys will facilitate identification of the adults and im- matures of all these species, with the exception of arizonica, whose female, pupa, and larva are unknown. KEY TO NORTH AMERICAN DIOPTOPSIS SPECIES ADULTS (Female of arizonica unknown) 1. General: Wing membrane infuscated. Palpus four segmented dismalea General: Wing membrane hyaline. Palpus five segmented. 2 1970 New Species of Net-Winged Midge 9 2. General: Upper eye division at least one-fifth to one-fourth the area of lower in males (at least 11-12 rows of ommatidia), greater than lower in females 3 General: Upper eye division greatly reduced, much smaller than lower in both sexes, a small area of 4-7 rows of ommatidia 4 3. Male genitalia: Paramere with conspicuous mesal spur. Mesal margin of IX tergite lobe produced into a nude flange (i.e. without micro- or macrochaetae) . Female genitalia : Spermathecae ovoid aylmeri (in part) Male genitalia: Paramere simple, without spur. Mesal margin of IX tergite lobe not produced. Female genitalia: Spermathecae elongate- ovoid sequoiarum 4. General: Antenna with 15 segments. Mid-tibial spur present though minute. Male genitalia: IX tergite lobe short, only 2 times width. Lobe of outer dististyle short and projecting dorsomesad 5 General: Antenna with 14 segments or less. Midtibial spur entirely absent. Male genitalia: IX tergite lobe elongate, length 5-6 times width. Lobe of outer dististyle elongate and projecting subparallel to dististyle face 6 5. Male genitalia: Tip of inner dististyle lobe recurved alpina Male genitalia: Tip of inner dististyle lobe with cup-shaped depression arizonica 6. General: Ultimate antennal segment smaller than penultimate. Female mandible reduced to absent. Male genitalia: Paramere with spur. Female genitalia: Spermathecae pear-shaped aylmeri (in part) General: Ultimate antennal segment longer than penultimate. Female mandible normal sized. Male genitalia: Paramere simple. Female geni- talia: Spermathecae ovate markii PUPAE (Unknown for arizonica) 1. Branchial sclerite strongly lobed, projecting well beyond cephalic sclerite in lateral aspect alpina Branchial sclerite weakly lobed, cephalic sclerite visible from lateral aspect 2 2. Branchial lamellae spreading, caudal pair strongly convergent, often overlapping markii Branchial lamellae all subparallel 3 3. Length/ width of anteriormost branchial lamella 1.1 or larger dismalea Length/ width of anteriormost branchial lamella 1 .0 or smaller 4 4. Interbranchial index ( =interbranchial distance at base of lamellae/width of anteriormost lamella) less than 0.7. Rugae of posterior region of scutum forming a line running parallel to hind margin of sclerite sequoiarum 10 Contributions in Science No. 178 Interbranchial index greater than 1.0. Rugae of posterior region of scutum random, not forming a line. aylmeri LARVAE (Unknown for arizonica) 1. Dorsal tubercles and plates present alpina Dorsal integument without tubercles or plates 2 2. Antenna indistinctly segmented, basal % membranous. Ventral gill tuft of segment 6 diminutive, with only 2 filaments. Dorsal pseudopods gently curving cephalolaterad markii Antenna clearly 3 segmented, middle segment much smaller than others. Ventral gill tuft of segment 6 subequal to others, with 5 filaments. Dorsal pseudopods angular, tips project ventrolaterad aylmeri, sequoiarum, dismalea Literature Cited Alexander, C. P. 1958. Geographical distribution of the net-winged midges (Blepharoceridae, Diptera). Proc. Tenth Int. Congr. Entomol. 1 (1956): 813- 838. Edwards, F. W. 1929. Blepharoceridae in Diptera of Patagonia and South Chile, Part II, fasc. II, pp. 33-75, 3 figs., 4 plates. Hogue, C. L. 1966. A new species of Dioptopsis from California (Diptera: Bleph- aroceridae). Los Angeles Co. Mus. Nat. Hist., Contrib. Sci., 120: 1-5. “. in press. The net-winged midges or Blepharoceridae of California. Bull. Calif. Insect Survey. Accepted for publication November 3, 1969. j ;los i ' ANGELES I ! 1 COUNTY l:' I MUSEUM 1" ftlBER 179 CONTRIBUTIONS IN SCIENCE February 25, 1970 GEOGRAPHIC VARIATION AND SYSTEMATIC STATUS OF MEXICAN LYRE SNAKES OF THE TRIMORPHODON TAU GROUP (COLUBRIDAE) By Roy W. McDiarmid and Norman J. Scott, Jr. Los Angeles County Museum of Natural History Los Angeles, California 90007 Exposition Park CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author oi divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor GEOGRAPHIC VARIATION AND SYSTEMATIC STATUS OF MEXICAN LYRE SNAKES OF THE TRIMORPHODON TAU GROUP (COLUBRIDAE) By Roy W. McDiarmid^ and Norman J. Scott, Jr." Abstract: Geographic variation in lepidosis and color pat- tern was studied in the nominal species Trimorphodon collaris, fasciolata, forbesi, latifascia, tan, and iipsilon. Specimens were grouped into seven geographically separated samples which were then compared. Ranges of most characteristics overlap broadly; others vary independently and discordantly. Ventral-subcaudal scales decrease from north to south. Snakes from the Balsas- Tepalcatepec Basin have fewer dorsal bands than do snakes from other areas. They also have a relatively uniform head pattern. Specimens from other areas have a complex head pattern con- sisting of prefrontal bars, interocular bars, and discrete parietal marks in various combinations. Head pattern and body band analyses indicate limited genetic exchange between the Balsas- Tepalcatepec Basin samples and those from other areas, appar- ently the result of secondary contact in southern Jalisco and adjacent Michoacan. The Transverse Volcanic Range and the Sierra Madre del Sur apparently are effective in restricting genetic exchange. The above nominal species are assigned to the synonymy of Trimorphodon tan. Trimorphodon tan tan is maintained for the foothills and plateau populations and T. tan latifascia (new com- bination) is assigned to the Balsas-Tepalcatepec Basin popula- tions. Trimorphodon tan is redescribed and its distribution delimited. Character variation in a brood from Puebla approaches the variation typical of the entire Puebla sample. Introduction In the years following 1935, rather extensive collections of amphibians and reptiles were made in Mexico. Early reports on many of these collections, culminating in the Mexican checklists by Hobart M. Smith and Edward H. Taylor (1945, 1948, 1950), are extremely useful in elucidating the distribution and relationships of the many components of this complex herpetofauna. Dur- ing the past 15 years, reports on collections from various parts of Mexico have continued to appear. Unfortunately, some of these reports have confused rather than clarified our knowledge of the distributions and systematics of certain of the Mexican species (see Duellman, 1966, for additional discussion). An example of such confusion is found within the snake genus Trimorphodon. ^Research Associate in Herpetology, Los Angeles County Museum of Natural His- tory and Collegiate Division of Biology, University of (Chicago. (Current address: Dept, of Zoology, University of South Florida, Tampa.) “Biological Sciences Group, University of Connecticut. 1 2 Contributions in Science No. 179 All 12 nominal species of Trimorphodon have been reported from Mexico (Smith and Taylor, 1945) . Smith ( 1941 ) divided the genus into two apparently natural groups. The biscutatus group, characterized by large chevron-shaped marks on the neck and head, included five nominal species {biscutatus, lambda, lyrophanes, paucimaculatus and vandenburghi) that range generally from the southwestern United States southward through the Pacific lowlands to Costa Rica. The tau group, characterized by a light, transverse nuchal collar of vary- ing width, included seven nominal species {collaris, fasciolata, latifascia, tau, upsilon and vilkinsonii) whose ranges encompass upland and mountainous portions of southeastern New Mexico, southwestern Texas and Mexico, north of the Isthmus of Tehuantepec. We became interested in the Trimorphodon tau species group in the course of an extensive study of the herpetofauna of Sinaloa, Mexico (Hardy and McDiarmid, 1969). In attempting to allocate specimens, we realized the need for a revision of this group of snakes and so began a study of the geographic variation and systematic relationships among the described forms, with empha- sis on the zoogeographic implications of the distribution of specific character- istics. Historical Review Cope (1869) described Trimorphodon tau from a single specimen col- lected by Sumichrast, supposedly from the western part of the Isthmus of Tehuantepec, and Trimorphodon upsilon from a specimen collected in Guada- lajara, Mexico by Major (see Zweifel, 1959 for discussion of this collection). Later in the same year, Peters (1869) described Dips as bis cut at a var. latifascia from specimens collected in Puebla by Berkenbusch. These descriptions were the first for representatives of the tau species group, though specimens of these species were known previously. Jan (1863) listed four specimens of Eteirodipsas biscutata, one from Mexcio. Nine years later this same specimen was illustrated in Jan and Sordelli (1872) and listed as part of the collection of “Westphal-Castelnau a Montepellier.” The Mexican specimen definitely is not Trimorphodon biscutatus (Dumeril, Bibron and Dumeril, 1854), nor is it representative of a species of the biscutatus group as presently understood. In fact the color pattern, especially of the head (Jan and Sordelli, 1872: vol. 3, livr. 39, pi. 1, no. 3) , is similar to the pattern of the holo- type of Cope’s Trimorphodon upsilon. Cope (1875) described a fourth species of the tau group, Trimorphodon collaris. This specimen, also collected by Sumichrast, was from Orizaba, Vera- cruz. In 1882 Sumichrast pointed out that the type specimen of Trimorphodon tau was collected near Quiotepec, between Techuacan and Oaxaca and was not from the western part of the Isthmus of Tehuantepec as stated by Cope (1869) in the original description. In this same publication, Sumichrast (1882) listed “los individuos tipicos” of Trimorphodon collaris from Tuxpango, near 1970 Geographic Variation of Mexican Lyre Snakes 3 Orizaba, Veracruz. However, in the description of T. collaris Cope (1875) made no reference to specimens other than the type. The location of any addi- tional specimens of T. collaris collected by Sumichrast is unknown. Gunther (1895) pointed out the similarities between Trimorphodon tan, collaris and upsilon, and referred his specimens to T. upsilon. Gunther con- sidered Peters’ latifascia synonymous with Trimorphodon biscutatus. Boulen- ger (1896) recognized Trimorphodon tan and T. upsilon as distinct species. Boulenger followed Gunther’s suggestion and included Trimorphodon collaris, as well as Trimorphodon biscutatus latifascia (Peters), in the synonymy of Trimorphodon upsilon. Taylor (1939) reviewed Mexican species of Trimorphodon and revived Peters’ name latifascia for specimens from Puebla and Morelos. Taylor recog- nized T. upsilon and T. tau as distinct species but considered T. collaris synony- mous with T. latifascia. In 1941 Smith surveyed the genus Trimorphodon and established the two species groups. He considered Trimorphodon latifascia, T. upsilon, T. tau and T. collaris distinct species. In addition Smith ( 1941 ) described two more forms in the tau group: Trimorphodon fasciolata from Tzararacua Falls, Michoa- can (one specimen) and Trimorphodon forbesi from San Diego, Puebla (one specimen) . After examining specimens of Trimorphodon from Michoacan, Schmidt and Shannon (1947) suggested that T. fasciolata might be conspecific with T. latifascia. Davis and Smith (1953) and Peters (1954) demonstrated that these two forms probably were representative of a single species. Smith and Darling (1952) examined some specimens from Tamaulipas and San Luis Potosi and regarded Trimorphodon upsilon as a subspecies of Trimorphodon tau. Martin (1958) confirmed the conspecificity of these forms. Duellman (1961) concluded that data were inadequate to support the retention of upsilon as a subspecies of Trimorphodon tau. He regarded Tri- morphodon tau as a monotypic species that has a highly variable color pattern. Thus, if the systematic changes since Smith’s revision (1941) are accepted, there are currently four species in the tau group, Trimorphodon tau, collaris, forbesi and latifascia. Jones and Findley (1963) suggested that T. vilkinsoni might be related to T. lambda, a member of the biscutatus group. Methods and Materials We have examined all known specimens, including holotypes, of the following nominal species of the Trimorphodon tau group: tau, upsilon, col- laris, latifascia, fasciolata and forbesi. Locality, sex (determined by dissection), and the following features were recorded for each of the 197 specmiens: num- ber of ventral scales (counted according to the method outlined by Dowling, 1951), number of subcaudal scales including the tail tip, total number of 4 Contributions in Science No. 179 ventral and subcaudal scales, nature of the anal plate (divided or entire), number of supralabials, supralabials that contact eye, number of infralabials, the infralabial that is first separated from the chin shields by small gular scales, number of loreal scales (all scales between the preocular and nasal scales), number of preocular and postocular scales, and the number of primary and secondary temporal scales. The dorsal scale rows were counted one head length behind the parietal scales, at midbody and one head length in front of the anus. The number of scales in the light collar counted on the midline was recorded, as was the number of body and tail bands or blotches. The first cross band that showed any indication of splitting (the presence of a light center on the midline) was recorded. For each band-interspace-band sequence the num- ber of scales on the midline was counted beginning with the first dark band- light interspace-second dark band, for the seventh dark band-light interspace- Figure 1. Distribution of Trimorphodon tan in Mexico. Each dot represents a locality or localities of one or more specimens. Solid dots indicate specimens examined; circles indicate additional specimens or literature records. Samples analyzed are lettered A through G. 1970 Geographic Variation of Mexican Lyre Snakes 5 eighth dark band, and at equivalent intervals along the body of the snake. In all instances the band-interspace-band sequence immediately anterior to the vent was counted. Detailed drawings of the dorsal head pattern were made for each specimen. Specimen localities were plotted on a topographical map of Mexico and grouped into seven geographical units (Fig. 1). These seven samples, here- after referred to as populations A through G, include specimens from the following areas : Population A (25 specimens, — from localities in the foothills of the Sierra Madre Occidental of southern Sonora and northern Sinaloa. These localities are separated from population B to the south by nearly 480 kilometers. Population B (15 specimens) — from localities in the upper drainage of the Rio Santiago and its tributaries in Nayarit and Zacatecas and from the general vicinity of Guadalajara and Lago de Chapala in Jalisco and adjacent Michoacan. These snakes probably are continuously distributed in suitable habitats northward through Sinaloa, as indicated by two specimens from western Durango, but are separated into two populations, A and B, because of the lack of adequate material from the intervening area. Population C (14 specimens) — from localities in Tamaulipas, San Luis Protosi and Hidalgo in the eastern drainage systems of the Rio Panuco and Rio Tamesi. There are six specimens known from within the 400 kilometer hiatus that separates population C from population B, the latter of which is restricted to western drainages. Population D {20 specimens) — from localities in central Oaxaca. This sample is situated about 480 kilometers southeast of population C with only a few specimens known from the intervening area. The uplifted southern section of the Mexican Plateau effectively separates populations C and D. Population E {51 specimens) — a large sample from the vicinity of Chil- pancingo, Guerrero located in the central part of the Sierra Madre del Sur. These localities are isolated from sample D by the eastern extension of the Sierra Madre del Sur and the mountains in northwestern Oaxaca. Population F (25 specimens) — from areas in northeastern Guerrero and adjacent Mexico and Morelos and from southwestern Puebla. These localities are in areas drained by the northeastern tributaries and headwaters of the Rio Balsas. Population F is separated from E by the arid Balsas Basin and from D by the mountainous areas of northwestern Oaxaca. Population F is separated from sample C south of the volcanic peaks and high southern portions of the Mexican Plateau. Population G (19 specimens)— from parts of Colima, southern Jalisco and western Michoacan. These localities are situated in areas drained by the Rio Armeria and Rio Coahuayana ( Autlan and Contla, Jalisco and Colima) or by the major northwestern tributary of the Rio Balsas, the Rio Tepalcatepec (Michoacan). This sample is isolated to the northwest of E by the arid Balsas 6 Contributions in Science No. 179 Basin and to the south of B by the Transverse Volcanic Range and western portions of the Mexican Plateau. Specimens removed from the geographic range or topographically iso- lated from these units were considered separately. In other words, we did not extend the geographical coverage of our units to accommodate individual specimens. After the seven groups were established, the meristic, scale and pattern characteristics were compared within each unit and among the seven units. Geographical Variation LEPIDOSIS Ventral and subcaudal scales. — Ventral scales vary from 20-231 in males and from 210-243 in females. Subcaudal scales range from 61-85 in males and from 55-80 in females. Females average 2-11 more ventrals than males in the various populations. However, males on the average possess 8-1 1 more sub- caudals than do females in the same population. The relatively shorter body and longer tail of males, as indicated by fewer ventrals and more subcaudals, appar- ently is the result of their more anteriorly placed cloacal opening. This condi- tion is probably an accommodation for the relatively long hemipenes in the tail. The converse probably is true of females which have a longer, more bulky body to accommodate eggs (Klauber, 1956:167). The number of ventral and subcaudal scales in snakes is indicative of the number of vertebral segments (Dowling, 1951:99). To test the assumption that there is little or no sexual dimorphism in the total number of vertebral elements in members of the same geographic population, we analyzed the number of ventral plus subcaudal scales for males and females in each of the seven samples using the Student’s t-test. No significant differences (95 per- cent level) were found. Therefore, for purposes of analysis, we combined the ventral and subcaudal counts of males and females in each population. The ventral-subcaudal scales range from 265-319. The higher numbers are characteristic of the northwestern population in Sonora and Sinaloa and the lower numbers are from the southern population in Oaxaca. Variation in numbers of ventral-subcaudal scales in the seven populations is illustrated (Fig. 2) . The relative geographic positions of specimens from localities outside of the geographic limits of the prescribed populations are plotted by number in Figure 2. There is a north to south reduction in the number of ventral-subcaudal scales within population A. The specimens from southern Sonora have 298- 319, X = 307 (N=ll). Specimens from the Rio Fuerte in northern Sinaloa have 300-306, x = 303 (N=4). Specimens from north central Sinaloa have 293-308, X = 303 (N=10). Two specimens from Ventanas, Durango (300 and 302 scales), and a specimen from the vicinity of San Bias, Nayarit (294 scales) are the only known specimens from the 480 kilometer distance between 1970 Geographic Variation of Mexican Lyre Snakes 7 populations A and B (Fig. 2, Nos. 1-3). Their counts are intermediate and suggest that additional material from the intervening area would show that the north to south reduction in the number of ventral-subcaudal scales ex- hibited in population A continues into B. A specimen from La Cumbre, Jalisco (Fig. 2, No. 4) has 292 ventral-subcaudal scales. Within population B, three high counts (299, 299, 303) are characteristic of specimens from northwestern localities of population B, at Barranquitas, Nayarit, and near Magdalena and Bolanos, Jalisco, respectively. Two speci- mens from the southeastern parts of population B near Emiliano Zapata, Michoacan have low counts (282, 287). Figure 2. Statistical analysis of the number of ventral plus subcaudal scales in popula- tions A through G of Trimorphodon tau. The number in parenthesis indicates sample size of each population. The horizontal lines indicate sample range and the vertical lines sample mean. The black rectangles represent one standard deviation on each side of the mean; the white rectangles represent 2.646 standard errors on each side of the mean (Eberhardt, 1968). Dots represent specimens from the following inter- mediate localities outside the geographic range of the samples: 1, 2, Ventanas, Durango; 3, N of San Bias, Nayarit; 4, La Cumbre de los Arrastrados, Jalisco; 5, 7 mi S Rincon de Romos, Aguascalientes; 6, 7, Tacicuaro, Michoacan; 8, 10, Guanajuato, Guanajuato; 9, Mt. Cubilete, Guanajuato; 11, Jalapa, Veracruz; 12, Tuxpango, Vera- cruz; 13, San Diego, Puebla; 14, Quiotepec, Oaxaca. 8 Contributions in Science No. 179 Specimens from population C in eastern Mexico have fewer ventral- subcaudal scales than specimens from B and in this character indicate a con- tinuation of a dine. The trend is continued within population C in that northern specimens have higher totals than do southern specimens. The highest counts (290, 300) are found in northern specimens from Tamaulipas, while the lowest counts (278, 277) are from snakes from more southern localities in San Luis Potosi and Hidalgo. Two specimens from Tacicuaro, Michoacan on the interior drainage of the Rio Grande de Morelia and three from the vicinity of Guanajuato, Guana- juato in the drainage of the Rio Lerma are situated between populations B and C. Four of these five specimens exhibit ventral-subcaudal counts that lie on or between the means for populations B and C, and thus reflect their intermediate geographic position (Fig. 2, Nos. 6-10). Specimens from population D have the lowest ventral-subcaudal counts and are from localities at the southern limit of the range. Most of these locali- ties lie within the Pacific drainage of the Rio Verde althought some are from localities in the Atlantic drainage of the Rio Papaloapan. These eastern speci- mens, when considered together with two other specimens from eastern locali- ties north of population D (Fig. 2, Nos. 11, 14), suggest a possible connection between samples C and D along the eastern edge of the Mexican Plateau. The specimen from Jalapa, Veracruz (Fig. 2, No. 11) and the type of Trimorpho- don tan from Quiotepec, Oaxaca (Fig. 2, No. 14) have ventral-subcaudal counts that are within the range of variation of population D but slightly higher than the average and thus approach populations C and F in this character. Snakes from the largest sample, population E, average about 13 more ventral-subcaudal scales than specimens in population D and, in this respect, are nearly identical to those in population C (Fig. 2). Sample C is located nearly 400 kilometers north of E and separated from it by the Balsas Basin and the Mexican Plateau. Population F averages slightly more ventral-subcaudal scales than E, but this difference is not significant at the 95 percent level. All known specimens in sample F are from localities in the upper Balsas Basin. The holotype of Trimorphodon collaris from western Veracruz (Fig.. 2, No. 12) and the holo- type of T. forbesi from southern Puebla (Fig. 2, No. 13) are both from locali- ties in eastern drainage systems. These two specimens have ventral-subcaudal counts similar to those of population F. Thus, specimens from southeastern Puebla and adjacent Oaxaca and from Veracruz have two patterns in the number of ventral-subcaudal counts. The holotypes of Trimorphodon collaris and T. forbesi (Fig. 2, Nos. 12, 13) are more similar to specimens from popu- lation F, while the holotype of T. tan and the Jalapa specimen (Fig. 2, Nos. 11, 14) are more similar to snakes from population D. Snakes from population G have more ventral-subcaudal scales than are found in either E or F. In this respect sample G approaches sample B which 1970 Geographic Variation of Mexican Lyre Snakes 9 lies immediately to the north. There is no significant difference (95 percent level) between samples B and G in the numbers of ventral-subcaudal scales. This analysis shows that there is a clinal trend in total ventral-subcaudal scales within population A and among populations A, B, C and D, with higher counts in the north and lower counts in the south. Populations E, F, and G are more closely related to each other than they are to either of the two geographically adjacent populations, B in the north and D in the south. How- ever, these three Balsas-Tepalcatepec Basin samples (E, F, G) have the same trend of higher totals in the north and lower totals in the south. In this instance, however, the trend is somewhat less marked. Anal scale. — The anal scale is divided in all specimens except one. A male from Morelos (TCWC 7390) has an entire anal scale. Loreal scales. — The number of loreal scales varies from two to five on each side of the head, with the majority of specimens having either two or three loreal scales (Table 1 ) . Two loreals are more common in specimens from populations B, C and D, while three loreals are typical of populations A, E, F and G. Preocular scales. — Although preocular scales may vary in number from one to four, the majority of snakes in all populations have three (Table 1). Specimens with two preocular scales are found in all populations except D. A single preocular is present in one specimen from population F. Postocular scales. — Three postocular scales are characteristic of most specimens (Table 1). All snakes from sample F have three postoculars. Two postocular scales occasionally occur in the other six samples. Four scales are known only from specimens in populations A and E. Temporal scales. — Most specimens have either two or three primary temporals and three or four secondary temporals. The majority of specimens in populations A, E, F and G have three primary and four secondary temporal scales, while specimens from population B, C, and D usually have two primary and three secondary temporals (Table 1). The differences may reflect the smaller sample sizes of these populations. A single primary temporal occurs only once, in population B; five primary temporals occur only once, in popula- tion E. In five instances snakes from sample D have two secondary temporals. A comparison of the total number of lateral head scales (loreals, oculars and temporals) for the specimens from each sample shows that the higher numbers and means are characteristic of populations A (28-37, x = 30.9, N=23), E (26-39, x = 30.8, N = 48) and F (26-34, x = 31.2, N = 25), while the lower totals and means are found in samples B (23-31, x = 28.2, N=14), C (25-31, X = 27.9, N=12) and D (24-32, x = 28.2, N=20). Population G (24-32, X = 29.0, N=19) is somewhat intermediate in total head scales but is slightly closer to the lower than to the higher values. There is considerable variation in the number of head scales within some populations and some variation among several populations. We do not consider the different head 10 Contributions in Science No. 179 scale characteristics to be of systematic importance in distinguishing any of the samples. Labial scales. — Most snakes have either eight or nine supralabials, except in population C where all specimens have eight (Table 1 ) . Eight is the more common number of supralabials in all populations except F, in which nine occurs more frequently. Only in sample D are specimens with seven supra- labials found. The type specimen of Trimorphodon tau has six supralabials on each side, the lowest number known. Although the fourth and fifth supralabials border the eye in the majority of specimens examined, some variation was noted. In populations A, D, E, and F occasional individuals have the fifth and sixth supralabial bordering the eye. One specimen from each of populations F and G has the third, fourth and fifth supralabials bordering the eye on one side while the fourth and fifth border the eye on the other. The fifth supralabial borders the eye on one side in two specimens from population F; the usual condition is found on the other side. There is considerable variation in the number of infralabials from popu- lation to population and within populations (Table 1). Twelve is the most common number, although ten, eleven, and thirteen infralabials are found in nearly all samples. Population D has the lowest recorded number of infra- labials (nine) and has a lower mean than do the other six samples. The small gular scales usually separate the infralabial series from the chin shields at the level of the fifth or sixth infralabial scale. In population A this separation occurs in the majority of snakes at infralabial six. In population D this separation occurs in most specimens at infralabial five. In the other five populations the separation occurs about equally at the fifth or sixth infra- labial. Occasionally the infralabial series may be separated from the chin shields at a level as far anterior as the fourth infralabial or as far posterior as the seventh. Dorsal scales. — The dorsal scales of Trimorphodon are smooth, with paired apical pits. The scales are arranged in diagonal rows and vary con- siderably in number. The anterior count at one head length behind the head ranges from 21-27 rows; 75 percent of the individuals have 22, 23, or 24 rows. The number of scale rows at midbody varies from 17-25; 80 percent of the snakes have 22 or 23 rows. At one head length in front of the vent, there are between 14 and 18 rows, and at this level 83.33 percent of the sample have 15 or 16 rows. The reduction in scale rows is similar to the reduction found in Lepto- deira (Duellman, 1958:20). The reduction involves changes in the para- vertebral row. The level (ventral scale) at which reduction in scale rows take place is highly variable with no geographic consistency. PATTERN AND COLORATION Cross bands. — The number of dorsal body bands (body and tail blotches) in individuals with complete tails ranges from 18-46, though evidence suggests 1970 Geographic Variation of Mexican Lyre Snakes 11 Table 1. Range and Percent Distribution of Various Head Scales in the Seven Populations of Trimorphodon No. of specimens A 23 B 15 C 14 D 20 E 51 F 25 G 19 2 6.5% 66.6% 75.0% 60.0% 18.0% 12.0% 44.7% 3 87.0 33.4 25.0 35.0 79.0 86.0 55.3 Loreals 4 6.5 — — 5.0 2.0 2.0 — 5 — — — — 1.0 — — 1 2.0 2 6.5 13.3 14.3 — 8.9 2.0 15.8 Preoculars 3 93.5 83.3 85.7 97.5 88.1 90.0 84.2 4 — 3.4 — 2.5 3.0 6.0 — 2 19.2 3.5 10.7 7.5 1.9 10.5 Postoculars 3 76.6 96.5 89.3 92.5 94.1 100.0 89.5 4 4.2 — — — 4.0 — — 1 3.5 2 19.6 55.2 39.3 60.0 23.5 24.0 31.6 Primary 3 78.3 41.3 57.2 37.5 71.6 74.0 68.4 Temporals 4 21.0 — 3.5 2.5 4.0 2.0 — 5 — — — — 0.9 — — 2 — — — 12.5 — — — Secondary 3 26.1 44.8 63.0 50.0 25.7 12.0 47.4 Temporals 4 60.9 55.2 37.0 35.0 72.3 86.0 52.6 5 13.0 — — 2.5 2.0 2.0 — 7 10.0 — 8 60.9 69.0 100.0 72.5 63.9 39.5 57.9 Supralabials 9 37.0 31.0 15.0 29.9 58.1 39.5 10 2.1 — — 2.5 6.2 2.4 2.6 9 — — 2.5 — — — 10 2.1 3.3 9.5 27.5 2.1 4.6 — 11 28.4 30.0 28.6 50.0 14.9 23.3 26.5 Infralabials 12 47.8 43.4 47.6 20.0 45.8 44.2 44.1 13 21.7 20.0 14.3 — 35.1 25.6 20.6 14 3.3 — — 2.1 2.3 8.8 12 Contributions in Science No. 179 that higher numbers do occur (see below). The higher values are found in samples A, B, C and D while the lower numbers are characteristic of the three Balsas-Tepalcatepec Basin populations, E, F and G. The numbers of body bands for the seven populations and for geographically intermediate specimens are shown in Figure 3. There is a decrease of dorsal body bands within population A from north to south. Specimens from Sonora have 27-36, x = 32 (N=ll); specimens from northern Sinaloa have 29-33, x = 30 (N=14); specimens from central Sinaloa have 25-31, x = 28 (N=10). Snakes from population B on the aver- age have more body bands than specimens from population A, and snakes of population C tend to have more bands than those from B. Specimens that are geographically intermediate between samples A and B have an intermediate number of bands. The same is true for two specimens from the area between samples B and C (Fig. 3, Nos. 8, 9) . In population C the Hidalgo specimens apparently have more bands than Figure 3. Statistical analysis of the number of cross bands in populations A through G of Trimorphodon tan. See legend to Figure 2 for explanation of diagrams and symbols. 1970 Geographic Variation of Mexican Lyre Snakes 13 do the Tamaulipas specimens. The only Hidalgo specimen with a complete tail has 43 body bands. Three other snakes from Hidalgo appear to have 40, 42 and 49 bands or more. These three specimens have damaged tails and, for this reason, are not included in Figure 3. The higher number of bands indicated for southern specimens in population C may account for the high numbers found in two of the intermediate specimens between B and C (Fig. 3, Nos. 7, 10). Both of these specimens (FMNH 105193, USNM 11370) are from localities slightly farther south than most of the localities in either B or C. Specimens from population D, like population B, have fewer dorsal blotches than those from population C. Three of the four snakes from locali- ties intermediate between C and D in eastern Mexico are more similar to D than to C. The fourth snake (Fig. 3, No. 12), the holotype of Trimorphodon collaris, is much closer in band number to specimens from sample F. Snakes from populations E and F average 11 to 13 fewer blotches than specimens from population D, the sample that is closest geographically. They are most similar to snakes from population A which is the farthest removed geographically. In band count, samples E and E are most similar to sample G. These populations (E, F, G) are significantly different from all of the other populations. Although populations B and G are in close geographic proximity, there is no overlap in the number of body bands. In fact these two samples are separated by a mean difference of 15 dorsal body bands. The lengths of the bands may vary considerably on a single individual and among individuals of the same or different populations. In order to evaluate the differences in band size, the lengths on the middorsal scale row (as indi- cated by the number of included scales) of three band-interspace-band se- quences were compared. The first sequence behind the collar (anterior) , a mid- body sequence, and the last sequence befor the vent (posterior) were used. These values are presented in Table 2. In all populations the first dark cross band usually is the longest. The second cross band is longer on the average than either of the midbody bands, and they, in turn, always average longer than the posterior body bands. The mean band length of specimens from population A is greater than that of specimens from B, and mean band length in B is greater than in C (Table 2). In all instances the bands on snakes from population D average longer than bands in C but are shorter than bands on snakes in B. Bands of specimens from populations E, F and G average longer than those of any of the other four samples. The only exception is the mean first band length of sample A which is longer than in sample E but never as long as in samples E or G. In general, specimens from population G have longer bands than specimens from E, and specimens from F have longer bands than specimens from E. The longest mean band lengths at all three body positions are characteristic of specimens from population G while the shortest band lengths are always found in population C. The Range and Mean of Band-Interspace-Band Lengths at Three Middorsal Positions on Specimens from the Seven Populations 14 Contributions in Science No. 179 >0 0^ ON ' O rN ri q q q ^ oo ^ vn ON ^ 2 ^ q ^ 00 ^ § o o • ^ . 0^0 ^ o o ^'^^ON^qcd^qrd^ioio 00 CO q q , 'f CO q q ' q cd >o id ri *0 o o q CO »o ri 9 lo a; ^ ^ q q in2^io'^NOio'T ooin ^j^o6'-"qND^»Ari^ .^ • >d q q ir^ O NO ^ ^ ® q «oi «/~i NO ri S q ^ ^ro^'^m*oON(Sl2*^ro ^ a. O c/^ H U 0^ ^ O Ci^ band R 5.5-11.0 4.0-8.0 3. 5-6.0 3.0-6.0 5.5-14.0 6.0-12.5 8.0-14.0 X 7.7 5.8 4.5 4.6 8.6 8.8 11.1 1970 Geographic Variation of Mexican Lyre Snakes 15 There is an average decrease in interspace length from anterior to posterior in all samples except F and G. In the former the midbody interspaces average shorter than the posterior interspaces while in the latter the midbody interspaces average slightly longer than either the anterior or posterior inter- spaces which are the same (Table 2). Specimens from populations A and B have the smallest mean interspace lengths. Specimens from population F have the longest interspace lengths. There is no consistent number or location of bands which show signs of splitting or fusion (indicated by light centers) in any population. However, there is some consistency within certain populations in the extent of the light central area. For example, in Oaxaca the light areas within the darker bands are relatively large and diffuse. In Sinaloa the central light areas often form discrete spots. Collar length. — The length of the light nuchal collar varies among indi- viduals from the same population. In some specimens the first body blotch contacts the posterior margin of the head cap or the parietal scales on the midline or is separated from them by only half a dorsal scale. Others may have Figure 4. Statistical analysis of the collar lengths in populations A through G of Trimorphodon tan. See legend to Figure 2 for explanation of diagrams and symbols. 16 Contributions in Science No. 179 as many as six dorsal scales separating the first body blotch from the head cap or the parietal scales. However, the collars among the seven samples have nearly the same average lengths as indicated by the number of light scales on the midline (Fig. 4). Only population F has a noticeably higher number of dorsal scales in the nuchal collar. In addition, snakes from population F have the greatest interspace length (Table 2). Most of the specimens from inter- mediate localities have intermediate collar lengths which fall within the range of variation of the two samples concerned. Notable exceptions, however, are the collar lengths for the type specimens of T. collaris and T. forbesi which are plotted between populations C and D (Fig. 4, Nos. 12, 13) . Both of these have collar lengths most similar to those characteristic of sample F. The type of Trimorphodon forbesi has four scales, very near the mean for F, while the type of T. collaris has five and is near the upper end of the range of sample F. Dorsal head cap. — The pattern of dark and light colors on the head generally is consistent within a geographic area. Because of this geographic consistency, a detailed analysis of the head patterns is extremely useful in determining the relationships among the seven samples. Although individual head patterns may be very complex (Figs. 5, 6), they can be analyzed in terms of the following components: snout — light or dark rostral and internasal scales; presence or absence of a prefrontal bar — light bar extending across the prefrontal scales with dark anterior and posterior borders; presence or absence of an interocular (interorbital) bar — a light bar extending between the eyes, usually including the supraocular scales and the frontal scale; presence or absence of a parietal mark — a light mark of various design that may or may not connect to the interorbital bar or the light collar; presence or absence of an occipital spot — a medial, dark rounded spot located on the posterior portion of each parietal scale and the adjacent dorsal scales in the collar; posterior border — the shape of the posterior margin of the dark cap where it is bordered by the light nuchal collar. Most snakes from population A have dark snouts (a few specimens from near Alamos, Sonora have light snouts) and a light prefrontal bar (blending into the snout of the light snouted forms) that is restricted to the anterior third of the prefrontal scales (Fig. 5A) . Occasionally (three specimens) the pre- frontal bar is connected with the interorbital bar by a straight or irregular medial area. The interocular bar is always present, usually has irregular bor- ders, and generally is continuous between the orbits (interrupted in two speci- mens) . A parietal mark is always present but extremely variable in shape and extent (Fig. 7). Frequently the parietal mark is shaped like an irregular-armed trident, a bell, a teardrop, or consists of two oblong parts arranged at right angles to form an inverted T. The parietal mark (connected medially to the interorbital bar in one specimen) begins at the posterior edge of the frontal scale and continue posteriorly to connect with the collar (56 percent) . In the remaining specimens the parietal mark is separated from the light collar. An 1970 Geographic Variation of Mexican Lyre Snakes 17 occipital spot is present in 12 of the 14 specimens with the parietal mark-collar connection. The posterior border of the head cap may have a single median indentation (16 specimens), a lateral indentation on either side of a median posterior extension (five specimens), or it may be nearly straight (three specimens) . The two specimens from Ventanas, Durango have complete interorbital Figure 5. Representative dorsal head cap patterns for specimens of Trimorpliodon tan from A) Terreros, Sinaloa in population A; B) between Hostotipaquillo and Magdalena, Jalisco in population B; C) S Ciudad Victoria, Tamaulipas in population C; D) intermediate locality at Jalapa, Veracruz. 18 Contributions in Science No. 179 bars and light snouts that include the area of the prefrontal bar (Fig. 7). The simple parietal marks are restricted to the suture lines between the parietal scales. The parietal mark is connected to the light collar in one specimen and separated in the other. An occipital spot is present in the snake wdth the parietal mark-collar connection. The specimen from near San Bias, Nayarit has a dark snout, a prefrontal bar, an interorbital bar, a parietal mark connected to the collar, and an occipital spot. It is very similar to several Sinaloan specimens. Most snakes from population B have dark snouts (light in the Zacatecas specimen and two Michoacan specimens) and distinct light prefrontal bars (66 percent of individuals) . The interorbital bar is present (faint in KU 67735) and complete in all specimens except one (UMMZ 118950), where it is inter- rupted medially. The arms of the parietal mark, which is usually Y-shaped, edge the frontal scale posteriorly (Fig. 5B; Cope, 1900: fig. 315). The upright base extending posteriorly along the parietal suture may connect to the collar (six specimens) or remain separate from it (five specimens). In one specimen (FMNH 105194) the arms of the parietal mark connect with the interorbital bar on the supraocular scales forming a triangle (Taylor, 1939: pL 35, fig. 2). The parietal mark is oblong in three snakes and heart-shaped in one. The occipital spot is less common than in population A, being found in only two specimens. The posterior border of the head cap is rounded in two specimens, forms a straight edge in one, is indented medially in five, and has a medial posterior projection in the remaining seven snakes. A snake from La Cumbre, a locality southwest of population B, has a head pattern very similar to many specimens from B in that it has an interocular bar and a medial posterior projection of the head cap. The snout is somewhat lighter than the rest of the head cap and includes the prefrontal bar. The parietal mark is V-shaped and outlines the frontal scale posteriorly. Two specimens from Tacicuaro, Michoacan (southeast of population B) have dark snouts, interorbital bars, and irregular Y-shaped parietal marks that connect to the light collar. An occipital spot is present in one specimen but absent from the other (Fig. 7) . Although these two snakes lack prefrontal bars, they are similar in head pattern to other snakes from population B. Three specimens from Guanajuato (east of population B) have inter- ocular bars and Y-shaped parietal marks characteristic of most specimens from B but resemble the Tacicuaro specimens in lacking the prefrontal bars. The parietal mark connects to the collar in two specimens and is separate from it in the third. The snout, including the prefrontal bar, is light in one specimen and dark in two. An Aguascalientes specimen from a locality north of population B has a light snout that includes the prefrontal. In this respect the Aguascalientes snake is similar to one of the Guanajuato specimens and to the Zacatecas specimen in population B. The interorbital bar is complete and the parietal mark, which 1970 Geographic Variation of Mexican Lyre Snakes 19 is roughly Y-shaped, does not connect to the light collar. The posterior border of the head cap is rounded. Specimens from sample C have a very complex head pattern. Half of the individuals examined have a light snout that encompasses the prefrontal area. The other specimens have a dark snout with either a well-developed or very faint prefrontal bar. Many specimens, especially those with the lighter snout have a patch of dark color in the center of each prefrontal scale. These patches may be joined medially or separated by a lighter suture line between the prefrontal scales. About half of the specimens have a second light bar on the posterior edges of the prefrontals. The interocular bar is always present and usually expanded on the frontal scale. In one-third of the specimens the inter- orbital bar is connected anteriorly to the prefrontal bars and posteriorly to the parietal mark. In these specimens the dark head color actually is restricted to six spots, one in the center of each prefrontal scale, one at the anterior edge of each supraocular and adjacent frontal scale, and one on the posterior part of each supraocular and adjoining frontal and parietal scales. In most of the specimens from Tamaulipas and San Luis Potosi, the arms of the parietal mark form a V and outline the posterior edges of the frontal scale. The posterior part of the parietal mark bifurcates and encloses a large occipital spot before joining the collar (Fig. 5C) . In these specimens the posterior border of the head cap is indented medially. The parietal marks of the four specimens from Hidalgo differ from the more northern representatives of sample C. One specimen (LACM 53023) has a Y-shaped mark outlining the frontal edges and connecting to the collar. It lacks an occipital spot and has a postero-medially indented head cap. In two other specimens (AMNH 93432 and ANSP 14770) the parietal mark is lyre- shaped and not connected to the light collar. The arms of the lyre extend to the frontal borders, connecting with the interorbital bar on one side in one speci- men. They do not outline the postero-medial edges of the frontal scale. In both specimens the occipital spot is incorporated into the head cap, thereby extend- ing the posterior border of the cap medially. In the fourth specimen (USNM 1 10401 ) the interocular bar extends posteriorly to the tip of the frontal scale. The parietal mark is fragmented into three parts. On the left side the mark borders the frontal anteriorly and extends posteriorly nearly to the light collar forming a shallow, laterally concave arc. On the right side only the anterior half of the arc is present. The posterior half consists of a light rounded spot on the posterior part of the right parietal scale, just above the light collar. The posterior border of the head collar is nearly straight medially but rounded laterally. A specimen from Jalapa, Veracruz has a light snout extending over the anterior half of the prefrontals. Each nasal scale is very faintly pigmented. The interorbital bar is complete and connects medially to the light snout along the prefrontal suture. The parietal mark is irregular in shape, restricted to the 20 Contributions in Science No. 179 parietal suture (Fig. 5D) and connected to the light collar. The occipital spot is well defined. The type specimen of Trimorphodon collaris lacks pigment on the inter- nasal scales. Its interocular bar is broken about the middle of each supraocular, and it has a more rectangular-shaped parietal mark connected to the collar. In other respects it is very similar to the Jalapa specimen. The type specimens of Trimorphodon tan and T. forbesi also are from localities in the Atlantic versant and generally are similar to the two specimens mentioned above. The type of T. tail has dark circular marks in the center of each internasal scale. The light snout is connected medially and laterally with the prefrontal bar which is also connected with the complete interocular bar along the suture between the prefrontals. The lighter color of the collar con- tinues onto the temporal parts of the head connecting with the postero-lateral portions of the interocular bar. As a result, the dark parietal portion of the head cap is greatly restricted and completely surrounded by the lighter colora- tion. The parietal mark is connected to the collar. Posteriorly the head cap is nearly straight. The occipital spot is present but small. The type of T. forbesi also has a lighter snout that includes the prefrontal area. The interocular bar is complete and connects along the prefrontal suture to the lighter snout. The parietal mark is rounded and broadly joined to the light collar. The occipital spot is small. The posterior border of the head cap is rounded laterally and in- dented medially. In general these four specimens from eastern Mexico (Vera- cruz, Puebla, Oaxaca) are very similar in overall head pattern. The head patterns of specimens from population D generally are more uniform than are those from population C. Most of the specimens (75 per- cent) have a dark snout and lack a prefrontal bar. Two of the five individuals with a light snout have dark pigment in the central parts of the internasal scales and a suggestion of a prefrontal bar. In this respect they are similar to the type of Trimorphodon tau. In several specimens, including light-snouted and dark-snouted individuals, the edges of the internasal and prefrontal scales are lighter than the central portions (Fig. 6A). Ten specimens have a light bar along the suture between the prefrontals. Often this bar is connected to the interocular bar (50 percent) forming the inverted T-pattern characteristic of Trimorphodon tau (Cope, 1869: 152) . The interocular bar is always present and frequently (50 percent) extends laterally to the orbits. In some specimens (eight of 20) the interocular bar forms a shallow arc, concave posteriorly. The parietal mark has a simple configuration, rarely touching the frontal scale, and is connected with the light collar in all but two specimens (Taylor, 1940: fig. 8). In one specimen (AMNH 1 0093) the mark is irregular in shape (two lateral arms on each side) and contacts the posterior part of the frontal scale. An occipital spot is present in only five specimens. The head cap is indented medially in all but two specimens from sample D. In these two specimens the parietal mark and the collar do not connect, and 1970 Geographic Variation of Mexican Lyre Snakes 21 thus the head cap extends postero-medially. The white collar extends anteriorly along the sides of the head in these specimens, together with 16 of the 18 indented head cap forms, giving the head cap a rounded appearance. In two (Fig. 7) the postero-lateral edges of the cap are nearly straight. We examined the head patterns of 48 snakes from population E. In only eight of these does the light snout include the prefrontal bar area. Although the Figure 6. Representative dorsal head cap patterns for specimens of Trimorphodon tan from A) 3 mi NE Mitla, Oaxaca in population D; B) Chilpancingo, Guerrero in population E; C) Coalcoman, Michoacan in population G; D) 4 mi S Autlan, Jalisco in population G. 22 Contributions in Science No. 179 snout in these few specimens is somewhat lighter than the rest of the cap, it is still well pigmented and not the white or very light tan characteristic of certain light-snouted specimens from populations A and C. None have a defined pre- frontal bar, although three individuals have a pair of light spots on the anterior prefrontal scales. Three other specimens have a medial light spot on the pre- frontal suture. An interorbital bar is lacking in 23 specimens, reduced to a median frontal spot or to two lateral supraocular spots in 21 specimens, and complete in four specimens. None has a connection between the interorbital bar and parietal mark. In all specimens, except four, the parietal mark is con- nected to the light collar (Fig. 6B) or is absent. Of these four, two have a small circular spot on the parietal suture, one has a small central and two small lateral spots, and the fourth has two lateral spots that appear to be remnants of a V- shaped mark. Of the remaining 44 specimens, eight have no indication of a parietal mark; the posterior margins of their head caps are rounded laterally (Fig. 7) . Most of the others have a median extension of the collar connecting to the parietal mark. Usually the parietal mark or anterior end of the collar extension is rounded, but occasionally it may be Y- or V-shaped or may narrow rapidly at its anterior apex near the frontal scale. Occasionally (six specimens) the collar has a double extension onto the parietal region, and the head cap extends posteriorly on the midline. The head cap is rounded laterally in nearly all of these specimens. An occipital spot is present in only nine. In twelve of 24 specimens from sample F, the snout is lighter than the rest of the head cap and includes the prefrontal bar area. In this respect they are similar to the light-snouted individuals of population E. Only two of the dark snouted specimens have a prefrontal bar. When a complete interorbital bar is present (three specimens), it is faint. In twelve other snakes, the bar is fragmented or reduced to one, two, or three circular spots or short bars. There is not interorbital bar in nine specimens (Taylor, 1940: pi. 52) . Four specimens lack parietal marks and have a head cap with a rounded posterior margin. Three specimens have shallow V-shaped parietal marks that outline the frontal scale and are not connected with the light collar. In these specimens the posterior edge of the head cap has a single medial indentation, two lateral indentations, or a postero-medial extension. In the remaining snakes (17) the parietal mark is connected to the collar, and thereby, the posterior edge of the head cap is indented medially. The extended collar-parietal mark may be broadly Y-shaped (Fig. 7), triangular, narrow and restricted to the parietal suture, or wishbone-shaped. There is a large distinct occipital spot in the collar extension in most (12 of 17) of these specimens. The head patterns are remarkably consistent in population G. None, ex- cept the type of T. fasciolata, have a light snout or show any trace of a pre- frontal bar. In the type specimen of T. fasciolata the snout and the edges of the prefrontal scales are somewhat lighter than the rest of the head cap. An inter- orbital bar is lacking in all specimens (Fig. 6C) except three. In two of these 1970 Geographic Variation of Mexican Lyre Snakes 23 (UMMZ 112516 and the type of T. fasciolata) the supraocular scales have light centers. The third specimen from Autlan, Jalisco (KU 27191) has two light circular spots in the interorbital area. Distinct parietal marks are lacking from all except the Autlan specimen (Fig. 6D) and a specimen from near Apatzingan (UMMZ 112515). In the latter there are two small spots on the posterior third of each parietal scale. The posterior margin of the head cap is indented in one specimen from Apatzingan (FMNH 37079) and the type of T. fasciolata and shallowly indented in a third (UMMZ 104696) . Occipital spots, lacking in other specimens, are present in these three snakes. The posterior margin of the head cap is doubly indented in nine specimens (Fig. 6C), rounded in four, nearly straight in two, and slightly convex in one. The geographic distribution of the different head patterns is illustrated in Figure 7. Coloration. — These snakes have considerable variation in coloration from sample to sample and within some samples. Because adequate color descrip- tions and pictures of specimens from many areas are available in the literature (Taylor, 1939: 364-367; 1940: 474-479; Smith, 1941: 161-167; Peters, 1954: 33; Davis and Dixon, 1957: 24; Duellman, 1961 : 111; Brown and Brown, Figure 7. Geographic distribution of representative head cap patterns for specimens of Trimorphodon tan from localities within the seven samples and selected inter- mediate localities. 24 Contributions in Science No. 179 1967: 325), no detailed color description will be included here. Instead we will summarize this information and present it together with color notes on living specimens from Sinaloa and Puebla. The ground color of adults ranges from various shades of brown to fawn, pale tan or gray, usually becoming darker posteriorly. Juveniles often have a more reddish or grayish ground color than do the adults. The cross bands of both age groups are darker than the ground color and often yellow-brown or gray-brown to black. In many specimens the center of each band is lighter. The edges of each band are usually bordered with dark brown or black. Juveniles apparently have darker bands than adults from the same area. Laterally, smaller dark blotches are present in the interspaces of some specimens. The head cap is brown, gray-brown or black and nearly the same color as the cross bands. The lips, chin and throat usually are tan or white. The venter may be tan or creamy white, and often with a reddish or salmon tinge posteriorly. In some snakes the venter is mottled by ventral extensions of the cross bands and interspace blotches; in others the venter may be spotted. Some have a nearly immaculate belly. In some juveniles the ventral surface of the tail may be nearly black. The iris may be yellow-brown, gray-brown or grayish-tan. Discussion CHARACTER ANALYSIS Much similarity exists among the seven samples. Some characters indicate independent, discordant patterns of variation. Even though most characters are not diagnostic, three are particularly useful in deducing the relationships among the seven samples: number of ventral plus subcaudal scales, number of dorsal body bands, and dorsal head cap pattern. The number of ventral-subcaudal scales decreases from north to south in samples A through C. The trend continues southward into population D, where the lowest counts are recorded. Two specimens (Fig. 2, Nos. 11, 14) from intermediate areas between C and D are more similar to D than to C in this character, while two other specimens from intermediate areas (Fig. 2, Nos. 12, 13) are more similar to C and F than they are to D. This suggests a con- nection among the samples through the high river valleys of southeastern Puebla and Oaxaca. Population G has the highest counts and is the most northern of the three Balsas-Tepalcatepec Basin samples. In this character sample G approached sample B, its closest geographic neighbor. The ventral- subcaudal scale counts will not allow for absolute distinction between speci- mens from geographically adjacent samples. An analysis of the number of dorsal cross bands (Fig. 3) gives a slightly different picture of the relationships among the seven samples. Among popula- tions A, B, and C there is an increase in number of dorsal blotches from north- west to southeast. The geographically intermediate specimens generally con- 1970 Geographic Variation of Mexican Lyre Snakes 25 firm this trend. Sample D has fewer bands than does C and thus has a slightly different relationship to A, B, and C in this character than it did in terms of ventral-subcaudal scales. In this respect, however, the number of dorsal blotches of sample D approaches those characteristic of samples E and F. Of the four specimens from eastern Mexico that are geographically intermediate among samples C, D and F, three (Fig. 3, Nos. 11, 13, 14) have counts most similar to the counts of D while the fourth (Fig. 3, No. 12) has a count most like those of sample F. In dorsal cross bands E, F and G are closely related to each other and well differentiated from sample D. The most striking thing about the dorsal band counts is the difference between samples G and B which do not approach each other even though the two populations are closely adjacent (Fig. 3). This is a notable exception to their similarity in the number of ventral-subcaudal scales. The same general relationship is present in the seven samples for both dorsal head patterns and dorsal cross bands. An interocular bar, usually com- plete, and a discrete parietal mark are always present in specimens from samples A, B, C, and D. The remnants of an interocular bar are present in 43 of 91 specimens from samples E, F and G. Only seven specimens from the Balsas-Tepalcatepec samples have a complete interocular bar. A discrete parie- tal mark, that is one not connected to the collar, is found only in nine of the 91 specimens in samples E, F and G. A prefrontal bar or light snout including the area of the prefrontal bar is found in most snakes from samples A, B and C. In sample D a prefrontal bar is usually lacking, as it is in most specimens from samples E and G. Approximately 50 percent of the specimens from sample F have light snouts that include the area of the prefrontal bar and thus are some- what intermediate between samples A, B, C and D, E, G. In summary, the Balsas-Tepalcatepec Basin samples (E, F and G) have a darker, more uniform head cap than is found in specimens from samples A, B, C and D. The Balsas-Tepalcatepec samples generally lack the prefrontal and interocular bars and the discrete parietal marks characteristic of the other samples. In general they lack the ornate head marking characteristic of the other four populations. Except for two snakes (KU 27191 and USNM 110400), there is no indication of any similarity between samples B and G in characters of head pattern. Some specimens from samples E and F show an apparent relationship to samples C and D by possessing remnants of an inter- ocular bar and parietal mark. The head patterns of specimens from areas geo- graphically intermediate between samples A, B, C and D have intermediate head patterns. TYPE SPECIMENS It seems appropriate to discuss briefly the six type specimens with refer- ence to our seven populations to clarify our systematic treatment of the various samples. 26 Contributions in Science No. 179 The type specimen of Trimorphodon tau is a small snake in relatively poor condition. It was adequately described by Cope (1869) and redescribed by Taylor (1939). Although from a locality northwest of sample D, in all char- acteristics, it is typical of specimens from that sample. The holotype of Trimorphodon upsilon is typical in all respects of speci- mens from sample B and was included in that unit. Zweifel (1959) noted that many of the specimens collected by Major probably did not come from Guadalajara, Jalisco as the locality information indicated but possibly were from Colima. However, Zweifel did point out that the type of Trimorphodon upsilon may actually have been from Guadalajara as additional specimens of this species are known from the same general area. Our analysis supports Zweifel’s contention that Guadalajara probably is the type locality as originally stated. The cotypes of Trimorphodon latifascia from Puebla have the character- istics typical of Puebla specimens from sample F. However, they were not included in sample F because the specific type locality is unknown (Peters, 1869). For purposes of nomenclatural stability and in accord with Article 74, International Commission on Zoological Nomenclature (1964), we designate ZMB 6652a, the larger of the two syntypes (588 mm total length), as the lectotype. The smaller specimen (417 mm total length), ZMB 6652b, is desig- nated the paralectotype in accord with recommendation 74E. Both specimens are males and have 211, 209 ventrals, 79, 74 subcaudals, and 21, 22 dorsal body bands. The snout is slightly lighter than the head cap in both specimens. In one the interocular bar consists of two short lateral spots and in the other of a median spot. The posterior border of the head cap is indented medially and rounded laterally in both. A faint occipital spot is present in the smaller paralectotype. The holotype of Trimorphodon collaris, like the type of T. tau, is a juvenile. Cope’s description (1875) is accurate for the characters listed. Some confusion exists concerning the type locality and number of specimens (see Historical Review and Smith, 1941 : 166) . This specimen, although from a locality somewhat intermediate among samples C, D and F, is typical of speci- mens from sample F in the important characteristics of ventral plus subcaudal scales, collar length, and number of dorsal blotches. The band-interspace-band lengths also are closest to those characteristic of sample F. The head pattern is similar to the patterns found on specimens from both D and F, although it is more similar to those from D. We consider the specimen to be most similar to snakes from sample F. The type specimen of Trimorphodon fasciolata is a large male collected from a locality on the northeastern periphery of sample G. The specimen was adequately described by Smith ( 1941 : 160-162) . In all characteristics except certain details of head pattern (see previous description), it is typical of rep- 1970 Geographic Variation of Mexican Lyre Snakes 27 resentatives of sample G, in which it was included. These differences in head pattern emphasize the marginal nature of the locality and may reflect its proximity to specimens from sample B and the intermediate specimens from Tacicuaro, Michoacan. Finally, the type of Trimorphodon forbesi is an exceptionally large male that is relatively well preserved but badly faded. In general our findings, on examination of the type, agree well with Smith’s description (1941: 163-165) with two exceptions. We recorded five more ventrals and two more subcaudals than listed in the type description. Unfortunately the ventral scales are dam- aged in some areas and accurate counts are difficult to make. This may account for the discrepancies. However, we believe our count of 297 scales to be accurate. The other point with which we disagree is Smith’s (1941: 164-165) dis- cussion of and reference to the “pattern reversal” characteristic of Trimorpho- don forbesi. It is quite obvious that this specimen is faded; apparently the brown color of the head pattern faded more rapidly than did the dark borders along the interocular bar and nuchal collar, thus giving the appearance of a light head pattern with a darker interocular bar and nuchal collar. We cannot accept the contention that this is anything other than a badly faded but originally nor- mally colored head pattern. The pattern itself is similar to the pattern of the type of Trimorphodon tau and other specimens in sample D. The remaining characters of forbesi that Smith (1941: 165) used to dis- tinguish it from other species include a divided anterior loreal, the fifth and sixth labials entering the eye, a faintly marked belly and white subcaudal sur- face, the number of cross bands, the narrow black borders of the cross bands, and the length of the cross bands on the mid-dorsal line. All of these charac- teristics are known from other samples (see previous description of geographic variation) and therefore are of little value in distinguishing this specimen. In ventral-subcaudal scales and collar length this specimen is more similar to sample F. In number of dorsal blotches and in head pattern, it is more similar to sample D. Certain characteristics of head scales and blotch length are inter- mediate between those for samples D and F. Thus, the type of T rimorphodon forbesi has some characters typical of specimens from sample D, some typical of specimens from sample F, and some that are intermediate. Because of its intermediate locality and character- istics, we consider the specimen to represent a population of Trimorphodon that apparently bridges the gap between samples D and F. In order to evaluate the usefulness of the three characters, ventral-sub- caudal scales, body bands and head cap pattern, as indications of phylogenetic relationship, we attempted to correlate their geographic variation with various environmental parameters. Klauber (1941 ) showed that desert specimens con- sistently tend to have more ventral scales than coastal specimens of the same species. He also found that specimens of Trimorphodon vandenburghi from 28 Contributions in Science No. 179 the desert had more dorsal blotches than those along the coast. Klauber at- tributed these disparities to differences in humidity, and possibly temperature. Fox (1948) showed how different temperature regimes can influence scutella- tion in developing embryos of Thamnophis. Gravid females from the same natural population kept at cooler temperatures give birth to young with significantly fewer ventral and subcaudal scales than young from females maintained at warmer temperatures. Stebbins (1949) suggested that physical factors of humidity, temperature and light are important considerations deter- mining the coloration and pattern of the salamander Ensatina. In our analysis we could find no consistent correlation between environ- mental factors of temperature, rainfall or elevation and the geographic varia- tion in ventral-subcaudal scales, dorsal cross bands or head pattern. In some samples there appeared to be a positive correlation between ventral-subcaudal scales and temperature; in other samples the same character indicated an inverse relationship to mean temperature. These same results generally were found in other comparisons. Unfortunately, the lack of precise microenviron- mental information for many areas prevents detailed analysis of the possible correlation between other environmental factors and morphology. Band lengths (Table 2) and band numbers (Fig. 3) have an inverse re- lationship. As the average band length increases, the number of bands de- creases, and the snakes have a greater dark to light ratio. This is found in samples E, F, and G and is exemplified also by their darker head pattern. The Balsas-Tepalcatepec Basin is characterized by both high mean annual tempera- tures and high mean annual rainfall (Vivo Escoto, 1964). This suggests that a detailed analysis might show a correlation between mean annual temperature, rainfall, and/or light intensity and the number of dorsal bands and head pattern. Details of vegetation type and density would also have to be considered. Variation in band numbers among the samples may be the result of splitting or fusion of some of the cross bands. For example, there is an average increase of five body bands from population A to B, but a decrease of about 15 scale lengths (difference between totals of six band lengths in Table 2). Although the light centers of many bands suggest splitting or fusion, there is no consistent location or number of bands which have light centers within any population. In addition, we were unable to derive the various band patterns from one another. It seems unlikely, therefore, that the light centers charac- teristic of many cross bands are indicative of fusion or division, a possible means of decreasing or increasing the band number. Differences in cross band numbers and lengths may represent adaptive responses to different selective pressures that increase or decrease the overall ratios between dark and light areas on the snakes. SYSTEMATIC TREATMENT Our data indicate overall similarities among the seven samples and the type specimens in nearly all characteristics examined. The analyses reveal 1970 Geographic Variation of Mexican Lyre Snakes 29 a geographic trend towards a southern reduction in ventral-subcaudal scales among samples A, B, C and D and among samples E, F and G, and also point to the similarities among samples A, B, C and D and samples E, F and G in number of body bands and head pattern. Specimens with characteristics that are intermediate between these two groups indicate that only a single species is involved. Therefore, we refer our seven samples, their intermediates, and the types of Trimorphodon tau, upsilon, collaris, Jatifascia, fasciolata and forbesi to one species. After careful consideration and consultation, we con- clude that Gunther’s (1895) use of Trimorphodon upsilon did not constitute the action of the first revisor as defined in Article 24, International Code of Zoological Nomenclature. Therefore, the name Trimorphodon tau is applicable because of the action of Smith and Darling (1952) who gave priority to tau rather than upsilon, by their action as first revisor. TRIMORPHODON TAU Cope Trimorphodon tau Cope, 1869, Proc. Amer. Phil. Soc., 11, p. 152 (type description; type locality “western part of the Isthmus of Tehuantepec, Mex- ico,” in error); 1875, J. Acad. Natur. Sci. Philadelphia, ser. 2, 8, p. 131. Sumi- chrast, 1892, La Naturaleza, 6(2), p. 44 (type from Quiotepec, between Tehu- acan and Oaxaca). Cope, 1886, Proc. Amer. Phil. Soc., 23, p. 286; 1887, Bull. U.S. Natl. Mus., 32 p. 68; 1892, Proc. U.S. Natl. Mus., 14* p. 678. Gunther, 1895, Biologia Centrali-Americana, Reptilia, p. 174. Boulenger, 1896, Cata- logue of the snakes in the British Museum, vol. 3, p. 56. Mocquard, 1899, Bull. Soc. Phil., ser. 9, l,p. 157. Cope, 1900, Rep. U.S. Natl. Mus. for 1898, p. 1101. Gadow, 1905, Proc. Zool. Soc. London, 1905, p. 224. Mocquard, 1908, in Dumeril and Bocourt, Mission Scientifique au Mexique et dans I’Amerique Centrale, 3, pp. 908, 912. do Amaral, 1929, Mem. Inst. Butantan, 4, p. 202. Taylor, 1939, Univ. Kansas, Sci. Bull., 25, pp. 366-67; 1940, Univ. Kansas Sci. Bull., 26, pp. 474-77, fig. 8, plate 51. Smith, 1941, Proc. U.S. Natl. Mus., 91, pp. 166-67; 1943, Proc. U.S. Natl. Mus., 93, p. 494. Smith and Taylor, 1945, Bull. U.S. Natl. Mus., 187, p. 148. Taylor, 1949, Univ. Kansas Sci. Bull., 33, pp. 210-11. Davis and Dixon, 1957, Southwestern Natur., 2, p. 24. Martin, 1958, Misc. Publ. Mus. Zool. Univ. Michigan, 101, p. 74. Duellman, 1961, Univ. Kansas Publ. Mus. Nat. Hist., 15, p. 112; 1965, Univ. Kansas Publ. Mus. Nat. Hist., 15, pp. 655, 657, 660-61, 666, 685. Lynch and Smith, 1965, Herpe- tologica, 21, p. 168. Brown and Brown, 1967, Texas J. Sci., 29, p. 325. Eteirodipsas biscutata (in part), Jan, 1863, Elenco sistematico degli ofidi . . ., p. 105. Jan and Sordelli, 1872, Iconographic generate des ophidiens, vol. 3, livr. 39, plate 1, fig. 3. Cope, 1892, Proc. U.S. Natl. Mus., 14, p. 679 (footnote). Gunther, 1895, Biologia Centrali-Americana, Reptilia, p. 175. Trimorphodon upsilon Cope, 1869, Proc. Amer. Phil. Soc., 11, p. 152 (type description; type locality “Guadalaxara, West Mexico”); 1875, J. Acad. 30 Contributions in Science No. 179 Natur. Sci. Philadelphia, ser. 2, 8, p. 131; 1886, Proc. Amer. Phil. Soc., 23, p. 286; 1887, Bull. U.S. Natl. Mus., 32, p. 68; 1892, Proc. U.S. Natl. Mus., 14, p. 678. Gunther, 1895, Biologia Centrali-Americana, Reptilia, pp. 174-75. Boulenger, 1896, Catalogue of the snakes in the British Museum, vol. 3, p. 55. Mocquard, 1899, Bull. Soc. Phil., ser. 9, 1, p. 157. Cope, 1900, Rep. U.S. Natl. Mus. for 1898, pp. 1 104-1105, fig. 315. Gadow, 1905, Proc. Zool. Soc. London, 1905, pp. 196, 231, 233; 1910, Zool. Jahrbuch, 29, pp. 697, 701. Mocquard, 1908, in Dumeril and Bocourt, Mission Scientifique an Mexique et dans I’Amerique Centrale, 3, pp. 908, 910-11. do Amaral, 1929, Mem. Inst. Butan- tan, 4, p. 202. Taylor, 1939, Univ. Kansas Sci. Bull., 25, pp. 365-66, plate 35, fig. 2. Smith, 1941, Proc. U.S. Natl. Mus., 91, pp. 162-163; 1943, Proc. U.S. Natl. Mus., 93, p. 494. Smith and Taylor, 1945, Bull. U.S. Natl. Mus., 187, p. 148. Zweifel and Norris, 1955, Amer. Midland Natur., 54, p. 245. Duellman, 1961, Univ. Kansas Publ. Mus. Nat. Hist., 15, p. 112. Dips as biscutatus var. latifascia Peters, 1869, Monatsber. Akad. Wiss. Berlin, p. 877 (type description; type locality, Puebla, Mexico) . Trimorphodon collaris Cope, 1875, J. Acad. Natur. Sci. Philadelphia, ser. 2, 8, p. 131 (type description; type locality “Orizaba, Vera Cruz”). Sumichrast, 1882, La Naturaleza 6(2), p. 44 (type(s) came from Tuxpango, near Ori- zaba) ; Cope, 1886, Proc. Amer. Phil. Soc., 23, p. 286; 1887, Bull. U.S. Natl. Mus., 32, p. 68; 1892, Proc. U.S. Natl. Mus., 14, p. 679. Gunther, 1895, Biologia Centrali-Americana, Reptilia, p. 174. Cope, 1900, Rep. U.S. Natl. Mus. for 1898, p. 1101. Taylor, 1939, Univ. Kansas Sci. Bull., 25, pp. 364-65. Smith, 1941, Proc. U.S. Natl. Mus., 91, pp. 165-66; 1943, Proc. U.S. Natl. Mus., 93, p. 493. Smith and Taylor, 1945, Bull. U.S. Natl. Mus., 187, p. 147. Sibon biscutatus, Garman, 1883, Mem. Mus. Comp. Zool. Harvard Univ., 8, pp. 16, 134 (part). Sibon biscutatum var. latifasciatum, Garman, 1883, Mem. Mus. Comp. Zool. Harvard Univ., 8, pp. 16-17, 134. Sibon upsilon, Garman, 1883, Mem. Mus. Comp. Zool. Harvard Univ., 8, p. 134. Trimorphodon biscutatus, Gunther, 1895, Biologia Centrali-Americana, Reptila, p. 174 (part). Trimorphodon upsilon var. collaris, Mocquard, 1908, in Dumeril and Bocourt, Mission Scientifique au Mexique et dans I’Amerique Centrale, 3, plate 74, fig. 2, a, b. Trimorphodon latifascia, Taylor, 1939, Univ. Kansas Sci. Bull. 25, pp. 364-65, plate 36, fig. 2; 1940, Univ. Kansas Sci. Bull., 26, p. 479, plate 52. Smith, 1941, Proc. U.S. Natl. Mus., 91, p. 160. Smith and Taylor, 1945, Bull. U.S. Natl. Mus., 187, p. 148. Schmidt and Shannon, 1947, Fieldiana Zool. Chi- cago Nat. Hist. Mus., 31, p. 83. Hall, 1951, Univ. Kansas Sci. Bull., 34, pp. 207-08. Davis and Smith, 1953, Herpetologica, 8, p. 140. Peters, 1954, Occ. Papers Mus. Zool. Univ. Michigan, 554, p. 32. Davis and Dixon, 1957, South- 1970 Geographic Variation of Mexican Lyre Snakes 31 western Natur., 2, p. 23. Webb and Fugler, 1957, Herpetologica, 13, p. 33. Duellman, 1958, Occ. Papers Mus. Zool. Univ. Michigan 589, p. 12. Davis and Dixon, 1959, Proc. Biol. Soc. Washington, 72, p. 88. Duellman, 1961, Univ. Kansas Publ. Mus. Natur. Hist., 15, pp. 111-12. Fouquette and Ross- man, 1963, Herpetologica, 19, p. 199. Duellman, 1965, Univ. Kansas Publ. Mus. Nat. Hist., 15, pp. 655-56, 659, 680, 699. Trimorphodon fasciolata Smith, 1941, Proc. U.S. Natl. Mus., (type de- scription; type locality “near Zararacua [sic] Falls, 6 kilometers southeast of Uruapan, Michoacan.”) 91, pp. 160-62; 1943, Proc. U.S. Natl. Mus., 93, p. 493. Smith and Taylor, 1945, Bull. U.S. Natl. Mus., 187, p. 147. Schmidt and Shannon, 1947, Fieldiana Zool. Chicago Nat. Hist. Mus., 31, p. 83. Davis and Smith, 1953, Herpetologica, 8, p. 140. Peters, 1954. Occ. Papers Mus. Zool. Univ. Michigan, 554, pp. 32-34. Davis and Dixon, 1957, Southwestern Natur. 2, p. 23. Trimorphodon forbesi Smith, 1941, Proc. U.S. Natl. Mus., 91, pp. 163-65 (type description; type locality “from San Diego [about 5 mites south of Tehuacan], Puebla”); 1943, Proc. U.S. Natl. Mus., 93, p. 493. Smith and Taylor, 1945, Bull. U.S. Natl. Mus., 187, p. 147. Trimorphodon tau tau, Smith and Darling, 1952, Herpetologica, 8, p. 85. Trimorphodon tau upsilon, Smith and Darling, 1952, Herpetologica, 8, p. 85. Zweifel, 1959, Amer. Mus. Novitates, 1949, pp. 3-4, 6-8. Description. — A species of Trimorphodon with 201-231 ventral and 61- 85 subcaudal scales in males; 210-243 ventrals and 55-80 subcaudals in fe- males; 265-319 total ventral-subcaudal scales; a divided anal scale; usually 3 (2-5) loreal scales, 3 (1-4) preocular, 3 (2-4) postocular, 3 (1-5) primary temporal, 4 (2-5) secondary temporal, 8 (7-10) supralabial with the fourth and fifth (third-sixth) bordering the eye, and 12 (9-14) infralabial scales usually separated from the chin shields at the sixth (fourth-seventh) infra- labial; dorsal scales usually in 22, 23, or 24 (21-27) rows at one head length behind the head, 22 or 23 (17-25) rows at midbody, and 15 or 16 (14-18) rows at one head length in front of the anus. The snakes are usually gray, tan or light brown in ground color and crossed by 18-46 body bands that usually reach the edges of the ventral and subcaudal scales. The cross bands are various shades of brown or black and often darkest along their margins. Most dark cross bands have lighter central areas that approach the ground color. These central areas may be large and diffuse or concentrated into discrete spots. They suggest the splitting of one band or fusion of two bands. The cross bands decrease in length on the midline from anterior to posterior. Often the first cross band is considerably larger than the second. The cross bands are widest dorsally and taper to a broad or narrow base laterally. Laterally, small, irregular interband spots may be present. Frequently the body bands continue onto the abdominal scales giving 32 Contributions in Science No. 179 the ventral surface an irregular blotched pattern. However, in some specimens the abdominal scales are irregularly spotted or nearly immaculate. The head pattern is extremely variable (see discussion). It consists of a dark head cap extending from the snout to the posterior parts of the parietals where it usually is separated from the first cross band by a light nuchal collar. In some specimens the snout is light. There may be a prefrontal bar, an inter- ocular bar, a parietal mark and an occipital spot or any combination of these components. In some specimens most or all of the head cap components may be lacking, and the head is uniformly dark brown. The nuchal collar sometimes extends medially onto the parietal area to connect with the parietal mark. In this instance the anterior border of the collar is chevron-shaped. Often the nuchal collar is straight or slightly indented anteromedially. The posterior border of the collar is nearly always straight. The first body blotch occasionally contacts the head cap or may be separated from it one-half to six dorsal scales. The collar is usually three scales long on the midline. Subspecies. — The numbers of cross bands, the cross band lengths, and the head pattern indicate that gene flow between samples B and G is limited. The geographic barrier of the Transverse Volcanic Range separating these two samples may be partly or wholly responsible for this apparent restriction of gene flow. Presumably these two populations have come into secondary con- tact after experiencing separate histories. In our opinion, these two samples do not represent a continuum that was split into two components by the uplift of the Transverse Volcanic Range. It seems more likely that population G evolved from ancestors that moved into the Tepalcatepec Basin from the south after the Transverse Volcanic Range reached its present height in Jalisco and Michoacan (see discussion below). The similarities between the Balsas- Tepalcatepec samples E, F and G and their differences from A, B, C and D support this view. The latter four samples are similar in several characteristics and probably are representative of a continuum, as indicated by specimens from intermediate localities. There appears to have been some genetic exchange between the two groups through samples F and D in some of the river valleys of southeastern Puebla, northeastern Oaxaca and eastern Veracruz. The similarities and differences between A B, C and D and between E, F and G have been thoroughly discussed. Consideration of these points, com- bined with the apparent geographic barriers between these two groups, at this time leads us to recognize them as subspecies. The specimens representative of samples A, B, C and D and their intermediates are referred to Trimorphodon tau tau. The specimens from samples E, F and G in the Balsas-Tepalcatepec Basin are referred to Trimorphodon tau latifascia (new combination). Distribution and ecology. — Trimorphodon tau is widely distributed along the coastal slopes and foothills of the Sierra Madre Occidental and Sierra Madre Oriental (Fig. 1 ) . Occasaional specimens have been taken on the coastal 1970 Geographic Variation of Mexican Lyre Snakes 33 plains. The species is found in suitable habitats to the north of the Transverse Volcanic Range and the high southern portions of the Mexican Plateau. Trimorphodon tau has also been recorded from scattered localities to the east and south of the Mexican Plateau, from the Valley of Oaxaca south and east of the Sierra Madre del Sur. The species is widely distributed in the foothills of the Balsas and Tepalcatepec Basins. Although the species is reported from localities ranging from 100 to 2600 meters in elevation, most of the range of the species lies between 1000 and 2100 meters. The lack of adequate material from localities between samples A and B, between samples D and E, and between samples E and G probably reflects the inaccessibility of these areas as well as inadequate sampling of suitable habitat. Only four specimens are known from the eastern versant between samples C and D. We suspect that local climatic factors (high annual precipitation and abundant cloud cover) probably restrict the distribution of Trimorphodon to drier, less accessible habitats in this area. Trimorphodon tau is a nocturnal species that occupies semi-arid to season- ally dry habitats. Specimens have been collected in arid tropical scrub, thorn woodland, tropical deciduous forest, mesquite grassland and dry pine-oak woodland (vegetation classifications follow Leopold, 1950). The species is terrestrial and commonly found in rocky areas of irregular relief. Individuals have been collected from beneath rocks, in holes and under bark on fence posts and dead trees during the day and in a flooded rice field at night. Most speci- mens were found on roads at night. The species feeds primarily on lizards, al- though frogs and small mammals probably are eaten also. Unidentifiable species of Sceloporus and Cnemidophorus were found in specimens from Guerrero, Michoacan and Oaxaca. Eggs and young. — Because very little is known concerning the eggs and young of oviparous snakes, the following data for Trimorphodon tau are ap- propriate. On July 18, 1966 a large female Trimorphodon tau (MVZ 81354) was collected 26 miles NW of Huajapan de Leon on Mexico Highway 190 in Puebla (Sample E) by J. L. LaPointe. Sometime between the date of capture and July 25, this snake deposited seven eggs. The eggs were incubated follow- ing the technique outlined by Zweifel (1961: 112-113). On September 25, one of the eggs was opened and found to be developing. On October 10, the eggs averaged 26.1 mm in length, 16.8 mm in width and 4.0 gm in weight. Eive of the remaining six eggs hatched between October 29 and November 1, at least 96 days after laying. The hatchlings made from 2-7 (x = 5.4) longitudinal slits in the egg shell to escape. The sixth egg did not hatch and was found, on November 4, to contain a well developed, but dead, female (MVZ 81361). Pertinent information for the mother and five hatchlings are presented in Table 3. The hatchling snakes are typical of sample E in the characteristics studied. 34 Contributions in Science No. 179 Table 3. Data for Adult Female and Five Offspring of Trimorphodon tau from Puebla, Mexico Specimen No. (MVZ) : Sex Weight - gms. Total length - mm. Snout-vent length - mm. tail length total length Ventrals Subcaudals Ventrals & Subcaudals Umbilical Scar * Body Bands Supralabials Infralabials First ventral scale with scar 81354 81359 81360 Ad. $ $ $ — 2.73 2.95 752 199 207 633 168 176 15.8 15.6 14.3 217 218 216 69 67 68 217 218 216 — 186 182 28 24 ? 8-9 8-8 9-9 13-13 13-12 13-13 81356 81357 81358 $ S 2.39 2.86 2.86 199 213 219 165 174 179 17.1 18.3 18.3 195 205 203 80 84 78 195 205 203 170 174 177 22 22 24 8-8 9-8 8-8 11-11 13-12 12-12 Five hatchlings have 275-289 (x = 282.8) total ventral-subcaudal scales. These counts are very similar to counts from other Puebla specimens from the same general area (275-290, x = 284.75, N = 8). The same similarities exist in comparing the total body band counts. Four hatchlings have 22-24 (x = 23.0) dorsal cross bands while other Puebla specimens have 19-26 (x = 22.3, N = 9) cross bands. The hatchlings also agree with other Puebla specimens of Tri- morphodon in their supralabial and infralabial counts (Table 2). Evolutionary History Some brief comments concerning speciation in the genus Trimorphodon seem appropriate. Because no fossil remains are known, our interpretation of the evolutionary history of the species is based primarily on an analysis of the distributions and relationships of living forms correlated with the geohistory and paleogeography of the area. Savage (1966) considered snakes of the genus Trimorphodon to be rep- resentative of the Middle American Element. Duellman (1958, 1966) sug- gested that Trimorphodon diverged early from the evolutionary stock that gave rise to Hypsiglena, Leptodeira, Eridiphas and Cryophis. We concur, and sug- gest that the ancestor of Trimorphodon diverged early in the Cenozoic and probably occupied areas of relatively low relief in northern and central Mexico. Axelrod (1958) has indicated that most of this region was dominated by a Neotropical Tertiary Geoflora. Based on indirect evidence several authors have suggested that a broad ecotone of Arcto-Tertiary Geoflora extended into the mountains of Mexico at this time (Axelrod, 1960; Brame and Wake 1963; Savage, 1966). 1970 Geographic Variation of Mexican Lyre Snakes 35 As a result of the general cooling trend in the early Tertiary, the northern borders of the tropics began to move towards the equator. In areas of north- western Mexico and the southwestern United States, elements of a Madro- Tertiary Geoflora, that developed in situ, began to spread. The Miocene was a period of major mountain building and general uplift (Maldonado-Koerdell, 1964). The Mexican Plateau, the Sierra Madre Occidental, the Sierra Madre Oriental, and the Sierra Madre del Sur were uplifted to their present height beginning in the Miocene and continuing to the present. This period of orogeny initiated major volcanic action, especially during the Pliocene and Recent, that formed the Transverse Volcanic Range. By the end of Miocene, in response to the gradual Tertiary temperature depression and increasing aridity, much of the tropical vegetation, except the more xeric type, was eliminated from cen- tral and western Mexico. In northern Mexico the derivatives of the Madro- Tertiary Geoflora were segregating into their component parts in the late Cenozoic. The mixed coniferous and deciduous derivatives of the Arco- Tertiary Geoflora were found throughout most of the mountainous areas of Mexico. We suggest that the ancestral Trimorphodon stock gave rise to three distinct lineages, each of which was markedly affected by the continuing trends of cooling and aridity and by the late Tertiary period of mountain building and volcanic activity. The first lineage probably occupied much of central Mexico and gave rise to Trimorphodon tail. As the climatic changes continued, this species adapted to dry habitats throughout much of the present Mexican Plateau per- haps as far south as Oaxaca. Tamayo and West (1964) suggested that much of this area, originally drained by the Lerma River included the lake basins drained by the present Rio Lerma as far southeast as the Valley of Mexico. With the late Tertiary volcanic activity that gave rise to the Mesa Central (West, 1964) and the Transverse Volcanic Range and the resultant cooler temperatures, this ancestral species was eliminated from the central part of its range and restricted to a horseshoe pattern of distribution north, east and south of the Mesa Central. One segment of this population gave rise to the subspecies latifascia in the eastern Balsas Basin. Subsequently, when conditions were suitable, the Balsas population apparently expanded into the Tepalcatepec drainage system and approached the range of the northern segment of the original population. These two ends of the horseshoe are now separated by the western portions of the Transverse Volcanic Range. At the same time that the Balsas-Tepalcatepec populations were expanding, the population ancestral to the subspecies tau was expanding northward along the slopes and foothills of the Sierra Madre Occidental and Sierra Madre Oriental. The second major lineage of Trimorphodon evolved in association with the developing Madro-Tertiary Geofloral derivatives and gave rise to the species found in the arid lowlands of the southwestern United States and north- 36 Contributions in Science No. 179 western Mexico today. We suspect that Trimorphodon vandenburghi, T. lambda and T. lyrophanes, if indeed they are distinct species, were derived from this second major lineage. Trimorphodon vilkinsonii, today found in eastern New Mexico, western Texas and adjacent parts of Mexico, likely was derived from the ancestral lyrophanes stock and isolated east of the Rocky Mountain- Sierra Madre Occidental mountain system, where it is found today in moun- tains and foothills on the margin of the Chihuahuan Desert. The third lineage gave rise to the species T. biscutatus and probably evolved in a xeric tropical lowland environment in the Pacific coastal lowlands of southern Mexico and northern Central America. Savage (1966) referred to this species in his discussion of the history of the Western Mesoamerican Complex. Trimorphodon biscutatus subsequently dispersed northward along the developing Pacific lowland route as far north as southern Jalisco. Areas of sympatry or near sympatry between Trimorphodon biscutatus and T. lambda occur in the southwestern corner of Jalisco. Trimorphodon tau is sympatric with Trimorphodon biscutatus at Apatzingan, Michoacan and Huajintlan, Morelos and with Trimorphodon lambda at Terreros, Sinaloa and about 10 miles west of Alamos, Sonora. Although T. tau has been reported from Chihuahua, Chihuahua (Cope 1900) near the type locality of Trimor- phodon vilkinsonii, this record needs to be verified. Acknowledgments We would like to thank the following people and institutions for allowing us to examine material in their care: Charles M. Bogert and Richard G. Zwei- fel, American Museum of Natural History, New York (AMNH) ; Edmond V. Malnate and James E. Bohlke, Academy of Natural Sciences, Philadelphia (ANSP); M. J. Fouquette, Arizona State University, Tempe (ASU); Alice G. C. Grandison, British Museum (Natural History), London (BM); Richard B. Loomis, California State College, Long Beach (CSCLB); Robert F. Inger and Hymen Marx, Field Museum of Natural History, Chicago (FMNH); William E. Duellman, Museum of Natural History, University of Kansas, Lawrence (KU); John W. Wright, Los Angeles County Museum of Natural History, Los Angeles (LACM); Douglas A. Rossman, Louisiana State Uni- versity, Museum of Zoology (LSUMZ); Ernest E. Williams, Museum of Comparative Zoology, Harvard University, Cambridge (MCZ); Walter W. Dalquest, Midwestern University, Wichita Falls, Texas (MU); Robert C. Stebbins, Museum of Vertebrate Zoology, University of California, Berkeley (MVZ); George S. Myers, Museum of Natural History, Stanford University, Palo Alto (SU — collections now housed at California Academy of Sciences, San Francisco); James R. Dixon, Texas Cooperative Wildlife Collection, Texas A & M University, College Station (TCWC); Charles H. Lowe, Uni- versity of Arizona, Tucson (UAZ) ; Walter Auffenberg, Florida State Museum, University of Florida, Gainesville (UF); Hobart M. Smith, University of 1970 Geographic Variation of Mexican Lyre Snakes 37 Illinois Museum of Natural History, Urbana (UIMNH); Charles F. Walker, University of Michigan Museum of Zoology, Ann Arbor (UMMZ) ; James A. Peters, United States National Museum, Washington, D.C. (USNM); and Gunther Peters, Zoologisches Museum Humboldt-Universitat, Berlin (ZMB). Robert L. Bezy and C. Jay Cole, and Richard D. Worthington also provided specimens for study. The information on eggs and hatchlings was made avail- able by Joseph L. LaPointe, Ray B. Huey and Ted Papenfuss. Evie Templeton illustrated the head patterns in Figures 5, 6, and 7. Mercedes McDiarmid and Richard D. Worthington provided technical assist- ance with the statistical analysis. Allene J. Brown typed the manuscript. David B. Wake and Hymen Marx read the manuscript in final form and made several helpful suggestions. Fred F. Gehlbach kindly relinquished his interest in this group of Trimorphodon. Carl L. Hubbs, W. I. Follett and Lillian Dempster offered their advice concerning the proper use of the name Trimorphodon tau. The Society of Sigma Xi provided financial assistance to the first author for field work in Sinaloa in 1962. To all of these people and institutions we extend our thanks. Specimens Examined MEXICO Aquascalintes: 7 mi S Rincon de Romos (UIMNH 27566). Colima: Colima (AMNH 12777); 4.3 mi SW Colima, 575' (UMMZ 114479). Durango: Ventanas (BM 83.4.16.68 and 69). Guanajuato: Guanajuato (SU 4412, USNM 11370), Mt. Cubilete (AMNH 93433). Guerrero: 15 mi NE Acapulco (LACM 7114); Acahuizotla (TCWC 22142); 3 mi WSW Axixintla (UMMZ 126541); Chilpancingo (FMNH 38412-32; KU 24100-03; MCZ 33657-58, 33663-5; UIMNH 34991-93); near Chilpancingo (UMMZ 85769- 74); vicinity of Chilpancingo (AMNH 72522-24); 4 mi W Chilpancingo, 5800' (TCWC 9575); 13 m S Chilpancingo, 3750' (KU 67733); 7 mi E Chilpancingo (UIMNH 19139); 40 mi N Chilpancingo (LACM 53026); 25 mi NNE Iguala, 3800' (TCWC 12608); Ojito de Agua, 2.5 mi S Almolonga, 5600' (TCWC 11602- 03); Ojito de Agua, 3 mi S Almolonga, 5400' (TCWC 11601); Omilteme, Sierra de Burro Mtns. (MCZ 42677); Palo Blanco, 4800' (TCWC 9574); 15 mi SE Tonatico (UMMZ 126542); Taxco (MCZ 33901). Hidalgo: 10 km N Jacala (USNM 1 10401 ); Zacualtipan (ANSP 14770); 5.6 mi N of road to Zimapan on Hwy 85 (LACM 53023 ) ; 5 mi S Zimapan (AMNH 93432). Jalisco: 2 mi ENE Acatlan, 5300' (KU 67735) ; 4 mi S Autlan (KU 27191); 2 mi E Bolanos, 3500' (KU 91428); 1 mi NE Contla, 3800' (KU 80762-63); La Cumbre de los Arrastrados, 8500' fTalpa Mascota] (BM 92.10.31.63); Guadalajara (MCZ 46891; USNM 31358-type of upsilon); between Hostotipaquillo and Magdalena (AMNH 19582) ; 10.4 km NNW Ixtlahuacan del Rio (KU 102986) ; Jamay (AMNH 38 Contributions in Science No. 179 19841); near Magdalena, 1300 m (FMNH 105194); 1 mi NW El Molino, 5100' (AMNH 96652). Mexico: Ixtapan de la Sal, 5250' (AMNH 71360). Miahoacan: Apatzingan, 1000' (FMNH 39078-79); 7.6 mi E Apatzingan, 1600' (UMMZ 112514); 14 mi E Apatzingan, 1700' (UMMZ 112516); 17.6 mi E Apat- zingan, 1600' (UMMZ 112515); 3.9 mi S Apatzingan (UlMNH 73730); Coal- coman, 945 m (UMMZ 104696-97); 6 mi E Emiliano Zapata, 5350' (UMMZ 118950); 7 mi E Emiliano Zapata, 5400' (UMMZ 118949); 4.3 mi N Lombardia, 2850' (UMMZ 118951); 8.7 mi S Lombardia (UMMZ 124039) ; Nueva Italia, 1250' (UMMZ 118952); 43 mi S Nueva Italia on Playa Azul Road, 1000' (UAZ 27060); Tacicuaro, 2000 m (FMNH 105193; UIMNH 19138); 5.5 m W Tangamandapio, 5800' (UMMZ 118948); 1.4 mi N Rio Tepalcatepec (UIMNH 73729); Tzararacua Falls, 6 km SE Uruapan (USNM 110400-type of fasciolata). Morelos: Alpuyeca, 3500' (TCWC 4127); 4 mi S Alpuyeca, 3600' (UMMZ 114478); Campo Agricola, Progreso (UIMNH 26073); camp near Cuernavaca (UIMNH 19135); Huajintlan (UIMNH 19137); km 128, near Puente de Ixtla (FMNH 105100); 12 mi S Puente de Ixtla (UIMNH 19134); Tepoztlan (UIMNH 19136); 6 mi W Yautepec (TCWC 7390). Nayarit: Barranquitas (AMNH 75585); 5 mi SE Ixtlan del Rio (TCWC 12609); N of San Bias (LACM 53025). Oaxaca: Juquila Mixes [^zXuquila Mixes] (AMNH 100649); 3 mi NE Mitla, 5800' (AMNH 97985); ca. 2 mi W Mitla, La Fortaleza (AMNH 89629); 17 mi SE Nochixtlan, 6900' (TCWC 12640); Oaxaca (FMNH 105305, 105350); 21 mi SE Oaxaca, 5350' (UMMZ 112522); 59 mi NW Ciudad Oaxaca, (UMMZ 121824); Quiotepec (USNM 30338-type of m^/); San Felipe del Agua, 6500' (AMNH 97988, 100932-33); 2 km W San Felipe del Agua, (UIMNH 60799); Cerro San Felipe (UIMNH 53121, 73654, 74462-64) ; San Lucas Camotlan (AMNH 89630); 3 mi N San Miguel del Valle, 6950' (AMNH 97986-87). Puebla: no specific locality (ZMB 6652-syntypes of latifascia); 11.4 mi NW Acatlan (LACM 38206); 16 mi N Acatlan (LACM 9509); 6 mi SE Acatlan (KU 31680); 10.1 mi NW Amatitlan (UF 11337); 10 mi NW Huajuapan de Leon on Hwy 190 (MVZ 81355); 26 mi NW Huajuapan de Leon on Hwy 190 (MVZ 81354); 3 km W Izucar de Matamoros (KU 39625); 17 mi SE Izucar de Matamoros (AMNH 89631); San Diego, ca. 5 mi S Tehuacan (USNM 110402-type of forbesi); 6.7 mi NNW Tehuitzingo (UF 11335); 6.3 mi SE Tehuitzingo (UF 11336). San Luis Potosi: 10.5 mi W Rioverde, 4000' (UMMZ 126193); 2.5 mi S Pendencia (LSUMZ 298). Sinaloa: 16 km NNE Choix, 1700' (KU 68754); 22.1 mi N Culiacan (LACM 53024); 4 mi NNE El Fuerte, 300' (FMNH 71531) 2-3 mi ESE Rio Fuerte Dam [8 mi NNE El Fuerte, 300'] (FMNH 71532-33) 8.3 mi N Guacamil [=Gua- muchil] (UF 12829); 1.6 mi S Guacamil [=Guamuchil] (UF 12830); Terreros (LACM 9510); near Terreros (LACM 7112); 2.5 mi NW Terreros (LACM 7108); 5 mi NW Terreros (LACM 7109); 7 mi NW Terreros (LACM 7111); 10 mi NW Terreros (LACM 7 110); 10.2 mi SE Terreros (LACM 7113). Sonora: La Aduana, 5 mi W Alamos (LACM 53028) ; Alamos (AMNH 75119); W 1970 Geographic Variation of Mexican Lyre Snakes 39 side Alamos (KU 24119); 3.5 mi W Alamos (LACM 9157); 9.7 mi W Alamos, 1300' (LACM 53029); 8 mi SSE Alamos on Rio Cuchujaqui, 900' (LACM 53030); Guirocoba (MVZ 50833-34); ca. 15 mi SE Navojoa (MVZ 76372); 18 mi SE Navojoa (MVZ 76373); 20 mi E Navojoa (LACM 53027). Tamaulipas: 10 km N Antiguo Morelos (MU 196); S Ciudad Victoria, on Hwy 85 between C. Victoria and Llera (UMMZ 1 1 1255) ; 13 mi S Ciudad Victoria (UIMNH 19284); 14 mi S Ciudad Victoria (UIMNH 27164); 23 mi S Ciudad Victoria (AMNH 72399) ; 24 mi S Ciudad Victoria (AMNH 72400) ; Gruta de Quintero, near Quintero (AMNH 58224) ; La Joya de Salas, 1550 m (UMMZ 110882). Vera Cruz: Jalapa (BM 81.10.31.65); Tuxpango, near Orizaba (USNM 26499-type of collaris). Zacatecas: Hacienda San Juan Capistrano (USNM 46334). Locality Unknown: “Mexico” (FMNH 42066; USNM 9911-12, 25361, 26138-9 [see Taylor, 1939: 366 for comments] ). “Southern Mexico” (BM 95.1.4.7 A and B) “Districto Federal” (AMNH 19718 [seeZweifel, 1959: 7 for comments]). Additional Specimens {examined but data not included) Colima: 4.3 mi SSW Colima, 1300' (UF 24783). Hidalgo: 10.4 mi NNW Ixmiquilpan, 7500' (UF 27378). Jalisco: 2 mi NE El Molino (UAZ 27023). Sonora: 6 mi W Alamos (ASU 6651, 6684); 9 mi W Alamos (ASU 6377); 11 mi W Alamos (ASU 6663, 6712); 14 mi W Alamos (ASU 6648). Literature Records (specimens not examined) Chihuahua: Batopilas (Cope, 1887: 68; Klauber, 1928: plate 22); city of Chihuahua (Cope, 1900: 1105). Jalisco: “Jalisco” (Mocquard, 1899: 157); Guadalajara [USNM 12419] (Taylor, 1939: 366; Smith, 1941: 163). Michoacan: San Salvador [BM 1914.1.28.136] (Peters, 1954: 32-34). Morelos: between Cuernavaca and Tepoztlan; near Huajintlan (Taylor, 1941: 479). Nayarit: Sierra del Nayarit (Mocquard, 1899: 157). Tamaulipas: 14 mi SW Jimenez (Brown and Brown, 1967: 325). Resumen La variacion geografica en lepidosis y tipo de coloracion han sido es- tudiadas en las siguientes especies nominales Trimorphodon collaris, fasciolata, forbesi, latifascia, tau y upsilon. Los ejemplares fueron agrupados en siete diferentes muestras geograficas que fueron luego comparadas entre si. La mayoria de las caracteristicas tienen una amplia distribucion y se confunden mas con otras; otras caracteristicas varian independientemente. Las escamas ventrales y subcaudales disminiiyen en numero de norte a sur. El numero de bandas dorsales en el cuerpo es menor en culebras del Valle de Balsas- 40 Contributions in Science No. 179 Tepalcatepec, las cuales tienen tambien coloracion uniforme en la cabeza. Ejemplares de otras regiones tienen una compleja coloraeion cefaliea que consiste en bandas prefrontales, bandas interoculares y marcas parietales discretas en varias combinaciones. Estudios de los bandas en el cuerpo y coloracion cefaliea indican un intercambio genetico limitado entre los ejem- plares del Valle de Balsas-Tepalcatepec y aquellos de otras regiones, posi- blemente como resultado de contacto secundario en el sur de Jalisco y la region adyascente de Michoacan. La Cadena Volcanica Transversa y la Sierra Madre del Sur aparentemente restringen el intercambio genetico. Estas especies nominates se asignan al sinonimo Trimorphodon tau. El nombre Trimorphodon tau tau se conserva para las poblaciones de las colinas y la meseta y T. tau lati fascia (nueva combinacion) se asigua a la pobla- cion del Valle de Balsas-Tepalcatepec. Trimorphodon tau es redescribo y se delimita su distribucion. La variacion en caracteristicas de un grupo procedente de Puebla es semejante a la variacion tipica que se encuentra en la muestra total de Puebla. Literature Cited DO Amaral, A. 1929. Estudos sobre ophidios neotropicos. XVIII. Lista remissiva dos ophidios da regiao neotropicos. Inst. Butantan, Mem., 4: 129-271. Axelrod, D. I. 1958. Evolution of the Madro-Tertiary Geoflora. Botan. Review, 24(7): 433-509. . 1960. The evolution of flowering plants. In Tax, S (ed.). Evolution after Darwin. Chicago, Univ. Chicago Press, Vol. 1, The Evolution of life: 227-305. Boulenger, G. a. 1896. Catalogue of the Snakes in the British Museum (Natural History). London, vol. 3, 727 p., 25 plates. Brame, a. H., and D. B. Wake. 1963. The salamanders of South America. Los An- geles County Mus., Contrib. Sci., 69: 1-72. Brown, B. C. and L. M. Brown. 1967. Notable records of Tamaulipan snakes. Texas J. Sci., 29(3): 323-26. Cope, E. D. 1869. Seventh contribution to the herpetology of tropical America. Amer. Phil. Soc., Proc., 11: 147-170, figs. 9-11. . 1875. On the Batrachia and Reptilia of Costa Rica. Acad. Natur. Sci., Philadelphia, J., ser. 2, 8(4): 93-154. . 1886. Thirteenth contribution to the herpetology of tropical America. Amer. Phil. Soc., Proc., 23: 271-287. . 1887. Catalogue of the batrachians and reptiles of Central America and Mexico. U.S. Natl. Mus., Bull., 32: 1-98. . 1892. A critical review of the characters and variations of the snakes of North America. U.S. Natl. Mus., Proc., 14(882): 589-694. . 1900. The crocodilians, lizards, and snakes of North America. U.S. Natl. Mus., Rep., 1898: 153-1270, figs. 1-347, plates 1-36. Davis, W. B. and H. M. Smith. 1953. Snakes of the Mexican state of Morelos. Her- petologica, 8(4) : 133-149. Davis, W. B. and J. R. Dixon. 1957. Notes on Mexican snakes (Ophidia). South- western Natur., 2(1): 19-27. 1970 Geographic Variation of Mexican Lyre Snakes 41 . 1959. Snakes of the Chilpancingo Region, Mexico, Biol. Soc. Washington, Proc., 72: 79-92. Dowling, H. G. 1951. A proposed standard system of counting ventrals in snakes, British J. Herpetol., 1(5): 97-99. Duellman, W. E. 1958. A monographic study of the colubrid snake genus Lepto- deir a. AmQV. Mus. Natur. Hist., Bull., 114(1): 1-152. 31 plates. , 1958. A preliminary analysis of the herpetofauna of Colima, Mexico. Mus. Zool. Univ. Michigan, Occ. Papers, 589: 1-22. . 1961. The amphibians and reptiles of Michoacan, Mexico. Univ. Kansas Publ. Mus. Natur. Hist., 15(1): 1-148, . 1965. A biogeographic account of the herpetofauna of Michoacan, Mexico. Univ. Kansas Publ. Mus. Natur. Hist., 15(14): 627-709. . 1966. Remarks on the systematic status of certain mexican snakes of the gQnus Leptodeira. Herpetologica, 22(2): 97-106. Dumeril, a. M. C., G. Bibron and A. Dumeril. 1854, Erpetologie generale au his- toire naturelle complete des reptiles. Paris, 7, 2nd part: 782-1536. Eberhardt, L. L. 1968. An approximation to a multiple-comparison test. Copeia, 1968(2): 314-319. Fouquette, M. J., Jr. and D, A. Rossman. 1963. Noteworthy records of Mexican amphibians and reptiles in the Florida State Museum and the Texas Natural History Collection. Herpetolgica, 19(3): 185-201. Fox, W. 1948. Effect of temperature on development of scutellation in the garter snake, Thamnophis elegans atratus. Copeia, 1948(4): 252-262. Gadow, H. 1905. The distribution of Mexican amphibians and reptiles. Zool. Soc. London, Proc., 1905(2): 191-244. -. 1910, The effect of altitude upon the distribution of Mexican amphibians and reptiles. Zool. Jahrbuch, 29(6) : 689-714. Garman, S. 1883. The reptiles and batrachians of North America. Mus. Comp, Zool. Harvard Univ., Mem., 8(3) : 1-185, plates 1-9. GiiNTHER, A. C. L. G. 1895. Biologia Centrali-Americana. Reptilia and Batrachia. 1885-1902. London, i-xx, 326 p,, 76 plates. Hall, C. W. 1951. Notes on a small herpetological collection from Guerrero. Univ. Kansas Sci. Bull., 34 (pt. 1), (4): 201-212. Hardy, L. M. and R. W. McDiarmid. 1969. The amphibians and reptiles of Sinaloa, Mexico. Univ. Kansas Publ. Mus. Natur. Hist., 18(3) : 39-252. International Commission on Zoological Nomenclature. 1964. International code of zoological nomenclature. London, 176 p. Jan, G. 1863. Elenco sistematico degli ofidi descritti e disegnati per ITconografia generale. Milano, 143 p. Jan, G. and F. Sordelli. 1872. Iconographia generale des ophidiens. Milan, vol. 3, livr. 39, plate 1, fig. 3. Jones, C. and J. S. Findley. 1963. Second record of the lyre snake Trimorplwdon vilkinsonii in New Mexico. Southwestern Natur., 8(3) : 175-177. Klauber, L. M. 1928. The Trimorplwdon (lyre snake) of California, with notes on the species of the adjacent areas. San Diego Soc. Nat. Hist., Trans., 5(11): 183-194, 2 plates. . 1941 . The correlation between scalation and life zones in San Diego County snakes. Zool. Soc. San Diego, Bull., 17(3) : 73-79. 42 Contributions in Science No. 179 . 1956. Rattlesnakes. Their habits, life histories and influence on mankind. 2 vols., Univ. California Press, Berkeley and Los Angeles, 1476 p. Leopold, A. S. 1950. Vegetation zones of Mexico. Ecology, 31(4) : 507-518. Lynch, J. D. and H. M. Smith. 1965. New or unusual amphibians and reptiles from Oaxaca, Mexico. I. Herpetologica, 21(3); 168-177. Maldonado-Koerdell, M. 1964. Geohistory and paleogeography of Middle Amer- ica. In Wauchope, R. (ed.). Handbook of Middle American Indians. Univ. Texas Press, Austin, Vol. I, Natural environment and early cultures, Chapt. 1 : 3-32. Martin, P. S. 1958. A biogeography of reptiles and amphibians in the Gomez Farias Region, Tamaulipas, Mexico. Mus. Zool. Univ. Michigan, Misc. Publ., 101: 1-102, 7 plates. Mocquard, M. F. 1899. Reptiles and batraciens recueillis au Mexique par M. Leon Diquet en 1896 et 1897. Soc. Philom. Paris, Bull., 9 ser., 1 (4) : 154-169, 1 plate. . 1908. In Diimeril, A. and M. Bocourt. Etudes sur les Reptiles. Recherches Zoologiques pour servir a I’historie de la faune de FAmerique Centrale et du Mexique. Mission Scientifique an Mexique et dans FAmerique Centrale. Paris, vol. 3, i-xiv, 1012 p., 77 plates. Peters, J. A. 1954. The amphibians and reptiles of the coast and coastal Sierra of Michoacan, Mexico. Mus. Zool. Univ. Michigan, Occ. Papers, 554: 1-37. Peters, W. C. H. 1869. Uber mexicanische Amphibien, welche Hr. Berkenbusch in Puebla auf Veranlassung des Hrn. Legationsraths von Schlozer dem zoolo- gischen Museum zugesandt hat. Akad. Wiss. Berlin, Monatsber., 1869: 874-881. Savage, J. M. 1966. The origins and history of the Central American herpetofauna. Copeia, 1966(4): 719-766. Schmidt, K. P. and F. A. Shannon. 1947. Notes on amphibians and reptiles of Michoacan, Mexico. Chicago Natur. Hist. Mus., Fieldiana Zool., 31(9) : 63-85. Smith, H. M. 1941. Notes on the snake genus Trimorphodon. U.S. Natl. Mus., Proc., 91(3130): 149-168. . 1943. Summary of the collections of snakes and crocodilians made in Mexico under the Walter Rathbone Bacon Traveling Scholarship. U.S. Natl. Mus., Proc., 93(3169): 393-504. Smith, H. M. and E. H. Taylor. 1945. An annotated checklist and key to the snakes of Mexico. U.S. Natl. Mus., Bull., 187: 1-239. Smith, P. W. and D. M. Darling. 1952. Results of a herpetological collection from eastern central Mexico. Herpetologica, 8(3): 81-86. Stebbins, R. C. 1949. Speciation in salamanders of the plethodontid genus Ensatina. Univ. California Publ. Zool., 48(6) : 377-526. SuMiCHRAST, F. 1882. Enumeracion de las especies de reptiles observados en la parte meridional de la Republica Mexicana. La Naturaleza, 6(2) : 31-45. Tamayo, J. L. and R. C. West. 1964. The hydrography of Middle America. In Wau- chope, R. (ed.) Handbook of Middle American Indians. Univ. Texas Press, Austin. Vol. 1, Natural environment and early cultures, Chapt. 3: 84-121. Taylor, E. H. 1939. On Mexican snakes of the genera Trimorphodon and Hyp- siglena. Univ. Kansas Sci. Bull., 25(16): 357-383, plates 35-38 (1938). . 1940. Some Mexican serpents. Univ. Kansas Sci. Bull. 26(14): 445-487, plates 49-52 (1939). . 1949. A preliminary account of the herpetology of the state of San Luis Potosi, Mexico. Univ. Kansas Sci. Bull., 33, pt. 1, (2) : 169-215. 1970 Geographic Variation of Mexican Lyre Snakes 43 Vivo Escoto, J. 1964. Weather and climate of Mexico and Central America. In Wauchope, R. (ed.) Handbook of Middle American Indians. Univ. Texas Press, Austin, Vol. 1, Natural environment and early cultures, Chapt. 6: 187-215. Webb, R. G. and C. M. Fugler. 1957. Selected comments on amphibians and reptiles from the Mexican state of Puebla. Herpetologica, 13( 1 ): 33-36. West, R. C. 1964. Surface configuration and associated geology of Middle America. In Wauchope, R. (ed.) Handbook of Middle American Indians. Univ. Texas Press., Austin. Vol. 1, Natural environment and early cultures, Chapt. 2: 33-83. ZwEiFEL, R. G. and K. S. Norris. 1955. Contribution to the herpetology of Sonora, Mexico: description of new subspecies of snakes {Micniroides euryxanthiis and Larnpropeltis getiilus) and miscellaneous collecting notes. Amer. Midland Natur., 54(1): 230-249. ZwEiFEL, R. G. 1959. The provenance of reptiles and amphibians collected in west- ern Mexico by J. J. Major. Amer. Mus. Natur. Hist., Novitates 1949: 1-9. . 1959. Additions to the herpetofauna of Nayarit, Mexico. Amer. Mus. Natur. Hist., Novitates, 1953: 1-13. . 1961. Another method of incubating reptile eggs. Copeia, 1961(1): 112- 113. Accepted for publication June 30, 1969. ■■ik. k i i LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE UMBER 180 February 27, 1970 STUDIES ON NORTH AMERICAN BEES OF THE GENUS HYLAEUS. 5. THE SUBGENERA HYLAEUS, S. STR. AND PARAPROSOPIS (HYMENOPTERA: COLLETIDAE) By Roy R. Shelling Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor STUDIES ON NORTH AMERICAN BEES OF THE GENUS HYLAEUS. 5. THE SUBGENERA HYLAEUS, S. STR. AND PARAPROSOPIS (HYMENOPTERA: COLLETIDAE) By Roy R. Swelling^ Abstract: The Nearctic representatives of the two subgen- era Hylaeus, s. str. and Paraprosopis are considered in this paper. Within Hylaeus, s. str,, eleven nominate species are recognized, one of these with a morphologically differentiated subspecies. One species {H. granulatus) is removed from synonymy, one new species {H. sejunctus) is described from the southwestern United States, and one name {H. gaigei) is transferred to the subgenus Prosopis. The distribution and morphological variation of several species are treated in detail, and the hypothesis is advanced that one species {H. stevensi) may be adventive from the Palearctic fauna. A key is given for all Nearctic species and facial views of the heads of both sexes of most species, and the male apical ven- trites are illustrated. Fourteen Nearctic species of Paraprosopis are recognized and included in the key to the species. Two new species {H. luni- craterius, H. timberlakei) are described from Idaho and Cali- fornia, respectively. The species H. hydrangeae is transferred from the subgenus Metziella to Paraprosopis as a synonym of H. georgicus. The distribution of most species is discussed and facial characteristics and male apical ventrites of all species are figured. The present paper is a continuation of those comprising the series in Literature Cited (Snelling, 1966 a, b, c; 1968), and includes all the species assigned to the nominate subgenus and to Paraprosopis. One additional paper, treating the subgenus Prosopis in greater detail than my earlier paper (1966a) is in preparation. Hopefully, the series will conclude with a treatment of the species of Mexico and Central America. HYLAEUS S. STR. The nominate subgenus is, in the Nearctic Region, one of the larger groups in numbers of species. Except for the ubiquitous H. cressoni (Cock- erell), however its component species are less abundant than those of Prosopis and Paraprosopis. As is the case of Prosopis, Hylaeus s. str., is Holarctic and most of the species are Palaearctic. Our fauna is closely allied to that of the Old World, and one species, H. ellipticus (Kirby) shows definite affinities with a small complex of Old World species. Its nearest relative seems to be ^ Entomology Section, Los Angeles County Museum of Natural History, Los Angeles, California 90007. 1 CMITUOomi A 2 Contributions in Science No. 180 H. cardioscapus Cockerell, of Siberia. The females, especially, are difficult to separate. Our species are definitely of boreal origin and seem to have only com- paratively recently invaded the more arid regions of the Southwest. Thus, of the eleven species here recognized, only one is a true xerophile. One other species has evolved a xeric subspecies. Of the remainder all save one are essentially northern and/or montane in distribution. The one exception seems to be restricted to the relatively cool sand dune areas along the California coast. That the invasion of the New World is a phenomenon of long standing is attested by the extensive southern distribution of the group and by the develop- ment of several rather distinctive groups of species. The most primitive of our species seems to be H. ellipticus which extends from as far north as the Firth River in the Northwest Territories south to California, Arizona and New Mexico. From Alaska on the west H. ellipticus ranges eastward across Canada to the Atlantic Coast in Nova Scotia. From there it ranges south as far as North Carolina and Tennessee at the high elevations in the Appalachians. Another essentially boreal species is H. verticalis (Cresson). The north- ern limit of this species is unknown to me although I have a few specimens from scattered localities in southern Alaska. It occurs as far south as Cali- fornia, Arizona and New Mexico in the western states, and North Carolina and Tennessee in the eastern. At its southern extremities H. verticalis occurs only at high elevations. Two species comprise the H. conspicuus group : H. conspicuus (Metz) and H. maritimus Bridwell. Both are highly modified from the usual pattern of the H. cressoni group, but seem, nonetheless, to be related to that assemblage through H. rudbeckiae (Cockerell and Casad) . Hylaeus conspicuus is the more common and widely distributed, extending along the Pacific Coast states from British Columbia south to southern California. Eastward the range includes northern Nevada and Utah, Idaho, Wyoming and Montana. Although often locally abundant H. conspicuus is of sporadic occurrence and is seldom rep- resented in collections by large series. Far more restricted in distribution, H. maritimus is known only from coastal sand dunes of California from Sonoma County south to Santa Barbara County. The one completely isolated species in the Nearctic fauna is H. stevensi. This species seems to be unrelated to any species in the Nearctic fauna. This species is most abundant in the north-central and northwestern states, but does not extend very far into Canada. In many samples which I have seen from these areas it is obviously the dominant species of Hylaeus. Beyond this region its occurrence is spotty but it may be locally very abundant. For example, it is quite common along the San Joaquin River, ten miles west of Turlock, Stan- islaus County, California, where it frequents Melilotus spp. At Turlock, where I collected intensively for about ten years, it appears to be very rare. The remaining group is apparently entirely Nearctic in distribution. This 1970 Studies on North American Bees 3 is the H. cressoni group which includes, in addition to H. cressoni itself, H. fedorica (Cockerell), H. labiatifrons (Cockerell), H. rudbeckiae (Cock- erell and Casad), H. saniculae (Robertson) and H. sejunctus, n. sp. The most common, and most widely distributed, member of this group is H. cressoni which extends transcontinentally from southern Canada to north- ern Mexico. Its elevational amplitude exceeds that of all other species in the subgenus, extending from sea level in California to altitudes in excess of 9500 feet in the same state and at the same latitude. Perhaps in response to this extensive distribution, H. cressoni has produced a welter of varietal forms, a number of which have been formally named. I recognize only the form which inhabits the southwestern deserts as a valid subspecies. Several additional species are very closely related to H. cressoni and their females are extremely difficult to separate from those of H. cressoni. Of these species, only H. rud- beckiae is at all common and widely distributed, and its range is nearly as extensive as that of H. cressoni, the species being absent only from very arid regions. The only other member of the group other than H. cressoni mesillae to occur in desert areas is the new species H. sejunctus; this species, obviously derived from H. cressoni, occurs sporadically in desert regions in southern California and Nevada, western Arizona and extreme northwestern Mexico. Three species, H. fedorica, H. labiatifrons and H. saniculae are found in the midwestern and eastern states. All are rare and very poorly known. It is not very difficult to visualize the evolution of the H. cressoni group from a more primitive bee such as H. ellipticus. Such requires only a reduction in size and in punctation, together with some modification of the male sec- ondary characters (scape configuration, extent and orientation of facial maculae). By a similar reasoning, the H. conspicuus group can be derived from the same source, through an intermediary form such as H. rudbeckiae. The larger size, larger and more clearly defined sculpturation and Holarctic distribution of the H. ellipticus group favor the conjecture that this is the more primitive of the two groups. The H. verticalis group appears to be still less specialized, and our one representative is structurally a more variable species. From the above sketchy commentary on our species of Hylaeus, s. str., it should be obvious that the species of this subgenus will prove difficult to identify. This particularly is true of those which I have assigned to the H. cressoni group. In the key which follows one should not experience any diffi- culty with the females until those couplets dealing with the H. cressoni group are reached (Nos. 7-12) . I have been forced to rely on characters which are, at best, exceedingly difficult to work with, and the results are far from satis- factory. During the preparation of this portion of the key, an AO-Spencer binocular microscope, with 18x ocular eyepieces and 6x objectives, was used. A five unit measuring reticle, divided into twenty subunits was used, this cali- brated against a stage micrometer for conversion into millimeters. Facial measurements were made with the head in full-face view, i.e., the head so 4 Contributions in Science No. 180 positioned as to produce maximum distance between the clypeal apex and the posterior head margin. Head length is a useful character although even slight overall size increase usually results in a disproportionate increase in the devel- opment of the vertex; this does not become apparent until one has examined many specimens. Far less variable, and hence more useful for some purposes, is the maximum measurable distance between the clypeal apex and the anterior margin of the anterior ocellus, along the mid-line of the face. This I have called the facial length. With the head positioned as indicated above, a measurement of the maximum width, across the compound eyes, is made to yield the head width. This measurement is subject to some allometric variation and hence is less useful than desired, but its alternative, facial width (maximum measurable distance between the eyes at the level of the antennal sockets) is less easily made with precision and is only slightly less variable. Characters of punctation have also been used to separate females within the H. cressoni group. These are subtle and difficult to express, and until one has become somewhat familiar with the various species, they are very difficult to appreciate. Unfortunately, there is a considerable amount of variation, especially in H. cressoni, and many (about 15% of those I have studied) fe- males cannot be keyed satisfactorily. The alternative was to bring all females of this group to a single couplet without any attempt to separate them. Others may wish to adopt this alternative and stop keying at couplet 7. Tentative determinations may then be made on the basis of associated males, although this is hazardous since there are few areas where but a single species is found, although in any area one species will be dominant, usually H. cressoni. The portion of the key separating the males should present few difficul- ties. Two forms (H. cressoni mesillae and H. sejunctus) are separated by differences in the shape of the ninth ventrite since no reliable external features have been discovered. All specimens from the southern California deserts, southwestern Nevada, the western third of Arizona and adjacent portions of northwestern Mexico must be separated in this manner. Specimens from the remaining areas of the southwestern deserts will all be H. cressoni mesillae or intergrades to the typical form (these are particularly common in Utah and Texas) . Males of the other species may also be identified simply by comparison with the figures of the faces; these are quite distinctive for each species. KEY TO HYLAEUS, S. STR. 1. Females 2 Males 13 2. Anterior coxae not distinctly angulate laterally; facial foveae ending no more than one-fourth of distance between eyes and lateral ocelli; pro- notal collar distinctly shorter medially than laterally; pronotal collar usually maculate 3 Anterior coxae distinctly angulate laterally; facial foveae ending about 1970 Studies on North American Bees 5 one-third of distance between eyes and lateral ocelli; pronotal collar usually as long medially as laterally; pronotal collar immaculate verticalis (Cresson) 3. Oblique propodeal carina evanescent or absent; lateral carina evanes- cent or absent above; mesopleura very densely tesselate, punctures obscure; if mesopleura densely tesselate above only, then genae broader than eyes 4 Oblique and lateral propodeal carinae usually well-defined; mesopleu- rae frequently punctate and shiny above, if densely tesselate and im- punctate, than genae narrower than eyes 5 4. Entire mesopleuron densely tesselate, dull, punctures very obscure; genae, seen from side, distinctly narrower than eyes; median length of basal zone of propodeum much shorter than median length of post- scutellum; clypeus usually extensively maculate conspicuus (Metz) Mesopleuron moderately shiny and punctate on lower half; genae, seen from side, distinctly broader than eyes; median length of basal zone of propodeum subequal to median length of postscutellum; clypeus immaculate maritimus (Bridwell) 5. Mesopleural punctures extremely coarse, several times diameter of those of vertex, distinctly larger than those of mesocutum, integument dull, densely tesselate; postscutellum with deep, distinct coarse punc- tures, frequently appearing rugose stevensi (Crawford) Mesopleural punctures little, if any coarser than those of vertex or mesocutum; punctures of postscutellum usually obscure, never coarse.... 6 6. Lateral and oblique carinae very strongly developed; transverse carina present, high; entire basal area coarsely rugose; mesopleura deeply, uniformly punctate; postscutellum dull, densely tesselate and rough- ened; larger species, length to apex of second tergite 5. 5-6.7 mm; cly- peus frequently with transverse, apical macula ellipticus (Kirby) Lateral and oblique carinae variable, but usually rather faint; transverse carina usually absent; basal area rarely entirely rugose; if rugose, then mesopleura tesselate on upper half, punctures sparse, obscure; post- scutellum usually slightly shiny, often with evident punctures; smaller, slender species, length to apex of second tergite 2. 7-4. 2 mm 7 7. Head width 1 .40 mm or more, usually about 1.5 mm; facial length 1.2 mm or more rudbeckiae (Cockerell and Casad) Head width not exceeding 1.31 mm; facial length not exceeding 1 .2 mm 8 8. Punctures of mesocutum fine, separated by 1.5 or more times a punc- ture diameter; interspaces very densely tesselated so that integument is dull 9 Punctures of mesocutum coarser, separated by about a puncture diam- eter or less; interspaces usually somewhat shiny; if densely tesselate, punctures quite close 10 6 Contributions in Science No. 180 9. Punctures of mesocutellum medially distinctly closer than those on either side, interspaces somewhat polished sejunctus Snelling Punctures of mesocutellum medially little, if any, closer than those on either side, interspaces slightly shining, distinctly tesselated - - cressoni mesillae (Cockerell) 10. Clypeus and lower half of supraclypeal area tesselate, but slightly shining with extremely fine longitudinal striolae ........11 Clypeus and lower half of supraclypeal area very densely tesselate, appearing almost granular, without fine longitudinal striolae fedorica (Cockerell) 11. Pronotal collar immaculate 12 Pronotal collar maculate; tegulae maculate; clypeus either maculate apically or strongly suffused with light reddish - cressoni cressoni (Cockerell) 12. Facial maculae well-developed; tubercules and basal two-thirds of front tibiae maculate cressoni cressoni (Cockerell) Facial maculae greatly reduced or absent; tubercules immaculate; front tibiae with small basal spot saniculae (Robertson) 13. Sides of face without conspicuous oval concavity; clypeus rarely marked with black 14 Sides of face, slightly above level of antennal sockets, with a conspicu- ous oval concavity; lateral face marks narrow, elongate; clypeus usually partially infuscated along lower margins saniculae (Robertson) 14. Antennal scape broad, its greatest width equal to one-half or more its length ....15 Antennal scape narrower, its greatest width not exceeding 0.47 times its greatest length 19 15. Scape more or less dilated, strongly flattened to concave beneath; lateral face marks variable... 16 Scape robust, thick, hardly, if at all, flattened beneath; lateral face marks curving away from eye margins over antennal sockets (Fig. 2H) verticalis (Cockerell) 16. Lateral face marks above well-removed from inner orbits and con- spicuously dilated or clavate above (Figs. 2C, D, J) ; transverse propo- deal Carina evanescent or absent ......17 Lateral face marks above not separated from eye margin, but with pointed median extension ending over antennal sockets (Fig. 2C) ; transverse propodeal carina sharply defined ....ellipticus (Kirby) 17. Scape no more than 0.58 times as broad as long; concave surface black, with at most a narrow yellow stripe along lower margin; mesopleura distinctly punctate 18 Scape at least 0.70 times as broad as long, concave under surface wholly yellow, with broad yellow stripe along lower margin which 1970 Studies on North American Bees 7 expands above onto convex surface, (Fig. 2J); mesopleura densely tesselate and roughened, without conspicuous punctures conspicuus (Metz) 18. First flagellar segment subequal to second; rugulae of basal area of propodeum not attaining posterior margin; the rugulae longitudinal, widely spaced; oblique carina absent; lateral face marks slightly con- stricted above leval of clypeus, separated from inner orbits by less than their width at constriction (Fig. 2D) maritimus Bridwell First flagellar segment half, or less, as long as second; basal area of propodeum rugoso-reticulate to posterior margin; oblique carina present, face marks strongly constricted, separated from inner orbits by much more than their minimum width at constriction (Fig. 2F) rudbeckiae (Cockerell and Casad) 19. Mesopleural punctures little, if any, coarser than those of frons; trans- verse and oblique carinae usually absent; lateral face marks constricted above (Figs. 2 A, G, I) ; smaller species, length to apex of second tergite, 2.7-3.5 mm 20 Mesopleural punctures much coarser than those of frons; transverse and oblique carinae present, distinct; lateral face marks not constricted above (Fig. 2B) ; larger species, length to apex of second tergite, 3. 0-4. 2 mm stevensi (Crawford) 20. Lateral face marks expanded above constriction, or continued upward and curving inward above antennal sockets (Figs. 2G, I) 21 Lateral face marks not expanded above constriction, rarely extending beyond level of upper margin of antennal sockets, never curving inward above sockets (Fig. 2A) fedorica (Cockerell) 21. Punctures of mesoscutum and mesoscutellum coarse, separated by one- half a puncture diameter or less, the interspaces densely tesselate, and dull; rugulae of basal area of propodeum usually extending to summit of declivity 22 Punctures of mesoscutum and mesoscutellum fine, mostly separated by one and one-half times a puncture diameter or more, interspaces lightly tesselated and moderately shiny; rugulae of basal area evanescent be- yond middle, rarely attaining summit of declivity 23 22. Tibiae and tarsi entirely yellow; tegulae maculate; scape largely yellow beneath labiatifrons ( Cockerell ) Middle and hind tibiae yellow on basal-half only; tegulae often black; scape often entirely dark beneath cressoni (Cockerell) 23. Apical lobes of ninth ventrite shorter than pedicle cressoni mesillae (Cockerell) Apical lobes of ninth ventrite distinctly longer than pedicle sejunctus Snelling 8 Contributions in Science No. 180 Hylaeus (Hylaeus) verticalis (Cresson) , NEW STATUS Figs. IG, 2H, 3B, 5B Prosopis verticalis Cresson, 1869. Proc. Boston Soc. Nat. Hist. 12:271. $ . Prosapis tridentulus Cockerell, 1896. Psyche 7 (sup.): 27. $ NEW SYNONYMY. Prosapis tridens Cockerell, 1896. Op. cit.\ 27. $ . Prosopis subtristis Swenk and Cockerell, 1910. Ent. News 21 :70 9. NEW SYNONYMY. H Figure 1. Hylaeus (//.) spp., female faces: A, H. fedorica; B, H. stevensi; C, H. ellipticLis', D, H. maritimus; E, H. riidbeckiae; F, H. cressoni cressoni', G, H. verticalis; H, H. sejunctus; I, H. conspiciius. Figures by Ruth A. DeNicola. 1970 Studies on North American Bees 9 Prosopis melitina Lovell, 1911. Ent. News 22:214. $. Prosopis tridentulus, Metz, 1911. Trans. Amer., Ent. Soc. 37:116-117. Prosopis episcopalis var. subtristis, Metz, 1911. Op. cit.\ 131. 9 {in part) . Hylaeus verticalis, Cockerell and Sumner, 1931. Amer. Mus. Nov. 490: 3-4. Hylaeus verticalis tridentulus, Linsley, In Mueseback, et ah, 1951. U.S.D.A., Agr. Monog. No. 2:1050. Hylaeus verticalis verticalis, Linsley, Op. cit.: 1052. Mitchell, 1960. N. C. Agr. Exp. Sta. Tech. Bui. 141 :76. With the exception of H. cressoni no species of Hylaeus is more widely distributed than H. verticalis. Throughout most of its range this species is quite common and highly variable. Most of the variation occurs in the females. This variation assumes several different forms. In western populations there is a strong tendency for the apical one-third of the clypeus to be suffused with reddish, rather than entirely black. This variant form is found together with females which have the clypeus entirely black, but there is a noticeable trend toward the maculate condition in those populations which occur at lower elevations in the Sierran and Cascade ranges, and it is a dominant feature in the populations of the Coast Ranges in California. The maculate clypeus is rare in northern and eastern populations. Very rarely are specimens found which have small yellow maculate on the pronotal collar. I have seen only three such specimens among the hundreds I have examined; all were females, one from Idaho, one from Colorado and one from California. Variation, too, occurs in the shape of the pronotal collar. In the usual condition the pronotal collar, when viewed from above is very nearly as long in the middle as at the sides. In profile it is convex and evenly rounded toward the anterior pronotal neck. Some females, however, have the middle portion flat and somewhat compressed so that it is much shorter here than at the sides. Such specimens, on the basis of the material I have seen, occur only in the western populations, and are especially common in samples from Oregon, California and Nevada. No specimens presenting an intermediate condition are known to me, and these individuals may actually represent a cryptic sibling of H. verticalis. I have not, however, been able to associate them with any male variant, so am content to assign them to H. verticalis for the present. I have examined the types of H. tridentulus, H. tridens and H. subtristis and can find no justification for their continued recognition. The first two, both based on males from Colorado, were named because of slight differences in the color of the maculae and slight differences in punctation from the typical form described from Pennsylvania. With large series of males from the entire range of the species available, it is evident that they fall well within the total range of variation exhibited by any one population. The females described by Swenk and Cockerell from Nebraska as H. subtristis are perfectly typical of the western populations of H. verticalis; they were merely the, until that time. 10 Contributions in Science No. 180 undescribed opposite sex of this species. The females which Metz (1911) had called H. episcopalis subtristis were a conglomerate association which in- cluded, in addition to H. verticalis, H. rugulosus rugulosus (Cockerell), H. r. episcopalis (Cockerell) and H. modestus citrinifrons (Cockerell). I have not seen the type of H. melitina, but as there is nothing in its description to indicate merit for its further consideration, I have accepted its synonymy as published by Mitchell (1960). Hylaeus {Hylaeus) cressoni cressoni (Cockerell) Figures IF; 2H; 6. Prosopis pygmaea Cresson, 1869. Proc. Boston Soc. Nat. Hist. 12:272. $ . Preoccupied by Schenck, 1853. Prosopis cressoni Cockerell, 1907. Annals and Magazine of Natural History (7) 20:131. New name for P. pygmaea Cresson, not P. pygmaea Schenk. Metz, 1911. Trans. Amer. Ent. Soc., 37:105. $ $. Prosopis pasadenae Cockerell, 1910. Annals and Magazine of Natural History (8) 5:30. $ $. Prosopis teleporus Lovell, 1911. Ent. News, 22:213. $ NEW SYN- ONYMY. Hylaeus mesillae race pasadenae, Meado-Waldo, 1923. In Wytsman, Gen. Insect., fasc. 181:30. Hylaeus laciniatus Cockerell and Sumner, 1931. Amer. Mus. Novitates, 490:9-10. $ . NEW SYNONYMY. Hylaeus repolitus Cockerell and Sumner, 1931. Op. cit., 13-14. $ . NEW SYNONYMY. Hylaeus {Hylaeus) cressoni cressoni, Linsley, 1951. IN Muesebeck, et al., Hymen. Amer. N. Mex., Synoptic Cat., Monogr. 2, U.S.D.A., p. 1050; Mitchell, 1960, N. C. Agr. Expt. Sta. Tech. Bull. 141:60, 62, 70. Hylaeus {Hylaeus) cressoni pasadenae, Linsley, 1951. IN Muesebeck, et al. Hymen. Amer. N. Mex., Synoptic Cat., Monogr. 2, U.S.D.A., p. 1050. Hylaeus {Hylaeus) teleporus, Linsley, 1951. IN Muesebeck, et al., Hymen. Amer. N. Mex., Synoptic Cat., Monogr. 2, U.S.D.A., p. 1052; Mitchell, 1960, N. C. Agr. Expt. Sta. Tech. Bull. 141:60, 62, 75; fig. 11. This species extends transcontinentally from southern Canada to north- ern Mexico. Over most of this range H. cressoni is a common species. The populations from eastern, northern and montane regions commonly have the clypeus and pronotal collars immaculate. Some individuals within these popu- lations may also lack maculae on the pronotal lobes and the tegulae. Clinal variation occurs uniformly from east to west, from north to south and with decreasing elevation. This clinal variation is manifested in two concordant characters, an increase in the tone and extent of the pale maculae and a de- crease in the size and density of metasomal punctation, so that populations at opposite poles of the dine are very different in appearance. Specimens from 1970 Studies on North American Bees 11 Turlock, California, for example have the face marks bright yellowish, the apical portion of the clypeus is strongly infused with reddish color and the pronotal lobes and collar and the tegulae are conspicuously maculate. The abdomen is smooth and shiny, with few or no distinct punctures. Such speci- mens contrast sharply with others from the New England area in which the maculae are dull yellowish-white, the clypeus is black and the thorax com- pletely immaculate; such specimens may have the metasomal tergites with obvious fine scattered punctures. The same is true if one compares the Turlock specimens with samples from Revelstoke, British Columbia, or with others from Sonora Pass in the Sierra Nevada of California. In all cases the two extremes are connected, through the intervening portions of the range, by a series of specimens of intermediate character. These series of intermediates form, in their entirety, a well-defined dine for the characters noted. Under these cricumstances continued recognition of H. pasadenae, de- scribed from Pasadena, California, as a subspecies of H. cressoni, a status accorded this form in the past, is untenable. This bee does not possess any of the characteristics of a definable subspecies and I agree with Metz in placing H. pasadenae in the synonymy of H. cressoni. The subsequent continued recognition of H. pasadenae as a subspecies of H. cressoni seems to be trace- able to Cockerell’s refusal to accept the synonymy published by Metz. How- ever persistent such a refusal has been, there appears to be no published indication why Metz’s proposal should not be adopted. Lovell’s H. teleporus was proposed for H. cressoni-\i\iQ. males in which the lateral face marks terminate acutely at the level of the lower margin of the antennal sockets. These males seem to occur sporadically within the popula- tions of H. cressoni in northern and montane areas. I have examined the terminalia of males of this form and find no distinctive characters which will serve to separate H. teleporus from H. cressoni. I believe that H. teleporus should be placed in the synonymy of H. cressoni until conclusive evidence for the specificity of this form can be advanced. Cockerell and Sumner described H. repolitus from a single female taken at Ogden, Utah. Dr. Rozen sent the type to me and I find nothing in any of its characters to justify separating this bee from H. cressoni. The nearly smooth basal zone of the propodeum is exactly like that of many females of H. cressoni available to me which also show a strong reduction in the longitudinal rugulae of the basal zone. This characteristic cannot be correlated with other features nor is it geographically consistent. Dr. Rozen has sent the type of H. laciniatus, and there is no doubt that this name, too, is a synonym of H. cressoni cressoni. Before I had the oppor- tunity to study the type, I was puzzled by the peculiar shape fo the seventh ven- trite as figured by Cockerell and Sumner (1930:10, fig. 2). With the type slide before me, it is clear that the apical lobes of this ventrite have been broken off. The poor preparation of the slide, on which everything is badly 12 Contributions in Science No. 180 flattened and distorted, is also responsible for the seemingly expanded gono- coxites, as these were illustrated by Cockerell and Sumner. Cockerell and Sumner cite the type as follows: “COLORADO— Boulder, May 24, 1913 (F. E. Lutz).” The labels on the specimens clearly read “Boulder, Colo., M. D. Ellis, May 24, 1913,” and “5u//x.” Since the specimen and slides are in accord with the original descrip- tion and figures, and the type is so marked in Cockerell’s handwriting, I have no reason to doubt that this is the true type. Hylaeus (Hylaeus) cressoni mesillae (Cockerell) Figures 3C; 5C; 6. Prosopis subtilis Cockerell, 1895. Trans. Amer. Ent. Soc., 22:295. 9. Preoccupied by Forster, 1871. Prosopis mesillae Cockerell, 1896. Canad. Ent., 28:42. new name for P. subtilis Cockerell, not H. subtilis Forster. Prosopis magniclavis Swenk and Cockerell, 1910. Ent. News, 21:68. 5 9 . NEW SYNONYMY. Prosopis cressoni, form mesillae, Metz, 1911. Trans. Amer. Ent. Soc., 37:107 (in part). Hylaeus cressoni, var. magniclavis, Meade-Waldo, 1923. IN Wytsman, Gen. Insect., fasc. 181:30. Hylaeus mesillae, Meade-Waldo, 1923. Op. cit. :30. Hylaeus {Hylaeus) magniclavis. Linsley, 1951. IN Muesebeck, et al., Hymen. Amer. N. Mex., Synoptic Cat., Monogr. 2, U.S.D.A., p. 1051. Hylaeus {Hylaeus) mesillae, Linsley, 1951. Op. cit. : 1051. The status of H. cressoni mesillae has been confused. Proposed originally as a full species, it was reduced by Metz to status as a “form” of H. cressoni. Crawford, two years later, elevated the name to specific status based on differ- ences in the shape of the eighth ventrite, and the bee has been recognized at this level since then. Unfortunately, while Crawford’s conclusions were sound, they were based on misidentified specimens. I have examined the genitalia and associated ventrites of many males from New Mexico, Arizona, California and Sonora, Mexico. With the exception of the California specimens all ex- hibited characters typical of H. cressoni. The California material, however, was represented by two very dis- tinctly different types of terminalia. The first of these was typically H. cressoni, identical to those specimens from New Mexico and Arizona. The second type agreed closely with the figure published by Crawford under the name H. mesillae. Crawford dissected the terminalia from at least three males, which I have seen; all were from California. The type locality of H. mesillae is Mesilla, New Mexico, and the terminalia of males from this area are con- sistently of the H. cressoni type, and I am certain that the interpretation 1970 Studies on North American Bees 13 offered here is the correct one: that H. mesillae is conspecific with H. cressoni, and that Crawford’s specimens represent another species, described below as new. I have reduced H. mesillae to subspecific rank under H. cressoni. The females of this subspecies are a little smaller, on the average, than those of I Figure 2. Hylaeus {H.) spp., male faces: A, H. fedorica; B, H. stevensi; C, H. ellipticus; D, H. maritimns; E, H. granidatus; F, H. riidbeckiae; G, H. cressoni cressoni', H, H. verticalis; I, H. sejunctiis; J, H. conspiciiiis. Figures by Ruth A. DeNicola. 14 Contributions in Science No. 180 H. cressoni, and the integument of the body is less densely tesselate, hence more shiny. The punctation, too, differs, being distinctly finer and sparser on the mesoscutum and scutellum, as indicated in the key. Some western material, such as the synonymous form of H. cressoni, H. pasadenae, are as extensively maculate as H. c. mesillae but differ in that the maculae are bright yellow in these, while in H. c. mesillae they are whitish. These highly maculate specimens also possess large maculae on the pronotal collars, usually lacking in H. c. mesillae. When such maculae are present in the desert subspecies, they are small and inconspicuous. Females from the desert areas commonly have the rugulae of the basal zone of the propodeum absent or restricted to the extreme anterior margin. In most material of H. cressoni cressoni the entire basal zone is distinctly longi- tudinally rugulose, except in some samples from Colorado and Utah. These appear to be intergrades between the two subspecies. This is somewhat puz- zling, since such specimens do not seem to occur in other areas where the ranges of these two forms overlap. Another intermediate form occurs in western Nebraska, eastern Wyoming and northeastern Colorado. This was described by Swenk and Cockerell as H. magniclavis. I have examined most of the type series of this bee and find no reason for its continued recognition. The males have the lateral face marks slightly expanded at their upper ends, but are otherwise similar to that sex of H. c. mesillae. I cannot find any differences to distinguish the females from those of H. c. mesillae. The male terminalia do not differ significantly from those of H. c. cressoni or H. c. mesillae, so there is no doubt that H. magni- clavis is conspecific with these. Since the pale maculae and fine, sparse thoracic punctation are in accord with those characters in H. c. mesillae, I consider H. magnicalvis a synonym of this subspecies. In figure 6 I have shown the results of a series of cephalic measurements which reflect differences between H. c. cressoni and H. c. mesillae. In 6A two phenomena are apparent. First that there are two concentration points on the basis of head width. There is some overlap, but the existence of two centers of concentration remains clear. A similar, but more obscure, pattern exists based on head length. These measurements of head length and head width are use- HL fully converted into a Cephalic Index (Cl = Hw X 100) which demonstrates their relationship as a ratio, figure 6B. A Cl value of 100, of course, indicates HL and HW being equal, values lower than 100 indicate increasing relative width. The Cl values of each form have been plotted in 4C as a frequency indication. From this figure it is obvious that while both forms most commonly possess Cl values ranging from 91 to 100, a much higher percentage (25%) of H. c. mesillae have the head longer than broad than is true of the nominate form (9% ). The second noteworthy feature in figure 6A is the displacement of the 1970 Studies on North American Bees 15 Figure 3. Hylaeiis (H.) spp., ventrite VIII of males: A, H. ellipticus; B, H. verticalis; C,H. cressoni mesillae; D, H. sejiinctiis; E, H. scmiciilae; F, H. fedorica; G, H. mari- timiis; H, H. stevensi. 16 Contributions in Science No. 180 Figure 4. Hylaeus {H.) spp., ventrites VIII and IX, respectively, of males: A, D, H. granulatus; B, E, H. rudbeckiae; C, F, H. conspicuus. 1970 Studies on North American Bees 17 Figure 5. Hylaeus (H.) spp., ventrite IX of males; A, H. ellipticus; B, H. verticalis; C, H. cressoni mesillae; D, H. sejunctus; E, H. saniculae\ F, H. fedorica; G, H. mari- timus; H, H. stevensi. 18 Contributions in Science No. 180 skew-line of H. c. mesillae to a level below that of H. c. cressoni. There is also an apparent, though slight, reorientation of the skew-line angle to that of H. c. cressoni. These suggest that perhaps H. c. mesillae might be best regarded as a full species rather than as a subspecies. I am not inclined, at this time, to do so since specimens of an apparently intermediate nature do exist. These specimens come from those areas where they may be expected to occur if H. c. mesillae is truly a subspecies, i.e., in the zones of sympatry. It is equally possible that these are hybrids of two incompletely separate species. However, since the male terminalia are, for practical purposes, so similar as to appear identical, I prefer to consider the desert form a subspecies of H. cressoni. This is also consistent with my treatment of other species. This subspecies extends from western Nebraska, Kansas and Texas west- ward to Nevada, California and Baja California, Mexico. Southward it ranges into the Mexican States of Nuevo Leon, Coahuila, Chihuahua and Sonora. Populations in Colorado and Utah are largely sympatric with those of the nominate subspecies and hybridization occurs in these areas. Hylaeus (Hylaeus) sejunctus Snelling, NEW SPECIES Figures IH; 21; 3D; 5D. Prosopis cressoni form mesillae, Metz, 1911, Trans. Amer. Ent. Soc. 37:107 (in part). Prosopis mesillae, Crawford, 1913. Canad. Ent. 45 : 154-155. Not of Cockerell, 1896. This is the species which Crawford called P. mesillae when he raised that name to specific level after Metz had reduced mesillae to a variety of H. cres- soni (Cockerell) . Diagnosis'. The male differs from that of H. c. mesillae in the less coarsely, closely punctate abdomen and in the shape of ventrite IX. The female differs in the much finer, sparser punctures of tergite II, the shorter, sparser pubes- cence of the abdomen, and the smaller size. Male. Very similar structually to that of H. c. mesillae but differs in hav- ing the punctures of the mesoscutum finer, sparser, usually separated by a puncture diameter or more; punctures of tergite I very fine, separated by two or more times a puncture diameter; discal pubescence of tergite III shorter than that of mesoscutum; apical lobes of ventrite IX broad, as long as, or longer than, basal pedicle. Female. Very similar structually to that of H. c. mesillae but differs as follows : punctures of median area of upper face of tergite I extremely fine, perceptible only under high magnification, very sparse; lateral areas of propo- deum roughened, densely tesselate, with scattered, but distinct, punctures; erect discal pubescence of tergite III little, if any, more than two-thirds an ocellar diameter in length (a few widely scattered hairs may attain this length, but are outnumbered by shorter hairs). 1970 Studies on North American Bees 19 Length, both sexes, to apex of tergite II, 3. 2-3. 7 mm. Holotype male, allotype female (Los Angeles County Museum of Natural History), Winterhaven, Imperial Co., California, March 25, 1956 (R. R. Snelling), Heliotropium curassavicum. Paratypes. — CALIF. Imperial Co.: 13 $ , 10 $ $ , same data as holo- type (LACM); 1 9 , 15.8 mi. NW Niland, April 11, 1963 (R. C. Dickson; UCR), on Prosopis’, \ , 2 mi. S. Travertine Rock, March 29, 1936 (F. R. Platt; UCR), on Prosopis) 1 9 , 5.2 mi. SE Kane Springs, April 15, 1949 (P. H. Timberlake; UCR), on Tamarix gallica) 1^,59 9 , Coachella Canal, 20 mi. SE Hwy 195, April 2, 1960 (R. C. Dickson, P. H. Timberlake; UCR), on Prosopis) 299, U.S.D.A. Exp. Farm, May 31, 1912 (J. C. Bridwell; USNM) ; 1 $ , 1 9 , same locality, June, 1912 (J. C. Bridwell; USNM) ; 1 $ , 1 9 , El Centro, April, 1911 (J. C. Bridwell; USNM) ; 2 $ $ , 1 9 , Imperial Valley, April, 1911 (J. C. Bridwell; USNM) . San Diego Co.: 1 $ , 1 9 , Borrego Valley, April 26, 1954 (J. G. Rozen; CIS) ; 1 $ , Palm Cyn., Borrego Valley, March 29, 1936 (P. H. Timberlake; UCR), on Prosopis) 1 9 , Borrego Valley, April 26, 1954 (M. Wasbauer; CIS) ; 1 9 , same locality and collector, April 30, 1954 (CIS) , on Croton calif or nica) 11 ^ $ , 19 9 9 , Coyote Cr., Borrego Valley, April 5, 1963 (F. D. Parker, R. M. Bohart, M. E. Irwin, A. Willink; UCD, IML); 2 $ $ , Borrego Springs, April 2, 1960 (M. Wasbauer; CDA), on Phacelia) 2 $ $ , Split Mt., Anza Desert State Park, April 1, 1955 (W. R. Richards; CNC) ; 1 9, Borrego Valley, April 20, 1961 (Rozen and Schrammel; AMNH), on Prosopis) 1 9 , Borrego Valley, April 11, 1962 (R. M. Bohart; UCD) . Riverside Co.: 10 $ $ , 5 ^ 9 , Andreas Cyn., March 26, 31, 1955 (W. R. M. Mason; CNC) ) \ $ , Willis Palms Oasis, April 9, 1955 (W. R. M. Mason; CNC) ; 2 $ $ , Indio, March 30, 1955 (W. R. Richards; CNC) ; 1 9 , Andreas Cyn., April 11, 1955 (W. R. M. Mason; CNC) ; 1 $ , 1000 Palms Cyn., April 3, 1944 (P. H. Timberlake; UCR) ; on Cryptantha barbigera) 3 $ $ , SVi mi. NW Indio, April 9, 1936 (P. H. Timberlake; UCR) , on Prosopis) 3 ^ ^ , Andreas Cyn., April 2, 1955 (P. H. Timberlake; UCR), on Hyptis enioryi) 1 9 , same locality and collector, April 1 1, 1936 (UCR), on Prosopis) 299, 1000 Palms Cyn., April 10, 1937 (P. H. Timberlake; UCR) , on Prosopis) 19,5 mi. W Indio, April 30, 1949 (E. G. Linsley, J. W. Mac- Swain, R. F. Smith; CIS) , on Melilotus. San Bernardino Co.: 10 $ $ , 1 2 9 , Cronise Valley, April 29, 1956 (J. Powell, M. Wasbauer; CIS) , on Prosopis) 5 $ $ , 1 9 , same locality, date, host (P. H. Timberlake; UCR), on Salix) 1 9 , 29 Palms, April 13, 1935 (P. H. Timberlake; UCR) , on Isomer is arbor ea) 7^^,1199, Needles, May 3, 1964 (P. Torchio & G. Bohart; USU), on Tamarix and Prosopis) 1 $ , Needles, 20 mi. S., May 3, 1964 (P. Torchio & G. Bohart; USU). Inyo Co.: 1 $ , Furnace Cr., Death Valley Natl. Mon., April 19, 1939 (E. G. Linsley; CIS), on Prosopis. NEVADA. 4 $ $ , Search- light, 30 mi. S., Clark Co., April 21, 1966 (Torchio, Rust, Yousef; USU). ARIZONA. 1 9 , Yuma, 21 mi. N., Yuma Co., April 4, 1963 (A. Willink; 20 Contributions in Science No. 180 IML) . MEXICO. 2 $ $ , 2 2 9 , Mexicali, 20 mi. W., Baja California del Norte, April, 1939 (C. D. Michener; CAS), on Prosopis. Hylaeus (Hylaeus) labiatifrons (Cockerell) Prosopis labiatifrons Cockerell, 1896. Psyche, 7 (suppl.) : 437. $ . Prosopis cressoni, Metz, 1911. Trans. Amer. Ent. Soc., 37 : 105 (in part) . Hylaeus {Hylaeus) labiatifrons, Mitchell, 1960. No. Car. Agr. Exp. Sta. Tech. Bull. 141:62,72. ^ . The status of this form, described from a single male from Georgia, is enigmatical. It is known only from the unique type, which I have not ex- amined. Metz (191 1 ) considered it a synonym of H. cressoni, but Mitchell (1960) has elevated it to species level and briefly redescribed the type. Un- fortunately, Mitchell’s description contributes nothing new to permit the recognition of H. labiatifrons. In all features it seems to agree quite closely with males of H. cressoni cressoni, and seems to differ from that species only in the entirely yellow tibiae. This difference seems to me to be of little value. Another Nearctic Hylaeus, H. illinoisensis (Robertson), is supposedly recognizable on the same basis. In that species, at least, males with entirely yellow tibiae are exceptional. Even those which have the tibiae wholly yellow show, under intense light, a slightly darker basal two-thirds which corresponds to the black basal area of most specimens. I suspect that the same is true in the case of H. labiatifrons, which I regard as a probable synonym of H. cressoni cressoni. The solution must wait until the type, especially ventrites eight and nine, can be examined and compared with those of the related species. Hylaeus {Hylaeus) fedorica (Cockerell) Figures lA; 2A; 3F; 5F. Prosopis digit at a var. fedorica Cockerell, 1909. Ann. Mag. Nat. Hist, (ser. 8) 4:27. $ . Prosopis grossicornis Swenk and Cockerell, 1910. Entomol. News, 21 :67. S . Metz, 1911. Trans. Amer. Ent. Soc., 37:118. NEW SYNONYMY. Hylaeus {Hylaeus) grossicornis, Linsley, 1951. IN Muesebeck, et al., U.S.D.A. Monogr. 2:1050; Mitchell, 1960. N. Car. Agr. Expt. Sta., Tech. Bull., 141:71-72. I have examined the type of Prosopis digit at a var. fedorica, now in the American Museum of Natural History, and have compared it with homotypes of H. grossicornis and there can be no doubt that these are conspecific, as first suspected by Metz (1911). The type locality of H. fedorica is Fedor, Lee County, Texas; H. fedorica has priority by one year over H. grossicornis. This species ranges from Michigan to Minnesota, then southward to Texas; an eastern extension exists through the southern United States to North Carolina. The male may be readily distinguished by the characters 1970 Studies on North American Bees 21 .5 .4 A □ □ □ □ □ □ □ □ □ □ □ □□□□□□ □ □ □ □ □ □ □ □ □ □ .3 5 I Q Q < UJ X .2 .1 1.0 o #• • • • <9 • • o o o o o O O o O O § O o o o o o o o oo o o o o o o o o o o o o oo o o o □ = rudbeckiae • : cressoni o= mesillae — I 1 1 ^ 1 \ 1 r— 1.0 .1 .2 .3 .4 HEAD LENGTH (MM.) Figure 6. A, scatter diagram plotting Head Width and Head Length of H. rudbeckiae (squares), H. cressoni cressoni (solid circles) and H. cressoni mesillae (open circles) ; B, Cephalic Index frequency for H. rudbeckiae (solid line), H. cressoni cressoni (broken line) and H. cressoni mesillae (dot-dash line). 22 Contributions in Science No. 180 given in the key. The female, similar to that of H. cressoni, is best recognized by the shape of the head in full face view. In all other species of the H. cressoni group, the inner margins of the eyes are strongly divergent above, with the eyes rather strongly bulging. In these species the widest part of the head is somewhat above the level of the antennal sockets. In the female of H. fedorica the inner margins of the eyes are only slightly divergent above; the eyes are much less strongly bulging above and the widest part of the head is at, or below, the level of the antennal sockets. The head is always slightly longer than broad. Hylaeus (Hylaeus) saniculae (Robertson) Figures 3E; 5E. Prosopis saniculae Robertson, 1896. Can. Entomol., 28:137. $ $ . Prosopis cressoni form saniculae, Metz, 1911. Trans. Amer. Ent. Soc., 37:106, 107. ^ 2. Hylaeus {Hylaeus) saniculae, Linsley, 1951. IN Muesebeck, et al, U.S.D.A. Monogr. 2:1052. Mitchell, 1960. N. Car. Agr. Expt. Sta. Tech. Bull., 141:73-74. This rare species, largely northern in distribution, is very poorly known. In the United States it extends from Minnesota to the New England states and southward, through the Appalachian Mountains, to Tennessee and Georgia. In Canada it is known from Ontario and Nova Scotia. Although the male may be readily recognized by the peculiar oval depressions on the sides of the face, the female appears to be indistinguishable from that of H. cressoni except for its more restricted maculae. Attempts to characterize the female of H. sani- culae are futile at this time because I have seen no females which I can confi- dently consider to be this species. Reared series, including both sexes, would be extremely useful in solving this problem. Hylaeus {Hylaeus) rudbeckiae (Cockerell & Casad) Figures IE; 2F; 3B; E; 6. Prosopis rudbeckiae Cockerell and Casad, 1895. Trans. Amer. Ent. Soc., 22:380. $ ; Metz, 1911. Trans. Amer. Ent. Soc., 37:111-114. Prosapis bakeri Cockerell, 1896. Psyche, 7 (suppl.) : 26. $ . Prosapis digit at us Cockerell, 1896. Op. cit., 30. $ . Prosapis rudbeckiae ruidosensis Cockerell, 1896. Op. cit., 30. $ . Prosapis rudbeckiae subdigitatus Cockerell, 1896. Op. cit., 31. ^ . Prosopis rudbeckiae form digitatus, Metz, 1911. Trans. Amer. Ent. Soc., 37:111, 112. Prosopis rudbeckiae var. bakeri, Metz, 1911. Op. cit., Ill, 112. Hylaeus rudbeckiae, Cockerell and Sumner, 1931. Amer. Mus. Novit., 490:7, 15. Hylaeus {H.) rudbeckiae, Linsley, 1951. IN Muesebeck, et al., U.S.D.A. 1970 Studies on North American Bees 23 Monogr. 2:1052; Mitchell, 1960. N. Car. Agr. Expt. Sta. Tech. Bull., 141:73. Although it is a simple matter to separate the males of this species from those of H. cressoni, I have found no reliable means to separate the females of the two species. There is a difference in the overall size of the females of these species which I have found most easily measured in terms of head width. However, the difference is a minute one and requires precise measurements. In figure 6A I have given the results of measurements of head width and head length of females of H. rudbeckiae compared with females of H. cressoni (represented by both subspecies). The range of head width is limited, varying from 1.31 mm to 1.50 mm, with the majority of the specimens having a head width in excess of 1.41 mm. In those specimens which have a head with less than 1.40 mm, the head length is 1.24 mm or more. In H. cressoni the total range of head width variation is much greater, extending from 1.12 mm to 1.37 mm. In those few specimens of H. cressoni which have a head width in excess of 1.34 mm the head length does not exceed 1.23 mm. These differences are slight but appear to be valid. They have been verified by checking against associated males and in reared series. Figure 6B contrasts the cephalic index of H. rudbeckiae with that of the nominate form of H. cressoni. It may be noted that while both species most frequently possess a Cl rating of 92, that is, with the head a little broader than long, a distinct tendency exists in H. rudbeckiae for the head to be broader still, and no specimens of this species seem to have a cephalic index over 100. While such high ratings are not common in H. cressoni, they do occur. In H. rud- beckiae, about 70% of the females examined had a cephalic index of 92 or less; in females of H. cressoni, only 33% have such a low rating. In other morphological features the females are very similar and I have noted none that are sufficiently constant to be reliable. The two species are almost completely sympatric, H. rudbeckiae absent only from those areas inhabited by H. c. mesillae. In most areas H. rudbeckiae is slightly more brightly maculate, the maculae as a rule both more extensive and of a brighter, yellower tone. The integument of the mesoscutum and scutellum is more densely tesselate and less shiny. Hylaeus (Hylaeus) granulatus (Metz) Figures 2F; 4A; D. Prosopis rudbeckiae var. granulatus Metz, 1911. Trans. Amer. Fnt. Soc., 37:114. $. Although this closely resembles H. rudbeckiae, it may be recognized by the features of the eighth and ninth ventrites (fig. 4A, D). External features do not appear to be reliable since both species are so variable. The following differences have been noted as tendencies which may be of value. The antennal and thoracic maculae of H. granulatus are less extensive than is usually true of H. rudbeckiae. In H. granulatus the scape may be black or with a small, basal / 24 Contributions in Science No. 180 spot on the lower margin; the pronotal collar may be immaculate or with a pair of minute yellowish spots at the extreme sides; the pronotal lobes are one- half, or more, black; the tegulae may be either black or with a small median spot. In H. rudbeckiae these areas are usually conspicuously maculate, except in specimens from northern or high altitude sites. The metanotum of both species is dull, but in H. granulatus it is rugosopunctate, while in H. rudbeckiae it has fine sparse rugulae and a few distinct scattered punctures. In H. rud- beckiae the basal zone of the propodeum is coarsely rugose, with the rugae in the median portion basically parallel. The basal zone of H. granulatus may be coarsely and irregularly rugulose, or with fine rugulae only on the anterior half, the posterior half appearing granulose. The status of this form is enigmatical. Metz described it from three males from Ormsby County, Nevada. No type was ever designated and so I am selecting a specimen from the original series, collected by Baker in July, as the lectotype. This specimen has been deposited in the American Museum of Natural History. Since the time of its description this bee has never been regarded as anything other than a variety of H. rudbeckiae, of which it has been considered a synonym. Its status here as a species is provisional, pending the acquisition of more material. This form may ultimately prove to be a hybrid resulting from a cross between H. rudbeckiae and H. cressoni. The extremes in variation in the sculpturing of the basal zone of the propodeum are suggestive of H. cressoni and the details of the eighth and ninth ventrites are intermediate between H. cressoni and H. rudbeckiae. If H. granulatus is a valid species, its female is currently unknown, but probably will be ex- tremely similar to that of H. rudbeckiae. In addition to the three males from Nevada I have seen others from scattered localities in the Sierra Nevada of California, from Lassen County to Tulare County, and from the San Gabriel Mountains in Los Angeles County. I also have a few males from Colorado which seem referrable to this form; these differ slightly, however, in the shape of the ninth ventrite which is very similar to that of H. cressoni. In areas where these specimens were taken, both H. cressoni and H. rudbeckiae are common. Hylaeus (Hylaeus) maritimus Bridwell Figures ID; 2D; 3G; 5G. Hylaeus maritimus Bridwell, 1910. Proc. Haw. Ent. Soc. 4: 159. $ $ . Hylaeus {Hylaeus) maritimus, Linsley, 1951. IN Musesebeck, et al., U.S.D.A. Monogr. 2:1051. This rare species appears to be largely restricted to the coastal sand dunes of central California. The two females taken by Timberlake at Mt. Pinos are the only exceptions, and it is difficult to explain the presence of H. maritimus in that area. 1970 Studies on North American Bees 25 Both sexes of this species, which is closely related to H. conspicuus, may be readily recognized by the characters given in the key. NEW RECORDS (all California) : Sonoma Co.\ 4 $ $ , Wright’s Beach, August 22, 1950 (J. Linsley; CIS), on Ehogonum. Marin Co.: 1 5 , Pt. Reyes, April 11, 1959 (G. I. Stage; LACM) ; 9 $ ^ , 2 $ 9 , same locality. May 16, 1958 (D. D. Linsdale; CIS) ; 2 $ $ , same locality and date (J. Powell; CIS) ; 1 9 , same locality. May 15, 1937 (R. C. Dickson; UCR) . San Francisco Co.: 1 $ , San Francisco (Land’s End), July, 1922 (F. X. Williams, UCR) ; 1 $ , San Francisco, April 16, 1960 (G. I. Stage and R. R. Snelling; LACM) ; 1 $ , San Francisco (Lobos Cr.), June 15, 1960 (G. I. Stage; GIS) ; 1 $ , San Fran- cisco (Lone Mtn.), July 4, 1920 (F. X. Williams; CAS) ; 1^,49 9 , San Francisco (sand dunes), April 9 to August 22, 1926 (C. L. Fox; CAS) ; 1 9 , Lake Merced, June 5, 1959 (T. R. Haig; UCD) . San Mateo Co.: 1 ^ , Salada Beach, April 21, 1918 (E. P. Van Duzee; CAS) . Monterey Co.: 1 $ , Pacific Grove, July 3 (T. D. A. Cockerell; UCR) ; 5 $ ^,109 9 , same locality. May 29, 31, 1954 (M. Wasbauer; CIS) ; 3 $ $ , same locality. May 16, 1959 (G. I. Stage; GIS, LACM) ; 1 9 , Carmel, July 21, 1923 (L. S. Slevin; CAS) ;2$ $ , 299, Asilomar, May 10, 16, 1959 (D. J. Burdick; GIS, LACM) ; 1 $ , Asilomar, 1 mi. S, May 17, 1959 (D. J. Burdick; LACM) ; 1 $ , Asilomar, 2 mi. S, September 26, 1959 (C. W. O’Brien; LACM) ; 1 5 , same locality. May 17, 1959 (G. I. Stage; GIS) . San Luis Obispo Co.: 1 $ <3,99 9 , Grover City, July 4, 1956 (E. G. Linsley, P. D. Hurd; CIS), “dunes Assoc.” Santa Barbara Co.: 1 $ , Refugio Cyn., July 4, 1959 (J. R. Russell: UCD) . Kern Co.: 1 9 , Mt. Pinos, August 17, 1935 (P. H. Timberlake; UCR), on Hemi- zonia corymbosa (var. bar clay i) ; 1 9 , same data, except, Grindelia (robusta) platyphylla (UCR). Hylaeus (Hylaeus) conspicuus (Metz) Figures II; 2J; 4C; F. Prosopis conspicua Metz, 1911. Trans. Amer. Ent. Soc. 37 : 1 14. S 9 . Hylaeus conspicuus, Cockerell and Sumner, 1931. Amer. Mus. Nov. 490:9. Cockerell and Sumner (1931) recorded a female from Jackson, Wy- oming, July 15, 1920, presumably collected by F. E. Lutz. I have on hand two females collected at Jackson on July 13-17, 1920, presumably by Lutz. Although these authors were surprised at finding this species in Wyoming, it is a widely distributed species in the northwestern states. In addition to the considerable material from California, I have seen specimens of H. conspicuus from the following states: Nevada (Ormsby Co., Gardnerville, Sparks, Para- dise V.; AMNH, USNM, UCD, Oregon (Medford, UCD), Washington (Pull- man; UCD), UC, CAS) , Idaho (Shoshone, Star, Murtaugh, Midvale, 5 mi. NE Mackay, Parma, Dixie, Boise, Moscow Mts., Sweet, Emery Can., Melba, 8 mi. N Twin Falls, 18 mi. S Moscow; UI), and Wyoming (Jackson; AMNH, CU). 26 Contributions in Science No. 180 Hylaeus (Hylaeus) stevensi (Crawford) Figures 5H. Prosopis stevensi Crawford, 1913. Can. Ent., 45; 155. $ $ . Hylaeus {Hylaeus) stevensi, Linsley, 1951. IN Muesbeck, et al, U.S.D.A. Agr. Monogr. 2:1052. Mitchell, 1960. N. Car. Agr. Exp. Sta. Tech. Bui. 141:61,62, 74-75; figs. 11, 12, 13, 14. This species is not closely related to any other in the Nearctic fauna, but is very similar to the Palearctic H. bisinuatus Forster. The two are so closely related that I know of no reliable external feautres which will separate them. The male terminalia of the two are very similar to one another. More material must be studied, however, to determine whether they are conspecific or not. This species has recently been introduced into Hawaii (Beardsley, 1959), and is evidently adventive in parts of its range where it is now quite common. I believe that this species has extended its range appreciably within the past 50 years. The earliest known capture of this species is in 1912 at Fargo, North Dakota; these specimens form the type series. The species has been collected in North Dakota on many occasions since then, but remained rare in Hylaeus samples taken until after 1925. After this date it became increasingly common, until it is now the species most frequently captured. The earliest records outside of North Dakota are 1915 (Colorado) and 1916 (Alberta, Canada). Earliest records for other states, proceeding west from North Dakota to the Pacific coast, are: Wyoming and Idaho (1920), Utah (1922), British Columbia and Washington (1925) and California (1933). This species is found in eleven states east of the Mississippi River; the earliest date of capture which I have seen is 1927 in Massachusetts. In none of his numerous papers on eastern bees does Robertson mention this bee; this species is so distinctly different from the remaining eastern Hylaeus that he certainly would have recognized it as different from any previously described. Within California the earliest record for this species is 1933 at Sacra- mento. There are few records in the 1930’s and 1940’s. But, beginning with 1950 there are abundant records, largely from the Central Valley area. My greatest field acquaintance with this species is in California, so the following observation must be considered with this point in mind. All of the California records are within, or on the periphery of, agricultural cultivated areas. I can affirm, further, that the bulk of my records from most of the area west of the Continental Divide come from similar situations, as nearly as I have been able to determine. This bee is of further interest in that it alone of all the species of Hylaeus, s. str., exhibits a pronounced tendency toward oligolecty on flowers of the family Leguminosae. Of 87 specimens bearing floral data, 56 were taken on Melilotus species. These legumes are introduced from the Palearctic region. Of the remaining 31 records, 13 are from introduced Palearctic plants in other families (Amaranthaceae, Umbelliferae, Cruciferaceae, Tamaricaceae, Legu- 1970 Studies on North American Bees 27 minosae) and eight of these are records of males. The remaining 18 records involve the families Polygonaceae, Asclepiadaccae, Salicaceae and Composi- tae, all belonging to genera represented in the Palearctic region; six of these records are of males. The above data would seem to indicate the possibility that this species was introduced into the area around Fargo, North Dakota sometime prior to 1912. Once established there it began to extend its distribution both east and west and by 1925 apparently reached both coasts; subsequent distribution has been extended southward, so that this species now occurs in southern California, Oklahoma and Georgia. It is now a well-established part of our fauna and appears to be oligolectic on the legume genus Melilotus. If this is, in fact, an introduced species, it may well prove to be a synonym of a described Palearc- tic species. This, however, may not be decided for many years yet due to the extremely chaotic status of the taxonomy of European Hylaeus. It is possible that this species is a native component which was, prior to the first quarter of the century, a very rare species. Subsequent to that time the increased abundance of a suitable food-source plant (the introduced genus Melilotus) and the conditions of an unbalanced ecology may have permitted this species first to thrive in its native region and then to spread out to other, similarly disturbed areas. Of numerous series of specimens reared from soda straw traps by A. R. Gittins in Idaho, many are referrable to H. stevensi. This species, then, is strongly apt to exploit artificial nesting possibilities, such as exist in and around farm buildings. Excluded Species In 1966 I included H. gaigei (Cockerell) in this subgenus largely on the conjectural statement of Mitchell (1960), who had seen the type, that it may prove to be the female of H. rudbeckiae. Since then I have examined the type, in the U.S. National Museum, and conclude that it was incorrectly associated by me with the subgenus Hylaeus. The type, a female, is a small, obscurely maculate individual. This species belongs to the subgenus Prosopis and may be a synonym of H. modestus Say. Paraprosopis The subgenus Paraprosopis in America north of Mexico is exceeded in number of species only by the subgenus Prosopis. This group is largely western in distribution and seems to be centered in the southwest. Although Holarctic, few species extend into Canada, and none of these very far. Only one species may be considered truly boreal; species of Paraprosopis are most commonly encountered in the Transition and Sonoran zones. This subgenus is a difficult one with which to work and this is particularly true of the females. The various species are very similar in appearance and structure and several of the most widely distributed species are quite variable. 28 Contributions in Science No. 180 Species groups are so poorly delimited that I see little point in attempting to define them except in very general terms. The largest group is that of H. wootoni (Cockerell) which includes the following species: H. color ad ensis (Cockerell), H. cookii (Metz), H. fiori- danus (Robertson) H. lunicraterius, n. sp., H. nevadensis (Cockerell), H. personatellus (Cockerell), H. polifolii (Cockerell) and H. wootoni (Cock- erell). There is considerable variation within the group, but all the species are superficially very similar in the females. In these the first tergite is densely and distinctly punctate (excepting H. fioridanus), the integument between the punctures usually smooth and somewhat shining, the transverse propodeal Carina is absent (except in H. cookii), the lateral and oblique propodeal cari- nae often reduced or absent, the facial foveae ending one-third or less of the distance between the eyes and ocelli, the mesopleurae densely and closely punctate and the head as broad, or broader, than long. Although the clypeus is typically immaculate, some females of one species {H. polifolii) may possess a small preapical spot. The males are similar to their females except for the usual sexual differences. Three groups include but a single species each. The H. sonorensis group contains only H. sonorensis Cockerell, a desert species recognizable by the head being longer than broad, the elongate pale macula on the female clypeus, the entire transverse macula on the pronotal collar and the extremely fine punctures of the second tergite. Another species in which the pronotal macula is entire is H. asininus (Cockerell and Casad), the sole member of the H. asininus group. In this group the facial foveae of the female end much nearer to the ocelli than the eyes, the transverse propodeal carina is present in both sexes, and the female mandibles are truncate apically. The only other species with an entire macula on the pronotal collar is H. megalot is. This unique member of the H. megalot is group may be easily separated from the other groups in the female by the elongate apical mandibular tooth, the absent or incomplete oblique propodeal carina and the broadly translucent apical mar- gins of the first four tergites. The male differs from that of most others by the finely, sparsely punctate first tergite, the broadly translucent tergal margins and the longer flagellar segments. All three of these groups, while apparently related to the H. wootoni group, seem not to be closely related to one another, although it is interesting that all occur in arid and semi-arid regions. Of the three, that of H. sonorensis seems farthest removed from the H. wootoni group as evidenced by the elongate face and the male apical sternites. The last group, that of H. calvus, includes four species: H. calvus (Metz), H. georgicus (Cockerell) and H. timberlakei, n. sp., and the Palearctic species, H. ater Saunders. In these species the facial foveae of the female end closer to the ocelli than the eyes, the anterior coxae tend to be angulate or sub- spinose laterobasally, the mesopleurae are shining and finely, sparsely punc- 1970 Studies on North American Bees 29 tate, the female genae are broader than the eyes in profile, the female clypeus is usually distinctly maculate and the first tergite of the female is finely and sparsely punctate. The females of several of the most common western species are very difficult to identify with certainty. The three most consistently difficult species are H. coloradensis (Cockerell) , H. polifolii (Cockerell) and H. wootoni (Cockerell) . Slight, but apparently consistent, differences in the shape of the eyes have been utilized. In the full frontal aspect the Eye Length (EL) is measured to the nearest tenth of a millimeter; the Frontal Eye Width (FEW) is measured across the eye to the inner margin at the level of the antennal sockets (fig. lOA) . Similar measurements are made in lateral view. Here, however. Lateral Eye Width (LEW) is measured at its greatest point. Con- siderable use has been made of the presence or absence of the oblique propodeal carina as a supportive feature. This carina is best seen when the dorsal surface of the propodeum is viewed obliquely from above and cephalad of the propodeum. The upper termination point of the facial foveae has also been frequently used in the key to the females. The position of termination is determined by viewing the top of the head from above. Measurement is made from the lateral ocellus to the nearest point of the inner eye margin and the position of the uppermost point of the fovea made on the basis of this line. The method of determining the Facial Width (FW) used in previous papers on this genus {i.e., the distance between the inner eye margins at the level of the antennal sockets) was found to be nearly useless within this sub- genus. A different method for determining facial width has been used below. In full frontal view, the inner margins of the eyes are seen to be slightly sinuate at a point a little above the level of the antennal sockets, at the lower end of the foveae. This sinuation probably corresponds to the ocular sinuses of many wasps. The measurement for facial width is made at the center of the slight sinuation. The Facial Length (FL) is measured as in previous papers {i.e., the distance from the anterior ocellus to the medio-apical clypeal margin). The Upper Facial Distance (UFD) is the maximal distance between the eye margins above, while the Lower Facial Distance (LED) is the minimal dis- tance between the eyes below. The Clypeo-Ocular Distance (COD) is meas- ured from the laterobasal clypeal angle to the nearest point on the eye margin; the Basal Clypeal Width (BCW) is the maximal width of the clypeal base; the Clypeo- Antennal Distance (CAD) is measured from the laterobasal clypeal angle to the nearest point on the rim of the antennal sockets. The Supraclypeal Quadrangel (SCQ) is bounded below by the clypeal base, on the sides by the subantennal sutures and above by an imaginary line connecting the juncture of the subantennal sutures with the rims of the antennal sockets. At several places in the key to the females reference has been made to punctures as being fine or coarse. The use of such poorly expressive terms has 30 Contributions in Science No. 180 been relegated to the role of supportive features, since such vague expressions are both subjective and relative. Punctures which appear to be fine on one bee may appear relatively coarse on another bee of smaller size. In a rough way, punctures which I refer to as “coarse” are those which exceed 0.025 mm in diameter and are usually about 0.03 mm in diameter. Fine punctures are usually less than 0.02 mm in diameter. Such measurements are nearly without value, however, since the punctures of the mesopleura may vary as much as 50% in diameter measurements. Statements relative to puncture size are to be interpreted as referring to the largest punctures on the pertinent sclerite. Ex- amination of females of several different species will soon permit one to adequately interpret the usage of such terms. The figures of the eighth and ninth sternites of the males require some comment. The eighth sternite consists of a complex arrangement of four membranous lobes, conveniently termed the lateral (or basal) and apical lobes, basad of which are the apodemal arms. The orientation on the plates places the apical lobes uppermost and the apodemal arms below; this is the same arrangement used in preceding papers and will be used in all subsequent papers on this genus. The lateral lobes of the eighth sternite are simple and usually possess a number of simple setae along their lower margins and at the apices. Some- times the apical portion of this lobe folds back upon itself and consequently assumes a shape very different from that which is considered normal. The apical lobes are much more apt to present difficulties because the membrane is much more delicate and complex, hence subject to more pronounced dis- tortion. These lobes may fold either along the inner margin or transversely at the middle. A well defined crease line is evident from the apex of the lateral arms to the outer margin and the transverse fold usually occurs at this point. Such folds will, of course, greatly alter the contour of the apical lobes and, because the membrane is so thin, this fact may easily pass unnoticed. An example is the figure by Metz (1911; pi. 5, fig. 60) of the eighth sternite of H. cookii; the distal half of the apical lobe has folded back upon itself, a fact unsuspected by Metz. The ninth sternite is a much simpler structure; it is elongate and apically bifurcate. The apical lobes are thin and are partially sclerotized. The normal plane of these lobes is more or less vertical, so that a true illustration of this segment in a normal position would show two thin, apically divergent process- es. However, since such a view would tend to conceal some characters of use taxonomically, the figures are traditionally based on slide mounted, and some- what flattened, specimens in which the apical lobes are even more divergent than is normal and the plane is angularly distorted. The alternative to such a procedure is to prepare two figures, one of the true ventral aspect and one of the lateral aspect. I have here abided by the traditional method of figuring this segment largely in order than a common ground for comparison between my 1970 Studies on North American Bees 31 figures and similar figures by other authors may be established. Thus, when comparing these figures with specimens, important features to note are the width of the basal shaft, the relative length of the lobes and the number and arrangement of the plumose setae. The shape and degree of divergence of the apical lobes should be viewed with caution. The key below includes all North American species described by previous authors, and recognized by me, as well as two new species described at this time. There exist a number of undescribed species from the western United States and Mexico but these are represented by too few specimens for me to wish to describe them at this time. If and when suitable series become available they will be described in a future part of this series, together with a new key. KEY TO NEARCTIC SPECIES OF PARAPROSOPIS 1. Females; antennae twelve-segmented 2 Males; antennae thirteen-segmented 15 2. Gaster without evident punctures, first tergite shining, polished; basal zone of propodeum as long as, or longer than, scutellum, much longer than metanotum; facial foveae ending much nearer ocelli than eyes georgicus (Cockerell) Gaster with distinct punctures, at least on first tergite, the tergite pol- ished or not; basal zone of propodeum shorter than scutellum, little, if any, longer than metanotum; facial foveae variable 3 3. Apical mandibular tooth elongate, as long or longer than its basal width, extending much beyond the well-defined inner tooth, apex of mandible thus acute; oblique propodeal carina absent or incomplete; facial foveae ending slightly nearer ocelli than eyes; first tergite polished, shining, punctures fine, separated by a puncture diameter or more; apical margins of tergites 1-4 broadly translucent yellowish megalotis Swenk and Cockerell Apical mandibular tooth much shorter than basal width, hardly ex- tending beyond the poorly defined inner tooth, apex of mandible thus truncate; oblique propodeal carina present or not; facial foveae vari- able; tergal surface variable; tergites usually without pallid apical margins 4 4. Facial length 1.4 times facial width, oblique propodeal carina absent, lateral carina absent or obsolescent; facial foveae ending near mid- point between eyes and ocelli; punctures of first tergite very fine, sep- arated by about a puncture diameter; second to fifth tergites duller, densely micropunctate and with fine appressed pubescence; clypeus with longitudinal macula; pronotal collar with complete transverse macula sonorensis Cockerell Facial length 1.3 or less, times facial width; oblique and lateral propo- deal carinae usually present; punctures of tergites variable, first tergite 32 Contributions in Science No. 180 frequently coarsely punctate; clypeus usually immaculate; pronotal collar usually with macula, when present, interrupted in middle 5 5. Outer margin of fore coxae abruptly produced near base, forming a triangular to subspinose process; facial foveae ending nearer ocelli than eyes; mesopleura distinctly shining, with well-separated fine punctures; first tergite finely, sparsely punctate; clypeus with longitudinal macula ...calvus (Metz) Outer margin of fore coxae not abruptly produced near base; remain- ing characters without the above combination 6 6. Basal zone of propodeum abruptly declivious behind, with distinct transverse carina separating the more or less horizontal anterior face from the vertical posterior face; metanotum densely tesselate, with scattered distinct punctures; wings pale whitish, veins yellowish to light brown 7 Without above combination of characters 8 7. Smaller species, facial length not over 1.2 mm.; second tergite sharply, closely punctate; facial foveae ending nearer eyes than ocelli, or mid- way between; macula on pronotal collar interrupted in middle cookii (Metz) Larger species, facial length 1,3 mm. or more; second tergite with very fine obscure punctures or nearly impunctate; facial foveae ending nearer ocelli than eyes; macula on pronotal collar usually entire - asininus (Cockerell and Casad) 8. First tergite smooth, punctures extremely fine, sparse to scattered; second tergite without obvious discernible punctures; mesopleura dull to moderately shining, with fine punctures separated by a puncture diameter or more; metanotum dull, without evident punctures fioridanus (Robertson) Without above combination of characters 9 9. Mesopleura moderately shining, with fine punctures separated by a puncture diameter or more; propodeal disc with coarse oblique rugae convergent toward mid-line, metanotum with coarse quadrate areolae at extreme base; clypeus with longitudinal median stripe timberlakei Snelling Mesopleura dull to slightly shining, punctures variable, usually sep- arated by about one-half a puncture diameter; propodeal disc usually without rugae, when rugae are present they are irregular; metanotum either not areolate basally or areolae extremely fine, appearing as a band of quadrate punctures; clypeus immaculate or with small pre- apical spot 10 10. Facial length 1.3 times facial width; oblique propodeal carina present, area above carina coarsely reticulate; transverse propodeal carina present lunicraterius Snelling Facial length no more than 1.2 times facial width; oblique propodeal 1970 Studies on North American Bees 33 Carina present or absent, area above usually not coarsely reticulate; transverse propodeal carina usually absent 11 11. Small species, facial length not exceeding 1.2 mm., usually about 1.1 mm.; facial length 1.1 times facial width; facial foveae ending about one-third of distance between eyes and ocelli; oblique propodeal carina usually present and area above coarsely VQ\.\cu\?iiQ..nevadensis (Cockerell) Larger species, facial length 1.3 mm. or more; facial length 1.2 times facial width; facial foveae ending about midway between eyes and ocelli; oblique propodeal carina and area above variable.... 12 12. Metanotum flattened, densely tesselate, but with distinct punctures nearly as large as those of scutellum; scutellum distinctly flattened, shinier than mesoscutum, more finely punctate, the punctures sep- arated by a puncture diameter or slightly more; lateral face marks re- duced, frequently absent; erect hairs of mesoscutum sparse, short, less than half as long as those of scutellum personatellus (Cockerell) Metanotum at least slightly convex, roughened, without evident punc- tures; scutellum slightly convex, not much duller than mesoscutum, punctures dense, nearly as large as those of mesoscutum; lateral face marks usually filling most of area between eye and clypeus; erect hairs of mesoscutum variable 13 13. Larger species, head width 1.6 mm. or more; oblique propodeal carina evanescent or absent; lateral propodeal carina usually evanescent above; clypeus always immaculate; EL 4. 5-4. 6 X ..coloradensis (Cockerell) Smaller species, head width 1.5 mm. or less; oblique and lateral propo- deal carina variable, but often present; clypeus often maculate; EL 3.3-4.5 X FEW 14 14. Oblique propodeal carina often absent, sometimes present but evanes- cent; clypeus immaculate wootoni (Cockerell) Oblique propodeal carina usually present and complete; clypeus often with preapical spot in middle polifolii (Cockerell) 15. Basal zone of propodeum twice as long as metanotum, as long or longer than scutellum; first tergite microtesselate, impunctate or with very fine scattered punctures; scape 1.2-1. 3 times longer than greatest width ..georgicus (Cockerell) Basal zone of propodeum usually less than twice as long as metanotum, usually much shorter than scutellum; first tergite polished or tesselate, as a rule with abundant punctures, scape variable 16 16. Head slightly longer than wide; EL 0.97 times HW, 1.17-1.20 times EL, 1.46-1.50 times UFW; oblique propodeal carina absent, lateral carina absent or evanescent; pronotal collar frequently with complete stripe; scape half as broad as long sonorensis Cockerell Head no longer than broad, usually broader than long; remaining characters not in above combination 17 34 Contributions in Science No. 180 17. Eyes rather strongly converging below, UFW at least 1.85 x LFW...=....18 Eyes less strongly convergent below, UFW no more than 1.80 x LFW, usually 1.75 or less 21 18. First two tergites with sparse, fine punctures, surface shining, with strong bluish iridescence; FL 2.6, or more, times LFW fioridanus (Robertson) First two tergites with coarse, distinct punctures mostly separated by a puncture diameter or less, surface variable, but never with bluish iridescence; FL 2.5, or less, times LFW. 19 19. Transverse propodeal carina absent, oblique carina poorly developed, basal zone narrowly rounded onto declivious face; second submarginal cell on marginal cell longer than adjacent portion of first submarginal cell; FL no more than 2.3 x LFW polifolii (Cockerell) Transverse and oblique propodeal carinae well-developed, basal zone abruptly declivious behind transverse carina; second submarginal cell on marginal cell shorter than adjacent portion of first submarginal cell; FL at least 2.50 x LFW 20 20. Smaller species, HW not exceeding 1.0 mm.; basal, horizontal portion of propodeum nearly twice as long as metanotum, little shorter than scutellum cookii (Metz) Larger species HW 1.28 mm. or more; basal, horizontal portion of propodeum little, if any, longer than metanotum, about half as long as scutellum asininus (Cockerell and Casad) 21. First tergite finely, sparsely punctate, the punctures often obscured by dense tesselation 22 First tergite coarsely, closely punctate, surface usually shining, punc- tures always clearly defined ..23 22. First tergite densely tesselate, rather dull, the punctures very obscure; lateral face marks ending abruptly at level of upper margin of antennal sockets (fig. 8K) or narrowly extended along inner orbits; labrum black or with median macula; flagellar segments beyond second slightly longer than broad... calvus (Metz) First tergite shining, the punctures distinct; lateral face marks slightly expanded above, and ending well above level of upper margin of antennal sockets (fig. 9F); labrum wholly yellowish; flagellar segment beyond second at least as broad as long ..megalot is (Swenk and Cockerell) 23. COD 0.85 X BCW; supraclypeal quadrangle distinctly wider than long; lateral face marks short, ending at or below upper margin of antennal sockets; supraclypeal area usually immaculate, but sometimes with an obscure, irregular macula nevadensis (Cockerell) COD at least equal to BCW; supraclypeal quadrangle often as long as wide; lateral face marks usually ending well above level of upper margin of antennal sockets; supraclypeal area usually with a large, quadrate 1970 Studies on North American Bees 35 macula 24 24. COD 1.45 X BCW or less; HW less than 1.2 mm.; BCW 1.20-1.25 x CAD wootoni (Cockerell) COD : : BCW variable; HW more than 1.3 mm.; BCW : : CAD variable 25 25. BCW less than CAD; COD more than BCW; sides of face usually strongly depressed below (fig. 8F) coloradensis (Cockerell) BCW equal to or exceeding CAD; COD : BCW; sides of face never depressed below 26 26. CAD more than BCW; supraclypeal mark always present distinctly quadrate; lateral face marks narrowed toward inner eye margin (fig. 8J) timberlakei Snelling CAD = BCW; supraclypeal mark often reduced or absent; lateral face marks ending broadly above (fig. 81) lunicraterius Snelling Hylaeus (Paraprosopis) wootoni (Cockerell) Figures 7A; 8A; lOB; HE. Prosapis wootoni Cockerell, 1896. Psyche 7 (sup.): 26. Prosapis divergens Cockerell, 1896. Op. cit.: 29. 5. NEW SYNONYM. Prosopis wootoni, Metz, 1911. Tarns. Amer. Ent. Soc. 37:144. Prosopis cookii Metz, 1911. Trans. Amer. Ent. Soc. 37:139-141. $, not <$ . Hylaeus wootoni, Cockerell and Sumner. Amer. Mus. Nov. 490:6. Hylaeus divergens, Cockerell and Sumner. Op. cit.: 7-8. Hylaeus perparvus Cockerell and Sumner, 1931. Op. cit.: 10-12. $. NEW SYNONYM. This is one of the more widely distributed species of the subgenus Paraprosopis. Throughout its range it varies considerably in both size and structure; one of these variants was named by Cockerell {H. divergens) . Metz reduced the name to the synonymy of H. wootoni, but twenty years later Cockerell and Sumner resurrected it on the basis of supposed genitalic differ- ences as well as a slight difference in the shape of the lateral face marks. These claimed differences will not hold up when one considers the range of variation in a long series of specimens from many localities. Thus, one finds specimens with the genitalic characters of H. wootoni and the face marks of H. diver- gens, and vice verse. Furthermore, these varying combinations are usually found to occur together in a single area. The differences noted are therefore considered merely to be due to variation in the range exhibited by a single species, H. wootoni. Unless there are other, more stable, characters which I have not discovered, there seems to be no reason for the continued recognition of H. divergens. I have examined the types of H. cookii, which are in the collection of 36 Contributions in Science No. 180 J K Figure 7. Hylaeus {Paraprosopis) spp., female faces: A, H. wootoni; B, H. flori- d anus', C, H. polifolii', D, H. nevadensis; E, H. cookii', F, H. coloradensis’, G, H. per- sonatellus; H, H. sonorensis', I, H. lunicraterius; J, H. timberlakei; K, H. calvus. Figures by Ruth A. DeNicola. 1970 Studies on North American Bees 37 J K L Figure 8. Hylcieus (Paraprosopis) spp., male faces: A, H. wootoni; B, H. fioridaniis', C, H. polifolii', D, H. nevadensis; E, H. cookii; F, H. coloradensis; G, H. personatel- liis; H, H. sonorensis; I, H. liinicrateriiis; J, H. timberlakei; K, H. calvus; L, H. poli- folii, melanic form. Figures by Ruth A. DeNicola. 38 Contributions in Science No. 180 P. H. Timberlake, and find that the female allotype cannot be separated from that sex of H. wootoni, although the type male represents a valid species (see below) . The holotype male of H. perparvus Cockerell and Sumner has been ex- amined. This specimen is a typical male of H. wootoni. The supposed differ- ences in the genitalia are due to distortion which resulted from extreme flat- tening during the preparation of the slide. The distal portions of the lateral wings of the eighth ventrite are curled upward on the mount; this segment is at the edge of the balsam and difficult to see clearly. The ninth ventrite falls well within the range of variation attributed to H. wootoni. Specimens have been examined from the following localities: UNITED STATES. Nebraska (Monroe Can. in Sioux Co.; UN), Colorado, New Mexi- co, Arizona, Utah (Emigration Can. and Big Cottonwood Can. in Salt Lake Co., Ogden; UCD) , Nevada, California, Oregon (Tumalo Res. and Sisters in Deschutes Co.; UCD, CIS), Idaho (Craters of the Moon, Moscow Mts., Starkey, Ahsahke, 8 mi. E., Whitebird, 7 mi. N., Rock Cr., 12 mi. S., in Twin Falls Co., Twin Spr., 2.5 mi. W., Emery Can. in Cassia Co.; UCD, UI), Wyoming (Little Goose Can. near Sheridan; AMNH, CU) . MEXICO. Chi- huahua (Madera, San Jose de Babicora; AMNH). CANADA. British Colum- bia (Revelstoke; CU). Hylaeiis (Paraprosopis) floridanus (Robertson) Figures 7B; 8B; IOC; IIF. Prosopis floridanus Robertson, 1893. Trans. Amer. Ent. Soc., 20:273. $ . Prosopis eulophi Robertson, 1905. Canad. Ent., 37:236. $ $. Prosopis modestus, Metz, 1911. Trans. Amer. Ent. Soc., 37:121. (in part) Hylaeus (Paraprosopis) packardi Mitchell, 1951. Jour. E. Mitchell Sci. Soc., 67:244. Mitchell, 1960. N. Car. Agr. Exp. Sta. Tech. Bui., 141:62, 77. NEW SYNONYMY. Hylaeus (Paraprosopis) floridanus, Mitchell, 1960. Op. cit., 61, 62, 77- 78; figs. 11, 12, 14. $ 9. I have examined the paratype of H. packardi and find that, aside from the immaculate pronotal collar and tegulae, it does not differ sufficiently from H. floridanus to justify specific status. Since such maculations are highly variable within this subgenus I do not feel that this form can be considered a subspecies either. While I have seen no other specimens with the immaculate pronotal collar, I have seen some with the tegular spot absent; it is commonly greatly reduced. This species is so far known from the eastern United States, from Maine to Florida. A westward extension occurs along the Great Lakes as far as Minnesota. 1970 Studies on North American Bees 39 Hylaeus (Paraprosopis) nevadensis (Cockerell) Figures 7D; 8D; lOD; IIH. Prosapis nevadensis Coc]f.QrQ\\, 1896. Psyche 7 (sup.): 32. Prosopis nevadensis, Metz, 1911. Trans. Amer. Ent. Soc. 37 : 148. Hylaeus oregonensis Bridwell, 1919. Proc. Haw. Ent. Soc. 4: 160-161. $ $ . NEW SYNONYMY. Bridwell based H. oregonensis on the lack of hairs on the eighth ventrite of the male, which he stated was like that of H. calvus (Metz). I have exam- ined the holotype specimen, which is in the United States National Museum, and the paratype male. An examination of the eighth ventrites of these speci- mens convinces me that there is no justification for continued recognition of this form. The presence or absence of hairs on this segment appears to be a variable feature in H. nevadensis and I have seen many degrees of expression in specimens from widely scattered areas. In all other features H. oregonensis agrees with material of a more typical nature. The paratype female is insep- arable from H. nevadensis. This species occurs at moderate elevations in California, Nevada, Oregon (Mt. Jefferson, Grayback Camp in Josephine Co., Union Cr. in Jackson Co.; USNM, UCD), Washington (Seattle; UC), Idaho (Troy, Craters of the Moon, Whitebird, 7 mi. N., Athol, Granite, Nampa; UCD, UI), and British Colum- bia (Lilleoet, Robson, Summerland; CU, CNC). Hylaeus (Paraprosopis) coloradensis (Cockerell) Figures 7F; 8F; lOF; IIJ. Prosapis coloradensis Cockerell, 1896. Psyche 7 (sup.) : 7. 5 . Prosopis coloradensis, Metz, 1911. Trans. Amer. Ent. Soc. 37 : 141. $ $ . Hylaeus coloradensis, Cockerell and Sumner, 1931. Amer. Mus. Nov. 490:6. $ (in part). Specimens examined are from the following states and Canadian prov- inces: UNITED STATES. Wyoming (Yellowstone Natl. Park; AMNH), Colorado, Utah, New Mexico, Arizona, Nevada, California, Idaho (White- bird, 7 mi. N., Leadore; UI) , Washington (Mt. Constitution, Pullman, Colfax; CAS, UC) . CANADA. British Columbia (Revelstoke; CU) . Hylaeus (Paraprosopis) personatellus (Cockerell) Figures 7G;8G; 101; 1 IK. Prosopis personatella Cockerell, 1915. Annals and Magazine of Natural History (ser. 8) 16:485. 9. This species was described from a single female from Corona, Colorado, a locality which I have been unable to locate. Cockerell remarked, however, that the collection was made above timber-line which suggests a high altitude. The species has remained virtually unknown since that time although (Linsley, 1951) added California to the distribution. During this study I have seen few 40 Contributions in Science No. 180 specimens which I can assign to this species, the type of which I have ex- amined at the U.S. National Museum. The females which I assign to this species all agree more or less closely with Cockerell’s original description and with the type. The lateral face marks are commonly absent; when present they are greatly reduced, restricted to an irregular stripe or blotch along the lower inner orbits. Since females of other species of Paraprosopis taken in the same area frequently also have reduced face marks, this character alone is of little value. The pronotal collar seems always to be immaculate; the pronotal lobes and tegulae are maculate or not. As a rule, only the hind tibiae have a yellow basal spot, although some have such a mark on the fore tibiae. The females most closely resemble those of H. coloradensis, to which this species appears to be related. They may be separated from that species by the distinctly flattened scutellum which is distinctly more shining than the mesoscutum and with fine, well-separated punctures. The metanotum, too, is flattened. As in H. coloradensis, it is dull and densely tesselate, but has distinct, well-separated punctures about equal to those of the scutellum. Females from areas where this species occurs together with H. coloradensis differ from that species also in having the erect hairs of the mesoscutum very sparse and short, so that they are hardly distinguishable from the reclinate hairs. These erect hairs are much shorter, usually no more than one-third as long, than those of the scutellum. In the sympatric females of H. coloradensis the erect mesoscutal hairs are abundant, and are usually as long, or nearly so, as those of the scutellum. The presumed male of this species is similar to that of H. coloradensis but the sides of the face are not depressed as in that species. There are tangible differences in head shape as may be seen upon comparing figures 8F and 8G. Although the eyes are apparently more strongly convergent below in H. personatellus this is not actually the case, as the ratios are very nearly identical in the two species. The greater eye length of H. personatellus may be used as a convenient factor. In this species the eye is 3.4 times its greatest width but twice as long as the lower interorbital distance. In H. coloradensis the eye is 3.8 times longer than its greatest width, but only about 1.7 times longer than the lower interorbital width. The few males of H. personatellus which I have seen are all more densely pubescent on the thoracic dorsum than is true of H. coloradensis. The difference, however, is a matter of degree, and I do not attribute much significance to this character. Otherwise, the males are very similar externally and no reliable differences between them have been dis- covered. In addition to the type I have seen the following additional specimens: COLORADO. 1 9, Ouray, 7500-8000', 1-15 July 1897 (H. F. Wickham; LACM). CALIFORNIA. 1 9, Strawberry Valley, El Dorado Co., 16 Aug. 1912 (E. C. Van Dyke; CAS) ; 1 9 , Sonora Pass, Tuolumne Co., 4 Aug. 1948 1970 Studies on North American Bees 41 (P. D. Hurd & J. W. MacSwain; CIS); 1 $, Blue Canyon, Sonora Pass, Tuolumne Co., 10 Aug. 1960 (J. W. MacSwain; CIS); 1 2, same locality, 30 Aug. 1960 (P. D. Hurd; CIS) ;9 S S , Tioga Pass, 7 mi. E., Mono Co., 15 July 1961 (H. V. Daly, G. I. Stage; CIS) ; I S , Tom’s Place, 1 mi. W., Mono Co., 13 Aug. 1957 (D. D. Linsdale; CIS), on Eriogonum elatum; A $ ^ \ 2 , near Mono Pass, 12,000', Inyo Co., 15 Aug. 1957 (C. D. MacNeill; CAS) ; 5 $ 2 , Mono Pass, Inyo Co., 13 Aug. 1957 (J. Powell, D. D. Linsdale; CIS), one on Pentstemon newberryi. Hylaeus (Paraprosopis) polifolii (Cockerell) Figures 1C; 8C; 8L; lOG; 1 IG. Prosopis polifolii Cockerell, 1901. Canad. Ent., 33:281. $ . Metz, 1911. Trans. Amer. Ent. Soc., 37 : 146-148. Hylaeus polifolii catalinensis Cockerell, 1938. Annals and Magazine of Natural History, ser. 11, 2:150. $ 2. NEW SYNONYMY. This common California species is closely related to H. wootoni and females are apt to be very difficult to separate in the absence of males. The two species are largely allopatric with overlaps occurring only in some areas of southern California and along the western portions of the Sierra Nevada. Females of H. polifolii which possess the preapical clypeal macula are easily recognized; I have never seen specimens of H. wootoni so maculated. The population of H. polifolii present on Santa Catalina Island, Los Angeles Co., Calif., was given a varietal name by Cockerell; in the females of this popula- tion, the clypeal macula seems always to be present. Populations from main- land California may or may not have a maculate clypeus, but this feature is subject to much variation within a given sample and is, moreover, not cor- related with distribution. I see nothing to be gained by recognition of this insular form and have synonymized it under the nominate form. The facial maculae of the males, too, are subject to considerable varia- tion. In what I consider to be the basic or “typical” pattern, the clypeus is entirely yellow, the supraclypeal area is black, and the lateral face marks terminate above at a point about equal to a socket diameter above the antennal sockets. One variant form, very rare, exhibits an obscure maculation of vari- able size on the supraclypeal area. In the second variant form, which is fairly common, the clypeus becomes darkened along the lateral margins; this in- fuscation frequently encroaches so extensively onto the clypeus that the maculation is reduced to an erect median stripe. Hylaeus {Paraprosopis) cookii (Metz) Figures 7E; 8E; lOE; 111. Prosopis cookii Metz, 1910. Trans. Amer. Ent. Soc. 37:139-141. not 2 . As indicated above, the female assigned by Metz to this species is H. 42 Contributions in Science No. 180 wootoni; the male type, now in the collection of the University of California at Riverside, is from Filmore Canyon, New Mexico, collected on Aug. 27 by T. D. A. Cockerell. Another male, with identical data is in the collections of the Los Angeles County Museum. Until the present study was undertaken H. cookii was known only from the type. It is doubtful if this species could have been recognized from the original description, or the figures of the genitalia and hidden ventrites as illustrated by Metz. When I was able to examine the type, I found it to be identical with a species which P. H. Timberlake and I had independently assumed to be undescribed. Because of the inadequacies of the original de- scription, the species is redescribed and the hidden ventrites illustrated. Diagnosis. — Male: Small species, head width not exceeding 1 mm.; scape less than half as wide as long; lateral face marks ending well above level of antennal sockets; frontal punctures coarse, distinct; mesoscutal punctures distinctly finer than those of frons; punctures of tergite I as coarse as those of mesoscutum, those of tergite II finer, dense on both segments. Female: Small, head width not exceeding 1.3 mm.; frontal punctures coarse, distinct; meso- scutal punctures equal to those of frons; metanotum not rugulose, with distinct punctures; abdominal punctation similar to that of male, but punc- tures finer, tergite III distinctly, though finely, punctate. Male. — Integument black, except the following pale creamy-white: cly- peus; supraclypeal mark; lateral face marks, ending acutely on eye margin well above level of antennal sockets; medially interrupted stripe on pronotal collar; pronotal lobe; spot on anterior half of tegula; small apical spot on all femora; stripe on anterior tibiae; elongate basal and small apical spots on middle and hind tibiae; basitarsal and mediotarsal segments. Apical tarsal segments light ferruginous; tibial spurs whitish. Wings whitish-hyaline, veins and stigma light ferruginous. Erect pubescence whitish. Head: UFD 0.76 x FL; LFD 0.52 x UFD; OCD 1.06-1.21 x TFD; basal width of clypeus 0.77 x distance from laterobasal angle to eye margin; distance between antennal sockets about equal to a socket diameter, distance from sockets to eye margin about 0.85 x a socket diameter. Maximum scape width about 0.42 x maximum length; minimum length of first flagellar segment about one-half length of second. Maculate portions of face slightly shining, finely punctate, those of clypeus separated by one-half to a full puncture diameter, those of lateral part of face somewhat elongate, linearly arranged, spacing variable; frons coarsely, contiguously punctate, the punctures distinct, the frons somewhat bulging on either side of midline. Vertex sharply raised above tops of eyes, coarsely punctate, the punctures crowded and less sharply defined than on frons. Thorax: Mesoscutum dull, closely coarsely punctate, the punctures slightly larger than those of frons, separated by one-third to one-half a punc- ture diameter; mesoscutellum slightly shining, punctures about equal to those 1970 Studies on North American Bees 43 of mesoscutum, separated by one-half to a full puncture diameter; metanotum dull, densely tesselate, with punctures finer than those of mesoscutum, sep- arated by about a puncture diameter; mesopleurae dull, tesselate, punctures about equal to those of mesoscutum, separated by about one-third to one-half a puncture diameter. Basal area of propodeum areolate; lateral areas coarsely punctate, the punctures poorly defined; transverse carina poorly developed, oblique carina absent; disc densely tesselate, with coarse punctures. Abdomen: Tergite I shining, smooth, densely punctate, the punctures smaller than those of mesoscutum, separated by one-half to a full puncture diameter; tergite II more finely closely punctate, apical one-third (approxi- mately) , depressed, with scattered micropunctures, apical margin slightly reflexed; tergite III more finely, sparsely punctate; apical ventrites as illustrated. Length, to apex of second tergite, 3. 0-3. 4 mm.; of forewings, 2.45-2.9 mm. Female. — Similar to male in essential features, except as follows: clypeus and supraclypeal area dark, lateral face marks more rounded above; UFD 0.72-0.80 X FL; LFD 0.55-0.61 x UFD; OCD 1.02-1.07 x TFD; basal width of clypeus equal to distance from laterobasal angle to eye margin; distance between antennal sockets about 1.5 x a socket diameter; distance from sockets to eye margin slightly exceeding a socket diameter. Length, to apex of second tergite, 4. 0-4. 3 mm.; of forewings, 3. 2-3. 5 mm. New records. — Arizona: 1 $ , Rincon Mtn., IX-16-1937 (E. D. Ball) ; $ $ , Stockton Peak, Pinaleno Mts., V-20-1954 (F. Werner), on Fallugia; 6 $ $ , Madera Can., Sta. Rita Mts., IX-28-1956 (F. Werner), swept from Aplopappus and Eriogonum; 2 $ $ , Sta. Catalina Mts., V-15-1955 (G. D. Butler) ; 1 9 , Ramsey Can., Huachuca Mts., IX-8-1955 (F. G. Werner and G. D. Butler) , on Petalostemum candidum\ 1 9 , west slope Patagonia Mts., V-21-1955 (G. D. Butler) , on Prosopis; 3 $ $ , 15 9 9 , same locality, IX-7- 1955 (G. D. Butler and F. G. Werner; all UA) , on Erigeron canadensis; 7 9 9 , Sycamore Can., Santa Cruz Co., IX-7, 8-1961 (G. I. Stage) ; 5 9 9 , Southwest Research Station, IX-3, 4-1959 (G. I. Stage; all LACM) ; 1 $ , same locality, V-14-1956, 1 9 , same locality, VI-14-1956, 1 9 , same locality, VII-19-1956 (all M. Statham) ; 699, Paradise, 4 mi. W., IX-9- 1950 (W. Gertsch and M. Cazier; all AMNH) ; 1 9 , 7 mi. W. Onion Saddle, Chiricahua Mts., Cochise Co., IV-10-1954, on Haplopappus gracilis; 299, Cave Cr., 5400 ft., Chiricahua Mts., Cochise Co., IX-8-1958, on Solidago; 499, Prescott, VI-27-1932, on Asclepias tuberosa; 1 $ , 5 9 9 , same lo- cality, VIII-30-1930, on Eriogonum; 1 $ , 1 9 , same locality and host, VII- 31-1930 (all P. H. Timberlake; all UCR) ; 1 $ , Painted Desert, Apache Co., VIII-18-1958 (G. I. Stage; UCR), on Eriogonum aureum. Nevada: 1 $ , Eastgate, 1 mi. W., Churchill Co., VIII-1 1-1958 (E. G. Linsley; CIS), at honeydew on Chrysothamnus sp. California: 12 $ $ , 2 9 9 , Paradise Camp, 44 Contributions in Science No. 180 Mono Co., V-7-1960 (A. E. Menke; UCD); 1 ^ , 1 $ , same locality and date (F. D. Parker; UCD); 1 $ , Cottonwood Spr., Riverside Co., IV- 12- 1950 (W. F. Barr; UI) ; 10 ^ ^ , Apple Valley, San Bernardino Co., V-10, 12-1955 (W. R. M. Mason; CNC); 5 $$,322, Ivanpah, 12 mi. SE., San Bernardino Co., V-1-1956 (M. Wasbauer) ; 1 $ , same data, on Salvia; 1 $ , same locality and date (B. J. Adelson), on Salvia; A $ $ , same locality and date (P. D. Hurd; all CIS), 2 on Sphaeralcea; 2 $ $ , same locality and date, on Rhus trilobata; 1 $ , same locality and date, on Salvia vitosa; 2 $ $ , same locality and date, on Sphaeralcea ambigua; 1 $ , 3 2 2 , Cactus Flat, 6000 ft., San Bernardino Mts., IX-1-1936, on Chrysothamnus stenophyllus; 1 2 , Morongo Valley, San Bernardino Co., IX-29-1944, on Eriogonum plumatella; 3 $ 9 , same locality, X-5-1934, on Gutierrezia lucida; 1 9 , same locality, X-14-1934, on G. lucida; 2 2 2, Yucca Valley, San Bernardino Co., IX-28- 1944, on G. lucida; 1 9 , same locality and host, IX-30-1944; 1 9 , Mt. Home Cr., San Bernardino Mts., 5000 ft., VIII-14-1934, on Phacelia ramosissima; 299, Andreas Cyn., Riverside Co., X-27-1934, on Solidago calif or nica; 1 9 , Pinon Flat, San Jacinto Mts., VI-4-1934, on Nolina perry i; 1 9 , same data except, on Eriogonum fasciculatum; 1 $ , Whitewater, Riverside Co., IV-22-1928, on E. fasciculatum; 1 2 2, Whitewater Cyn., Riverside Co., X-14-1951, on Lepidospartum; 899, Lone Pine Cyn., 6000 ft., San Gabriel Mts., VII-21-1936, on Tetradymia canescens; 1 9 , same locality and date, on E. fasciculatum; 1 $ , Surprise Cyn., Panamint Mts., Inyo Co., IV-25-1957 (all P. H. Timberlake; all UCR), on Eriogonum inflatum. Hylaeus (Paraprosopis) sonorensis Cockerell Figures 7H; 8H; lOK; IIL. Hylaeus sonorensis Cockerell, 1924. Proc. Calif. Acad. Sci. (4) 12:530- 531. 2. Hylaeus sonorensis var. melanorhinus Cockerell, 1924. Op. cit. :531. 2. NEW SYNONYMY. Diagnosis: Although the superficial appearance of this species is quite similar to that of H. as minus (Cockerell & Casad), the shape of ventrites VIII and IX of the male ally it to H. wootoni (Cockerell) . Both sexes differ from H. asininus in the more elongate face, denser and finer thoracic punctures and microscopic punctures of tergite II. From H. wootoni this species differs in the longer, narrower face, more finely punctate abdomen and more extensive maculae. Male. — Integument black, except the following pale creamy-white; elongate spot at base of mandibles; labrum, clypeus; elongate supraclypeal mark; paraocular areas and along inner orbits, ending acutely on level mid- way between anterior ocellus and antennal sockets; entire transverse stripe on pronotal collar; pronotal tubercules; anterior half of tegulae; apical spot on all femora; entire fore tibiae except irregular blotch on inner surface; basal half 1970 Studies on North American Bees 45 of middle tibiae; basal three-fifths of hind tibiae; all basitarsi and first hind mediotarsal segment. Underside of flagellum yellowish-ferruginous. Medio- and distitarsal segments brownish; wings clear hyaline, veins and stigma brownish. Pubescence all pale whitish, on abdomen suberect and obscure. Head: UFD 0.64-0.65 x FL; LFD 0.59-0.60 x UFD; OCD 1.19-1.21 x TFD; clypeus distinctly narrower basally than distance from latero-basal angle to eye margin; distance between antennal sockets less than socket diameter, distance from sockets to eye margins equal to socket diameter; clypeus, supra- clypeal area and maculate facial area dull, densely tesselate, with scattered obscure punctures; nonmaculate areas with large, subcontiguous punctures, the densely tesselate interstices slightly shining; genae shining, punctures small, separated; minimum length of first flagellar segment about one-half length of second. Thorax: Outer margin of fore coxae slightly convex; mesopleural punc- tures about equal to those of frons, mostly separated by about one-half a puncture diameter, the tesselate interstices more shining than frons; mesoscu- tal punctures equal to those of frons, subcontiguous, interstices slightly shin- ing; mesoscutellar punctures larger than mesoscutal, spacing variable, but denser medially; metanotum not at all shining, very densely tesselate, with a few, coarse, deep punctures. Basal area of propodeum with a few irregular striae, the lateral ones transverse; latero-basal area roughened, shining, im- punctate; lateral area slightly shining, with coarse, close punctures; oblique Carina absent; lateral carina feebly developed, absent above lower half of disc. Abdomen: Tergite I micro-tesselate, slightly shining, with fine punctures mostly separated by about a puncture diameter; tergite II more distinctly tesse- late, punctures very fine, obscure, separated by about a puncture diameter; remaining tergites tesselate, slightly shining, with a few, scattered, obscure punctures; ventrites VIII and IX as in figures lOK and 1 IL. Measurements: Body length (front of vertex to apex of second tergite), 3.5 to 3.8 mm.; forewing length, 2.9 to 3.1 mm. Female. — Similar to male in sculpture and color, except as follows : mandibles and labrum entirely dark; clypeus with _L-shaped macula; supra- clypeal area with small irregular mark; lateral face marks more rounded above; facial foveae ending about one-third of distance between eyes and lateral ocelli; UFD 0.69-0.70 x FL; LFD 0.68-0.70 x UFD; OCD 0.94-0.96 x TFD; distance between latero-basal clypeal angle and inner eye margin 0.75 times basal width of clypeus; distance between antennal sockets equal to a socket diameter and to distance between sockets and eyes. Fore basitarsus con- colorous with medio- and distitarsi, hind basitarsus dark at apex. Punctures of tergite I a little finer and more separated than in male. Measurements: Body length (front of vertex to apex of second tergite), 4.5 to 4.7 mm.; forewing length, 3.4 to 3.7 mm. Both forms of H. sonorensis were described by Cockerell from single 46 Contributions in Science No. 180 females from Guaymas, Sonora, Mexico; I have examined the types of both at the California Academy of Sciences and can affirm that they are nonspecific. Although Cockerell claimed this species to be related to those of tropical Mexico, this is not true. On the basis of currently available specimens H. sonorensis ranges from the Death Valley region of California and Nevada into southern Arizona, thence south into Sonora, Mexico at least as far as Guaymas. Another extension intrudes into Baja California at least as far as Bahia de Los Angeles. The variety melanorhinus was distinguished by the lack of a supraclypeal mark and the reduction of the clypeal macula to a transverse apical band with a median upward extension. In the “typical” form the apical third of the clypeus was pale yellowish with a broad, cuneate upward extension. Such variants are common in the series of females available and there seems to be no good reason for the recognition of these variant forms, since the variations are not correlated with distribution. Distribution. — MEXICO. Baja California del Norte : 2 $ $ , Bahia de los Angeles, May 14, 15, 1949 (R. C. Dickson; UCR), one on Prosopis glandulosa. Sonora: 1 $ , Guaymas, April 8, 1921 (E. P. Van Duzee; CAS, type of H. s. melanorhinus) ; 1 $ , Guaymas, April 10, 1921 (E. P. VanDuzee; CAS, type of H. sonorensis) . UNITED STATES. Arizona: 4 $ $ , 2 2 $ , Alamo Cyn., Organ Pipe Cactus Natl. Mon., April 15, 1955 (G. D. Butler and F. G. Werner; UA), on mesquite; 1 2 , same data, except on Encelia (UA) ; 1 2 , Sabino Cyn., Santa Catalina Mts., May 26, 1962 (F. D. Parker and L. A. Stange; UCD). Nevada: 1 $ , Aurora, 7500', Mineral Co., August 9, 1939 (W. D. Pierce; LACM) , on Scrophularia californica. California : 3 $ $ , 6 2 $ 5 Surprise Cyn., Panamint Mts., Inyo Co., April 22, 1957, on Erio- gonum inflatum; 2 $ $ , 9 2 $ , same locality and host, April 23, 1957; 1 $ , same locality, April 23, 1957, on Stanleya pinnata; 3 $ $ , 16 2 2, same locality, April 25, 1957, on E. inflatum; 1 $ , same locality, April 25, 1957, on Prosopis; 2 $ $ , same locality, April 25, 1957, on S. pinnata; 2 $ $ same locality, April 28, 1957, on S. pinnata; 2 $ $ , same locality, April 29, 1957, on S. pinnata {all P. H. Timberlake; UCR) ; 14 2 2,9 $ $ , same locality, April 24, 1957 (P. D. Hurd and G. I. Stage) , on E. inflatum; 2 $ $ , same locality, April 28, 1953 (P. D. Hurd) ; 3 $ $ , same locality and date (G. A. Marsh), one on Eucnide urens, two on Eriogonum sp.; 1 2, Furnace Cr., Death Valley, Inyo Co., March 3, 1953 (J. W. MacSwain) ; 1 2 , Cronise Valley, San Bernardino Co., April 29, 1956 (M. Wasbauer), on Prosopis; 2 $ $ , Plaster City, 15 mi. N., Imperial Co., March 27, 1961 (C. A. Toschi;^// CIS). Hylaeus (Paraprosopis) lunicraterius, new species Figures 71; 81; lOH; 12A. This species is very closely related to the following species. The males 1970 Studies on North American Bees 47 are especially similar in details of the hidden ventrites and genitalia. Diagnosis’. Female with head as broad as long; clypeus immaculate; facial foveae ending about one-third of distance between eyes and ocelli; mesopleural punctures separated by less than one-third a puncture diameter, uniform in size; propodeal disc roughened and with scattered oblique rugulae; punctures of first tergite dense, coarser than those of second. Male with head as broad as long; lateral face marks ending abruptly a little above level of antennal sockets; supraclypeal mark present; lower part of face coarsely punctate; apical portion of tergites II-VI broadly impressed, margins slightly reflexed. Male. — Integument black, with pure white maculae as follows: clypeus; quadrate supraclypeal mark; lateral marks to a point ending about an antennal socket diameter above upper margin of antennal sockets; medially interrupted stripe on pronotal collar; pronotal lobe; anterior tegular spot; minute apical spot on all femora; complete stripe on anterior tibiae; basal and apical spots on mid and hind tibiae; all basitarsi, except small ferruginous apical spot; tibial spurs. Wings hyaline, veins and stigma brown. Scape entirely dark, flagellum light ferruginous beneath. Erect pubescence whitish. Head’. UFD 0.71-0.79 x FL; LFD 0.58-0.61 x UFD; OCD 1.05-1.17 x TFD. Distance from laterobasal clypeal angle to eye margin 1.3 times basal clypeal width; distance between antennal sockets slightly less than a socket diameter, distance from sockets to eye margins about equal to a socket dia- meter. Facial integument somewhat shining, maculate areas moderately coarsely punctate, punctures separated by one-third to three-fourths a punc- ture diameter on sides of face, the punctures somewhat elongate, linearly arranged; clypeal punctures more rounded, separated by about one-half a puncture diameter; frontal area slightly swollen, coarsely, closely punctate, the punctures for the most part clearly defined; vertex coarsely punctate laterally, the punctures a little larger than those of frons; ocellar triangle and occiput in middle dull, densely tesselate, contiguously punctate. Scape distinctly flattened, greatest width about one-half its length, median flagellar segments about as long as broad. Thorax’. Mesoscutal punctures a little smaller than those of clypeus, sep- arated by half a puncture diameter or less; mesoscutellar punctures a little coarser, separated by about half a puncture diameter; metanotum densely tesselate and roughened, with a few obscure punctures; mesopleural punctures about equal to those of mesoscutellum, separated by about one-fourth a punc- ture diameter. Basal area of propodeum coarsely areolate, the areolae finer laterally; transverse and oblique carinae obsolescent; lateral area roughened, with some indistinct punctures, a few fine rugae below at the sides. Abdomen’. Tergite I shining, coarsely punctate, punctures separated by about a puncture diameter, with scattered micropunctures; tergite II more finely punctate, duller, punctures separated by one-half to three-fourths a 48 Contributions in Science No. 180 puncture diameter; apical one-third depressed, very finely and sparsely punc- tate, apical margin slightly reflexed; ventrites VIII and IX as illustrated (figures lOH and 12A) . Length, to apex of second tergite, 3. 8-4. 3 mm.; of forewings, 3.15-3.50 mm. Female. — Integument black, except yellowish-white maculae as follows: lateral face marks, ending slightly above level of antennal sockets; inter- rupted stripe on pronotal collar; pronotal lobe; spot anteriorly on tegulae; minute apical spot on femora; basal spot on all tibiae; tibial spurs. Underside of flagellum light ferruginous. Erect pubescence whitish. Wings hyaline, veins and stigma brown. A very minute ferruginous to yellowish preapical spot sometimes present on clypeus. Head: UFD 0.71-0.78 x FL; LFD 0.62-0.69 x UFD; OCD 0.81-0.92 x TFD. Basal margin of clypeus about 1.8 times as wide as distance from laterobasal angle to eye margin; distance between antennal sockets about 1.4 times a socket diameter; distance between sockets and eye margin slightly exceeding a socket diameter. Clypeus reticulate-lineolate, with fine punctures separated by a puncture diameter or more; supraclypeal area lineolate, with punctures separated by about a puncture diameter; maculate portions of face finely punctate, punctures somewhat elongate, linearly arranged; frontal punc- tures coarse, contiguous, becoming finer toward midline; occiput and ocellar triangle densely tesselate between coarse, close punctures; genae striatopunc- tate. Median flagellar segments broader than long. Thorax: Mesoscutum dull, finely tesselate with punctures subequal to those of clypeus, separated by one-third a puncture diameter or less; meso- scutellar punctures about equal to those of mesoscutum, separated by about one-half a puncture diameter; metanotum somewhat granulose, punctures subequal to those of mesoscutum, separated by a puncture diameter or more; mesopleural punctures finer than those of clypeus, separated by about one- half a puncture diameter; propodeal basal area coarsely areolate, the areolae finer laterad; transverse carina present by poorly indicated, oblique carina barely evident; lateral areas strongly roughened; disc obliquely striolate. Abdomen: Tergite I smooth, shining, finely punctate, punctures sep- arated by a puncture diameter or more; tergite II smooth, shining, very finely punctate, punctures separated by one and one-half to two times a puncture diameter. Length, to apex of second tergite, 3.95-4.35 mm., of forewings, 3.70- 3.95 mm. Holotype male, allotype female (Los Angeles County Museum of Natural History) ; Craters of the Moon National Monument, Idaho, Aug. 21, 1964 (D. S. Horning, Jr.) , on Chrysothamnus naiiseosus. Paratypes, all from the type locality, collected by D. S. Horning, Jr., dates ranging from July 3 to Sept. 5: 23 $ $ , 12 $ $ . Host associations: Chryso- 1970 Studies on North American Bees 49 Figure 9. Hylaeus (Paraprosopis) spp., faces: A, H. peorgiciis, female; B, same, male; C, H, asininiis, female; D, same, male; E, H. megalotis, female; F, same, male. Figures by Ruth A. DeNicola. 50 Contributions in Science No. 180 thamnus viscidiflorus, C. nauseosus, Chamaebatiaria millefolium, Senecio serra, Eriogonum ovalifolium, Phacelia leucophylla and Populus tremuloides; two males taken on Aug. 17, 1965 are marked “prey of Ph-2.” Paratypes in United States National Museum, University of Idaho and the Los Angeles County Museum of Natural History. The specific name for this species is derived from the type locality. Hylaeus (Paraprosopis) timberlakei Snelling, NEW SPECIES Figures 7J; 8J; lOJ; 12B. This species bears a strong resemblance to H. calvus in most features. The male differs from that of H. calvus externally in having the punctures on the dorsal surface of the first tergite sharply defined, the interspaces smooth and shining, while in H. calvus the punctures are obscured due to the strongly tesselated and roughened interspaces. The female of H. timberlakei has the declivity of the propodeum with coarse, oblique rugulae (absent or poorly defined in H. calvus) and the punctures of the first tergite are moderately dense and distinctly larger than those of the second tergite (sparse, finer than those of second tergite in H. calvus). Diagnosis'. Female with head broader than long; clypeus maculate; facial foveae ending slightly less than halfway between eyes and ocelli; punctures of mesopleura separated by more than a puncture diameter, uniform in size; rugulae of propodeal declivity coarse, obliquely directed toward mid-line; punctures of first tergite dense, distinctly larger than those of second tergite. Male with head slightly broader than long; lateral face marks ending rather broadly on eye margin about midway between level of antennae sockets and median ocellus; supraclypeal mark present; punctures on lower part of face coarse; apical margin of second tergite broadly impressed, very finely and sparsely punctate. Male. — Integument black, with creamy-white maculae as follows: cly- peus; triangular supraclypeal mark; lateral face marks, ending broadly on inner eye margins, about midway between level of antennal sockets and middle ocellus; pronotal collar, interrupted medially; spot on pronotal lobe; anterior spot on tegulae; small apical spot on all femora; complete outer stripe on anterior tibiae; basal and apical spots on middle and hind tibiae; all basitarsi (usually with distinct apical infuscation) . Anterior medio- and distitarsi ferruginous, middle and hind medio- and distitarsi fuscous to black. Tibial spurs yellowish. Wings, hyaline, veins and stigma brownish. Scape all black; pedicel with reddish-yellow apical spot beneath; flagellum yellowish beneath, reddish-brown above. Erect pubescence whitish. Head'. UFD 0.70-0.77 x FL; LED 0.50-0.61 x UFD; OCD 1.00-1.11 x TED. Distance from latero-basal clypeal angle to eye margin 1.3 times basal clypeal width; distance between antennal sockets slightly more than a socket diameter, distance from sockets to eyes slightly more than a socket diameter. 1970 Studies on North American Bees 51 Facial integument dull, maculate areas coarsely punctate, punctures sep- arated by a puncture diameter or less; frontal area slightly swollen, coarsely rugosopunctate; vertex and genae coarsely continguously punctate. Scape somewhat distinctly flattened, greatest width about one-half its length; median flagellar segments about as long as broad, or slightly longer. Thorax: Mesoscutal punctures finer than those of vertex, separated by half a puncture diameter or less, interspaces tesselate and dull; punctures of mesoscutellum coarser than those of mesoscutum, about equal to those of vertex, separated by about half a puncture diameter on disc, apical one-fifth of the segment dull, densely rugosopunctate; metanotum dull, rugosopunctate; mesopleural punctures about equal to those of mesoscutellum, separated by one-fourth a puncture diameter or less. Basal area of propodeum areolate, the ridges sharply defined by very irregular; lateral area rugulose; disc rugulose laterally, with a few poorly defined oblique rugae medially; transverse carina present across basal area, oblique carina poorly defined. Outer margin of fore coxa subangulate basally to evenly convex. Abdomen: Tergite I shining, densely punctate, punctures a little smaller than those of mesoscutum, separated by one-half to three-fourths a puncture diameter; punctures of tergite II a little finer, separated by about one-half a puncture diameter or less; apical one-third to one-fourth of tergite II de- pressed, brownish, transversely lineolate, with scattered very fine punctures; ventrites VIII and IX as illustrated (figures lOJ and 12B). Length, to apex of second tergite, 3.8 to 4.3 mm.; of forewings, 2.6 to 3.0 mm. Female. — Integument black, except for yellow maculae as follows: median clypeal macula, varying from small preapical spot to broad vertical stripe, broad lateral face marks, ending slightly above antennal sockets; medially interrupted stripe on pronotal collar; large spot on pronotal lobe; triangular spot anteriorly on tegulae; basal spot on all tibiae. Underside of flagellum light reddish. Erect pubescence whitish. Wings hyaline, veins and stigma brownish. Head: UFD 0.77-0.84 x FL; LFD 0.61-0.66 x UFD; OCD 0.81-0.91 x TFD. Basal margin of clypeus 1.5 times as wide as distance from laterobasal angle to inner margin; distance between antennal sockets 1.5 times a socket diameter; distance from socket to eye margin 1.3 times a socket diameter. Clypeus finely reticulate, with small punctures over most of the disc separated by about a puncture diameter, becoming denser, coarser, more elongate toward apical margin; supraclypeal area lineolate, somewhat roughened above, with scattered obscure punctures; maculate portions of sides of face lineolate, with fine punctures separated by one-half to a full puncture diameter; frons coarse- ly, contiguously punctate, becoming finely, but very densely punctate on vertex, especially in ocellar triangle and behind; genal punctures obscure, a little larger than those of clypeus. Flagellar segments broader than long. 52 Contributions in Science No. 180 Thorax: Mesoscutum dull, finely tesselate with punctures finer than those of frons, separated by about one-half a puncture diameter on disc; mesoscutel- lum a little shinier, punctures variable, but mostly about same size as those of mesoscutum, some a little larger, mostly separated by about a puncture dia- meter; metanotum dull, rugulose; mesopleura bipunctate, i.e., with punctures of two sizes, the larger ones equal to those of mesoscutum, punctures variably spaced, but most separated by about a puncture diameter. Basal zone of propodeum coarsely areolate; transverse, oblique and lateral carinae distinct; disc finely rugulose above, the rugulae becoming coarser, oblique, below. Outer margin of fore coxa evenly convex, without basal angulation or spine. Abdomen: Tergite I shining, smooth, finely punctate, the punctures sep- arated by a puncture diameter or more; tergite II shining, very finely reticulate, punctures finer than those of tergite I, separated by from one to one and one- half times a puncture diameter; remaining tergites increasingly more distinctly reticulate, with scattered micropunctures. Length, to apex of second tergite, 4.5-5. 1 mm.; of forewings, 3.5- 4.0 mm. Holotype: Male (Los Angeles County Museum of Natural History) ; Forest Hill, 12 mi. E., Placer Co., Calif., June 24, 1962 (R. R. Snelling), on Ceanothus. Allotype : Female (University of California, Riverside) ; Tyler, 4.5 mi. W., Nevada Co., Calif., Sept. 27, 1956 (P. H. Timberlake), on Haplopappus. Paratypes (all California) : 8 $ $ , same data as allotype (UCR) ; 1 $, Goodyear’s Bar, Sierra Co., Calif., Aug. 11, 1963 (E. J. Montgomery; UCD) ; 1 $ , Buckhorn, Amador Co., July 25, 1955 (E. I. Schlinger; UCD) ; S $ $ , Boca, Nevada Co., June 25, 1961 (F. D. Parker; UCD) ; 1 $ , same locality, June 26, 1960 (F. D. Parker; UCD) ; 1 $ , same locality, June 26, 1960 (A. S. Menke; UCD) ; 1 $ , Pollock Pines, Eldorado Co., Aug. 9, 1957 (E. I. Schlinger; UCD) ; 1 ^ , Long Barn, 2.5 mi. W., Tuolumne Co., July 29, 1960 (G. I. Stage; GIS) ; 1 $ , Camp Bob MacBride, 5400', near Pinecrest, Tuo- lumne Co., Aug. 1, 1952 (R. R. Snelling; LACM), on Solidago californica; 10 ^ ^ , Shasta Co., May 2, 1913 (F. W. Nunenmacher; CIS, LACM) ; 1 $ , 1 $ , Eagle Creek, Trinity Co., Aug. 12, 1952 (H. L. Hansen; CIS) . This species is dedicated to Professor P. H. Timberlake, in recognition of his eminence in apoid systematics, and of the generous assistance he has given to me over many years. Professor Timberlake had recognized this species as new and with characteristic generosity permitted me to describe it. VARIATION : In the females, the clypeal maculation varies a great deal. Some have the spot so minute as to be barely perceptible and other have it so developed as to occupy as much as 75% of the total area. The Eagle Creek female has the rugulae of the propodeal declivity confined to the lower one- third. The males of this species are remarkably uniform structurally. They do possess a melanic form which strongly resembles that of such species as 1970 Studies on North American Bees 53 Figure 10. Hylaeiis (Paraprosopis) spp. A, female face, illustrating methods of obtaining measurements used in text: BCW = Basal Clypeal Width; CAD=:Clypeo- Antennal Distance; COD = Clypeo-Ocular Distance; EL = Eye Length; EL — Facial Length; LW — Facial Width; FEW = Frontal Eye Width; LEW = Lower Facial Width; SCQ = Supra-clypeal Quadrangle; UFW = Upper Facial Width. B-K, sternite VIII of males: B, H. wootoni; C, H. floridaniis; D, H. nevadensis; E, H. cookii; F, H. coloradensis; G, H. polifolii; H, H. limicrateriiis; I, H. personatellus; J, H. timber- lakei; K, H. sonoreusis. 54 Contributions in Science No. 180 Figure 11. Hylaeiis (Paraprosopis) spp., sternite VIII of males: A, H. megalotis; B, H. asininiis; C, H. calviis', D, H. georgiciis. E-L, sternite IX of males: E, H. wootoni; F, H. fioridaniis', G, H. polifolii', H, H. nevadensis; I, H. cookii; J, H. color- adensis; K, H. personatellus; L, H. sonorensis. 1970 Studies on North American Bees 55 Figure 12. Hylaeiis (Paraprosopis) spp., sternite IX of males: A, H. liinicraterius', B, H. timberlakei', C, H. megalotis; D, H. asiuinus', E, H. calviis; F, H. georgiciis. 56 Contributions in Science No. 180 H. calvus and H. polifolii, but may be separated from these by the characters in the key. Hylaeus (Paraprosopis) calvus (Metz) Figures 7K; 8K; IIC; 12E. Prosopis rudbeckiae, Metz, 1911. Trans. Amer. Ent. Soc., 37:111-114. 9 (in part). Prosopis calvus Metz, 1911. Op. cit. : 143-144. 5. This common species ranges from southern Oregon to northern Baja California. A disjunct population exists in north central Arizona. In California H. calvus occurs through much of the Upper Sonoran area, along the coast ranges and the eastern foothills of the Sierra Nevada. Although in flight from early April until the middle of July it is most abundant during May and June. The females are often taken in large numbers at flowers of various chaparral shrubs such as Rhamnus, Eriodictyon, Rhus and Ceanothus. The following are the marginal localities known to me for H. calvus: MEXICO. Baja California del Norte: 1 2 , Encinas, 4 mi. S., Sierra San Pedro Martir, VI-2-1958 (J. Powell; CIS). UNITED STATES. Oregon: 2 $ $, Griffin Creek, Jackson Co., VI-6-1957 (C. Fitch; UCD). Nevada: 1 $ , 1 $ , Verdi, Washoe Co., VI-25-1961 (F. D. Parker; UCD) . Arizona: 1 $ , Parker Creek, Sierra Ancha, V-4-1947 (H. and M. Townes) ; 4 I $ , Oak Creek Canyon, Coconino Co., V-16-1947 (H. and M. Townes) ; 1 $ , same locality and collectors, V-19-1947; 1 $ , 2 $ $ , same locality and collectors, V-20-1947; 1 9, same locality, VI-26-1950 (L. D. Beamer; all UK). This is one of the least variable of our species of Paraprosopis and one of the most easily recognized. In the female the combination of maculate clypeus, angularly produced laterobasal angle of the anterior coxae, shining, sparsely punctate mesopleurae and sparsely punctate abdomen are diagnostic. Typical males may be recognized by the lateral face marks truncate at the level of the antennal sockets, the finely, sparsely punctate shining mesopleurae and dull, densely tesselate abdomen. Males of this species, as well as those of H. neva- densis and H. polifolii, may have the clypeal borders broadly infuscated and the lateral face marks strongly narrowed above, ending acutely on the eye margin. Such males of this species may be recognized by the shining meso- pleurae and dull abdomen with the punctures obscured by the dense tes- selation. Hylaeus (Paraprosopis) georgicus (Cockerell) Figures 9A; 9B; 1 ID; 12F. Prosopis georgica Cockerell, 1898. Psyche, 7 (suppl.) : 438. $ . Prosopis georgica var. leeana Cockerell, 1909. Annals and Magazine of Natural History, (ser. 8) 4:27. 5. Hylaeus (Metziella) hydrangeae Mitchell, 1951. Jour. E. Mitchell Sci. 1970 Studies on North American Bees 57 Soc., 67:244. 9; Mitchell, 1960. N. Car. Agr. Exp. Sta. Tech. Bui. 141:60, 79; Snelling, 1968. L. A. Co. Mus. Nat. Hist., Contrib. Sci., 144:3, 4. NEW SYNONYMY. A careful reappraisal of the characters of H. {Metziella) hydrangeae has convinced me that this is actually the previously unrecognized female of H. georgicus. Both sexes were associated by me on the basis of a series of speci- mens from Texas before I realized that the females were conspecific with H. hydrangeae. In the Texas material two characteristics are outstanding: the greatly elongated basal zone of the propodeum and the extremely fine abdomi- nal punctation. The angulate laterobasal angle of the fore coxa is much more pronounced in the female than in the male, but this is true also of these sexes of H. calvus. The propodeal structure of the female, while aberrant for a Paraprosopis is much more similar to that of other Paraprosopis than it is to that of the subgenus Metziella. In my paper on that subgenus (Snelling; 1968) I included H. hydrangeae because at that time there seemed no good reason not to do so. This is now obviously incorrect inasmuch as the male of this species is, aside from propodeal structure, a perfectly typical Paraprosopis. The distribution of this species is limited. From New York it extends south to Florida and west to south central Texas. Mitchell (1960) recorded males from Ithaca, New York and Raleigh and Wake Co., North Carolina. The female was recorded (as H. hydrangeae) from Cruso, North Carolina. Snelling (1968) recorded a female (as H. hydrangeae) from Ft. Gordon, Georgia. Cockerell described H. georgicus leeana from males taken by Birk- mann in Fee County, Texas. Additional specimens, including both sexes, have been examined which were collected in Fee County by Birkmann. In addition to the above I have examined the following: 1 $ , Augusta, Richmond Co., Georgia, 1 l-IV-1959 (R. R. Snelling; FACM) , on Crataegus’, 1 9, Gadsden Co., Florida, 1 l-IV-1960 (H. V. Weems; FPB) , on Crataegus’, 1 9 , Hwy. 269, 8 mi. S. Chattahoochee, Gadsden Co., 6-IV-1955 (D. C. L. Gosling & party; MSU) ; 1 9 , Torreya State Park, Fiberty Co., Florida, 12-IV-1960 (H. A. Denmark; FPB), on Crataegus’, 1 S, Lexington, Lee Co., Texas, 24- III- 1951 (C. D. Michener; KU), on Salix’, 2 $ s. Tilling, Caldwell Co., Texas, 30-III-1951 (L. D. Beamer; KU), on Salix. Hylaeus {Paraprosopis) megalotis (Swenk and Cockerell) Figures 9E; F; 11 A; 12C. Prosopis megalotis Swenk and Cockerell, 1910. Ent. News, 21:69-70. $ 9. This very distinctive species was originally described from a series of specimens from Glen, Sioux County, NEBRASKA and a single female from Fort Garland, COLORADO. The types were taken on Cleome serrulata. In addition to the types I have seen specimens of both sexes taken at the type locality on Aug. 28, 1959 (W. E. LaBerge and O. W. Isakson; UN) on 58 Contributions in Science No. 180 Solidago. The following are new records for this species. MONTANA: 13 $ 0, Whitehall, Aug. 13, 1931 (J. Nottingham; KU). COLORADO: 1 2, Blue Mtn., Motfatt Co., Aug. 14, 1966 (G. F. Knowlton; USU). UTAH: 1 $ , Castle Dale, Emery Co., Aug. 10, 1957 (C. D. Michener; KU) , on Cleome serrulata; 2 $ $ , 1 2 9 , Clear Creek Canyon, Box Elder Co., June 11, 1954 (G. F. Knowlton; KU); 1 5, Hanksville, Wayne Co., Sept. 11, 1938 (G. F. Knowlton, F. C. Harmston; USU) ; 1 2, Eureka, Juab Co., Aug. 6, 1958 (G. E. Bohart; USU), on Chrysothamnus; 8 $ $ , Manila, Daggett Co., Aug. 11, 1962 (G. E. Bohart; USU) . ARIZONA: 1 2 , Painted Desert, Apache Co., Aug. 18, 1958 (G. I. Stage; LACM), on Eriogonum aureus. The female of this species is easily recognized by its complete pronotal macula and the elongate apical mandibular tooth. This tooth is subject to wear, however, and in older individuals may not be noticeably elongate. The male is very similar to that of H. asininus but may be separated by the finely and sparsely punctate first tergite, the head about as broad as long (broader than long in H. asininus) and by the gradually, rather than abruptly, depressed apico-lateral portion of the second tergite. Species of Uncertain Position Hylaeus (Paraprosopis) suffusus (Cockerell) Prosapis suffusus Cockerell, 1896. Psyche, 7 (supl.) : 32 $ . This name is based on a single male from an unknown locality in Nevada. Although the type should be in the collections of the Academy of Natural Sciences, Philadelphia, it is apparently lost or destroyed. If the type cannot be found, it seems best to drop the name altogether since the species cannot be determined from the original description alone. The present assignment to the subgenus Paraprosopis is also tentative. Hylaeus {Paraprosopis) tuertonis (Cockerell) Prosopis tuertonis Cockerell, 1906. Bui. Amer. Mus. Nat. Hist., 22: 423-424. 2. This species was described from two females, one from Tuerto Moun- tain, near Santa Fe, New Mexico, and one from Florissant, Colorado. I have not seen these specimens, and the original description offers little information of use in identifying the species. It is probably correctly assigned to Para- prosoips and seems nearest to H. wootoni, of which it may be a synonym. The description applies equally well, however, to H. cookii. Until the types can be studied, the status of this name must remain uncertain. Acknowledgments The material used during the preparation of this paper has been made available to me through the kindness of many individuals who have con- tributed specimens from the collections in their care. I wish to thank each of 1970 Studies on North American Bees 59 the following, not only for the original loan of specimens, but for their patience and cooperation during the course of this study: G. E. Bohart, Utah State University (USU); G. D. Butler and F. G. Werner, University of Arizona (UA) ; R. L. Fischer, Michigan State University (MSU) ; A. R. Gittins, Uni- versity of Idaho (UI) ; P. D. Hurd, California Insect Survey, University of California, Berkeley (CIS) ; K. V. Krombein and G. I. Stage, United States National Museum (USNM) ; U. N. Lanham, University of Colorado (UC) ; H. B. Leech and G. I. Stage, California Academy of Sciences (CAS) ; A. T. McClay and R. O. Schuster, University of California, Davis (UCD) ; C. D. Michener, University of Kansas (KU) ; H. E. Milliron, Canadian National Collections (CNC) ; J. G. Rozen, American Museum of Natural History (AMNH) ; G. I. Stage, personal collection (GIS) ; P. H. Timberlake, Univer- sity of California, Riverside (UCR) ; M. S. Wasbauer, California Department of Agriculture (CDA) ; A. Willink, Institute Miguel Lillo, Tucuman, Argen- tina (IMF) . The facial figures of the various species were prepared by Ruth A. DeNicola. Literature Cited Beardsley, J. W. 1959. Proc. Haw. Ent. Soc., 17:23 (note presented at meeting, Oct. 13, 1958). Bridwell, j. C. 1919. Miscellaneous notes on Hymenoptera, with descriptions of new genera and species. 22. North American Hylaeiis. Hawaiian Entomol. Soc., Proc., 4:156-162. Cockerell, T. D. A. 1937. Bees of Alberta. V. Canad. Entomol., 69:126-127. Cockerell, T. D. A. and R. Sumner. 1931. Rocky Mountain bees. III. The genus Prosopis. Amer. Mus. Nov., 490: 1-15. Crawford, J. C. 1913. Notes on some species of the genus Prosopis. Canad. Entomol., 45; 154-156. Linsley, E. C. 1951. In Muesebeck, et al., Hymenoptera of America North of Mexico, Synoptic Catalog. Monog. 2, U. S. Dept. Agric., 1-1420. Metz, C. W. 1911. A revision of the genus Prosopis in North America. Amer. Entomol. Soc., Trans., 37: 85-156. Mitchell, T. B. 1960. Bees of the Eastern United States. N. Car. Agric. Exp. Sta., Tech. Bui. 141, 538 pp. Snelling, R. R. 1966a. Studies on North American bees of the genus Hylaeiis. 1. Distribution of the western species of the subgenus Prosopis with descrip- tions of new forms. Los Angeles Co. Mus., Contrib. Sci., 98: 1-18. 1966b. Studies on North American bees of the genus Hylaeiis. 2. Description of a new subgenus and species. Biol. Soc. Wash., Proc., 79: 139-143. 1966c. Studies on North American bees of the genus Hylaeiis. 3. The Nearctic subgenera. So. Calif. Acad. Sci., Bull., 65: 164-175. 1968. Studies on North American bees of the genus Hylaeiis. 4. The subgenera Cephalylaeiis, Metziella and Hylaeana. Los Angeles Co. Mus. Nat Hist., Con- trib. Sci., 144: 1-6. SwENK, M. H. AND T. D. Cockerell. 1910. New bees of the genus Prosopis. Entomol. News, 21 : 67-71. Accepted for publication December 15, 1969. 4‘. « P'» ■ ■■ W : ■» 'l ' 'll Bun^ i[' 'l; K-* "^1 > ?> ^ .r:i!i, ' ? (. ;'T I LOS * ANGELES COUNTY MUSEUM CONTRIBUTIONS iTzg. IN SCIENCE juMBER 181 February 27, 1970 TYPE SPECIMENS OF FOSSIL INVERTEBRATA IN THE LOS ANGELES COUNTY MUSEUM OF NATURAL HISTORY, EXCLUSIVE OF PALEOENTOMOLOGY By Edward C. Wilson and Donald E. Bing Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor TYPE SPECIMENS OF FOSSIL INVERTEBRATA IN THE LOS ANGELES COUNTY MUSEUM OF NATURAL HISTORY, EXCLUSIVE OF PALEOENTOMOLOGY By Edward C. Wilson' and Donald E. Bing- Abstract: Type specimens of fossil Porifera, Coelemerata, Brachiopoda, Bryozoa, Molliisca, Arthropoda, and Echinoder- mata in the Los Angeles County Museum of Natural History are listed by the original published names. Paleoentomological types are not included. The original reference, the museum type number, the age, the formation, and the locality are given for each specimen. Figured specimens not of the holotype or paratype groups are included. Introduction This catalogue is published in compliance with Recommendation 72D of the International Commission on Zoologieal Nomenclature (1964, p. 75). It includes all the type and figured specimens, as defined below, of fossil invertebrates in the Los Angeles County Museum of Natural History, except for the paleoentomological types of W. D. Pierce and Georg Statz. A list of these will be published separately. The catalogue is arranged systematically by phyla and classes. Within these, the genus and species group names, as originally published, are listed alphabetically. Each entry also contains the author and bibliographic reference for the lowest species-group name, the type category and number, the age assignment, the formation (if known), and the locality. We assign a separate number to every type specimen, unless it is not feasible to distinguish indi- viduals, as in a rock containing many specimens, in which case the number refers to a lot. Figure references are included only if our specimen was figured. A question mark after the figure citation indicates some doubt that the figure was made from the specimen in our collections. “Missing” means that the specimen cannot now be found. Definitions of type categories by Schenk, McMasters, Keen, and Muller (1956: 5-9) and the International Commission on Zoological Nomenclature (1964) are followed. In this list, hypotype refers to a specimen figured later than the original description of the species to which it is referred, or figured without a species designation. Casts of type specimens are classified like the originals from which they were made, with the prefix “plasto.” Figured refer- ences for these refer to the original specimens. ^Curator of Invertebrate Paleontology, Los Angeles County Museum of Natural History. -Student Professional Worker, Los Angeles County Museum of Natural History. 1 SMITHSONIAN INSTITUTION MR 0 4 1970 Contributions in Science No. 181 Catalogue PORIFERA: HYALOSPONGEA Chancelloria eros Walcott, 1920 MASON, 1935, p. 100, pi. 15, fig. 1. Hypotype 1290; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. COELENTERATA: ANTHOZOA Astrangia cf. A. insignifica Nomland, 1916 HERTLEIN and GRANT, 1960, p. 79, pi. 19, figs. 1-4, 7. Hypotypes 1731 (figs. 1, 7), 1732 (figs. 2-4); Pliocene, San Diego For- mation; hills south of Tia Juana River, San Diego County, California. Balanophyllia elegans Verrill, 1864 HERTLEIN and GRANT, 1960, p. 84, pi. 19, figs. 14, 19-21; pi. 24, figs. 4, 5,8, 13, 15-17. Hypotypes 1738 (fig. 14), 1739 (fig. 19), 1740 (fig. 20), 1741 (fig. 21), 1742 (fig. 4), 1743 (fig. 5), 1744 (fig. 8), 1745 (fig. 13), 1746 (fig. 15), 1747 (fig. 16), 1748 (fig. 17) ; Pliocene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. Bergaueria sp. ARAI and McGUGAN, 1969, p. 93, pi. 18. Hypotype 1232 (fig. 1), 1233 (figs. 2, 7), 1234 (figs. 3, 6), 1235 (fig. 4), 1236 (fig. 5) ; Cambrian; near Goldfield, Esmeralda County, Nevada. Dendrophyllia cf. D. oldroydi Faustino, in Oldroyd, 1925 HERTLEIN and GRANT, 1960, p. 82, pi. 19, figs. 5, 6, 15. Hypotype 1737; Pliocene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. Paracyathus stearnsii Verrill, 1869 HERTLEIN and GRANT, 1960, p. 80, pi. 19, figs. 8-13. Hypotypes 1733 (figs. 8, 9), 1734 (figs. 10, 11), 1735 (fig. 12), 1736 (fig. 13) ; Pliocene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. BRACHIOPODA: INARTICULATA Glottidia albida (Hinds, 1844) HERTLEIN and GRANT, 1960, p. 89, pi. 19, figs. 18, 24, 25. Hypotypes 1749 (fig. 18), 1750 (fig. 24), 1751 (fig. 25) ; Pliocene, San Diego Formation; hills south of Tia Juana River (1749) ; quarry at end of Arroyo Drive, San Diego (1750, 1751 ), all San Diego County, California. 1970 Type Specimens of Fossil Invertebrata 3 BRACHIOPODA: ARTICULATA Laqueus californianus (Koch), in Kuster, 1844 HERTLEIN and GRANT, 1960, p. 96, pL 20, figs. 1-3, 5-7. Hypotypes 1753 (figs. 1-3), 1754 (figs. 5-7); Pliocene, San Diego For- mation; quarry at end of Arroyo Drive, San Diego, San Diego County, California. Laqueus vancouveriensis cliegensis Hertlein and Grant HERTLEIN and GRANT, 1960, p. 97, pi. 20, figs. 4, 8, 12. Paratypes 1755 (figs. 4, 8, 12), 1756; Pliocene, San Diego Eormation; end of Coring Street (1755), 0.2 miles north of Harbor Boulevard and Tourmaline Street (1756), both San Diego, San Diego County, California. Or this decipiens Phleger PHLEGER, 1933, p. 17, pi. 1, fig. 2. Holotype 1258; Ordovician, Barrel Spring Formation; Barrel Spring Canyon, Inyo County, California. Or this minusculus Phleger PHLEGER, 1933, p. 7, pi. 2, figs. 6, 7. Syntypes 1 239 (fig. 6?) , 1240 (fig. 7) ; Ordovician, Mazourka Eorma- tion; Mazourka Canyon, Inyo County, California. Plectambonites angulatus Phleger PHLEGER, 1933, p. 18, pi. 1, fig. 1. Syntypes 1259 (fig. 1), 1260; Ordovician Barrel Spring Formation; Barrel Spring Canyon, Inyo County, California. Plectorthis mazourkaensis Phleger PHLEGER, 1933, p. 9, pi. 2, figs. 3-5. Syntypes 1244 (fig. 3), 1245 (fig. 5) , 1246, 1247 (fig. 4) , 1 248 ; Ordovi- cian, Mazourka Eormation; Mazourka Canyon, Inyo County, California. Plectorthis patulus Phleger PHLEGER, 1933, p. 10, pi. 2, figs. 1, 2. Syntypes 1241 (fig. 1), 1242 (fig. 2) , 1 243 ; Ordovician, Mazourka Eor- mation; Mazourka Canyon, Inyo County, California. Terehratalia hemphilli Dali, 1902 HERTLEIN and GRANT, 1960, p. 93, pi. 19, fig. 26. Hypotype 1752; Pliocene, San Diego Formation; end of Coring Street, San Diego, San Diego County, California. BRYOZOA: GYMNOLAEMATA Ceriocava eastoni Woollacott WOOLLACOTT, 1966, p. 226, figs, la-lc. Syntypes 1148; Cretaceous, Ladd Formation; Williams Canyon Road, 0.5 mile east of Santiago Canyon Road, Santa Ana Mountains, Orange County, California. 4 Contributions in Science No. 181 MOLLUSCA: PELECYPODA Anatina (?) sp. PACKARD, 1922, p. 423, pi. 31, fig. 5. Plastohypotype 2425; Cretaceous; west side of Baker Creek, Orange County, California. B.[arbatia] Morsel Gabb GABB, 1864, p. 216, pi. 32, fig. 286. Plastoholotype 2421; Cretaceous [Eocene]; San Diego, San Diego County, California. Bornia cooki Willett WILLETT, 1937b, p. 389, pi. 25, figs. 2, 3. Holotype 1047; Pleistocene; Baldwin Hills, Los Angeles, Los Angeles County, California. Cardita hilli Willett WILLETT, 1944, p. 19, pi. 7. Holotype 1067; Pleistocene, Palos Verdes Sand; Newport Bay mesa. Orange County, California. Cardium {Protocardium) marysvillensis Dickerson, 1916 CLARK and WOODFORD, 1927, p. 94, pi. 15, fig. 12. Plastoparatype 2433; Eocene, Meganos Formation; north side of Mt. Diablo, Contra Costa County, California. Cardium sorrentoensis Hanna HANNA, 1927, p. 285, pi. 41, fig. 10. Plastoholotype 2430; Eocene, La Jolla Formation; east side of Rose Canyon, 2 miles east of La Jolla, San Diego County, California. Chione picta Willett WILLETT, 1944, p. 21, pi. 8, figs. B. Paratypes 1068 (pi. 8, figs. B, destroyed, see SPHON, 1962, p. 57) , 1954-1975, 2313-2414; Pleistocene, Palos Verdes Sand; Newport Bay mesa. Orange County, California. Exputens Uajasensis Clark CLARK, 1934, p. 270, pi. 37, fig. 14. Plastoparatype 2432; Eocene; branch of Llajas Canyon, Ventura County, California. Leda rosa Hanna HANNA, 1927, p. 271, pi. 25, fig. 9. Plastoholotype 2428; Eocene, La Jolla Formation; Rose Creek, San Diego County, California. Lucina packi Dickerson DICKERSON, 1916, p. 484, pi. 36, fig. 12. Plastoholotype 2422; Eocene; Parson’s Peak, Fresno County, California. 1970 Type Specimens of Fossil Invertebrata 5 Liicina packi Dickerson, 1916 TURNER, 1938, p. 52, pi. 9, fig. 11. Plastohypotype 2423; Eocene; Simi Valley, Ventura County, California. Nucula (Acila) semirostrata Grant and Gale GRANT and GALE, 1931, p. 113. Paratype 1976; Pliocene; \ Va miles west of San Martinez Grande Can- yon, boundary of Los Angeles and Ventura Counties, California. Perna goniglensis Hanna HANNA, 1927, p. 275, pi. 27, fig. 13 (?). Plastoholotype 2429; Eocene, La Jolla Formation; sec. 12, La Jolla Quadrangle, San Diego County, California. Rochefortia reyana Willett WILLETT, 1937b, p. 388, pi. 25, figs. 1. Syntypes 1046 (fig. 1, left) , 1703 (fig. 1, right) , 1704-1728; Pleistocene; Baldwin Hills, Los Angeles, Los Angeles County, California. T.[ellina] ooides Gabb GABB, 1864, p. 157, pi. 22, fig. 135. Plastolectotype 2420; Cretaceous; west of Martinez, Contra Costa County, California. Tellina santana Packard PACKARD, 1922, p. 426, pi. 33, fig. 4. Plastoholotype 2424; Cretaceous; crest of Santiago-Aliso divide. Orange County, California. MOLLUSC A: GASTROPODA Acmaea mitchelli Lipps LIPPS, 1963, p. 3, figs. 2a-d, 3a-z, 3aa-kk, 4a-z, 4aa-kk. Holotype 1126 (figs. 2a-d, 3a, 4a) , paratypes 1 127 (figs. 3b, 4b) , 1781 (figs. 3c, 4c), 1782 (figs. 3d, 4d), 1783 (figs. 3e, 4e), 1784 (figs. 3f, 4f), 1785 (figs. 3g, 4g), 1786 (figs. 3h, 4h), 1787 (figs. 3i, 4i), 1788 (figs. 3j, 4j), 1789 (figs. 3k, 4k), 1790 (figs. 31, 41), 1791 (figs. 3m, 4m), 1792 (figs. 3n, 4n), 1793 (figs. 3o, 4o), 1794 (figs. 3p, 4p), 1795 (figs. 3q, 4q), 1796 (figs. 3r, 4r), 1797 (figs. 3s, 4s), 1798 (figs. 3t, 4t), 1799 (figs. 3u, 4u), 1800 (figs. 3v, 4v), 1801 (figs. 3w, 4w), 1802 (figs. 3x, 4x), 1803 (figs. 3y, 4y), 1804 (figs. 3z, 4z), 1805 (figs. 3aa, 4aa), 1806 (figs. 3bb, 4bb), 1807 (figs. 3cc, 4cc), 1808 (figs. 3dd, 4dd), 1809 (figs. 3ee, 4ee), 1810 (figs. 3ff, 4ff), 1811 (figs. 3gg, 4gg), 1812 (figs. 3hh, 4hh), 1813 (figs. 3ii, 4ii), 1814 (figs. 3jj, 4jj), 1815 (figs. 3kk, 4kk), 1816-1861 ; Pleistocene; San Nicolas Island, Ventura County, California. A cteocina anomala Willett WILLETT, 1945, p. 29. Holotype 1082, paratypes 1083, 2179-2181 ; Pleistocene (?) ; near shore of Salton Sea, Imperial County, California. 6 Contributions in Science No. 181 Alabina effiae V^xWeii WILLETT, 1939, p. 202, pi. 54. Holotype 1061 (pi. 54), paratypes 1729, 1730; Pleistocene, San Pedro Formation; Hilltop Quarry, San Pedro (holotype); Pleistocene, Timms Point Formation; Timms Point (paratypes), both Los Angeles County, California. Architectonica llajasensis Sutherland SUTHERLAND, 1966, p. 1, figs. 1, 2. Holotype 1140; Eocene, Llajas Formation; Las Llajas Canyon, Santa Susana Mountains, Ventura County, California. Balds {Balds) tersa Berry BERRY, 1954, p. 261. Paratype 2182; Pleistocene, Lomita Formation; Hilltop Quarry, San Pedro, Los Angeles County, California. Balds (Vitreolina) ebriconus Berry BERRY, 1954, p. 265. Paratype 2188; Pleistocene, Lomita Formation; Hilltop Quarry, San Pedro, Los Angeles County, California. Balds {Vitreolina) incallida Berry BERRY, 1954, p. 264. Paratypes 2186, 2187; Pleistocene, Lomita Formation; Hilltop Quarry, San Pedro, Los Angeles County, California. Balds {Vitreolina) obstipa Berry BERRY, 1954, p. 262. Paratypes 2183-2185; Pleistocene, Lomita Formation; Hilltop Quarry, San Pedro, Los Angeles County, California. Boetica hertleini Kanakoff KANAKOFF, 1966, p. 3, fig. 1. Holotype 1 145 (fig. 1), paratypes 1 146, 2176-2178; Pliocene, Pico For- mation; mile south of Humphreys railroad station, Los Angeles County, California. Calyptraea calabasasensis Nelson NELSON, 1925, p. 419, pi. 54, figs. 8a, 8b. Plastoholotype 2437; Eocene, Martinez Formation; Simi Hills, Ventura County, California. Caly[p]traea filosa (Gabb, 1866) CLARK, 1915, p. 560, pi. 65, figs. 23, 24. Plastohypotype 2438; Miocene, San Pablo Group; Mt. Diablo quad- rangle, Contra Costa County, California. Chedvillia steward Clark CLARK, 1942, p. 117, pi. 19, fig. 8. Plastoparatype 2439; Eocene, Llajas Formation; Ventura County, California. 1970 Type Specimens of Fossil Invertebrata 7 Diodora constantiae Kanakoff KANAKOFF, 1953, p. 67, pis. 12, 13. Holotype 1089 (pi. 12), paratypes 1094 (pi. 13, missing?), 2191-2312; Pleistocene, Palos Verdes Sand; Vermont Avenue, south of Sepulveda Boulevard, Wilmington, Los Angeles County, California. Exilia taliaferroi Vokes VOKES, 1939, p. 125, pi. 17, fig. 21. Plastoholotype 2435; Eocene, Arroyo Hondo Formation; north side of Domengine Creek, Fresno County, California. Fusinus {Exilia) lincolnensis (Weaver, 1916) CLARK, 1918, p. 178, pi. 23, fig. 10. Plastohypotype 2440; Oligocene, Kreyenhagen Shale; 15 miles north of Coalinga, Fresno County, California. Fusinus ucalius Vokes VOKES, 1939, p. 137, pi. 18, fig. 15. Plastoholotype 2436; Eocene, Domengine Formation; first large canyon south of Big Tar Canyon, Kings County, California. Homalopoma berryi McLean McLEAN, 1964, p. 132, pi. 24, figs. 7-8. Holotype 1138 (figs. 7-8), paratypes 1139, 1657-1702; Pleistocene, Timms Point Formation; Timms Point, Los Angeles County, California. Kelletia vladimiri Kanakoff KANAKOFF, 1954, p. 114, pis. 29-31. Holotype 1097 (pi. 29), paratypes 1098 (pi. 30), 1099 (pi. 31); Plio- cene, Pico Formation; Vi mile south of Humphreys railroad station, Los Angeles County, California. Leptothyra suhobsoleta Willett WILLETT, 1937a, p. 63, pi. 25. Holotype 1054; Pleistocene, Timms Point Silt; San Pedro, Los Angeles County, California. Macrarene diegensis McLean McLEAN, 1964, p. 131, pi. 24, figs. 9-14. Holotype 1136 (figs. 9, 10), paratypes 1137 (fig. 11), 1300 (figs. 12- 14) — 1656; Pliocene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. Marginella jewettii nanella Oldroyd OLDROYD, 1925, p. 24. Paratypes 2152-2154; Pleistocene, San Pedro Formation; Nob Hill, San Pedro, Los Angeles County, California. Mediargo mediocris (Dali, 1909) TERRY, 1968, pi. 4, fig. 2. Hypotype 1190; reworked Pliocene in Pleistocene Lomita Marl; Western Avenue cut, Palos Verdes Hills, Los Angeles County, California. 8 Contributions in Science No. 181 Megistostoma caminoensis Hanna HANNA, 1927, p. 330, pi. 57, figs. 9, 10. Plastoholotype 2431; Eocene, La Jolla Formation; 0.4 mile south of Sorrento Station, San Diego County, California. Micrarionta (Xerarionta) agnesae Kanakoff KANAKOFF, 1950, p. 85, pi. 27. Holotype 1088 (pi. 27), paratypes 1864-1953; Pleistocene (?); China Point, San Clemente Island, Los Angeles County, California. Mitromorpha galeana Berry BERRY, 1941, p. 12. Paratypes 2189, 2190; Pleistocene, Lomita Formation; Hilltop Quarry, San Pedro, Los Angeles County, California. Nassarius (Caesia) fossatus forma coiloterus (Woodring), in Woodring, Bramlette, and Kew, 1946 ADDICOTT, 1965, p. 8, pi. 2, figs. 30, 31. Hypotypes 1129 (fig. 30), 1130 (fig. 31); Pleistocene, Palos Verdes Sand; Playa del Rey, Los Angeles County, California. Nassarius (Catilon) hildegardae Kanakoff, 1956 ADDICOTT, 1965, p. 16, pi. 3, figs. 11, 12. Hypotypes 1131 (fig. 11), 1132 (fig. 12) ; Pliocene, Pico Formation (Towsley Formation) ; Vi mile south of Humphreys railroad station, Los Angeles County, California. Nassarius hildegardae Kanakoff KANAKOFF, 1956, p. 113, pi. 31. Holotype 1 110 (pi. 31 ), paratypes 2097-2151 ; Pliocene, Pico Forma- tion; Vi mile south of Humphreys railroad station, Los Angeles County, California. Nassarius stocki Kanakoff KANAKOFF, 1956, p. 110, pi. 30A-C. Holotype 1109 (pi. 30A, C), paratypes 1977 (pi. 30B)— 2096; Plio- cene, Pico Formation; Vi mile south of Humphreys railroad station, Los Angeles County, California. Neadmete schwartzi Mount MOUNT, 1970, p. 3, fig. 1. Holotype 1779 (fig. 1), paratype 1780; Pliocene, Fernando Formation; Newport Bay mesa. Orange County, California. Neadmete siitherlandi Kanakoff and McLean KANAKOFF and M3LEAN, 1966, p. 4, figs. 1, 2. Holotype 1150 (figs. 1, 2), paratypes 1151, 1862, 1863; Pleistocene, Lomita Marl; San Pedro, Los Angeles County, California. Odostomia (Amaura) menzola Oldroyd OLDROYD, 1925, p. 33. 1970 Type Specimens of Fossil Invertebrata 9 Paratype 2173; Pleistocene, San Pedro Formation; Nob Hill, San Pedro, Los Angeles County, California. Odostomia (Amaura) timessa Oldroyd OLDROYD, 1925, p. 35. Paratypes 2174, 2175; Pleistocene, San Pedro Formation; Nob Hill, San Pedro, Los Angeles County, California. Odostomia (Chrysallida) elsiae Willett WILLETT, 1948, p. 19, pi. 4, fig. 3. Holotype 1071 (fig. 3), paratype 2415; Pleistocene, Palos Verdes Sand; Newport Bay mesa. Orange County, California. Odostomia (Evalea) manca Oldroyd OLDROYD, 1925, p. 32. Paratypes 2170-2172; Pleistocene, San Pedro Formation; Nob Hill, San Pedro, Los Angeles County, California. Odostomia (Menestho) effiae Willett WILLETT, 1948, p. 19, pi. 4, fig. 4. Holotype 1070 (fig. 4), paratypes 2416-2419; Pleistocene; Palos Verdes Sand; Newport Bay mesa. Orange County, California. Surcula (Surculites) andersoni Dickerson DICKERSON, 1914, p. 149, pi. 16, fig. 11. Plastoholotype 2434; Eocene, Martinez Group; southeast of Muir Sta- tion, Contra Costa County, California. Triphora fossilis Willett WILLETT, 1937a, p. 62, pi. 24. Holotype 1053; Pleistocene, Timms Point Silt; San Pedro, Los Angeles County, California. Triphora kanakoffi Willett WILLETT, 1948, p. 20, pi. 4, fig. 2. Holotype 1072; Pleistocene, Palos Verdes Sand; Newport Bay mesa. Orange County, California. Turbonilla (Mormula) epiphanea Oldroyd OLDROYD, 1925, p. 28. Paratype 2169; Pleistocene, San Pedro Eormation; Nob Hill, San Pedro, Los Angeles County, California. Turbonilla (Pyrgolampros) collisella Oldroyd OLDROYD, 1925, p. 25. Paratypes 2157, 2158; Pleistocene, San Pedro Formation; Nob Hill, San Pedro, Los Angeles County, California. Turbonilla {Pyrgolampros) idae Oldroyd OLDROYD, 1925, p. 26. Paratypes 2159-2168; Pleistocene, San Pedro Formation; Nob Hill, San Pedro, Los Angeles County, California. 10 Contributions in Science No. 181 Turbonilla (Strioturbonilla) pecora Oldroyd OLDROYD, 1925, p. 24. Paratypes 2155, 2156; Pleistocene, San Pedro Formation; Nob Hill, San Pedro, Los Angeles County, California. Turbonilla {Turbonilla) grouardi WILLETT, 1948, p. 17, pi. 4, fig. 1. Holotype 1069; Pleistocene, Palos Verdes Sand; Newport Bay mesa. Orange County, California. Turritella andersoni subsp. lawsoni Dickerson, 1916 MERRIAM, 1941, p. 77, pi. 12, fig. 4. Plastohypotypes 2427; Eocene, Llajas Formation; Las Llajas Canyon, Simi Valley, Ventura County, California. Turritella schencki Merriam MERRIAM, 1941, p. 81, pi. 10, fig. 10. Plastoholotype 2425 (fig. 10) , plastoparatype 2426; Eocene, Tejon For- mation; Teeny a Creek, Kern County, California. ARTHROPOD A: TRILOBITA Acrocephalites ? trifossatus Mason MASON, 1935, p. 102, pi. 15, figs. 2, 3. Syntypes 1279 (fig. 2), 1280 (fig. 3) ; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. Alokistocare linnarssoni (Walcott, 1884) MASON, 1935, p. 104, pi. 15, fig. 4. Hypotype 1281 ; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. Amecephalus piochensis (Walcott, 1886) MASON, 1935, p. 105, pi. 15, figs. 5-7. Hypotypes 1282 (fig. 5), 1283 (fig. 6), 1284 (fig. 7?) ; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. Amecephalus strangulatus Mason MASON, 1935, p. 106, pi. 15, figs. 8, 9. Holotype 1285 (fig. 8), paratype 1286 (fig. 9) ; Cambrian, Cadiz For- mation; Marble Mountains, San Bernardino County, California. Anomocarella ? spatha Mason MASON, 1935, p. 107, pi. 15, fig. 10. Holotype 1287; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. A noria lodensis (Clark, 1921 ) MASON, 1935, p. 109, pi. 15, fig. 11. Hypotype 1288 (fig. 11), 2442 (fig. 12, missing) ; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. 1970 Type Specimens of Fossil Invertebrata 11 Bathyiiriscus howelli Walcott, 1886 MASON, 1935, p. 110, pi. 15, fig. 13. Flypotype 1289; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. Bathyiiriscus maximiis Mason MASON, 1935, p. 1 1 1, pi. 15, figs. 14-16. Syntypes 1291 (fig. 14), 1292 (fig. 15), 1293 (fig. 16) ; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. Biimastus billingsi Raymond and Narraway RAYMOND and NARRAWAY, 1908, p. 250, pi. 62, figs. 1, 2. Plastoholotype 2441; Ordovician, Trenton Limestone; Hull, Quebec, Canada. Ceraurus infrequens Phleger PHLEGER, 1933, p. 15, pi. 2, fig. 12. Holotype 1255; Ordovician, Mazourka Eormation; Mazourka Canyon, Inyo County, California. Corynexochiis ? sp. MASON, 1935, p. 113, pi. 15, fig. 17. Hypotype 1294; Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. Cybeloides calliteles Phleger PHLEGER, 1933, p. 14, pi. 2, fig. 8. Holotype 1254; Ordovician, Mazourka Eormation; Mazourka Canyon, Inyo County, California. Dorypyge quadriceps (Hall and Whitfield, 1877) MASON, 1935, p. 1 16, pi. 15, figs. 18-22. Hypotypes 1295 (fig. 18), 1296 (fig. 19), 1297 (fig. 20), 1298 (fig. 21), 1299 (fig. 22); Cambrian, Cadiz Formation; Marble Mountains, San Bernardino County, California. Encrinurus hastula Phleger PHLEGER, 1933, p. 12, pi. 2, figs. 13, 14. Syntypes 1251 (fig. 13), 1252 (fig. 14); Ordovician, Mazourka Eorma- tion; Mazourka Canyon, Inyo County, California. Encrinurus octonarius Phleger PHLEGER, 1933, p. 13, pi. 2, fig. 9. Holotype 1253; Ordovician, Mazourka Eormation; Mazourka Canyon, Inyo County, California. Isotelus gigas Dekay, 1824 PHLEGER, 1933, p. 3, pi. 1, figs. 5, 6. Hypotypes 1277 (fig. 5), 1278 (fig. 6); Ordovician, Barrel Spring Eor- mation; Barrel Spring Canyon, Inyo County, California. Isotelus spurius Phleger PHLEGER, 1933, p. 20, pi. 1, fig. 7. 12 Contributions in Science No. 181 Syntypes 1275 (fig. 7), 1276; Ordovician, Barrel Spring Formation; Barrel Spring Canyon, Inyo County, California. Lloydia obsoletus Phleger PHLEGER, 1933, p. 11, pi. 2, fig. 15. Syntypes 1249 (fig. 15, left), 1250 (fig. 15, right); Ordovician, Mazour- ka Formation; Mazourka Canyon, Inyo County, California. PUomerops barrandei (Billings, 1865) PHLEGER, 1933, p. 16, pi. 2, figs. 10, 11. Hypotypes 1256 (fig. 10), 1257 (fig. 11); Ordovician, Mazourka Eor- mation; Mazourka Canyon, Inyo County, California. Remopleurides occidens Phleger PHLEGER, 1933, p. 18, pi. 1, figs. 3, 4. Syntypes 1261 (fig. 3?), 1262 (fig. 4), 1263-1274; Ordovician, Barrel Spring Formation; Barrel Spring Canyon, Inyo County, California. ARTHROPODA: CRUSTACEA Balaniis gregarius (Conrad, 1856) ZULLO, 1969, p. 6, figs. 3-7, 45. Hypotypes 1197 (fig. 3, missing), 1198 (fig. 4, missing), 1199 (fig. 5), 1200 (fig. 6), 1201 (fig. 7, missing), 1202 (fig. 45) ; Pliocene, San Diego Formation; Market Street, east of Euclid Avenue, San Diego, San Diego County, California. Balanus kanakoffi. Zullo ZULLO, 1969, p. 7, figs. 11-12, 15-17, 19-20, 22. Holotype 1209 (fig. 19), paratypes 1205 (fig. 11), 1206 (fig. 12), 1207 (fig. 15), 1208 (figs. 16-17), 1210 (fig. 20), 1211 (fig. 22); Pliocene, San Diego Eormation; hills south of Tia Juana River, San Diego County, California. Balanus nubilus Darwin, 1854 ZULLO, 1969, p. 8, figs. 8-10. Hypotype 1 203 (figs. 8-9) , 1204 (fig. 10) ; Pliocene, San Diego Forma- tion; hills south of Tia Juana River, San Diego County, California. Balanus proinus Woodring, in Woodring and Bramlette, 1950 ZULLO, 1969, p. 16^figs. 26-28, 30-32. Hypotypes 1214 (figs. 26, 27), 1215 (fig. 28), 1216 (figs. 30-32); Plio- cene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. Balanus sp., cf. B. pacificus Pilsbry, 1916 ZULLO, 1969, p. 10, figs. 23, 25. Hypotypes 1212 (fig. 23 ) , 1213 (fig. 25) ; Pliocene, San Diego Forma- tion; hills south of Tia Juana River, San Diego County, California. 1970 Type Specimens of Fossil Invertebrata 13 Balanus wilsoni Zullo ZULLO, 1969, p. 10, figs. 33-34, 36, 38-39, 43-45. Holotype 1224 (fig. 45), paratypes 1217 (fig. 33), 1218 (fig. 34), 1219 (fig. 36), 1220 (fig. 38), 1221 (fig. 39), 1222 (fig. 43), 1223 (fig. 44); Pleistocene (?), end of Loring Street (1222); Pliocene, San Diego For- mation; Market Street, east of Euclid Avenue (1217-1221, 1223, 1224), both San Diego, San Diego County, California. Cancer branneri Rathbun, 1 926 MENZIES, 1951, p. 170, figs. 2a-2d. Hypotypes 2448 (figs. 2a, 2b), 2449 (figs. 2c, 2d); Pleistocene, Timm’s Point Formation; Timm’s Point, Los Angeles County, California. Cancer jordani Rathbun, 1 900 MENZIES, 1951, p. 169, figs. 5a-5d. Hypotypes 2446 (figs. 5a, 5b), 2447 (figs. 5c, 5d); Pleistocene; San Pedro, Los Angeles County, California. Cancer productus Randall, 1839 MENZIES, 1951, p. 168, figs. 4a-4d. Hypotypes 2444 (figs. 4a, 4b, missing), 2445 (figs. 4c, 4d, missing); Pleistocene, San Pedro Formation; San Pedro, Los Angeles County, California. Cetolepas hertleini Zullo ZULLO, 1969, p. 17, figs. 47, 49-50, 53, 55, 58-59, 61-63, 66, 68, 71-72. Holotype 1230 (figs. 49, 61-62), paratypes 1225 (figs. 47, 59), 1226 (figs. 58, 71-72), 1227 (figs. 50, 63), 1228 (figs. 53, 66), 1229 (figs. 55, 68); Pliocene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. Coronula barbara Darwin, 1854 (?) ZULLO, 1969, p. 21, figs. 73-75. Hypotype 1231 ; Pliocene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. Lepas sp. ZULLO, 1969, p. 4, figs. 1,2. Hypotype 1 196; Pliocene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. ECHINODERMATA: STELLEROIDEA Inyoaster bradleyi Phleger PHLEGER, 1936, p. 82. Paratypes 1237, 1238; Ordovician, Barrel Spring Formation; Barrel Spring Canyon, Inyo County, California. 14 Contributions in Science No. 181 ECHINODERMATA: ECHINOIDEA Briaster townsendi woynari Hertlein and Grant HERTLEIN and GRANT, 1960, p. 132, pi. 25, fig. 5; pi. 26, figs. 1-3. Holotype 1778 (figs. 1-3), paratype 1777 (fig. 5); Pliocene, San Diego Formation; quarry at end of Arroyo Drive, San Diego, San Diego County, California. Dendr aster ashleyi (Arnold), in Arnold and Anderson, 1907 HERTLEIN and GRANT, 1960, p. 117, pi. 21, figs. 2, 6; pi. 22, fig. 1; pi. 26, fig. 9. Hypotypes 1762 (fig. 2), 1763 (figs. 6, 9), 1764 (fig. 1); Euclid Avenue between Federal Boulevard and Home Avenue (1762, 1764); 2400 block of Euclid Avenue (1763), both San Diego, San Diego County, California. Dendraster ashleyi ynezensis Kew, 1919 HERTLEIN and GRANT, 1960, p. 118, pi. 21, figs. 4, 5; pi. 22, fig. 2; pi. 26, fig. 7. Hypotypes 1765 (fig. 4), 1766 (figs. 5, 7), 1767 (fig. 2); Pliocene, San Diego Formation; quarry at end of Arroyo Drive (1765) ; Market Street, 14 mile south of Euclid Avenue (1766); 3550 Dove Street (1767), all San Diego, San Diego County, California. Dendraster casseli Grant and Hertlein, 1938 HERTLEIN and GRANT, 1960, p. 120, pi. 22, fig. 7. Hypotype 1768; Pliocene, San Diego Formation; quarry at end of Arroyo Drive, San Diego, San Diego County, California. Encope tenuis Kew, 1914 HERTLEIN and GRANT, 1960, p. 126, pi. 23, figs. 7, 13; pi. 24, fig. 11. Hypotype 1775; Pliocene, San Diego Formation; quarry at end of Arroyo Drive, San Diego, San Diego County, California. Eucidaris cf. E. thouarsii (Valenciennes) , in Agassiz and Desor, 1846 HERTLEIN and GRANT, 1960, p. 103, pi. 24, figs. 19-22. Hypotypes 1757 (fig. 19), 1758 (fig. 20), 1759 (fig. 21), 1760 (fig. 22); Pliocene, San Diego Formation; hills south of Tia Juana River, San Diego County, California. Lovenia hemphilli Israelsky, 1923 HERTLEIN and GRANT, 1960, p. 130, pi. 24, fig. 18. Hypotype 1776; Pliocene, San Diego Formation; end of Loring Street, San Diego, San Diego County, California. Mellita kanakoffi Durham DURHAM, 1961, p. 5, pi. 2, fig. 2; text-fig. ID. Holotype 1121 (fig. 2, text-fig. ID), paratype 1122 (missing) ; Pleisto- cene, Palos Verdes Sand; Newport Bay mesa. Orange County, California (1121); Lomita Boulevard and Main Street, Wilmington, Los Angeles County, California (1122). 1970 Type Specimens of Fossil Invertebrata 15 Merriamaster cf. M. israelskyi (Jordan and Hertlein, 1926) HERTLEIN and GRANT, 1960, p. 122, pi. 23, figs. 12, 14. Hypotypes 1769 (fig. 12), 1770 (fig. 14); Pliocene, San Diego Forma- tion; under Snyder Continuation School, San Diego, San Diego County, California. Merriamaster pacificus (Kew, 1920) HERTLEIN and GRANT, 1960, p. 124, pi. 23, figs. 2-4, 6, 11; pi. 25, fig. 6; pi. 26, fig. 11. Hypotypes 1771 (pi. 23, figs. 2, 3, 11), 1772 (pi. 23, fig. 4), 1773 (pi. 23, fig. 6) , 1774 (pi. 25, fig. 6; pi. 26, fig. 11); Pliocene, San Diego For- mation; 0.2 mile north of Harbor Boulevard and Tourmaline Street (1771, 1772); end of Coring Street (1773, 1774), both San Diego, San Diego County, California. Strongylocentrotus purpiiratus (Stimpson, 1857) HERTLEIN and GRANT, 1960, p. 112, pi. 19, fig. 29. Hypotype 1761; Pliocene, San Diego Eormation; Pacific Beach, San Diego, San Diego County, California. Literature Cited Addicott, W. O. 1965. Some western American Cenozoic gastropods of the genus Nassariiis. U.S. Geol. Surv. Prof. Paper 503B: 1-24, pis. 1-3. Agassiz, Louis and P. J. E. Desor. 1846. Catalogue raisonne des families, des genres, et des especes de la classes des echinodermes. Ann. Sci. Natur., Zool. 6:305-374, pis. 15-16. Aral M. N. and Alan McGugan. 1969. A problematical Cambrian coelenterate (?). J. Paleontol. 43:93-94, pi. 18. Arnold, Ralph and Robert Anderson. 1907. Geology and oil resources of the Santa Maria Oil District, Santa Barbara County, California. U.S. Geol. Surv. Bull. 322:1-161, pis. 1-26. Berry, S. S. 1941. New Mollusca from the Pleistocene of San Pedro, California-II. Bull. Amer. Paleontol. 27:1-18, pi. 1. . 1954. New Californian Pleistocene Eulimidae. Bull. Amer. Paleontol. 35:257-270, pi. 24. Billings, E. 1861-1865. Palaeozoic fossils: v. 1. Containing descriptions and figures of new or little known species of organic remains from the Silurian rocks. Geol. Surv. Canada. 426 pp. Clark, B. L. 1915. Fauna of the San Pablo group of middle California. Univ. Calif. Pub., Bull. Dep. Geol. 8:385-572, pis. 42-71. . 1918. The San Lorenzo Series of middle California. Univ. Calif. Pub., Bull. Dep. Geol. 11:45-234, pis. 3-24. . 1934. A new genus and two new species of Lamellibranchiata from the Middle Eocene of California. J. Paleontol. 8:270-272, pi. 37. . 1942. New Middle Eocene gastropods from California. J. Paleontol. 16: 116-119, pi. 19. Clark, B. L. and A. O. Woodford. 1927. The geology and paleontology of the type section of the Meganos Formation (Lower Middle Eocene) of California. Univ. Calif. Pub., Bull. Dep. Geol. Sci. 17:63-142, pis. 14-22. Clark, C. W. 1921. Lower and Middle Cambrian Formations of the Mohave Desert. Univ. Calif. Publ., Bull. Dept. Geol. Sci. 13:1-7. 16 Contributions in Science No. 181 Conrad, T. A. 1856. Descriptions of three new genera; twenty-three new species of Middle Tertiary fossils from California, and one from Texas. Acad. Natiir. Sci. Philadelphia, Prcc. 8:312-316. Dale, W. H. 1902. Illustrations and descriptions of new, unfigured, or imperfectly known shells, chiefly American, in the U.S. National Museum. Proc. U.S. Nat. Mus. 24:499-566, pis. 27-40. . 1909. Contributions to the Tertiary paleontology of the Pacific coast. I. The Miocene of Astoria and Coos Bay, Oregon. U.S. Geol. Surv., Prof. Paper 59:1-278, pis. 1-23. Darwin, Charles. 1854. A monograph of the sub-class Cirripedia. 2. Balanidae, Verrucidae. Ray Soc. London. 684 pp., 30 pis. Dekay, J. E. 1824. Observations on the structure of trilobites and descriptions of an apparently new genus. Ann. Natur. Hist. New York. 1 : 174-189. Dickerson, R. E. 1914. Eauna of the Martinez Eocene of California. Univ. Calif. Pub., Bull. Dep. Geol. 8:61-180, pis. 6-18. . 1916. Stratigraphy and fauna of the Tejon Eocene of California. Univ. Calif. Pub., Bulk Dep. Geol. 9:363-524, pis. 36-46. Durham, J. W. 1961 . The echinoid Mellita in the Pacific coast Cenozoic. Los Angeles Co. Mus., Contrib. Sci. 48: 1-12, pis. 1-2. Gabb, W. M. 1 864. Description of the Cretaceous fossils. Calif. Geol. Surv., Paleontol. 1:55-236, pis. 9-31. Grant, U.S. IV and H. R. Gale. 1931. Catalogue of the marine Pliocene and Pleistocene Mollusca of California. San Diego Soc. Nat. Hist., Mem. 1 : 1-1036, pis. 1-32. Grant, U.S. IV and L. G. Hertlein. 1938. The west American Cenozoic Echi- noidea. Publ. Univ. Calif. Los Angeles Math. Phys. Sci. 2: 1-226, pis. 1-30. Hael, James and R. P. Whitfieed. 1877. Palaeontology. U.S. Geol. Exploration 40th Parallel 4(2) : 199-302, pis. 1-7. Hanna, M. A. 1927. An Eocene invertebrate fauna from the La Jolla Quadrangle, California. Univ. Calif. Pub., Bull. Dep. Geol. Sci. 16:247-398, pis. 24-57. Herteein, L. G. and U.S. Grant IV. 1960. The geology and paleontology of the marine Pliocene of San Diego, California. San Diego Soc. Nat. Hist., Mem. 2(2a):73-133, pis. 19-26. Hinds, R. B. 1 844. The zoology of the voyage of H. M. S. Sulphur under the com- mand of Capt. Sir Edward Belcher . . . during 1836-1842. London. Mollusca. 72 pp., 21 pis. International Commission on Zoological Nomenclature. 1964. International code of zoological nomenclature, adopted by the XV International Congress of Zoology. Inst. Trust Zool. Nomenclature, London. 176 p. IsRAELSKY, M. C. 1923. Some new forms of west coast fossil Echinoidea. Univ. Calif. Pub., Bull. Dept. Geol. Sci. 14:377-396, pis. 69-74. Jordan, E. K. and L. G. Hertlein. 1926. Contribution to the geology and paleon- tology of the Tertiary of Cedros Island and adjacent parts of Lower California. Proc. Calif. Acad. Sci. 15:409-464, pis. 27-34. Kanakoff, G. P. 1950. Some observations on the land snails of San Clemente Island. Bull. So. Calif. Acad. Sci. 49:79-89, pis. 26-31. . 1953. A new fossil shell from the Palos Verdes Sand. Bull. So. Calif. Acad. Sci. 52:67-70, pis. 12-13. . A new Kelletia from the Pliocene of California. Bull. So. Calif. Acad. Sci. 53:114-117, pis. 29-31. . 1956. Two new species of Nassariiis from the Pliocene of Los Angeles County, California. Bull. So. Calif. Acad. Sci. 55:110-114, pis. 30-31. 1970 Type Specimens of Fossil Invertebrata 17 . 1966. A new species of Boelica from the Pliocene of California. Los Angeles Co. Mus., Contrib. Sci. 103:1-4, 3 figs. Kanakoff, G. P. and J. H. McLean. 1966. Recognition of the cancellariid genus Neadmete Habe, 1961, in the west American fauna, with description of a new species from the Lomita Marl of Los Angeles County, California. Los Angeles Co. Mus., Contrib. Sci. 116:1-6, figs. 1-2. Kew, W. S. W. 1914. Tertiary echinoids of the Carrizo Creek region in the Colorado Desert. Univ. Calif. Pub., Bull. Dept. Geol. 8:39-60, pis. 1-5. . 1919. Geology of a part of the Santa Ynez River District, Santa Barbara County, California. Univ. Calif. Publ., Bull. Dept. Geol. 12:1-21, pis. 1-2. . 1920. Cretaceous and Cenozoic Echinoidea of the Pacific coast of North America. Univ. Calif. Publ., Bull. Dept. Geol. 12:23-236, pis. 3-42. Kuster, H. C. 1844. Mollusca, Brachiopoda, Terebratulacea. Conchylien Cabinet 7(1 ):19-49, pis. 2-4. Lipps, J. H. 1963. A new species of Acmaea ( Archaeogastropoda) from the Pleisto- cene of San Nicolas Island, California. Los Angeles Co. Mus., Contrib. Sci. 75:1-15, figs. 1-6. Mason, J. F. 1935. Fauna of the Cambrian Cadiz Formation, Marble Mountains, California. Bull. So. Calif. Acad. Sci. 34:97-1 19, pi. 1 5. McLean, J. H. 1964. New species of Recent and fossil west American aspidobranch gastropods. Veliger 7: 129-133, pi. 24. Menzies, R. j. 1951. Pleistocene Brachyura from the Los Angeles area: Cancridae. J. Paleontol. 25:165-170, 13 figs. Merriam, C. W. 1941. Fossil turritellas from the Pacific coast region of North America. Univ. Calif. Pub., Bull. Dep. Geol. Sci. 26: 1-214, pis. 1-41. Mount, J. D. 1970. A new species of Neadmete (Neogastropoda) from the Pliocene of California. Los Angeles Co. Mus., Contrib. Sci. 177: 1-4, 1 figure. Nelson, R. N. 1925. A contribution to the paleontology of the Martinez Eocene of California. Univ. Calif. Pub., Bull. Dep. Geol. Sci. 15:397-466, pis. 49-61 . Nomland, j. O. 1916. Corals from the Cretaceous and Tertiary of California and Oregon. Univ. Calif. Pub., Bull. Dept. Geol. 9:59-76, pis. 3-6. Oldroyd, I. S. 1925. The marine shells of the west coast of North America. Stanford Univ. Pub. Geol. Sci. 1:1-247, pis. 1-57. Oldroyd, T. S. 1925. The fossils of the Lower San Pedro fauna of the Nob Hill cut, San Pedro, California. Proc. U.S. Nat. Mus. 65(22) : 1-39, pis. 1-2. Packard, E. L. 1922. New species from the Cretaceous of the Santa Ana Mountains, California. Univ. Calif. Pub., Bull. Dep. Geol. Sci. 13:413-462, pis. 24-38. Phleger, F. B., Jr. 1933. Notes on certain Ordovician faunas of the Inyo Moun- tains, California. Bull. So. Calif. Acad. Sci. 32: 1-21, pis. 1-2. . 1936. An Ordovician auluroid from California. Bull. So. Calif. Acad. Sci. 35:82-83, pi. 20. Pilsbry, H. a. 1916. The sessile barnacles (Cirripedia) contained in the collections of the U.S. National Museum, including a monograph of the American species. U.S. Nat. Mus., Bull. 93:1-366, pis. 1-76. Randall, J. W. 1839. Catalogue of the Crustacea brought by Thomas Nuttall and J. K. Townsend, from the west coast of North America and the Sandwich Islands. Acad. Natur. Sci. Philadelphia, J. 8: 106-147. Rathbun, M. j. 1900. Synopses of North American invertebrates, VII, the cyclo- metopous of cancroid crabs of North America. American Natur. 34: 131-143. . 1926. The fossil stalk-eyed Crustacea of the Pacific slope of North Amer- ica. U.S. Nat. Mus. Bull. 138: 1-155, pis. 1-39. 18 Contributions in Science No. 181 Raymond, P. E. and J. E. Narraway. 1908. Notes on Ordovician trilobites: Illaeni- dae from the Black River Limestone near Ottawa, Canada. Ann. Carnegie Mus. 4:242-255, pis. 60-62. Schenk, E. T., H. H. McM asters, A. M. Keen, and S. W. Muller. 1956. Procedure in Taxonomy. Stanford Univ. Press, Stanford. 149 p. Sphon, G. G., Jr. 1962. Destroyed type material. Veliger 5:57. Stimpson, William. 1857. On the Crustacea and Echinodermata of the Pacific shores of North America. Boston J. Natur. Hist. 6:444-532, pis. 1-6. Sutherland, J. A. 1966. A new species of Architectonica from the Santa Susana Mountains, Ventura County, California. Los Angeles Co. Mus., Contrib. Sci. 117:1-4, figs. 1-2. Terry, J. S. 1968. Mediargo, a new Tertiary genus in the family Cymatiidae. Veliger 11:42-44, pi. 4. Turner, F. E. 1938. Stratigraphy and Mollusca of the Eocene of western Oregon. Geol. Soc. Amer., Spec. Paper 10: 1-130, pis. 1-22. Verrill, a. E. 1864. List of the polyps and corals sent by the Museum of Compara- tive Zoology to other institutions in exchange, with annotations. Bull. Mus. Comp. Zool. 1:29-60. . 1869. On new and imperfectly known echinoderms and corals. Proc. Boston Soc. Natur. Hist. 12:381-396. VoKEs, H. E. 1939. Molluscan faunas of the Domengine and Arroyo Hondo For- mations of the California Eocene. Ann. N.Y. Acad. Sci. 38:1-246, pis. 1-22. Walcott, C. D. 1884. Paleontology of the Eureka District. U.S. Geol. Surv. Monog. 8:1-298, pis. 1-24. . 1886. Second contribution to the studies on the Cambrian faunas of North America. U.S. Geol. Surv. Bull. 30: 1-369. . 1920. Cambrian geology and palentology. IV. Middle Cambrian Spongiae. Smithsonian Misc. Coll. 67:261-364, pis. 60-90. Weaver, C. E. 1916. Tertiary faunal horizons of western Washington. Univ. Wash. Publ. Geol. 1:1-67, pis. 1-5. Willett, George. 1937a. Additions to knowledge of the fossil invertebrate fauna of California. Bull. So. Calif. Acad. Sci. 36:61-64, pis. 24-25. . 1937b. An Upper Pleistocene fauna from the Baldwin Hills, Los Angeles County, California. Trans. San Diego Soc. Nat. Hist. 8:379-406, pis. 25-26. . 1949. A new species of mollusk from the San Pedro Pleistocene. Bull. So. Calif. Acad. Sci. 38:202-203, pi. 54. . 1944. Two new west American pelecypods. Bull. So. Calif. Acad. Sci. 43:19-22, pis. 7-8. . 1945. The Acteocina of Salton Sink, Colorado Desert, California. Bull. So. Calif. Acad. Sci. 44:28-29. . 1948. Four new gastropods from the Upper Pleistocene of Newport Bay mesa. Orange County, California. Bull. Soc. Calif. Acad. Sci. 47 : 17-21, pi. 4. Woodring, W. P. and M. N. Bramlette. 1950. Geology and paleontology of the Santa Maria District, California. U.S. Geol. Surv., Prof. Paper 222: 1-185, pis. 1-23. Woodring, W. P., M. N. Bramlette, and W. S. W. Kew. 1946. Geology and paleontology of Palos Verdes Hills, California. U.S. Geol. Surv. Prof. Paper 207:1-145, pis. 1-37. Woollacott, R. M. 1966. Ceriocava eastoni, sp. nov., the first described species of Bryozoa from the Cretaceous of California. Bull. So. Calif. Acad. Sci. 65:225- 228, fig. 1. ZuLLO, V. A. 1969. Thoracic Cirripedia of the San Diego Formation, San Diego County, California. Los Angeles Co. Mus., Contrib. Sci. 159:1-25, figs. 1-77. 1970 Type Specimens of Fossil Invertebrata 19 Species Index agnesae, Micrarionta (Xerarionta), p. 8 albida, Glottidia, p. 2 andersoni, Surcula (Surculites), p. 9 angulatus, Plectambonites, p. 3 anomala, Acteocina, p. 5 ashleyi, Dendraster, p. 14 barbara, Coronula, p. 13 barrandei, Pliomerops, p. 12 branneri, Cancer, p. 13 berryi, Homalopoma, p. 7 billingsi, Bumastus, p. 1 1 bradleyi, Inyoaster, p. 13 calabasasensis, Calyptraea, p. 6 californianus, Laqueus, p. 3 calliteles, Cybeloides, p. 1 1 caminoensis, Megistostoma, p. 8 casseli, Dendraster, p. 14 coiloterus, Nassarius (Caesia) fossatus, p. 8 collisella, Turbonilla (Pyrgolampros), p. 9 constantiae, Diodora, p. 7 cooki, Bornia, p. 4 decipiens, Orthis, p. 3 diegensis, Laqueus vancouveriensis, p. 3 diegensis, Macrarene, p. 7 eastoni, Ceriocava, p, 3 ebriconus, Balcis (Vitreolina), p. 6 effiae, Alabina, p. 6 effiae, Odostomia (Menestho), p. 9 elegans, Balanophyllia, p. 2 elsiae, Odostomia (Chrysallida), p. 9 epiphanea, Turbonilla (Mormula), p. 9 eros, Chancelloria, p. 2 filosa, Calyptraea, p. 6 fossilis, Triphora, p. 9 galeana, Mitromorpha, p. 8 gigas, Isotelus, p. 1 1 goniglensis, Perna, p. 5 gregarius, Balanus, p. 12 grouardi, Turbonilla (Turbonilla), p. 10 hastula, Encrinurus, p. 1 1 hemphilli, Lovenia, p. 14 hemphilli, Terebratalia, p. 3 hertleini, Boetica, p. 6 hertleini, Cetolepas, p. 13 hildegardae, Nassarius, p. 8 hilli, Cardita, p. 4 howelli, Bathyuriscus, p. 1 1 idae, Turbonilla (Pyrgolampros), p. 9 incallida, Balcis (Vitreolina), p. 6 infrequens, Ceraurus, p, 1 1 insignifica, Astrangia, p. 2 israelskyi, Merriamaster, p. 15 jordani. Cancer, p. 13 kanakoffi, Balanus, p. 12 kanakoffi, Mellita, p. 14 kanakoffi, Triphora, p. 9 lawsoni, Turritella andersoni, p. !0 lincolnensis, Fusinus (Exilia), p. 7 linnarssoni, Alokistocare, p. 10 llajasensis, Architectonica, p. 6 llajasensis, Exputens, p. 4 lodensis, Anoria, p, 10 manca, Odostomia (Evalea), p. 9 marysvillensis, Cardium (Protocardium), p. 4 maximus, Bathyuriscus, p. 1 1 mazourkaensis, Plectorthis, p. 3 mediocris, Mediargo, p. 7 menzola, Odostomia (Amaura), p. 8 minusculus, Orthis, p. 3 mitchelli, Acmaea, p. 5 morsei, Barbatia, p. 4 nanella, Marginella jewettii, p. 7 nubilus, Balanus, p. 12 obsoletus, Lloydia, p. 12 obstipa, Balcis (Vitreolina), p. 6 occidens, Remopleurides, p. 1 2 octonarius, Encrinurus, p. 1 1 oldroydi, Dendrophyllia, p. 2 ooides, Tellina, p. 5 pacificus, Balanus, p. 12 pacificus, Merriamaster, p. 15 packi, Lucina, p, 4 patulus, Plectorthis, p. 3 pecora, Turbonilla (Strioturbonilla), p. 10 picta, Chione, p. 4 piochensis, Amecephalus, p. 10 productus. Cancer, p. 13 proinus, Balanus, p. 12 purpuratus, Strongylocentrotus, p. 15 quadriceps, Dorypyge, p. 1 1 reyana, Rochefortia, p. 5 rosa, Leda, p, 4 santana, Tellina, p, 5 schencki, Turritella, p. 10 schwartzi, Neadmete, p. 8 semirostrata, Nucula (Acila), p. 5 20 Contributions in Science No. 181 sorrentoensis, Cardium, p. 4 spatha, Anomocarella, p. 10 spurius, Isotelus, p. 1 1 stearnsii, Paracyathus, p. 2 stewarti, Chedvillia, p. 6 stocki, Nassarius, p. 8 strangulatus, Amecephalus, p. 10 subobsoleta, Leptothyra, p. 7 Sutherland!, Neadmete, p. 8 taliaferroi, Exilia, p. 7 tenuis, Encope, p. 14 tersa. Balds (Balds), p. 6 thouarsii, Euddaris, p. 14 timessa, Odostomia (Amaura), p. 9 trifossatus, Acrocephalites, p. 10 ucalius, Fusinus, p. 7 vladimiri, Kelletia, p. 7 wilsoni, Balanus, p. 13 woynari, Briaster townsendi, p. 14 ynezensis, Dendraster ashleyi, p. 14 Accepted for publication November 18, 1969. LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE fUMBER 182 April 17, 1970 AQUATIC ADAPTATIONS OF THE WATER MOUSE, RHEOMYS UNDERWOODI By Andrew Starrett and George F. Fisler Los Angeles County Museum OF Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor AQUATIC ADAPTATIONS OF THE WATER MOUSE, RHEOMYS UNDERWOODI By Andrew Starrett and George F. Fisler^ Abstract: An adult male Rheomys imderwoodi caught alive along the Rio Poasito, at 2000 m on Volcan Poas, Alajuela Prov- ince, Costa Rica, in April, 1966, was maintained in captivity for a short time. Limited observations on maintenance behavior and locomotion of this animal are presented. In addition, morphologi- cal features (body form, tail shape and foot structure) related to the high degree of specialization for aquatic locomotion shown by this rodent are described, and some ideas concerning the evolution of the semi-aquatic habit are discussed in relation to Rheomys and other small mammals. Introduction Information available on the biology of the genus Rheomys is quite limited. These animals live primarily in and around swift flowing streams of the Neotropics. Howell (1930) described certain aquatic features of Rheomys but did not give a full account of the high degree of adaptation to a semi- aquatic existence which the species R. underwoodi, in particular, exhibits. Hooper (1968) briefly mentioned some of the aquatic characteristics of Rheomys in a comparison of R. underwoodi and R. hartmanni, the two species which occur in Costa Rica, and also included ecological information and stomach analyses for both. Our capture of a living animal has prompted us to record some observations on the behavior and morphology of the water mouse with additional comments on aquatic adaptations in this and other species. On the basis of information and materials provided for us by Emmet T. Hooper, we set 31 traps (18 museum specials, 13 National single door live traps) along the banks of the Rio Poasito, 2000 m, Volcan Poas, in Alajuela Province, Costa Rica, on the evening of 9 April 1966. This same area was trapped on 6 and 7 April 1966 by Hooper and James H. Brown and is described by Hooper (1968) . Instead of using the method of trapping Rheomys de- scribed by Stirton (1944), we set seven live traps along relatively quiet eddy pools just outside white-water areas. Our Rheomys was taken near such a pool in a trap which was placed on rocks about 7.5 cm above the water line. The mouth of the trap was about 20 cm from the pool, one meter from any white water, and 15-20 meters below a small waterfall. Bait used was peanut butter with sardine meat and oil spread over it, placed on the treadle inside of the trap. Some sardine oil was smeared on the trap entrance but no bait was placed outside of the trap. No other mammals were taken in the live traps ■■Both authors: Research Associates, Los Angeles County Museum of Natural History; and Department of Biology, San Fernando Valley State College, North- ridge, California 91324. 1 9 Contributions in Science No. 182 but three Reithrodontomys creper and one Peromyscus nudipes were caught in the snap traps. About half of the snap traps were placed next to the water. The water mouse was an adult male having abdominal testes (10 x 18 mm) with epididymides well developed and caudae protruding into the inguinal passages. The standard measurements are 300-158-39-7 (mm). General Behavior The actions of the Rheomys when first observed were slow and deliberate, as if it had been in the trap for some time and was weakened, but later it became more active. It darted about the trap hurriedly, in the manner of a shrew, yet when quiet, it sat mostly hunched over like a muskrat {Ondatra zibet hicus) . Stirton (1944) described it as a miniature model of a beaver ( Castor canadensis) . When seated, its hind feet were spread out at an angle of some 30-40 degrees from the body with its fore feet usually tucked along its chest. A striking, and probably adaptively significant reaction, was its tendency to bite at objects with which it came into contact. Even while being carried in the trap, it nipped at fingers whenever it encountered them at the edges of the trap as it dashed about. When handled, however, it was docile and did not bite, and attempts to induce it to nibble fingers were successful only if the animal blundered into them, as it did not seem to seek out such prey. The biting reaction on contact with moving objects was also later shown when the animal was observed swimming. In the capture of prey under water, where vision and olfaction may be of minimal use, it would be of great ad- vantage for an animal to be adapted to biting at any moving object with which it came in contact. The eyes of this rodent are very small and presumably are of little use, particularly under water, and especially in fast moving, turbulent streams. The face is rather well endowed with vibrissae, probably advan- tageous in hunting and orientation, but it was difficult to tell if they were used as sensors. When the trap was first observed, all of the bait had been eaten, or, less likely, washed away. Our captive animal was fed small pieces of sardines which it ate from the container rather than manipulating the food while sitting up as do many of the more familiar rodents. It occasionally manipulated larger chunks of sardine (10-12 mm) with its fore feet, so perhaps the size and con- sistency of the food determines the manner of feeding. It probably handles insect larvae with its fore feet. In general, the animal would eat for a minute or so, pause for 15-20 seconds, and then resume eating. It was first fed about thirty minutes after the trap was picked up and it ate roughly a small thimbleful of sardines in the succeeding hour. Attempts at feeding it green vegetation (leaves and herbs from its habitat) while it was still hungry were unsuccessful. Stirton (1944) found no evidence that Rheomys was piscivorous but thought that it was omnivorous and even cannibalistic. Goldman (1920) suggested that one species may feed on snails. Hooper’s (1968) analysis of contents from six stomachs of Rheomys from Costa Rica and Guatemala (one stomach 1970 Adaptations of a Water Mouse 3 empty) shows a diet of primarily aquatic insect larvae. Fish are present in the Rio Poasito but are not native. Our captive was observed drinking sparingly. It lapped with its tongue in the familiar manner of many small rodents, directly off the water surface. When first seen in the trap, the Rheomys was dry except for its tail, un- derparts, feet and muzzle. After removal from the stream area, it dried quickly. Grooming was accomplished in the manner of many rodents, except that the enlarged hind feet serve as very effective combs for the body fur. When wet or grooming, or both, as grooming is evidently an important part of the drying process, the animal moved the skin of the back in a peculiar fashion, appearing to erect the guard hairs, a grooming as well as a drying aid. This movement appeared to pass in waves over the hunched back antero-posteriorly and the reverse. Ail of the time the animal sat much like a muskrat. After leaving the water, the animal groomed extensively. Limited swimming observations of the captive Rheomys were made in a small 27 x 36 cm wash basin. The water was about 8 cm deep and for part of the time the trap was used as an island although the basin could barely ac- commodate it. The small size of the container restricted the movement of the animal. It tried to climb the sides, violently treading water with its hind feet, maintaining its body about half to two-thirds out of the water. The large hind feet made tremendous splashes as it attempted to escape, very strikingly pointing out their propulsive power. Observations of diving were limited but the process was extremely quick. While swimming on the surface, the mouse had its head out of the water, the rump under, while the tail was curved con- cavely with the tip at the water surface. Most of the swimming motion was with the hind feet with little body undulation except when sculling (noted only 2 or 3 times). These latter motions seemed to give the animal added impetus for quick acceleration or turning. The fore feet could not be seen while swimming, indicating that they are not used other than to hold on to objects when landing, or perhaps to fend off objects while swimming. In all swimming motions, the animal was extremely quick, presumably an adaptive feature for feeding under water as well as for swimming in swift moving streams. Hooper (1968) has reported on the swiftness in movement and swimming of an animal, undoubtedly Rheomys underwoodi, observed in the wild. While in captivity, the animal lived in the trap. Soft tissue and news- paper given to the Rheomys was slightly shredded but no real nest building occurred. Excelsior was ignored. Rather, the animal slept mostly on its side rolled in the typical rodent ball, under the paper. Stirton (1944) mentioned that water mice probably nest in dead brush and logs lodged in the stream or under large boulders or other rocks. After the swimming session on the evening of capture, the animal appeared to be drying rather slowly. However, by the following morning ( 1 1 April) it was dry and appeared to be in good health. It ate readily (sardines) . It was placed in a light cloth sack for transport to Los Angeles, California, 4 Contributions in Science No. 182 but it died in transit. The specimen was preserved as a skin and skull with the body preserved in fluid and is now in the collection of the Los Angeles County Museum of Natural History (LACM 28260) . Morphology Because of the absence of such information from the literature, and in order to lay the background for a discussion of the adaptations of this animal to a semi-aquatic existence, a detailed account of pertinent morphological characters is here provided. The body is streamlined to a high degree with essentially no projections other than the hind feet and tail when in the water. The head is somewhat depressed and the mouth opening is small. The eyes are also reduced in size. The snout is truncated in dorsal view and the rhinarium is prominent. The nostrils are situated laterally and open posteri- orly behind flaplike valves. The numerous vibrissae are situated dorsally and laterally on the snout and cheeks posterior to the rhinarium and around the mouth. They reach maximum lengths of 32 to 34 mm, although most are 14 to 26 mm long. The ear pinnae are reduced essentially to small, pointed posterior lobes, about 4 mm in length, which do not extend above the surface of the fur. The pelage is generally dense and soft, the type characteristically asso- ciated with the smaller aquatic mammals. In the mid-dorsal region the underfur is dense with fine hairs which range from 5 to 7 mm in length. Abundant but less dense and grosser hairs, about 12 to 14 mm long, form an outer layer. The ventral pelage is similar to that of the dorsum, but the underfur is shorter. The fore limbs are relatively short and do not project much from the body, even when walking. The proximal portions of the hind limbs blend similarly into the body contours, leaving the remarkable hind feet as the only limb segments which project obviously from the furred outline of the body. The tail also is flared into the body outline over the proximal 15 mm of its length. The front feet are much smaller than the hind feet with long, curved, sharply pointed claws. On the hind feet, digit I is the shortest, then in order of increasing length, digits V, II, III, and IV. The greatest differences in length are between digits I and II and, to a lesser degree, between digits V and IV. The first and, to a lesser extent, the fifth digits appear to have the greatest amount of independent movement. The hind feet have laterally compressed digits interconnected by webbing. The sides of the feet and toes are lined with single rows of closely spaced stiff hairs which reach maximum lengths of three and four mm on the medial and lateral sides of the feet, respectively, and two to three mm on the toes and webbing, with the shortest hairs being those near the middle of each web. The claws are deep and essentially as broad as the tip of the terminal phalanx, that on digit I being the deepest and the claw on digit II the broadest. The tips are pointed in lateral profile, but somewhat rounded and scooplike in ventral aspect. The massive appearance of the hind claws is very striking when they 1970 Adaptations of a Water Mouse 5 are compared with those of other rodents of similar size such as Neotoma, Nelsonia, Xenomys, Tylomys, Nyctomys, and Rattus. When the digits are abducted, the webbing reaches the level of the pennultimate interphalangeal joint between digits II, III, IV, and V, but is less extensive between digits I and II. Each web originates at the base of the claw and extends, as a widening flap, to the levels just mentioned. There the margin turns and a similar but narrower flap continues to about the same claw level on the next toe. On the lateral (post-axial) side of each digit, the webbing arises along the dorsolateral surface, whereas on the medial side of the succeeding digit it connects along the medial surface. When the foot is extended, the digits rotate so that their ventral surfaces turn mediad. The metatarsophalangeal joints are somewhat loose, apparently allowing the digits to rotate even more than do the metatarsals. The metatarsals (except III and IV) and free digits abduct and, since they are rotated, the latter also extend at the metatarso- phalangeal joints in a more or less horizontal plane. These actions spread the foot and toes while turning the digits so that they present their broad com- pressed surfaces and spread the webbing to its greatest extent. Part of the rotation of the digits may be passively effected by the tension of the webbing. When fully expanded, the hind foot of our specimen measured 14 mm across the plantar surface at the level of the distal ends of the metatarsals, and the expanded hair fringes added another six mm at that level, resulting in greatest dimensions of 20 x 39 mm. Planimetric measurements of the best of a number of inkpad prints of the right hind foot, including hair fringes, showed areas of 3.8 to 4.0 cm^. The two hind feet, then, form impressive propulsive surfaces during extension in the backward stroke while swimming. In the re- covery stroke, it is probable that the foot is flexed and the digits flexed, ab- ducted, and possibly rotated somewhat laterad again so that the compressed toes are curled slightly and lie close together in a staggered sequence (cross sectional), with digit I leading, and the hair fringes flattened against the sides of the foot and toes. Thus the pes becomes narrowed with the digit alignment presenting a laterally compressed, rather than an anteroposteriorly flattened unit, approaching the characteristics of a hydrofoil, and offering minimum resistance to the water. The free tail, from the point at which it leaves the furred base which flares into the body, measures 143 mm to the tip. The proximal 30 mm is circular in cross section and approximately 7 mm in diameter. At the end of this basal portion, there is a constriction to about 6.5 mm diameter and the tail gradually tapers and becomes compressed. In the last 15 mm the tail narrows more rapidly to almost a point in dorsal view and, in lateral aspect, reaches its shallowest dimension (a little less than 5 mm), then expands slightly ventrally and ends in a somewhat rounded tip which gives the appearance of turning a little ventrad. Along the ventral surface of the free tail, particularly from the level of the basal constriction where the cross section of the tail begins to change shape, the hairs arising from the ventral surface of the tail, represent- ing approximately the tail circumference arc, become abruptly longer (to 6 Contributions in Science No. 182 about 5 mm) and more dense, forming a kind of broad fringe extending along the undersurface of the tail to the point where it expands ventrally near the tip. The hairs on the rest of the free tail are shorter and relatively sparsely but evenly distributed. Internally the most notable feature of the tail structure is the presence of numerous heavy tendons attaching firmly to the vertebrae, fascia and skin, distal to the basal constriction. The extraction of the tail beyond this point during skinning was difficult. (James H. Brown, personal communi- cation, recalled the same difficulty with the specimens of R. underwoodi he prepared from the same locality.) This clearly suggests well developed mus- cular control of the tail, further evidence of which was obtained by selective pulling of the tendons, while the tail was partially extracted, which caused bending in various directions and at different points in the more distal portions of the tail. The streamlining provided by reduction and blending of projecting parts of the body and the generally fusiform shape and flared tail base make Rheomys a hydrodynamically efficient body. The large hind feet provide the main propulsive force and in recovery strokes tend to give minimum resistance by the combined characters of compressed digits and the foot action mech- anism. The tail apparently serves a stabilizing and a supplementary propulsive function, being adapted for both by compression, by effective control through the tendon arrangement, and probably also by the paddle-shaped tip. Dorsal- ventral bending of the tail was noted above, during surface swimming. The slightly ventrally oriented distal portion of the tail (the paddle) probably is significant when the tail is thus curved in such a way that the tip approaches, or is at, the water surface while the rest of the tail is below. The tip then would be more nearly horizontally oriented. This curving might also make the ventral hair fringe stand out more and so contribute to the compressed tail shape. This fringe may serve other functions since the hairs are not particularly stiff. Possibly it protects the ventral surface of the tail from abrasion by streamside rocks and gravel, or it may serve as a drip line for rapid water runoff. The thoracic vertebrae of the skinned carcass were capable of rather acute kyphotic curvature. The panniculus carnosus is a rather broad, thin sheet posteriorly (lumbosacral region), but gives rise to a number of long, narrow extensions running up over the back to the head and shoulder region. Contraction of this muscle pulls the skin back and forth over the thoracic region. This action separates the hairs of the back when the thoracic region is curved during grooming. Discussion In attempting to understand aquatic adaptation of Rheomys underwoodi, we surveyed certain literature on semi-aquatic mammals. (Anthony, 1921, 1923, 1929; Bauchot and Stephan, 1968; Conaway, 1960; Dickey, 1928; Enders, 1938; Goldman, 1912, 1920; Goodwin, 1959; Guth et al., 1959; Handley and Mondolfi, 1963; Hershkovitz, 1944, 1955; Hooper, 1947, 1968; Howell, 1930; Lonnberg, 1921; Malzy, 1965; Peyre, 1956; Stirton, 1944; 1970 Adaptations of a Water Mouse 7 Svihla, 1934; Thomas, 1893, 1897, 1906a, 1906b, 1924a, 1924b; De Winton, 1896). Most of our comparative information was taken from Walker ( 1968). Although the survey was not exhaustive, we found that relatively little is known about semi-aquatic small mammals, especially the more highly special- ized forms. However, limited discussion of the position which R. iinderwoodi occupies in the evolution of specializations in these mammals is possible. Rheomys iinderwoodi is among the most highly hydrodynamically spe- cialized mammals, particularly when compared with species of comparable size and habits, the smaller semi-aquatic rodents and insectivores, and the marsupial Chironectes. On the basis of what we could determine, it is quite likely that most aquatic features of R. iinderwoodi are not unique, but a comparable total combination of specific features is apparently limited to members of only eight other genera: three of rodents (in addition to certain other Rheomys species), Anotomys, Ichthyomys (both closely related to Rheomys) , and the hydromyine Crossomys; five of insectivores, Limnogale, and Micropotamogale (M. ruwenzori) (Tenrecidae) , Desmana and Galemys (Talpidae) and Nectogale (Soricidae). As might be expected, the details of specialization are most similar among the rodents. Limnogale and Nectogale seem to be the most closely comparable of the insectivores. Davis (1964:323) considered two factors to be “primarily responsible for adaptive modifications in the morphology of mammals: locomotion, and foods and feeding.” In the case of forms which are more highly adapted for aquatic living, the locomotor requirements strongly influence the morphologi- cal modification related to obtaining food. In those particularly adapted for rapid swimming, morphological feeding specializations not in accord with the hydrodynamic requirements imposed by locomotor specializations are not to be found. Although this applies to semi-aquatic, or amphibious mammals, as well as to the more completely aquatic, the degree of overall modification in the former is limited by the requirements of terrestrial locomotion, and a structural-functional balance is struck according to the relative significances of aquatic and terrestrial locomotion in the lives of these animals. Further- more, not all aquatic adaptations contribute directly to aquatic locomotion, and the ability to swim does not necessarily involve much in the way of mor- phological modification (Howell, 1930). Rheomys iinderwoodi must be con- sidered a semi-aquatic rodent with a high degree of morphological specializa- tion reflecting primarily selection for effective, rapid aquatic locomotion as the major influence. Davis (1964:323) further stated that “In mammals the major forces impinging on locomotion are escape from enemies, pursuit of prey and wan- dering in search of food or water.” These categories can be modified to include as potentially significant functions (forces or factors) : 1) simple translocation or unforced movement (wandering for whatever reason); 2) escape from predators; 3) escape in the event of changing environmental conditions; 4) locating and feeding on plants (or mollusks; i.e. stationary food); 5) pur- suing and capturing other animals (mobile food). All of these can play a part 8 Contributions in Science No. 182 in determining the degree of adaptation for aquatic locomotion in the evolu- tion of semi-aquatic forms. One other important factor remains which has a basic influence on the selective requirements related to the five functions just listed: namely, the nature of the aquatic environment. Ability to swim in still water requires less in the way of morphological specialization than does the ability to swim in a strong current (function 1 or 4 above), and if forced locomotion (functions 2 and 5) is required, then increased specialization becomes a necessity under both types of aquatic conditions. Escape by swim- ming from predators that do not swim puts more stress on the ability to get into the water rapidly than it does on swimming. The high degree of hydro- dynamic specialization in Rheomys underwoodi appears to have evolved as an adaptation of this semi-aquatic rodent for unforced movement, escape from predators and, probably most significantly, for obtaining moving food in the sometimes torrential streams along which it lives. The adaptive features re- lated to these factors doubtless have survival value, as well, when the streams rise rapidly and become swifter and more turbulent during heavy rains (function 3). The types of food (discussed in more detail by Bauchot and Stephan, 1968) and the type of aquatic environment appear to be the most important factors in the evolution of the semi-aquatic radiation among small mammals. Most semi-aquatic herbivorous mammals occur in level or lowland and coastal regions in which graminoid and low herbaceous vegetation forms a major source of food and where marshes, ponds and river overflows are to be found but rapid streams are unusual. In such situations, water edge species probably took to water, mainly, in order to take advantage of the additional (similar) plant food thus made available and also to survive during periods of flooding. In this category apparently fall most of the semi-aquatic rodents, of which many are little specialized aquatically; others show varying degrees of spe- cialization, but all are still limited by the importance of terrestrial locomotion as a means of dispersal and of relocation during times of change in their aquatic habitats. Among those showing little or no specific adaptations for aquatic loco- motion, but which more or less routinely enter water and swim, can be men- tioned species of Thryonomys (Thryonomyidae) , Bandicota, Dasymys, Hy- bomys (Muridae), Microtus, Oryzomys and Reithrodontomys (Cricetidae) , to name only a few in just these three families (see Walker, 1968; Fisler, 1961, 1965). Species of other genera have evolved definite aquatic locomotor spe- cializations and apparently utilize more of the food occurring in the water: the oryzomine Nectomys, the sigmodont Holochilus, and the microtine Arvicola are obvious examples. Those genera (such as Hydrochoerus, Myocastor, Ondatra, and Neofiber), which have become more dependent on semi-aquatic or aquatic plant food, have evolved more aquatic locomotor specializations and are capable of swimming in relatively slow and steady currents present in small streams and rivers. It seems significant that these last mentioned genera are among the larger rodents, suggesting that there is a relationship between 1970 Adaptations of a Water Mouse 9 the size of the animal and the ability to enter larger bodies of water and stronger currents where more sustained swimming is required. As must be the case with all aquatic and semi-aquatic mammals, their respective degrees of specialization have evolved in relation to the most severe aquatic conditions (currents, distances to be covered in water) in the bodies of water which they frequent or inhabit. Many of the rodents mentioned may be found in a variety of aquatic situations, but few, if any, characteristically occur in water with rapid currents. None feeds to any extent on active aquatic food although some do utilize inactive aquatic animals at least occasionally. The water opossum, Chironectes, seems to be comparable with this group of rodents in terms of degree of aquatic specialization, but it also fits in with the next group of small mammals in that it both feeds on aquatic animals and probably evolved in a forest environment. Apparently the semi-aquatic insectivores and the more highly specialized semi-aquatic rodents have evolved, for the most part, in hilly or mountainous regions. Here forest is more prevalent than open grassland or other vegeta- tional types which provide much in the way of leafy food, and moving water characterizes the aquatic habitats (streams and small rivers) . Some of these conditions, and/or other coincident factors, have apparently been favorable to insectivore radiations, and the presence of good sources of food along the water edge and in the water itself has provided the setting for the evolution of the semi-aquatic members of the order. The more highly specialized insecti- vores already mentioned are characteristically found associated with streams or small rivers, usually in forested areas. The other known semi-aquatic in- sectivores show somewhat wider and more variable ecological characteristics but probably have evolved under similar conditions. Potamogale, one of the largest insectivores, is unique among semi-aquatic members of the order in having a highly specialized tail as its sole locomotor organ, a feature which probably makes it one of the most efficient swimmers of the group (more nearly comparable to the otters) . This African tenrecid is found in all types of water, from lakes to torrential mountain streams. The remaining semi- aquatic members of this order are not highly specialized for aquatic locomo- tion showing only relatively slight modifications of the terrestrial locomotor apparatus. They do not possess swimming abilities comparable to those of the insectivores already mentioned. The relatively large hind feet, long tail and more streamlined body of the star-nosed mole, Condylura, (as compared to typical moles) are of positive value in aquatic locomotion, if not direct spe- cializations for this function, as well may be the less fossorially modified fore limb of this animal. The water shrews belonging to the soricid genera Neomys and Chimarrogale and species Sorex palustris and S. bendihi probably can be considered ecological equivalents (roughly) in their respective separate ranges of distribution, having approximately the same morphological adaptations (fringed feet and tail) . They appear to be most similar in aquatic adaptations to the less specialized species of the Neotropical rodent genera Holochilus and Nectomys. Finally, Micropotamogale lamottei, characteristically found in 10 Contributions in Science No. 182 slower waters, seems to be morphologically adapted for aquatic locomotion to about the same extent as the water shrews although it is larger and shows more aquatic specializations. The insectivores mentioned previously as being adaptively comparable to Rheomys underwoodi are more highly specialized for aquatic locomotion and are, in fact, more aquatic than those just discussed, with the exception of Potamogale, and they characteristically live in rivers and rapid to torrential streams. Their adaptations have evolved in response to the requirements im- posed by more or less active pursuit of food in the currents and turbulence of the aquatic environment. In comparison with this last group of insectivores, rodents living in regions characterized by forested streams and rivers have less potential food available in and around the water if they depend on plant sources. Those entering the streams have had to assume a diet of moving animal food (most likely before taking to the water), the capturing of which, under the aquatic conditions presented by the active waters, has been a significant factor in the evolution of their locomotor specializations. Two groups of rodents have shown some- what similar radiations in such situations: the Neotropical ichthyomyine cricetids and the New Guinea- Australian hydromyine murids. The ten genera of hydromyines (with some 16 species) are diversified and include small mouse and shrew-like forms (Microhydromys, Parahydro- mys) and other apparently terrestrial herbivores, as well as three genera which have taken to the water and include animals in their diets. These three, Xeromys, Hydromys and Crossomys, have followed different lines of speciali- zation representing the extremes of the spectrum of aquatic modifications found in other groups of rodents (Lidicker, 1968; Mahoney, 1968; Tate, 1951). Xeromys, the only endemic Australian genus of hydromyines, feeds on aquatic vegetation and inactive aquatic mollusks in swampy situations and has little, if any, apparent morphological adaptations for aquatic locomotion. The more widely distributed genus, Hydromys, represents a more muskrat-like type of semi-aquatic specialization, with the widespread H. chrysogaster being associated with a wide range of aquatic environments and feeding on a variety of animal and plant foods. Finally, the monotypic New Guinean genus Crossomys, found along mountain streams and very likely feeding on aquatic animals, appears to be most similar to Rheomys underwoodi in ecological characteristics and in both degree and detail of morphological adaptation for aquatic locomotion. The five genera of ichthyomyines are less diversified than the hydro- myines, rather they show differing degrees of specialization for life along mountain streams. Hershkovitz (1962) proposes that the ichthyomyines are derived from pastoral forms, primarily on the basis of dentition; we believe that, in any case, the ancestral forms most likely had to enter the forest to be exposed to the stream habitat, since gallery forest, at least, is almost always associated with permanent streams in the New World Tropics. The monotypic genera Neusticomys and Daptomys exhibit only slight morphological modifi- 1970 Adaptations of a Water Mouse 11 cation for aquatic habits, whereas the various species (16 named forms; eight to 14 species, depending on the authority consulted) of Rheomys and Ichthy- omys carry aquatic locomotor modifications to varying extents of specialization (Cabrera, 1961; Goodwin, 1959; Handley and Mondolfi, 1963; Hooper, 1968). Anotomys (monotypic), Rheomys imderwoodi, probably R. mexicanus (according to the informative description provided by Goodwin, 1959) and several other species of Rheomys (Hooper, 1968) and of Ichthyomys (An- thony, 1929) appear to have attained more or less similar high levels of aquatic locomotor adaptation, along several parallel evolutionary pathways. As already pointed out by Hooper (1968), the other ichthyomyine species found in Costa Rica, Rheomys hartmanni, shows less specialization for aquatic locomotion than does R. imderwoodi. The LACM specimen of R. hartmanni (LACM 25418) from north of San Isidro del General, in the Cordillera de Talamanca (Hooper’s 5200-ft. station, Rio Buena Vista tributaries, Prov. San Jose: Hooper, 1968) shows the lesser degree of specialization of the fur, ears, hind feet and tail indicated by Hooper (1968) and, furthermore, shows much less (negligible) webbing of the hind feet, and a tail which appears to be round and to lack the ventral hair fringe. This specimen also lacked the specialized tail muscle-tendon complex described above in R. imderwoodi. Differences such as those between Rheomys imderwoodi and R. hart- manni, which have been outlined, are not always readily explainable. A more thorough interpretation of the evolution of the adaptations of small mammals to the semi-aquatic habit awaits a detailed knowledge of these adaptations. Most references we have seen have been taxonomic in nature and have dis- cussed mainly diagnostic or comparative features of the mammals described, presenting little useful information on other aspects of their morphology. More ecological data, such as kinds of food consumed and detailed characteristics of habitats where the animals are found, along with information on behavioral adaptations, will provide bases for interpreting the significance of morphologi- cal specializations. Acknowledgments The LACM specimens and original field data reported in this paper were collected in 1963 and 1966 during work carried on in Costa Rica as part of the Los Angeles County Museum of Natural History mammalian ectoparasite project (LACM-USACR; see Starrett and Casebeer, 1968) under U.S. Army Medical Research and Development Command grants No. DA-MD-49-183- 62-G54 and DA-MD-49-193-63-G94 (F. S. Truxal and C. A. McLaughlin, Jr., principal investigators). It should also be mentioned that both LACM specimens of Rheomys were previously recorded in relation to this project by Geest and Loomis (1968). In addition, in 1966, facilities and various services were provided by The Organization for Tropical Studies (O.T.S.) and the Universidad de Costa Rica, San Jose. The authors wish to thank, in particular, Norman J. Scott, Jr., O.T.S. , for his help. We also here express our apprecia- tion to Emmet T. Hooper and James H. Brown for their invaluable courtesy in loaning traps to us and providing information and advice on several occasions. 12 Contributions in Science No. 182 Resumen Un macho adulto de Rheomys underwoodi fue capturado vivo, en Abril, 1966, en la orilla del Rio Poasito, a 2000 m en el Volcan Poas, Provincia de Alajuela, Costa Rica. En captividad, este animal comio sardinas de lata vorazmente, pero reuso varios tipos de plantas de la region de captura que le fueron ofrecidas. En general, sus acciones al moverse en la trampa, como limpiandose y comiendo, fueron rapidas y parecidas a las acciones de musa- ranas. Basado en observaciones limitadas de este roedor semi-acuatico en el agua de un recipiente restringido, su comportamiento mostro tambien acciones rapidas al entrar y salir del agua, nadando en la superficie o zambullendose. La propulsion era efectuada por movimientos alternados de las anchas patas traseras. Entre las especializaciones morfologicas adaptadas a la locomocion acu- atica, que este raton presenta, las siguientes ban sido notadas: un cuerpo hidrodinamico del cual solo las patas traseras y la cola protruden durante la natacion, la cola ligeramente compresa longitudinalmente con arreglo complejo de tendones para su control; pies traseros grandes con tela inter- digital y crestas de pelos que dan una superficie total de cada pie de aproxi- madamente 3.9 cm-. Ademas, las patas traseras pueden voltear y doblarse de tal manera que presentan una estructura de “hydrofoil,” con una resistencia minima contra el agua durante el movimiento hacia adelante (de recobro) al nadar. Finalmente, en una forma general y especulativa, una discusion de la evolucion de habitos semi-acuaticos en pequenos mamiferos es presentada. Dos cursos principales se pueden proponer: 1) en areas de tierras bajas, donde el habitat acuatico consiste principalmente de aguas quietas o lentamente corredizas y la vegetacion acuatica es abundante, los pequenos mamiferos semi-acuaticos son en su mayoria roedores que comen, principalmente, plantas, y que tienen tendencia a producir especializaciones morfologicas limitadas para locomocion acuatica; 2) en areas de montana bien forestadas, donde los arroyos y riachuelos tienen aguas rapidas que contienen mas animates ac- uaticos que plantas acuaticas, los habitos semi-acuaticos se han desarrollado en varios insectivoros y algunos grupos de roedores que han podido adaptarse a una dieta de pequenos animates. Rheomys underwoodi, asi como otros roedores ichthyomyines Neotropicales y los roedores hydromyines, Crossomys, de Nueva Guinea, pertenecen al ultimo grupo y han desarrollado un alto grado de especializacion morfologica a la locomocion en aguas rapidas, a veces tor- rentosas, en que ellos viven. 1970 Adaptations of a Water Mouse 13 Literature Cited Anthony, H, E. 1921. Preliminary report on Ecuadorean mammals. No. 1. Amer. Mus. Nov., 20: 1-6. . 1923. Preliminary report on Ecuadorean mammals. No. 3. Amer. Mus. Nov., 55: 1-14. . 1929. Two new genera of rodents from South America. Amer. Mus. Nov., 383: 1-6. Bauchot, Roland, and Heinz Stephan. 1968. Etude des modifications encephali- ques observees chez les insectivores adaptes a la recherche de nourriture en milieu aquatique. Mammalia, 32: 228-275. Cabrera, A. 1961. Catalogo de los mamiferos de America del Sur. II. Buenos Aires, Mus. Argent. Cienc. Nat. “Bernardino Rivadavia,” Rev., Cienc. Zool., 4: 309-732. Conaway, C. H. 1960. The water walker. Nat. Hist., 69(6) : 21-25. Davis, D. Dwight. 1964. The giant panda, a morphological study of evolutionary mechanisms. Eieldiana: Zool. Mem., 3: 1-339. De Winton, W. E. 1896. On some mammals from Ecuador. Zool. Soc. Lond., Proc., pp. 507-513. Dickey, Donald R. 1928. Five new mammals of the rodent genera Sciurus, Ortho- geomys, Heteromys, and Rheomys, from El Salvador. Biol. Soc. Wash., Proc., 41: 7-14. Enders, Robert K. 1938. A new rodent of the genus Rheomys from Chiriqui. Acad. Nat. Sci. Phila., Proc., 90: 295-296. Fisler, George F. 1961. Behavior of salt-marsh Microtus during winter high tides. J. Mammal., 42: 37-43. . 1965. Adaptations and speciation in harvest mice of the marshes of San Francisco Bay. Univ. Calif. Publ. Zool., 77 : 1-108. Geest, Julius C., and Richard B. Loomis. 1968. Chiggers of the genus Pseiido- schoengastia (Acarina: Trombiculidae) from Costa Rica. Los Angeles Co. Mus. Nat. Hist., Contrib. Sci., 150: 1-49. Goldman, Edward A. 1912. New mammals from eastern Panama. Smiths. Misc. Coll., 60 (2): 1-18. . 1920. Mammals of Panama. Smiths. Misc. Coll., 69(5): 1-309. Goodwin, George G. 1959. Descriptions of some new mammals. Amer. Mus. Nov., 1967: 1-8. Guth, C., H. Heim de Balsac, and M. Lamotte. 1959. Recherches sur la morpholo- gie de Micropotamogale lamottei et 1’evolution des Potamogalinae. I. Ecologie, denture, anatomie cranienne. Mammalia, 23 : 423-447. Handley, Charles O., Jr., and Edgardo Mondolfi. 1963. A new species of fish- eating rat, Ichthyomys, from Venezuela (Rodentia Cricetidae) . Acta Biol. Venezuelica, 3: 417-419. Hershkovitz, Philip. 1944. A systematic review of the Neotropical water rats of the genus Nectomys (Cricetinae). Univ. Mich., Mus. Zool., Misc. Publ., 58: 1-101. . 1955. South American marsh rats, genus Holochilus, with a summary of sigmodont rodents. Eieldiana: Zool., 37: 639-688. 14 Contributions in Science No. 182 . 1962. Evolution of Neotropical cricetine rodents (Muridae) with special reference to the Phyllotine group. Fieldiana; ZooL, 46: 1-524. Hooper, Emmet T. 1947. Notes on Mexican mammals. J. Mammal., 28: 40-57. . 1968. Habitats and food of amphibious mice of the genus Rheomys. J. Mammal., 49: 550-553. Howell, A. Brazier. 1930. Aquatic mammals. Chas. C. Thomas, Baltimore 338 p. Lidicker, W. Z., Jr, 1968. A phylogeny of New Guinea rodent genera based on phallic morphology. J. Mammal., 49: 609-643. Lonnberg, Einar. 1921. A second contribution to the mammalogy of Ecuador, with some remarks on Caenolestes. Ark. for Zook, 14(4) : 1-104. Mahoney, J. A, 1968. Baiyankamys Hinton, 1943 (Muridae, Hydrornyinae) a New Guinea rodent genus named for an incorrectly associated skin and skull (Hydro- myinae, Hydromys) and mandible (Murinae, Rat t us). Mammalia, 32: 64-71. Malzy, P. 1965. Un mammifere aquatique de Madagascar: le limnogale. Mammalia, 29: 400-411. Peyre, a. 1956. Ecologie et biogeographie du desman {Galemys pyrenaicus G.) dans les Pyrenees frangaises. Mammalia, 20: 405-418. Starrett, Andrew, and Richard S. Casebeer. 1968. Records of bats from Costa Rica. Los Angeles Co. Mus. Nat. Hist., Contrib. Sci., 148: 1-21. Stirton, R. a. 1944. Tropical mammal trapping I: the water mouse Rheomys. J. Mammal., 25: 337-343. SviHLA, Arthur. 1934. The mountain water shrew. Murrelet, 15: 44-45, Tate, G. H. H. 1951. Results of the Archbold Expeditions. No. 65. The rodents of Australia and New Guinea. Amer. Mus. Nat. Hist., Bulk, 97 : 183-430. Thomas, Oldfield. 1893. On some mammals from central Peru. Zook Soc. Lond., Proc., pp. 333-341. . 1897. Descriptions of four new South-American mammals. Ann. Mag. Nat. Hist., (6) 20: 218-221. . 1906a. A new aquatic genus of Muridae discovered by Consul L. Soder- strom in Ecuador. Ann. Mag. Nat. Hist., (7) 17: 86-88. . 1906b. A third genus of the Ichthyornys group. Ann. Mag. Nat. Hist., (7) 17: 421-423. . 1924a. On a new fish-eating rat from Bogota. Ann. Mag. Nat. Hist. (9) 13: 164-165. . 1924b. A new fish-eating rat from Ecuador. Ann. Mag. Nat. Hist. (9) 13: 541-542. Walker, Ernest P. 1968. Mammals of the World. 2nd Ed., rev. (John L. Paradiso, ed.), Vols. I and II. Johns Hopkins Press, Baltimore, 1500 p. Accepted for publication December 1, 1969 contributions = IN SCIENCE Number 183 April 17, 1970 OBSERVATIONS AND RECORDS OF MYOTIS (PIZONYX) VIVESI MENEGAUX (Chiroptera: Vespertilionidae) By Donald R. Patten and Lloyd T. Findley Los Angeles County Museum of Natural History Los Angeles, California 90007 Exposition Park CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor OBSERVATIONS AND RECORDS OF MYOTIS (PIZONYX) VIVESI MENEGAUX (Chiroptera: Vespertilionidae) By Donald R. Patten' and Lloyd T. Findley- Abstract: A large number of Myotis (Pizonyx) vivesi Menegaux were observed at a minimum distance of four nautical miles from land from the research vessel, Velero IV, trawling in 800 fathoms of water in the Canal de Salsipuedes, Gulf of Califor- nia, at a speed of four knots. Bats accompanied the vessel for a minimum of three hours flying into a headwind of 16 knots. Flight patterns in relation to the vessel and apparent feeding behavior are described. Contents of the intestinal tracts of three specimens included finely masticated muscle tissue, fish scales, at least one piece of a caridean shrimp and a structure thought to be the ab- domen of an unidentified polychaete larva. New records for Myotis vivesi are Isla de Alcatraz, Sonora, and Bahia Rosario, Baja California Sur, the latter being the southernmost distribu- tional record. Detection of localized surface disturbances of the water are discussed as a possible means of food detection. The corres- pondence between the distribution of M. vivesi and areas of large scale upwellings and subsequent phytoplankton blooms in the Gulf of California and adjacent Pacific Ocean is noted. Three roundworms, genus Trichuris, the mites Spintiirnix mexicanus Rudnick and Steatonyssus leptus Radovsky and ticks of the genus Ornithodoros were found as parasites on M. vivesi. All known ectoparasite records from M. vivesi are presented. Similarity in morphological and behavioral characteristics (including piscivorous habits) between M. (Pizonyx) vivesi and other species of Myotis is noted. The taxonomy of Pizonyx Miller is discussed, and the taxon is considered a subgenus of Myotis Kaup. In the course of a midwater trawl in the Gulf of California on the evening of November 30, 1967 the research vessel, Velero IV, was nearly engulfed by a large aggregation of bats. The circumstances and observations of this incident merit reporting. Observations and distributional records: At 11:30 pm three non-gravid, non-lactating females were caught in large dip nets as they flew past the fantail and were subsequently prepared and identified as Myotis (Pizonyx) vivesi Menegaux (LACM 28273-28275). When the bats were caught, the vessel was trawling in a northwesterly ^Curator of Mammalogy, Los Angeles County Museum of Natural History. ^Assistant Curator of Fishes, Department of Biological Sciences, University of Arizona, Tucson, Arizona 85721. Formerly, Section of Ichthyology, Los Angeles County Museum of Natural History. 1 9 Contributions in Science No. 183 direction in 800 fathoms of water in the Canal de Salsipuedes (sometimes written Sal si Puedes) at a speed of four knots. The exact location was 4.2 nautical mi. W (263° Radar Target) of Isla Salsipuedes, 28° 43' 58" N, 113° 04' 00" W. Barometric pressure was 29.97 in. Hg, sea temperature was 21.1° C (70° F), and the wet and dry bulb air temperatures were 16.1° C (61° F) and 19.7° C (67.5° F), respectively. At approximately 1 1 pm a single bat was noticed when it flew up the port passageway outside the gallery. This was the first time that any of the scientists were cognizant of the situation. Personnel on the bridge, however, reported that the bats first appeared at 10 pm and were continuously present until about 2 AM of December 1 when they all left abruptly. A minimum of 17 nautical miles was traversed during this four hour time interval. At 10 pm the vessel was a minimum of four nautical miles from the closest land mass, Isla las Animas, while at 2 am Isla Partida was 4.7 nautical miles to the east, and Punta de las Animas (mainland Baja California Norte) was 2.5 nautical miles to the west. The vessel was never closer than four nautical miles to any land during the first three hours. Due to its specialized habitat, it is unlikely that M. vivesi is ubiquitously distributed throughout the islands of the Gulf of California and adjacent land areas. Maya (1968) found Pizonyx present in January, spring and summer on the following four islands in the vicinity of the Canal de Salsipuedes: Partida (= Isla Cordonosa, the type locality) , Raza, Salsipuedes and San Lorenzo. For a summary of distributional records see Maya (1968), Reeder and Norris ( 1954) , and Banks (1964) . In addition to the Canal de Salsipuedes record, there are specimens in the Los Angeles County Museum of Natural History (LACM) mammal collection from two unpublished localities as follows: Isla de Alcatraz, Bahia Kino, Sonora, Mexico (33 specimens, LACM 13109- 13140, collected by A. L. Gardner, June 25 through July 9, 1960) and Bahia Rosario, Baja California Sur, Mexico (9 specimens, LACM 28276-28286, collected by K. E. Stager, March 23, 1963) . The latter specimens were taken from under rocks on a small island at the south end of Bahia Rosario (24° 15' N, 1 10° 09' W), ca. 1 mi. SE Punta Coyote, and constitute the southernmost record for this species. Reeder and Norris (1954) reported a probable sight record of this species at sea two miles off Punta Malarrimo, Bahia de Sebastian Vizcaino, Pacific side of Baja California. Our specimens from Salsipuedes appear to be the first conclusive evidence of M. vivesi occurring at considerable distances from land. According to the ship’s log the wind was from the southeast at 6 knots at 9 PM, “calm” at 10 pm, and thereafter from the northwest at an average velocity of 16 knots. Several observers noted that the bats first appeared during the period of calm and for the remaining four hours flew alongside the vessel into a 16 knot headwind. 1970 Observations of MYOTIS (PIZONYX) VIVESI 3 Despite the absence of moonlight, observation was made possible by the vessel’s running lights, a portable spotlight, and special lights on the fantail and A-frame which supported the trawl. Extent of visibility is estimated to have been 13.7 m (45 ft.) from the vessel. No phosphorescence or biolumi- nescence was observed in the water. Approximately two hundred individuals were estimated to be visible around the fantail at any one moment. The total number is conservatively estimated to have been at least twice that number. It is not known if the same individual bats were present during the entire four hours or even if those visible at any one moment were just part of a much larger, and unobservable number. Bats were observed flying all around the vessel except directly in front of and under the bow, but they appeared to be concentrated around the stern, presumably because of the lights and wake of the vessel. Most individuals re- mained more than 2.4 m (8 ft.) from the vessel, but some bats flew diagonally across the stern of the vessel and maneuvered to avoid the superstructure of the vessel and the attempts of observers to catch them in dip nets. Those alongside the vessel appeared to fly repeatedly in an oval pattern, the long axis parallel to the ship. From the vicinity of the bow individuals would fly adjacent to the sides of the ship until reaching the stern where they would disperse over the wake of the ship, reorient into the wind, drop down to the water surface, and fly forward toward the bow. In the forward flight bats were observed just above the sea surface until reaching the vicinity of the bow where they abruptly rose in altitude. Flight speed in the sweep tow'ard the stern was discernibly faster than that toward the bow. The bats appeared to be definitely interested in the area of crests produced by the wake of the vessel as well as the choppy sea surfaces lateral to the vessel. Bats at the stern were estimated to be flying 2 in. to 1 ft. above the surface and were definitely observed to dip downward touching the water. Several observers reported that they actually saw the bats touching the water surface but could not distinguish which body part made contact. Robert J. Lavenberg states, that while operating the spotlight, he observed individual bats swoop down until just above the water surface and then alternately lower and raise their feet, making contact with the water. Lavenberg’s observation is pertinent in light of the controversy over the manner in which food is procured. Miller and Allen (1928) and Burt (1932) reasoned that the elongated, laterally compressed and recurved feet func- tioned to capture small fish or cmstacea. On the basis of several observations, Reeder and Norris (1954) stressed that the tail, uropatagium and calcar cartilages appeared to be the primary mechanism. Walker (1950) observed a V-shaped ripple mark remaining after a low swoop by a bat and believed that this was made by the animal’s chin. To our knowledge no one has recorded observing M. vivesi actually obtaining prey from the sea, although Maya 4 Contributions in Science No. 183 (1968, p. 65) observed several individuals . . with fish in their claws.” In addition to the feet, the tail and uropatagium were considered by Maya (1968) to be used in capturing prey because these were often wet and frequently bore fish scales. Food: After examination of both guano and stomach contents, Burt (1932, p. 364) found definite evidence of “. . . fish remains, chiefly scales with an occasional small fish bone.” Burt further reported (p. 363) that one of the bat stomachs was found to contain fish “. . . belonging to the sardine tribe.” On the basis of circumstantial evidence. Walker (1950) believed the diet to consist of small shrimp. He further theorized that upon approaching the water surface the white venter (atypical for the genus Myotis) served to agitate and frighten schools of crustaceans, thus facilitating their capture. Reeder and Norris (1954) recorded in stomach contents small crustaceans and, in one specimen, many insect fragments. Norris (Reeder and Norris 1954, p. 85) observed one bat in flight “. . . holding a shining little fish in its hind feet.” Of an undisclosed number of fecal pellets obtained from roosting sites. Carpenter (1968, p. 956) found “. . . 62.5 per cent contained remains of only Crustacea, 26.3 per cent contained only fish remains and 11 per cent contained remains of both.” Maya (1968) found the presence of crustaceans, fish, insects, and algae in the stomachs of 22 specimens. In droppings he found some indication of seasonal variations in the relative frequences of the above food items, but concluded that crustaceans were the major food source. Examination of the digestive tracts of our three specimens disclosed finely masticated pieces of muscle tissue, fish scales, at least one piece of a shrimp (Decapoda: Caridea) and a structure thought to be the abdomen of an unidentified polychaete larva. Analysis of the fish scales revealed that the majority were from the family Myctophidae (lantern fishes), comparing well in size, shape and other scale characters to those of the genera Hygophum, Diaphus and Triphoturus. Thir- teen of the 65 scales showed a definite, raised center portion which appeared to be the lens remnant of a myctophid photophore scale whereas 50 additional scales were non-photophore myctophid scales. Two scales were small and ctenoid and were not myctophid in origin, those of the latter being larger and cycloid. Many myctophids are vertical migrators from great depths at night when they would occur at the surface. From roosts of M. vivesi Maya (1968) found the dry and partially chewed remains of the following fishes: File stolifera, Ophisthonema liboleta [sic, Opisthonema libertate], bothids, and perhaps flying fish, with bothid remains being the most common. Adult flatfishes of the family Bothidae are benthic whereas the larvae and prejuveniles live in plankton. Scale develop- ment is probably slight, if at all, during the planktonic stages. The degree of scale development of the bothid remains mentioned by Maya (1968, fig. 15) would indicate that they were probably recently developed juveniles which 1970 Observations of MY OTIS {PIZONYX) V IV ESI 5 inhabit shallow waters near shore. LUe stolifera and Opisthouema lihertate are herrings (Clupeidae) and are surface schooling fishes. When removed from the dip net, one of the three bats was discovered to have a small unidentified shrimp adhering to the surface of the uropatagium. During the frantic efforts to capture the bats as they flew close to the sides of the vessel and across the fantail, one of the specimens was caught in a dip net which subsequently touched the water before being raised to the deck. It is not known whether or not the specimen bearing the crustacean was the one in- volved nor is it known if the shrimp was present when the bat was initially caught or if it was originally adhering to the dip net and subsequently trans- ferred to the bat. Food detection: The bats emitted audible sounds described as short, high pitched squeals, which were readily discernible. The overall intensity of sound varied in direct proportion to the height of flight, being loudest when the bats were closest to the water. No noticeable change in pitch was reported by the observers whereas Maya (1968, p. 66) states that “As they approach the surface, their vocal chatter, a series of short bursts, becomes high pitched, more rapid, and finally inaudible.” Slithers (1965) found that in captivity Noctilio leporinus Linnaeus, a neotropical bat known to be piscivorous, could readily detect small wires projecting vertically above the water as well as localized minor disturbances in the water surface. M. vivesi probably detects prey in the same manner as N. leporinus although in a subsequent study Slithers (1967) reported a differ- ence in the orientation sounds produced by these two species. He was unable to induce captive M. vivesi to take pieces of shrimp from the surface of an experimental pool. The behavior and apparent feeding forays by M. vivesi in the areas ad- jacent to the stern and sides of the vessel may have resulted from disturbances of the water by the wind and the passage of the vessel. The choppy sea pro- duced by the wind was similar to the wake of the vessel. Once agitated, any small fish or crustacean could well have produced small surface disturbances, thus facilitating their capture by the bats. Maya (1968, p. 67) mentions that M. vivesi cannot select only small fish, pointing out that “. . . many [of the fish] caught are frequently too large for effective feeding.” On another occasion he witnessed a bat “. . . hovering above and actively pursuing an object” (1968, p. 66). Maya recovered the object which was an unidentified shrimp (1968, pp. 61 and 66, fig. 16). Judging from the scale in the figure the shrimp appears about 6 cm. long and too large for effective feeding by M. vivesi. This would seem to be a logical consequence if M. vivesi was reacting to a localized surface disturbance rather than dis- criminating as to the kind and size of organism producing the disturbance. The known distribution of M. vivesi conforms in general with areas of the Gulf of California and adjacent areas of the Pacific Ocean subject to large 6 Contributions in Science No. 183 scale upwellings and subsequent phytoplankton blooms, e.g. the Canal de Salsipuedes (Parker, 1964, p. 35). Perhaps M. vivesi actively seeks out those areas where their prey species, higher trophic level consumers, would be rela- tively more abundant. Parasites: Three female roundworms of the genus Trichuris (Dorylai- mida: Trichinellidae) were found as intestinal parasites in one of our three specimens. In addition, three types of ectoparasites were found: one mite, Spinturnix mexicanus Rudnick (Acarina: Spinturnicidae) ; two protonymphs of the mite, Steatonyssus leptus Radovsky (Acarina: Macronyssidae) ; and four specimens of a tick of the genus Ornithodoros (Acarina: Argasidae), which apparently are not O. dyeri Cooley and Kohls (Glen M. Kohls, per- sonal communication). The following ectoparasites have been recorded from M. (Pizonyx) vivesi: Ornithodoros dyeri Cooley and Kohls (1940) (Acarina: Argasidae); Whartonia sonorensis Hotfmann (1960) (Acarina: Trombiculidae) ; Trom- bicula thompsoni Brennan (1966) (Acarina: Trombiculidae); Speleocola cortezi Loomis and Webb (1969) (Acarina: Trombiculidae); and the batfly, Basilia pizonychus Scott (1939) (Diptera: Nycteribiidae) . The latter has also been reported by Reeder and Norris (1954) and Maya (1968). Rupert L. Wenzel (personal communication) reports that there are no Streblidae known from M. vivesi. Whartonia sonorensis, Speleocola cortezi and Basilia pizony- chus are known only from M. vivesi. Apart from the one record on M. vivesi, Trombicula thompsoni is known only from marine birds in the Caribbean. At one time, Ornithodoros dyeri was known only from guano and rock crevices of bat caves, mines, and tunnels from several localities in the western United States and Mexico, but it is now known from both M. vivesi and the sac- winged bat, Balantiopteryx plicata (Kohls, Sonenshine, and Clifford, 1965). Because it is incomplete, current parasitological information can furnish little evidence concerning the taxonomic relationships between M. vivesi and the other species of Myotis. Taxonomy: We have followed the suggestions of Ellerman and Morrison- Scott (1951, p. 138) and subsequent investigators that Pizonyx Miller and Myotis Kaup be regarded as congeneric. These two have been shown to be essentially similar in hair structure (Benedict, 1957), baculum (Hamilton, 1949), and karyotype (Baker and Patton, 1967). Characteristics used to distinguish Pizonyx from Myotis have been summarized by Baker and Patton (1967) and are primarily associated with a piscivorous habit (Miller, 1906 and 1907). To support the generic retention of Pizonyx, Maya (1968) stressed pis- civorous behavioral adaptations and specializations in water metabolism. However, other species of Myotis are now known, or strongly suspected, to be piscivorous and exhibit many of the same specialized morphological and be- havioral characteristics formerly regarded as unique to Pizonyx. These species 1970 Observations of MYOTIS (PIZONYX) VIVESI 1 are as follows: M. {Leuconoe) megalopus (Dobson), Africa, (Aellen and Brosset, 1968); M. (Rickettia) ricketti Thomas in Asia [discussed as Ves~ pertilio pilosus Peters by Allen, 1936]; M. (Leuconoe) macrotarsus Water- house, Asia, and M. (Leuconoe) daubentoni Kuhl, Europe (Brosset and Deboiitteville, 1966). The subgeneric designations follow Tate (1941) and are distinguished by enlarged feet, Leuconoe, versus enormous feet, Rickettia, in addition to other characteristics similar to those of Pizonyx. The big-footed condition according to Tate (1941), at least in Leuconoe, is due to phalangeal elonga- tion (as in Pizonyx) and not to metatarsal elongation as stated by Maya (1968). In addition to the aforesaid species, Myotis dasycneme Boie (Europe and Asia), Myotis macrodactylus Temminck (Japan), and Myotis stalkeri Thomas (Kei Islands) also have large feet. Although a white venter is not common in the genus Myotis, it cannot be considered unique to Pizonyx. Specimens in the American Museum of Natural History of M. ricketti, M. daubentoni and M. dasycneme all possess a con- spicuous and extensive white frosting to the venter. Braun (1965) and Carpenter (1968) have shown that kidney structure and water metabolism, respectively, of M. vivesi are highly specialized, but without equivalent information for the other species of Myotis, it is difficult to interpret the taxonomic significance, if any, of this information. Surely the desert species of Myotis, if not the others, are also specialized in renal physi- ology, morphology and water metabolism. Dalquest (1948) reports that both Myotis californicus (Audubon and Backman) and Myotis yurnanensis (H. Allen) apparently drink saltwater. Whether or not the characteristics common to both M. vivesi and the several old world species indicate convergence or a common phylogenetic origin is a question yet to be resolved. Certainly the present taxonomic arrange- ment appears illogical and inconsistent, and pending a revision of all the taxa concerned, we have considered Pizonyx a subgenus of Myotis. Acknowledgments For identification and information concerning the ectoparasites we are indebted to Deane P. Furman of the Division of Parasitology, Agricultural Experiment Station, University of California, Berkeley; Glen M. Kohls, Rocky Mountain Laboratory, Hamilton, Montana; Richard B. Loomis, Department of Biology, California State College (Long Beach); Frank J. Rodovsky, George Williams Hooper Foundation, University of California, San Francisco Medical Center; and Rupert L. Wenzel, Field Museum of Natural History, Chicago. The nematodes were identified by Murray D. Dailey, Department of Biology, California State College (Long Beach). Donald B. Bright, De- partment of Biology, California State College (Fullerton) kindly identified the invertebrate material in the stomach contents. Robert J. Lavenberg, Los 8 Contributions in Science No. 183 Angeles County Museum of Natural History, Kristian Fauchald, Allan Han- cock Foundation, University of Southern California, and crew members of the Velero IV, Fred Ziesenhenne, Captain, John Falcone, Second Mate, and Fred Schmidt, Chief Engineer, contributed observations and technical data. To Andrew Starrett, Charles L. Hogue, Ronald H. Pine, James S. Findley and Karl F. Koopman we owe a special debt of thanks for suggestions con- cerning the manuscript. To all of the above individuals we express our appreciation. Literature Cited Aellen, V., AND A. Brosset. 1968. Chiropteres du sud du Congo (Brazzaville). Rev. Suisse Zook, 75: 435-458, 1 pi. Allen, G. M. 1936. The status of Vespertilio pilosus Peters. Jour. Mamm., 17; 168-169. Baker, R. J., and J. L. Patton. 1967. Karyotypes and karyotypic variation of North American vespertilionid bats. Jour. Mamm., 48: 270-286. Banks, R. C. 1964. Range extensions for three bats in Baja California, Mexico. Jour. Mamm., 45: 489. Benedict, F. A. 1957. Hair structure as a generic character in bats. Univ. Calif. Pub. Zook, 59: 285-549. Braun, E. 1965. The gross and microscopic renal anatomy of the fish-eating bat, Pizonyx vivesi. M.S. Thesis. Univ. Ariz,, 31 pp. Brennan, J. M. 1966. New records of chiggers (Acarina: Trombiculidae) from Baja California and islands of the Gulf of California. Jour. Parasitok, 52: 772-775. Brosset, A., and C. D. Deboutteville. 1966. Le regime alimentaire du vespertilion de Daubenton Myotis daubentoni. Mammalia, 30: 247-251. Burt, W. H. 1932. The fish-eating habits of Pizonyx vivesi (Menegaux). Jour. Mamm., 13: 363-365. Carpenter, R. E. 1968. Salt and water metabolism in the marine fish-eating bat, Pizonyx vivesi. Comp. Biochem. Physiol., 24; 951-964. Cooley, R. A., and G. M. Kohls. 1940. Two new species of Argasidae (Acarina: Ixodoidea). U.S. Pub. Health Rpk, 55: 925-933. Dalquest, W. W. 1948. Mammals of Washington. Univ. Kans. Pubis., Mus. Nat. Hist., 2: 1-444. Ellerman, j. R., and T. C. S. Morrison-Scott. 1951. Checklist of Palaearctic and Indian mammals, 1758 to 1946. 2 ed. Brit. Mus. (Nat. Hist.), London, 810 pp. Hamilton, W. J., Jr. 1949. The bacula of some North American vespertilionid bats. Jour. Mamm., 30: 97-102. Hoffmann, A. 1960. Contribuciones al conocimiento de los Trombiculidos Mexi- canos (Acarina, Trombiculidae) (9a. parte). Acta Zook Mexicana, 4(4): 1-11. Kohls, G. M., D. E. Sonenshine, and C. M. Clifford. 1965. The systematics of the subfamily Ornithodorinae (Acarina: Argasidae). II. Identification of the larvae of the western hemisphere and descriptions of three new species. Entomok Soc. Amer., Annals, 58: 331-364. Loomis, R. B., and J. P. Webb, Jr. 1969. A new species of Speleocola (Acarina: Trombiculidae), off a bat, Pizonyx vivesi, from Baja California, Mexico. So. Calif. Acad. Sci., Bulk 68: 36-42. Maya, J. A. 1968. The natural history of the fish-eating bat, Pizonyx vivesi. Ph.D. Thesis. Univ. Arizona. 106 pp. Univ. Microfilms. Ann Arbor, Mich. (Diss. Abstr. 69-2016). 1970 Observations of MYOTIS (PIZONYX) VIVESI 9 Miller, G. S., Jr. 1906. Twelve new genera of bats. Biol. Soc. Wash., Proc., 19: 83-86. . 1907. The families and genera of bats, U.S. Natl. Miis., Bull., 57 : 1-282. Miller, G. S., Jr., and G. M. Allen. 1928. The American bats of the genera Myotis and Pizonyx. U.S. Natl. Mus., Bull., 144: 1-214. Parker, R. H. 1964. Zoogeography and ecology of some macroinvertebrates, par- ticularly mollusks, in the Gulf of California and the continental slope off Mexico. Dansk Naturh. Foren., Vidensk. Medd., 126: 1-177. Reeder, W. G., and K. S. Norris. 1954. Distribution, type locality, and habits of the fish-eating bat, Pizonyx vivesi. Jour. Mamm., 35: 81-87. Scott, H. 1939, A new species of Nycteribiidae (Diptera Pupipara) from islands in the Gulf of California. Allan Hancock Foundation, Allan Hancock Pacific Expeditions, 2: 167-172. SuTHERS, R. A. 1965. Acoustic orientation by fish-catching bats. Jour. Exp. Zool., 158: 319-347. . 1967. Comparative echolocation by fishing bats. Jour. Mamm., 48: 79-87. Tate, G. H. H. 1941. Results of the Archbold Expeditions. No. 39. A review of the genus Myotis (Chiroptera) of Eurasia, with special reference to species occur- ring in the East Indies. Amer. Mus. Nat. Hist., Bull., 78: 537-565. Walker, L. W. 1950. The fish bats of Pescadora. Audubon Mag., 52: 294-299. Accepted for publication February 16, 1970 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE ^^UMBER 184 April 17, 1970 i CONOCARDIUM LANGENHEIMI SP. N. i; (MOLLUSCA: BIVALVIA) IN THE LOWER PERMIAN SERIES I OF THE McCLOUD LIMESTONE, NORTHERN CALIFORNIA i 1 ! I By Edward C. Wilson I ;t li I' I Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author oi divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor CONOCARDIUM LANGENHEIMl SP. N. (MOLLUSC A: BIVALVIA) IN THE LOWER PERMIAN SERIES OF THE McCLOUD LIMESTONE, NORTHERN CALIFORNIA By Edward C. Wilson^ Abstract: Conocardiiim langenheimi sp. n, (Mollusca: Bi- valvia) occurs in the Wolfcampian Stage, Lower Permian Series of the McCloud Limestone in Shasta County, California. It is a large shell with a non-carinate posterior end, three pairs of perfo- rate “shelves” within the gaping anterior end, a tube enclosed dor- sally in the shell layers of each valve, unique internal hinge structures, and microstructure unlike most other bivalves. The functions of these and other characters is unknown. The entire Order Conocardidoida needs investigation to determine the signi- ficance of such characters, especially for placing the group in higher categories of classification. Introduction At the suggestion of R. L. Langenheim, Jr., Professor of Geology at the University of Illinois, I began a study of the paleontology of the McCloud Limestone in Shasta County, California, in the spring of 1963. Two previous papers (Wilson, 1967a, 1967b) resulted from this study and this paper is another contribution. Acknowledgments I am grateful for opinions on the morphology and systematic position of the Conocardidoida from J. Wyatt Durham of the University of California at Berkeley, John Pojeta, Jr. of the U.S. National Museum, and Norman D. Newell of the American Museum of Natural History. Professor Durham kindly visited the type localities with me. Most of the photographs are by Lawrence S. Reynolds of the Los Angeles County Museum of Natural History. Field work in 1963 and 1964 was financed in part by the University of California Museum of Paleontology (UCMP) and the University of California Committee on Research, both at Berkeley. The Los Angeles County Museum of Natural History (LACM) provided funds for further field work in 1968. ^Curator of Invertebrate Paleontology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 184 SYSTEMATIC DESCRIPTION Phylum MOLLUSCA Class BIVALVIA Linnaeus, 1758 Order CONOCARDIDOIDA Neumayr, 1891 Family CONOCARDIIDAE Miller, 1889 Genus Conocardium Bronn, 1834 Remarks: Terminology has not been standardized for the unique char- acters of the Conocardidoida. I have used where possible the terminology for other bivalve mollusks, although this is partially unsatisfactory. The structures called “auricles” are probably not homologous with the auricles of other bivalves. The “shelves” in the gaping anterior “auricle” are without parallel in the rest of the Bivalvia. The openings in the “shelves” and the tubes en- closed in the shells have no formal names. The kind of “teeth” on the ventral and posterior edges should probably be named. It is not even possible to use uniformitarianism to orient the shell with certainty. It may seem reasonable that dorsal and ventral are accurately delineated, but there is no such certainty about anterior and posterior. The figures and legends indicate usage of mor- phological terms in this paper. Conocardium langenheimi, sp. n. Fig. 1 Description {external) : Valves equivalve, inequilateral, opisthogyrate, alate (?), not ankylosed; main body of valves inflated, rounded (not carinate posteriorly); posterior “auricle” small, sloping smoothly into main body of valves, not gaping; anterior “auricle” separated by constriction from main body of valves, large, rostrate, gaping anteriorly and ventrally, with large trigonal aperture having scalloped margins; hinge line straight, except anterior part of anterior “auricle” may dip to 25° (preservation?), occupied by groove, shallow in anterior part of anterior “auricle,” deep in posterior part of an- terior “auricle” and main body of shell; commissure straight on exterior of ventral parts of shells; holotype shell length 6.5 cm, height 4.3 cm, width 2.9 cm (largest specimen, LACM paratype 2443, is 4 mm. higher, but incomplete Figures 1-4. Conocardium langenheimi sp. n. UCMP holotype 10589, UCMP locality D-831, all figures X 1.3; (1) dorsal view, anterior “auricle” at left, posterior “auricle” at right, main body of shell inflated, openings in umbos caused by post- burial weathering, double wall caused by dissolution in HCl of calcareous layer between silicified layers, note absence of posterior carina; (2) ventral view, showing straight exterior commissure of main body, gape of anterior auricle, internal ridge at hinge line; (3) lateral view of left valve, sculpture on main body possibly dissolved in part during etching; (4) ventral view, with piece of shell removed to show dorsal interior, dorsal ridge along hinge line of main body broken with fragments lying in right valve, note trigonal cross sections of rdiges remaining in shell wall after etching. 1970 Conocardium langenheimi Sp. N. (Mollusca: Bivalvia) 3 4 Contributions in Science No. 184 in other dimensions) ; sculpture concentric or cancellate (?) on main body of valves, with about 4 evenly spaced concentric riblets in one mm., with possible traces of radial riblets; sculpture radial on “auricles” (UCMP paratype 10590, juvenile, shows some concentric), about 17 ribs on anterior “auricle” of each holotype valve, fewer on posterior “auricle” (UCMP paratype 10590, juvenile, has 9 ribs on each posterior “auricle”) ; ribs straight to very gently sigmoid, sloping down from hinge line at angles varying from 65° on posterior part of anterior “auricle” to 25° on anterior part, flat topped with straight to rounded sides sloping down and out at less than 45°; rib interspaces about equal to rib widths, lirate, with about 20 dorsally convex lirae in 1 mm, rarely occupied by single riblet. Description (internal) ; Main body of valves smooth; pallial line, muscle scars not observed; ventral and posterior commissure zig-zag, formed by series of trigonal “teeth” and intervening sockets in each valve, 3 occurring every 5 mm each side of both valves (holotype) , sockets showing plainly only in section (“teeth” are surficial extensions of radial rods, trigonal in cross sec- tion, apex pointing inward, that dorsally form part of the internal shell layer) ; hinge line a complex ridge, low in anterior end, becoming higher posteriorly, developing horizontal distal bar (inverted T-shaped in cross section) within main body cavity of valves, where narrow trough borders each side (marking lateral limits of apparent separate hinge plate in this area), becoming lower and broader posteriorly, troughs continuing alongside into posterior “auricle,” where termination not observed because of preservation; 3 pairs of complex “shelves” lining lateral interior of anterior “auricle,” sloping in and up same angles as exterior ribs, each pair meeting in midline, dividing “auricle” roughly horizontally into 4 unequal areas; dorsal pair of shelves beginning about 15 mm back of “auricle” anterior edge, about 8 mm below top of “auricle,” form- ing large heart shaped opening in cross section above “shelves,” turned up along midline by lip-like ridge, sloping upward, separating along midline below Figures 5-9. Conocardium langenheimi sp. n. (5) holotype, anterior view into gaping “auricle” showing 3 pairs of “shelves,” X 1.3; (6) same specimen, posterior view, umbos partially weathered away, note zig-zag commissure of “auricle,” straight com- missure of main body, X 1.3; (7) same specimen, same view as figure 5 with part of auricle removed, missing parts of wall dissolved during etching, note exterior groove and interior ridge of hinge line with grooves bordering latter, dorsal “shelves” meet- ing along midline with opening into main body visible just above, middle shelves similar but extending anteriorly to form keyhole shaped opening, partly obscured small ventral “shelves” shaped like middle “shelves,” X 3.6; (8) LACMNH paratype 2443, LACMNH locality 1 133, fragment showing zig-zag interior ventral commis- sure formed of series of teeth and sockets, note left valve showing original smooth interior, right valve showing etched layer of shell wall with internal trigonal ridges which terminate ventrally as teeth, X 0.6; (9) UCMP paratype 10590, UCMP local- ity D-831, lateral view of right valve of juvenile specimen, note concentric sculpture on main body and anterior auricle, offset posterior “auricle,” outline of anterior “auricle” varying from figure 3, X 1.4. 1970 Conocardium langenheimi Sp. N. (Mollusca: Bivalvia) 5 6 Contributions in Science No. 184 hinge line, leaving opening here into main part of shell, shelves continuing dorsally and posteriorly with lips incurving to form tubes passing into shell layers before leaving “auricle,” tubes continuing posteriorly in shell above main body of shell, lateral to hinge line, becoming closer posteriorly and apparently dropping into external groove above hinge line, with posterior course un- known due to preservation (no apparent natural external opening observed in posterior “auricle” or hinge line); middle “shelves” also large, paralleling dorsal ones, similar in shape and slope, but extending to anterior edge of “auricle,” forming key-hole shaped opening in anterior ends as well as heart- shaped opening posteriorly into main body of shell, with anterior lips as in dorsal “shelves,” extending into grooves in “auricle” walls under dorsal “shelves” posteriorly, ending at posterior termination of dorsal “shelves,” without continuing internal tube; ventral “shelves” small, shaped like middle “shelves.” Description {microstructure) : Preservation very imperfect, shell par- tially silicified, some layers separated by silica, some missing (erosion?) in places; shell calcium carbonate, 1.5 to 3.5 mm thick, with 3 layers throughout plus 3 highly restricted layers; exterior layer 0.33 to 1.0 mm thick, prismatic, distal crystal tips reflected ventrally, elsewhere normal to shell surface; middle layer 0.2 to 0.3 mm thick (generally replaced by silica and difficult to measure) , lamellar, showing ventrally convex concentric growth lines crossed by finer radial lines, separated in places from prismatic layer by dark line (conchio- lin?); inner layer 2 to 3 mm thick, much wider than other layers, formed of large calcium carbonate crystals (apparently recrystallized), showing residual (?) growth lines in some areas, etching in dilute hydrochloric acid in main body of shell into trigonal ridges (see above), evidence of which not observed in polished or thin sections; myostracum (?) in 3 bands in each valve, extend- ing from opposed axial edges of “shelves” to umbos, straight to V-shaped in Figures 10-18. Conocardium langenheimi sp. n. UCMP paratype 10591, UCMP locality D-831, specimen ground down from anterior to posterior to show successive cross sections, all figures X 2.2; (10) section near anterior of anterior “auricle,” note great shell thickening (left) from thin anterior edge (right); (11) dorsal “shelves” begin to form, note rod like myostracum (?) protruding from “shelves,” internal scalloped outline of outer shell layers; (12) three pairs of “shelves” forming, note deepening grooves between dorsal and middle “shelves”; (13) three pairs of “shelves” formed here, each with associated myostracum (?); (14) upper “shelves” move dorsally, curl inwards, middle “shelves” pass into two grooves, lower “shelves” have disappeared, internal dorsal ridge along hinge line begins to lengthen; (15) dorsal “shelves” move dorsally, other “shelves” disappear (groove extending from middle “shelf” on right still present), rods of myostracum (?) move dorsally in shell wall, teeth and sockets appear ventrally marking ventral beginning of main body of valves; (16) dorsal ridge lengthens, pre-burial erosion noticeable on dorsal edge of valve at left; (17) dorsal ridge lengthens; (18) dorsal ridge lengthens, dorsal edge of eroded valve thins, dorsal “shelves” move higher, accompanied by their myostracum (?) rods and those from two lower “shelves.” 1970 Conocardium langenheimi Sp. N. (Mollusca: Bivalvia) 7 8 Contributions in Science No. 184 cross section (varies), about 0.2 mm thick, apparently exposed at “shelf” edges (indicates muscles joining opposing “shelves”?), showing concentric and radial lines as in middle layer. Name: This species is named for R. L. Langenheim, Jr. in recognition of his contributions to our knowledge of the paleontology and stratigraphy of North America and for his inspiration and training of many students in this work. Type specimens: Holotype UCMP 10589, paratypes UCMP 10590- 10591, paratype LACM 2443, all UCMP locality D-831 (=LACM locality 1133);paratype UCMP 10592, UCMP locality D-819. Localities: UCMP locality D-819. 1140 feet stratigraphically above the lowest exposure of the McCloud Limestone on the first spur south of the eastern end of the McCloud River bridge as shown in the northwest 14 of the southwest Vat of section 32, township 35 north, range 3 west of the United States Geological Survey topographic quadrangle (15 minutes, 1957) of Bol- libokka Mountain, Shasta County, California. Series of one foot high ledges of limestone, fine to medium grained, black to medium gray, weathering medium to light gray; little chert; many silicified fossils, including large cani- noid corals, Conocardium, Omphalotrochus, fusulinids. UCMP locality D-831 (= LACM locality 1133). Estimated 200 yards south of locality D-819 along strike and about 15 feet higher in the section. Narrow bench of thin-bedded, black, silty limestone below a sheer massive cliff. Abundant syringoporid corals, very large solitary rugose corals, fusu- linids, bryozoans, Conocardium. Age: The fusulinids have the most refined stratigraphic ranges of the fossils found with Conocardium langenheimi. Two species were identified: Pseudofusulina soluta Skinner and Wilde, 1965 and Triticites mulleri Skinner and Wilde, 1965, both having type localities in the McCloud Limestone nearby the Conocardium localities. Skinner and Wilde (1965, pi. 4) considered the former fusulinid to indicate definite Permian age (the genus is a Wolfcampian index fossil) and the latter to be of either Pennsylvanian (Virgilian) or Per- Figures 19-24. Conocardium langenheimi sp. n. Continuation of sections of same specimen as figures 10-18, same magnification; (19) dorsal “shelves” have incurved and passed into tubes in shell wall, marking dorsal beginning of main body of valves, myostracum (?) rods now wholly within shell wall, dorsal ridge greatly lengthened and now having horizontal distal end (not clear in this photograph but see fig. 34), possibly developing into separate plate along hinge line, with lateral grooves inter- nally, midline groove externally, tube at left opened by erosion; (20) dorsal tubes narrow, note ventral teeth showing how ventral commissure can be exteriorly straight yet interior zig-zag, also shows shell not ankylosed; (21 ) dorsal tubes narrow, dorsal ridge shortens; (22) dorsal tubes narrow and move toward midline, dorsal ridge shortens; (23) dorsal tubes narrow, valve at left badly eroded; (24) dorsal tubes smaller, dorsal ridge shorter, dorsal part of valves move ventrally, deepening external groove. 1970 Conocardium langenheimi Sp. N. (Mollusca: Bivalvia) 9 10 Contributions in Science No. 184 mian (Wolfcampian) age. It seems clear, therefore, that the interval of the McCloud Limestone from which C. langenheimi was collected is definitely occupied by Lower Permian Series, Wolfcampian Stage rocks. Diagnosis’. C. langenheimi can be separated readily from most other species because it lacks a carina on the posterior part of the main body of the shell. It is also larger than most species and the anterior “auricle” is propor- tionately longer in relation to the rest of the shell than in many other species. Discussion: C. langenheimi resembles closely only C. ouralicum (Ver- neuil, in Murchison, Verneuil, and Keyserling, 1845, p. 301, pi. 20, figs. 11a, 1 lb) from Upper Carboniferous rocks of the Ural Mountains. Both species lack posterior carinae and have elongate anterior “auricles.” However, C. ouralicum is smaller (length 36 mm) and has an anterior “auricle” that is much more elongate in proportion to its height than that of C. langenheimi. Internal structures and shell microstructure of C. ouralicum are unknown as yet. The microstructure of C. /ang^n/z^/m/ apparently lacks “roofed-over pits” like those figured by Pannella and MacClintock (1968, pi. 8, fig. 5) in C. sp. from Upper Pennsylvanian rocks of Oklahoma. The “outer complex-prismatic shell layer” shown by Pannella, MacClintock, and Thompson (1968, fig. 1 A) of C. herculeum Konick from Lower Carboniferous rocks of Belgium re- sembles a similarly positioned layer in C. langenheimi, but lacks the reflected tips that are present in the latter species. The presence of complex internal structures in Conocardium was early reported by Hind (1900, p. 451, pi. 51, fig. 11a) in some detail, although he apparently worked from specimens somewhat worn internally. Curiously, no further refined work on them seems to have been published, although Decha- seaux (1952, p. 319, fig. 162a) and Branson, LaRoque, and Newell (1969, p. 859, fig. Gl) cursorily discussed and figured some internal characters. Future workers should examine the internal structure of specimens of Con- ocardium with great care. I have tried to find clues that would help determine the habitat and functional morphology of C. langenheimi. The associated fossils show that the salinity of the water must have been whatever was normal for the open oceans in Permian time. Permian near-shore faunas with restricted access to the sea in western North America generally lacked corals and fusulinids, both of which Figures 25-29. Conocardium langenheimi sp. n. Continuation of sections of same specimen as figures 10-24, same magnification; (25) dorsal ridge shorter but hori- zontal expansions of distal end still present, dorsal tube openings smaller (not seen in valve at left), some of dorsal part of valve at right shows erosion; (26) dorsal tube moves inward and downward, dorsal ridge shortens, distal expansions disappear, erosion of both valves and distortion of valve at left apparent; (27) continuation of movements seen in figure 26; (28) continuation, myostracum (?) rod (dark chevron) shows beside eroded tube in valve at right; (29) dorsal tubes move ventrally, hinge line plate broadens. 1970 Conocardium langenheimi Sp. N. (Mollusca: Bivalvia) 11 12 Contributions in Science No. 184 are abundant at the C. langenheimi localities. The shell itself is sturdy, es- pecially the thick walled anterior “auricle,” which suggests exposure to some strong current or wave action. Seemingly eroded parts of some of the speci- mens indicates that they may have been moved by currents from their original habitats. However, some fine silty layers in the limestone indicate that sedimen- tation was generally not rapid. None of the specimens shows evidence of having been buried in a burrow, which seems to me the most likely habitat by analogy with Recent bivalves of the most similar external form, size, and sturdiness (Pholadidae). Branson (1969, p. 860, figs. 3a, 3b, in Branson, LaRoque, and Newell) figured, but did not discuss, a restoration of a Cono- car dium apparently occupying a burrow, but having the anterior “auricle” innermost, which is unlike the Pholadidae. Nicol (1970, p. 70) also compared Conocardium with Recent bivalves, but reached the different conclusion that it “was attached by a short, stout byssus . . .” All the specimens of C. langen- heimi appeared to be lying randomly on surfaces parallel to the bedding planes. They are all tightly closed pairs, which is unlike other Bivalvia, if one assumes because of the eroded areas that they were buried after death. Reconstruction of the soft part morphology of C. langenheimi will not be simple. I have been unable to create a satisfactory model for incurrent and excurrent siphons. The complex form of the “shelves” indicates that some extremely specialized soft part morphology was present. The apparent absence of muscle scars in the anterior end may be explained by the existence of muscles between the edges of opposing “shelves,” as suggested by the myo- stracum (?) exposed there. One would expect posterior muscles also, which seemingly are not represented by muscle scars or a posterior myostracum. Perhaps the function of the unique hinge line structures and the ventral and posterior “teeth” rendered posterior muscles unnecessary. The internal ridge along the hinge line looks greatly like a support, but one wonders for what, since the other bivalves seem to function well without such a rigid support for the ctenidia, alimentary canal, or other systems. It is tempting to postulate a Figures 30-34. Conocardium langenheimi sp. n. (30) continuation of sections of same specimen as figures 10-29, same magnification, dorsal tubes drop into exterior grooves bordering hinge line, dorsal ridge disappears; (31) same specimen, same magnification, dorsal tubes not present (preservation?), external grooves coalesce, with narrowed opening above, hole in shell at left caused by erosion, shell not pre- served posterior to this section and nature of structures there unknown; (32) thin section made from polished section shown in fig. 11, note especially exposed end of myostracum (?) at edge of “shelf” at left and indeterminate nature of hinge articula- tion, X 3.2; (33) thin section made from polished section shown in fig. 31, showing nature of hinge and apparent absence of dorsal tubes seen in figs. 19-30, X 3.2; (34) thin section made from polished section shown in fig. 21, showing dorsal tubes, myostracum (?) layers, elongate internal hinge line ridge with horizontal distal bar, exterior ligament (?) groove, and apparent separate nature of plate along hinge line, X 3.2. 1970 Conocardium langenheimi Sp. N. (Mollusca: Bivalvia) 13 14 Contributions in Science No. 184 function such as brood chambers to the tubes in the dorsal parts of the valves, but, again, close analogies in living bivalves seem absent. Some of the grooves in the shell suggest ciliated tracts, but there is no certainty about this. Conclusions C. langenheimi is a member of the Conocardidoida, which is externally like the Bivalvia, has a world wide distribution, and ranges from Ordovician through Permian (Triassic?) rocks. The degree of complexity shown by the shell microstructure definitely appears to be molluscan. The paired valves sug- gest that the Order belongs in the Class Bivalvia, however the internal mor- phology and shell microstructure is unlike others in that class. If typical of the order, these suggest need for a reconsideration of the systematic position at the class or higher level. Literature Cited Branson, C. C., A. LaRocque, and N. D. Newell. 1969. Order Conocardioida, in Treatise on invertebrate paleontology, R. C. MOORE, ed., Bivalvia: Part N. Geol. Soc. America and Univ. Kans. Press, Lawrence, Kansas. 2:491-951, figures. Dechaseaux, Colette. 1952. Etude systematique, in Traite de Paleontologie, J. PIVETEAU, ed. Masson et Cie., Paris. 2:261-364, figs. 35-215. Hind, W. 1900. A monograph of the British Carboniferous Lambellibranchiata, part V. Paleont. Soc., London. 361-476, pis. 40-54. Linnaeus, Carolus. 1758. Systema naturae. Editio decima. Laurentii Salvii, Stock- holm. 1-824. Miller, S. A. 1889. North American geology and palaeontology for the use of amateurs, students, and scientists. Western Methodist Book Concern, Cincinnati. 664 pp., 1 194 figs. Murchison, R. I., E. De Verneuil, and A. De Keyserling. 1845. Geologic de la Russie d’Europe et des Montagnes de I’Oural. Paleontologie. John Murray, London. Bertrand, Paris. 2(3) : 1-512, 50 pis. Neumayr, M. 1891. Beitrage zur morphologischen Einteilung der Bivalven. Denk- schr. Acad. Wiss. Wien Math.-Nat. Cl. 58:701-801. Nicol, David. 1970. Mode of life of Conocardiiim, a Paleozoic pelecypod. Nautilus 83:77-80, figs. 1-5. Pannella, G. and C. MacClintock. 1968. Biological and environmental rhythms reflected in molluscan shell growth. Jour. Paleontol. 42:64-80, pis. 1-9. Pannella, G., C. MacClintock, and M. N. Thompson. 1968. Paleontological evi- dence of variations in length of synodic month since Late Cambrian. Science 162:792-796, figs. 1, 2. Skinner, J. W. and G. L. Wilde. 1965. Permian biostratigraphy and fusulinid faunas of the Shasta Lake area, northern California. Univ. Kans. Paleontol. Contr., Protozoa 6:1-98, pis. 1-65. Wilson, E. C. 1967a. Fusulina gracilis Meek, 1864 (Foraminiferida) : proposed invalidation of neotype-material, designation as type-species of Eoparafusulina Coogan, 1960, and related actions. Bull. Zool. Nomencl. 24:234-236, pi. 1. . 1967b. Primary type-material of the Early Permian archaeogastropod Omphalotrochus whitneyi (Meek, 1864). Jour. Paleontol. 41 : 1 113-1 1 15, pi. 142. Accepted for publication February 17, 1970 ! roH.es CONTRIBUTIONS z::z IN SCIENCE Dumber 185 April 17, 1970 : FIVE NEW EPITONIID GASTROPODS FROM THE |i WEST COAST OF THE AMERICAS i| ji By Helen DuShane ii ii !i Los Angeles County Museum of Natural History • Exposition Park :j Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— ( 1 ) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS.- — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author oi divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor FIVE NEW EPITONIID GASTROPODS FROM THE WEST COAST OF THE AMERICAS By Helen DuShane^ Abstract: Five epitoniids (Mollusca: Gastropoda) are de- scribed as new, including {Asperiscala) longinosanum from the west coast of Baja California, Mexico; Epitonium (Asperiscala) macleani from Ventana Bay, Baja California, Mexico; Epitonium (Acirsa) cerralvoensis from Cerralvo Island, Baja California, Mexico; Epitonium (Acirsa) murrha from Panama; and Amaea contexta from Guerrero, Mexico. The epitoniid collections of the Los Angeles County Museum of Natural History (LACM), including those now on loan from the Allan Hancock Foundation, contain five new species described here. Two of the new species were dredged by the R/V Valero III in the 1930’s and are from the Hancock Foundation Collection. Two are from a 1966 Museum expedition to the Cape San Lucas area of Baja California, sponsored by Mr. Richard Dwyer of Corona del Mar, on the R/V Sea Quest. One other new species was collected by Fred E. Lewis and remained unidentified until now. Acknowledgments I am grateful to the Los Angeles County Museum of Natural History for the opportunity to study their collections and for the help rendered by Dr. James H. McLean and Mr. Gale Sphon of the Invertebrate Zoology section. I also thank the photography department of the Museum for the illustrations. Epitonium (Asperiscala) longinosanum, sp. nov. Fig. 1 Diagnosis: An elongate species with 1 1 low varices differing from all other epitoniids from the Panamic molluscan province in having a brown shell with white ribs and unevenly spaced minute spiral striae. Description: Shell large, tall and slender, thin but strong, medium brown in color, shining, with 1 1 gradually enlarging, rounded whorls, nuclear whorls missing; suture distinct, varices white, 11, low and slightly reflected with a small, sharp spine on shoulder of whorl, varices continuous from whorl to whorl, an occasional costa thicker than others; spiral striations uneven, visible only under 10 X magnification; columellar lip curved, peritreme continuous, reflected over the ends of the varices, outer lip thickened by last varix; um- ^Research Assistant in Invertebrate Zoology, Los Angeles County Museum of Natural History, Los Angeles, California 90007. 1 2 Contributions in Science No. 185 Figures 1-2. 1, Epitonium (Asperiscala) longinosanum sp. nov., holotype, LACM 1 196. X 3; 2, Epitonium (Asperiscala) macleani sp. nov., holotype, LACM 1 197. X 9. bilious lacking, operculum missing in holotype. Dimensions (in mm) : length 16.5, width 5.5. Type Locality: Rancho Inocentes, Baja California del Sur, Mexico, 23° 45' N, 1 10° 40' W. Collected by Captain Fred E. Lewis prior to 1945, one specimen. Type Material : Holotype, LACM 1196. Discussion: The coloring and 1 1 varices with sharp, shoulder spines distinguish E. longinosanum from all other west American epitoniids. It differs from E. appressicostatum Dali, 1917 (p. 482) , which it most resembles, in having fewer costae, a less glassy surface, and fine spiral striations between the costae. Epitonium longinosanum is named for Jose Longinos Martinez who es- tablished a museum of natural history in Mexico City in 1791 and the follow- ing year collected shells from La Paz to San Francisco (Longinos Martinez, 1938). 1970 Five New Gastropods 3 Epitonium (Asperiscala) macleani, sp. nov. Fig. 2 Diagnosis: A small species differing from all others in having rounded whorls, numerous low axial costae of differing strengths and numerous fine spiral ribs. Description: Shell small, white, fragile, with approximately 32 fine costae of varying size, nuclear whorls two, smooth, rounded, light brown, first postnuclear whorl with axial striations; remaining whorls six; suture distinct, deeply impressed; umbilicus small and nearly covered by peritreme; surface between whorls with raised spiral threads, approximately 28 on the body whorl; whorls rapidly expanding, rounded, basal disk or cord lacking; aperture oval (lip broken in holotype) ; operculum unknown. Dimensions (in mm) : length 7.0; width 3.5 (holotype) . Type Locality: Off Punta Ventana, Baja California del Sur, Mexico (opposite Cerralvo Island), 24° 4' N, 109° 49' W, 15-25 fms, sand bottom, LACM Invertebrate Zoology locality 66-23, James H. McLean, Peter M. Oringer, and Louis Marincovich, R/V Sea Quest, 8 April 1966, ten specimens. Type Material: Holotype, LACM 1197; 8 paratypes, LACM 1198; 1 paratype, DuShane Collection. Discussion: This species differs from all others known from the Panamic province. Less globose than the brownish Epitonium (Asperiscala) huffmani DuShane and McLean, 1968 (fig. 1), it is white and has somewhat similar sculpture. Epitonium (Asperiscala) macleani is named for Dr. James H. McLean, Curator of Invertebrate Zoology, Los Angeles County Museum of Natural History. Epitonium (Acirsa) cerralvoensis, sp. nov. Fig. 3 Diagnosis: A slender, pale brown species differing from other epitoniids from the Panamic molluscan province in having reduced axial costae and unevenly spaced spiral striae. Description: Shell of medium size, tall and slender, thin but strong, light brown in color, with 10-12 gradually enlarging, rounded whorls, two opaque nuclear whorls; suture moderately impressed; axial ridges pale brown, 16 on the fourth whorl, gradually becoming obsolete on succeeding whorls, spiral ridges wider than the interspaces, 19 on last whorl; peritreme incomplete, lip thin, columella arched; no basal ridge; operculum horn colored, paucispiral. Dimensions (in mm) : length 1 1.8, width 3.7 (holotype) . Type Locality : West side, Cerralvo Island, Baja California del Sur, Mexico, 24° 11' N, 109° 55' W, 7 fms, sand bottom, LACM Invertebrate Zoology locality 66-24, James H. McLean, Peter M. Oringer, and Louis Marincovich, R/V Sea Quest, 9 April 1966, two specimens. 4 Contributions in Science No. 185 Figures 3-5. 3, Epitonium (Acirsa) cerralvoensis sp. nov., holotype, LACM 1200. X 5; 4, Epitonium (Acirsa) murrha sp. nov., holotype, LACM 1232. X 3; 5, Amaea (s.l.) contexta sp. nov., holotype, LACM 1 199. X 4. Type Material : Holotype, LACM 1200; 1 paratype, LACM 1201. Al- though larger (15.3 mm in length), the paratype is worn and has several holes in the shell. Referred Material : LACM 66-17, 10-20 fms off Cape San Lucas, Baja California, 1 specimen; LACM 66-17, 10-20 fms off Rancho Palmilla, Baja California, 1 specimen; LACM 66-20, 4 fms, Pulmo Bay, Baja California, 1 specimen; AHF 1051-40, 21 fms off Puerto Refugio, Angel de la Guardia Island, Baja California, 1 specimen. Discussion: No other epitoniid from the west coasts of the Americas has sculpture like this species. Epitonium (Acirsa) cerralvoensis is named for Cerralvo Island, its type locality. Epitonium (Acirsa) murrha, sp. nov. Fig. 4 Diagnosis: A large species with 19 low axial ribs over which the spiral sculpture crosses; differing from other Acirsa from the west coasts of the Americas in having more pronounced axial sculpture on the body whorl. Description: Shell of medium size, white, texture of opaque china; 1970 Five New Gastropods 5 nuclear whorls two, opaque (first nuclear whorl missing in holotype), first postnuclear whorl with faint indication of axial sculpture; second postnuclear whorl with incised spiral lines; remaining convex whorls 10, gradually en- larging; 19 retractive axial ridges on body whorl giving a sinuous appearance; penultimate whorl with 15 spiral cords, uneven in width, separated by narrow, incised lines riding over the axial ridges; suture moderately impressed; basal ridge with 10-11 flat cords separated by incised lines; aperture oval, peritreme incomplete, umbilicus lacking, operculum unknown. Dimensions (in mm) : length 19.1, width 6.9 (holotype). Type Locality: Ten miles southwest of Secas Islands, Panama, 7° 5T 10" N, 82° 12' 05" W, 30 fms, gray sand, R/V Velero III station 944-39, 27 March 1939, one specimen. Type Material: Holotype, LACM-AHF 1232. Discussion: No other epitoniid from the west American coast has the color or sculpture of this species. The north Atlantic species, Acirsa costulata Mighels and Adams, 1842, figured by Clench and Turner (1950, p. 230) is similar in proportions but has weaker axial and spiral sculpture. The name murrha, a Latin noun, meaning porcelain clay, refers to the chinalike surface of the shell. It is intended as a noun in apposition. Amaea (s.L) contexta, sp. nov. Fig. 5 Diagnosis: An epitoniid differing from all other species in having many whorls, a dark brown color, numerous low axial ribs, and coarse spiral ribs. Description: Shell of medium size, slender, dark brown, with four smooth, white, glassy nuclear whorls (missing in holotype), first postnuclear whorl axially ribbed; the following 10 whorls dark brown, convex; suture deeply impressed; axial sculpture of thin, strong ribs, about 24 on the body whorl, with occasionally a very heavy rib; these ribs continue to the columellar wall; spiral sculpture of about 18 ridges per whorl, beaded on the anterior face of the ribs; basal ridge present; outer lip formed by the extension of the last varix, aperture oval, umbilicus lacking; operculum lacking in holotype. Dimen- sions (in mm) : length 15.4, width 4.5 (holotype). Type Locality: Petatlan Bay, Guerrero, Mexico, 17° 31' 45" N, 101° 27' 34" W, 5-10 fms, hard sand and shell bottom, R/V Velero III station 265-34, 3 February 1934, one specimen. Type Material: Holotype, LACM 1199. Referred Material : Two smaller, damaged specimens with intact nuclear whorls were dredged by Laura and Carl Shy off Manzanillo, Colima, Mexico, in 5-10 fms, November 1967. These two specimens are uniformly dark brown, suggesting that the color of the holotype has faded. Discussion : No other epitoniid from the west American coast is related 6 Contributions in Science No. 185 to this species. Sculpture is unique, making the generic allocation uncertain. Since the general features suggest Amaea H. and A. Adams, 1853, it is placed here provisionally. The Latin name is derived from contextus, meaning interwoven, and refers to the woven texture of the axial and spiral sculpture. Resumen Cinco epitonidos (Mollusca: Gastropoda) se describen como nuevas especies, incluyendo (Asperiscala) longinosanum de la costa oeste de la Baja California, Mexico; Epitonium {Asperiscala) macleani de la Bahia de Ventana, Baja California; Mexico; Epitonium (Acirsa) cerralvoensis de la Isla Cer- ralvo, Baja California, Mexico; Epitonium (Acirsa) murrha procedente de Panama; y Amaea contexta de Guerrero, Mexico. Literature Cited Clench, W. J., and R. D. Turner. 1950. The genera Sthenorytis, Cirsotrema, Acirsa, Opalia and Amaea in the western Atlantic. Johnsonia, 2: 221-248. Dale, W. H. 1917. Notes on the shells of the genus Epitonium and its allies of the Pacific coast of America, U.S. Nat. Mus., Proc., 53: 471-488. DuShane, Helen, and J. H. McLean. 1968. Three new epitoniid gastropods from the Panamic Province. L.A. County Mus. Nat. Hist., Contrib. Sci., 145: 1-6. Longinos Martinez, Jose. 1938. California in 1792. The expedition of Jose Longi- nos Martinez, transl. by L. B. Simpson. Huntington Library, San Marino, Calif. 1 1 1 pp. Accepted for publication February 11, 1970 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE UMBER 186 April 17, 1970 A GENERIC REVISION AND CATALOG OF THE WESTERN HEMISPHERE GLENURINI WITH THE DESCRIPTION OF A NEW GENUS AND SPECIES FROM BRAZIL (NEUROPTERA : MYRMELEONTIDAE) By L. A. Stance Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23,. 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM. — ( 1 ) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3 ) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS.— All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author oi divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor A GENERIC REVISION AND CATALOG OF THE WESTERN HEMISPHERE GLENURINI WITH THE DESCRIPTION OF A NEW GENUS AND SPECIES FROM BRAZIL ( NEUROPTER A : M YRMELEONTID AE ) By L. A. Stance^ Abstract: Studies of the type species of described glenurine genera have led to considerable new synonymy as follows: Elach- yleon Esben-Petersen 1927 {—Sericoleon Esben-Petersen 1932); Eremoleon Banks 1901 (-^zlncamoleon Banks 1913, —Sosa Navas 1914, =Cortesiiis Navas 1914, =Dobla Navas 1926, =Joergenia Esben-Petersen 1932, =Antilloleon Banks 1943); Psammoleon Banks 1899 {=Diaziis Navas 1914). Pachyleon alvarengai, new genus and new species, is described from Mato Grosso, Brazil. Eight genera are now recognized and a key is given for their iden- tification. Each genus is characterized and briefly discussed, in- cluding new synonyms. A complete bibliography and catalog is provided for the Western Hemisphere Glenurini. New informa- tion contained in this catalog includes the following new species synonyms: Psammoleon caiitiis (Walker) 1853 {—Feinenis neb- ulosus Navas 1922) ; Psammoleon serrei (Navas) 1920 (=For- micaleo chaperi Navas 1922); Gleniinis peculiaris (Walker) 1859 {—Glenunis brasiliensis Navas 1920); Eremoleon macer (Hagen) 1861 (=Hesperoleon atomariiis Navas 1933). Also a number of new combinations are indicated. The tribe Dimarellini Markl 1954 is now considered a synonym of the Glenurini Banks 1928. The tribe Glenurini is now referred to the subfamily Myr- meleontinae. Introduction The tribe Glenurini is one of the largest tribes of antiions in the world in numbers of described genera. Markl (1954:246) lists 30 genera of Glenurini from all major zoogeographical regions of which sixteen are restricted in dis- tribution to the Western Hemisphere. However, no modern comprehensive generic study has ever been made of the Glenurini. Furthermore the definition of the tribe is probably deficient judging from the paucity of taxonomic char- acters given by Markl (1954, p. 245) and earlier authors. The present account deals only with the glenurine genera of the Western Hemisphere. An attempt is made to signal taxonomic characters of importance and to point out the variation that exists in others that have been used in the classification of the tribe. Terminology is adopted from Stange ( 1970) . ^Department of Zoology, Instituto Miguel Lillo, San Miguel de Tucuman, Argentina. 1 2 Contributions in Science No. 186 Acknowledgments Study of nearly all of the types of the type species of glenurine genera by me was made possible only by the cooperation of numerous individuals and organizations. I am especially indebted to the Society of Sigma Xi for a Grant- in-Aid of research which enabled me to study type material in Europe during the summer of 1964. I would also like to thank Dr. S. L. Tuxen, Copenhagen University, and Dr. Max Beier, Vienna Museum, for the loan of type material. The following list of collections mentioned in the present study also includes the names of persons who facilitated my studies: AMNH, American Museum of Natural History, Dr. Jerome Rosen; Barcelona, Museo de Zoologia, Barce- lona, Dr. Francisco Espahol Coll.; Berlin, Institut fur spezielle Zoologie und zoologisches Museum, East Berlin, Dr. E. Konigsmann; BM(NH), British Museum (Natural History), London, Mr. D. E. Kimmins; CAS, California Academy of Sciences, Dr. Edward Ross; Copenhagen, Universitets Zoologiske Museum, Copenhagen, Dr. S. L. Tuxen; Greifswald, Zoologisches Institut und Museum, Ernst-Moritz-Arndt-Universitat, Greifswald. Dr. Ilselotte Groth. Halle, Zoologisches Institut, Martin-Luther-Universitat, Halle-Saale, Dr. J. O. Hiising; Havana, Institute de Biologia, La Habana; LACM, Los Angeles County Museum of Natural History, Dr. Charles Hogue; La Plata, Museo de La Plata, Facultad de Ciencias Naturales, La Plata, Prof. Luis De Santis; MACN, Museo Argentine de Ciencias Naturales, Buenos Aires, Dr. Manuel Viana; MCZ, Museum of Comparative Zoology, Harvard University, Dr. Howard Evans; Paris, Museum National d’Histoire Naturelle, Paris, Miss S. Kelner-Pillault; Hamburg, Zoologisches Staatsinstitut und Zoologisches Mu- seum, Hamburg; Turin, Institute e Museo di Zoologia, Universita di Torino, Turin; USNM, United States National Museum, Dr. Oliver Flint; Vienna, Naturhistorisches Museum, Vienna, Dr. Max Beier; Zaragoza, Colegio del Salvador, Zaragoza. Systematic Position of the Glenurini Banks (1943, p. 166) classified genera of the Glenurini in the subfamily Macronemurinae. Markl (1954) published a comprehensive world classifica- toin of the Myrmeleontidae but eliminated subfamilies and only dealt with tribes. Stange (1968) placed the Glenurini in the subfamily Dendroleontinae. My attempts to provide a satisfactory definition of the Dendroleontinae from the Myrmeleontinae have now convinced me that this separation can not be maintained. Therefore I consider the subfamilies Dendroleontinae and Macro- nemurinae to be synonyms of the Myrmeleontinae. Essentially only two vena- tional characters have been employed in the past for separating these subfami- lies. One is the condition of vein 2A of the forewing. In some groups {Den- droleon and Brachynemurus) vein 2A is evenly curved from its base to its termination whereas it is strongly angled in Myrmeleon L, Macronemurus (an old world genus), and in nearly all of the Glenurini. However the close struc- 1970 Catalog of the Western Hemisphere Glenurini 3 Figures 1-7: Pachyleon alvarengai. 1. Head of holotype. 2. Base of forewing. 3. Male terminalia (right ectoproct removed). 4. Male genitalia (a) side view; (b) posterior view. 5. Female terminalia, ventral view. 6. Foreleg of holotype, apex of femur, tibia and tarsus. 7. Hindleg of holotype, apex of tibia and tarsus. 4 Contributions in Science No. 186 tural resemblance between Dendroleon and some genera of the Glenurini (i.e. Glenurus) argues against such an extreme separation. Furthermore vein 2A shows some variation in the Glenurini. In a new genus to be described later in this paper, vein 2A is closely associated with 3A and not with lA as is typical in the tribe. Also in an undescribed species (Mystroleon praedator Sensu Banks) vein 2 A is intermediate in condition between typical Glenurini and Dendroleontini. For these reasons I consider this character insufficient for separating subfamilies. The second venational character that has figured prom- inently in subfamily classifications is the number of hindwing presectoral cross veins. In the Myrmeleontinae there are four or more such crossveins whereas the typical number in the Dendroleontinae is one. However, within the tribe Brachynemurini, which I consider closely related to the Glenurini, there are sometimes several crossveins depending upon the point of origin of the radial sector. This indicates that the number of presectoral crossveins is of limited value as a subfamily character. Markl (1954) recognizes numerous tribes that are referrable to the Myr- meleontinae. Probably many of these are invalid. Markl defines the Glenurini almost entirely on wing venational characters some of which I have found to be variable. Markl emphasizes the forking of forewing veins 2A and 3 A. In the Glenurini, Nyutini (restricted to Africa) and Dimarellini (restricted to the Neotropical Area), vein 2A is forked and 3 A is simple. Other related tribes, all restricted to the Eastern Hemisphere, have vein 2A simple and 3A forked. This condition is given for the Formicaleontini, Creoleontini, Obini, Nemo- leontini and Protoplectrini. However this character is variable, even at the specific level in some instances. In the glenurine genus Eremoleon, most spe- cies fit the concept of Markl (vein 2 A forked, 3 A simple in insipidus, longior, etc.) but other species show the opposite condition (as in some nigribasis) . Therefore this character is of limited taxonomic importance. Another vena- tional character that Markl uses in differentiating the Glenurini (from the Dimarellini and Nyutini) is the presence of the anterior banksian line. This prominent longitudinal line is found in either wing and is formed by the bend- ing of the branches of the radial sector. It is highly variable in its development in the genus Brachynemurus, a member of the related tribe Brachynemurini. In the tribe Glenurini, nearly all of the Western Hemisphere species that I have examined show no development of the banksian lines. It is apparent that Markl was in error employing this character in the diagnosis of the Glenurini (at least the Western Hemisphere component). Markl (1954, p. 247) defined the tribe Dimarellini on venational char- acters which Stange (1963, p. 810) criticized. Stange offered the character of the pretarsal claws as a possible means of separating the Dimarellini from the Glenurini. With this new criterion the genera Elachyleon and Navasoleon were removed from the Glenurini and placed with Dimarella in the Dimarellini. In these three genera the pretarsal claws are capable of closing against the distal 1970 Catalog of the Western Hemisphere Glenurini 5 tarsomere whereas the claws are ordinary in other genera. However it is now apparent that this specialized condition of the pretarsal claws have evolved several times in unrelated groups of Myrmeleontidae as evidenced by the occurrence of this character in the Dendroleontini {Tricholeon Esben-Petersen from Africa; Froggattisca Esben-Petersen from Australia) and in the Glenu- rini {Megistopus Rambur from Europe). The genus Elachyleon, possessor of the specialized pretarsal claws, agrees with the glenurine genus Eremoleon in all generic level taxonomic characters except the condition of the pretarsal claws. Thus, I now consider the pretarsal claw character to be a generic charac- ter. In search of a possible new means of maintaining the tribe Dimarellini I have reconsidered all the specialized features of the genus Dimarella. The mem- bers of this genus differ from other Glenurini in the following characters: (1) wing venation (CuP + lA run parallel with the posterior fork of forewing CuA) ; (2) antennal fossa widely separated from eye; (3) male ectoproct with elongate postventral lobe; (4) form of the male gonarcus (large lateral expan- sions) and shape of the parameres (not in the form of flat plates). A new genus to be described later in this paper also shares these characters (but has ordinary pretarsal claws). However recent study of the poorly known species Psammoleon-cautus has revealed that this species is intermediate between the genus Dimarella and the glenurine genus Psammoleon. This species appears to be a rather ordinary Psammoleon in regards to the wing venation, head structure and legs but possesses the elongate postventral lobe of the male ecto- proct as in Dimarella as well as comparable development of the male gonarcus and parameres. From the foregoing discussion it is apparent that we are confronted with a weak tribal classification of the Myrmeleontinae. More study is needed of the Old World genera of the Glenurini as well as of the related Old World tribes Formicaleontini, Creoleontini, Obini, Nemoleontini, Protoplectrini and Nyutini before a satisfactory world-based classification can be given. The new world tribe Dimarellini can not be maintained and I now consider Dimarella as a rather specialized genus of the Glenurini. In the Western Hemisphere there is less diversity in the Myrmeleontinae, at least from the point of view of described tribes. The following key will dis- tinguish the Glenurini from other Western Hemisphere groups of antiions. 1. Sensory pit of the distal labial palpomere slitlike; hindleg with femoral sense hair or femoral sense hair absent Palparinae and Acanthaclisinae Sensory pit of the distal labial palpomere oval-shaped; femoral sense hair present on foreleg and midleg (except few species of Brachynemurus) but not on hindleg ’ (Myrmeleontinae) 2 2. Forewing vein 2A evenly curved from base to hind margin, widely separated from 3A at least at basal one-half Dendroleontini and Brachynemurini Forewing vein 2A runs close to vein CuP + 1 A for a short distance, then 6 Contributions in Science No. 186 abruptly angles toward hind margin or veins 2A and 3A closely associated.... 3 3. Radial sector of hindwing arises beyond cubital fork, four or more presec- toral crossveins present; pilula axillaris present Myrmeleontini and Porrerini Radial sector of hindwing arises well before cubital fork, one (rarely two or three) presectoral crossveins present; pilula axillaris absent Glenurini Western Hemisphere Genera of the Glenurini All of the glenurine genera known from the Western Hemisphere are restricted to that area. It is possible that this reflects the regionalism of former workers on this group since no modern world study has been made. Even the Nearctic and Neotropical faunas of the Glenurini have not been treated to- gether. Banks provided keys to the Nearctic genera (1928) and the Neotropical genera (1943). The Nearctic fauna is relatively small. The three genera repre- sented in this region are also found in both the North American and South American Neotropical areas. All of the new world glenurine genera are found in South America and three of these are restricted to that continent. Twenty- four generic names have been proposed for the Western Hemisphere Glenu- rini. Some of these have been placed in synonymy. My study of nearly all of the type species of Western Hemisphere genera of the Glenurini has led to the discovery of further generic synonymy. In the following account of the glen- urine genera this synonymy will be discussed. Araucaleon Banks Araucaleon Banks, 1938. Proc. Ent. Soc. Wash. 40:127. Type-species: Araucaleon inca Banks, by original designation. Description: Antennal fossa separated from ocular rim by less than great- est diameter of pedicel; antenna long and slender; pronotum longer than wide; mesoscutellum smaller than metanotum; all legs about equal in length; tibial spurs longer than forecoxal length; pretarsal claws not capable of closing against distal tarsomere; forewing with anterior margin abruptly angled and swollen near coalescing of subcostal and radial veins, costal area at this point nearly twice as high as at middle of wing; forewing radial sector originates well basad to forking of CuA; posterior fork of forewing vein CuA at an oblique angle to hind margin; forewing vein 2 A widely separated from normal 3 A be- fore strong angle; hindwing vein CuA ends well before midpoint to forking of MP9; male ectoproct simple, without postventral lobe; male paramere in form of rigid plate; male gonarcus not greatly expanded laterally. Discussion: This genus is rare in collections. I have seen only two males of the type-species. In addition, Glenurus withycombei Esben-Petersen, de- scribed from Trinidad, belongs to Araucaleon. This species is known only from the holotype female. This genus appears to be allied with Eremoleon Banks 1970 Catalog of the Western Hemisphere Glenurini 7 but differs from that genus and other known glenurine genera by the swelling of the forewing costal margin near the point of coalescing of the subcostal and radial veins and by the termination of vein CuA of the hindwing well before the midpoint between the wing base and forking of MP2. Dimarella Banks Dimarella Banks, 1913. Trans. Amer. Ent. Soc. 39:229. Type-species: Eremoleon angiistus Banks, by original designation. Description: Antennal fossa separated from ocular rim by more than greatest diameter of pedicel; antenna variable in length; pronotum wider than long; mesoscutellum as long as metanotum; legs different in lengths, often midleg shortest; tibial spurs much shorter than forecoxal length; pretarsal claws capable of closing against distal tarsomere; forewing with anterior margin evenly curved toward apex, costal area at point of coalescing of subcostal and radial veins usually lower than at middle of wing; forewing radial sector orig- inates well beyond forking of CuA; forewing vein CuP + 1 A runs parallel with posterior branch of CuA for a long distance; forewing vein 2A usually widely separated from normal 3A before strong angle toward posterior margin; hind- wing vein CuA ends near forking of MP._,; male ectoproct with elongate post- ventral lobe; male paramere with hook; male gonarcus expanded laterally as a round plate; female ectoproct with strong digging setae ventrally, shorter than on lateral gonapophyses. Discussion: Dimarella is limited to the Neotropical Region. Synonymical notes on the genus were given by Stange (1963; 1968). The Mexican species were revised by Stange (1963). This genus stands apart from other described genera by the prominent postventral lobe of the male ectoproct, the parallel course of veins CuP + lA and CuA of the forewing and the wide space be- tween the ocular rim and antennal fossa. All of these features are shared with a new genus to be described later in this paper. The specialized pretarsal claws is a morphological trait shared only with Navasoleon and Elachyleon. Elachyleon Esben-Petersen Elachyleon Esben-Petersen, 1927. Ann. Mag. nat. Hist. (Ser. 9) 20:348. Type- species: Elachyleon punctipennis Esben-Petersen, by original designation. Sericoleon Esben-Petersen, 1932. Vidensk. Medd. Dansk. naturh. Foren 94:114. Type-species: Sericoleon paessleri Esben-Petersen, by original designation. New Synonymy. Description: Antennal fossa separated from ocular rim by less than great- est diameter of pedicel; antenna long and slender; mesoscutellum shorter than metanotum; all legs about equal in length; tibial spurs shorter than fore- coxal length; pretarsal claws capable of closing against distal tarsomere; forewing with anterior margin evenly curved toward apex, costal area at point 8 Contributions in Science No. 186 of coalescing of subcostal and radial veins either somewhat lower or higher than at middle of wing; forewing radial sector originates only somewhat before forking of CuA or well beyond; posterior fork of forewing vein CuA at an oblique angle to hind margin; forewing vein 2 A widely separated from normal 3 A before strong angle toward posterior margin; hindwing vein CuA extends nearly to forking of MP2; male ectoproct simple, without postventral lobe; male paramere in form of rigid plate; male gonarcus not greatly expanded laterally; female ectoproct usually with weak digging setae; posterior gonapophysis longer than wide, digitiform. Discussion: I have studied one of the paratypes of the type-species of Sericoleon, S. paessleri Esben-Petersen, described from Chile. This species possesses all of the generic characters of Elachyleon and therefore I am placing Sericoleon in synonymy with Elachyleon. Elachyleon can be distinguished from all other Western Hemisphere Myrmeleontidae by the specialized pre- tarsal claws, except for Navasoleon (which lacks tibial spurs) and Dimarella (which has forewing veins CuP + 1 A and CuA parallel) . This genus agrees closely with Eremoleon in both morphological and biological characters. I have reared an undescribed species from Mexico and from Argentina. Both species are found in the silt floors of caves. This is the same habitat for various species of Eremoleon. Elachyleon is now constituted of four described species. The most widespread species, E. punctipennis (ranges from Mexico to Argentina) has been illustrated in part by Stange ( 1963) . Additional generic synonymy is given by Stange ( 1 968 ) . Eremoleon Banks Eremoleon Banks, 1901. Trans. Amer. Ent. Soc. 27:366. Type-species: Myr- meleon macer Hagen, by original designation. Incamoleon Banks, 1913. Trans. Amer. Ent. Soc. 39:229. Type-species: Psammoleon punctipennis Banks, by original designation. New Syn- onymy. Sosa Navas, 1914. Broteria 12:218. Type-species: Sosa conspicuus Navas, by original designation. New Synonymy. Cortesius Navas, 1924. Broteria 21 : 107. Type-species: Cortesius genini Navas, by original designation. New Synonymy. Dobla Navas, 1926. Deutsch. ent. Zeitschr. :428. Type-species: Dobla arcuata Navas, by original designation. New Synonymy. Joergenia Esben-Petersen, 1932. Vidensk. Medd. Dansk. naturh. Foren 94:118. Type-species: Joergenia pulchra Esben-Petersen, by original des- ignation. New Synonymy. Antilloleon Banks, 1 943. Bol. ent. venezolana 2: 168. Type-species: Glenurus cerverai Navas, by original designation. New Synonymy. Description: Antennal fossa separated from ocular rim by less than great- 1970 Catalog of the Western Hemisphere Glenurini 9 est diameter of pedicel; antenna long and slender; pronotum variable, usually longer than wide; mesoscutellum shorter than metanotum; legs about equal in length except usually hindlegs longer; tibial spurs variable in length, usually shorter than forecoxal length; pretarsal claws not capable of closing against distal tarsomere; forewing with anterior margin evenly curved toward apex, costal area at point of coalesing of subcostal and radial veins usually lower than at middle of wing; forewing radial sector originates somewhat before forking of CuA or well beyond; posterior fork of forewing vein CuA at an oblique angle to hind margin; forewing vein 2A widely separated from normal 3A before strong angle toward posterior margin; hindwing vein CuA extends nearly to forking of MP2 or somewhat beyond; male ectoproct simple without postventral lobe; male paramere in form of rigid plate; male gonarcus not greatly expanded laterally; female ectoproct usually with weak digging setae; posterior gonapophysis longer than wide, digitiform. Discussion: There is considerable new generic synonymy in addition to that reported by Navas (1916) Banks (1928) & Adams (1957, 1958). These new synonyms will be discussed in chronological order. Incamoleon Banks 1913 — The type-species of this genus, punctipennis Banks, was originally described in the genus Psammoleon. Banks separated this species in its own genus apparently based on the longer, more slender legs of punctipennis. Also the wing shape is more slender than in Psammoleon. All of these features are found in the genus Eremoleon. Banks (1943:166) dis- tinguishes Incamoleon from Eremoleon {—Glenopsis Banks) by the relative length of the tarsomeres. In Incamoleon the distal tarsomere is about equal to the other tarsomeres together whereas in Eremoleon the distal tarsomere is shorter. However my studies on Incamoleon puftctipennis have revealed that this tarsomere proportion character is not constant and although useful for identification it is not important enough for generic distinction. Further studies have revealed two characters found in Incamoleon that are unknown in other Eremoleon. First, there is marked sexual dimorphism. The female wings usually have conspicuous dark spots, whereas the wings are not spotted (or only weakly so) in the males. Also the female genitalia appear somewhat distinct in that the posterior gonapophysis is more inflated than in other members of the genus Eremoleon. However I am of the opinion that these two characters are not sufficient to maintain Incamoleon as a distinct genus so that I am now referring punctipennis to the genus Eremoleon. Sosa Navas 1914 — I studied the holotype female of the type-species, Sosa conspicuus, in 1962 from the collections of the Vienna Museum. Navas gave few notes in his original description as to the basis of this new genus but did point out that it was related to Eormicaleon (an old world genus) but with longer and more slender legs and that the genus might belong to the Dendro- leontini. However my studies of the holotype revealed no important differences from the current generic concept of Eremoleon Banks. Sosa conspicuus re- 10 Contributions in Science No. 186 sembles Eremoleon macer (Hagen) except that the wings are more slender. This species is now placed in Eremoleon. Cortesius Navas 1924 — I studied the two syntypes of the type-species, Cortesius genini Navas, in the Paris Museum in 1964. Navas (1924:107) did not give much information in his original description as to the relationships of this genus except to point to a similarity with the genus Glenurus Hagen. My studies revealed that Cortesius genini agrees with all generic characters of Eremoleon. The only unusual feature of this species is that the forewing costal area is biareolate to triareolate. Interconnected crossveins in the costal area are found otherwise only in the genus Psammoleon among the Glenurini. Dobla Navas 1926 — The holotype female of the type-species, Dobla ar- cuata Navas, is located in the Paris Museum. My studies of the holotype indi- cate that it is a synonym of Cortesius genini Navas. Therefore the genus Dobla is a synonym of Eremoleon Banks. Joergenia Esben-Petersen 1932 — I have just recently received for study the holotype female of the type-species, Joergenia pulchra Esben-Petersen, on loan from the Copenhagen Museum. Although this species is quite distinctive by virtue of the unusual narrowing of the wings, especially of the costal area, I have not found characters that would warrant generic separation from Ere- moleon. There is considerable range in the relative broadening of the costal area in Eremoleon and I am of the opinion that J. pulchra represents one ex- treme. The other extreme, broadening of the costal area, is exemplified by E. genini (Navas) and E. anomalus (Rabur), whereas intermediate conditions are found in E. conspicuus (Navas), E. gracile Adams and many others. I have also seen an undescribed species of Eremoleon from Venezuela which has the costal area nearly as narrow as in /. pulchra. Antilloleon Banks 1943 — I have studied the type material of the type- species, Glenurus cerverai Navas, in the collections of the Museum of Com- parative Zoology, Harvard University. Banks (1943:167) pointed out the unusual venational feature present in the Caribbean species. The radial sector originates only a little beyond the level of forking of CuA in the forewing. In most other species of the Glenurini the radial sector originates well beyond the forking of CuA. However in the genus Araucaleon Banks, the radial sector originates well before the forking of CuA, a peculiarity in the Glenurini. In other respects A. cerverai Navas agrees well with species of the genus Eremo- leon. Considerable variation is present in the genus Eremoleon in the relative position of the origin of the radial sector and forking of CuA. I am, therefore, of the opinion that Glenurus cerverai belongs to the genus Eremoleon, Antil- loleon thus falling as a synonym of Eremoleon. Revisionary studies are needed of the genus Eremoleon which appears to be a complex taxon. Adams (1957) presented a key to most of the Nearctic species. There is no one key character of the genus which permits easy recog- nition, although the genus differs considerably from other glenurine genera 1970 Catalog of the Western Hemisphere Glenurini 11 except Elachyleon Esben-Petersen, which has specialized pretarsal claws. I have reared four species of Eremoleon (from Arizona, Mexico and Argentina) and all exhibit similar biological characteristics. The larvae live in the silt of caves or rock crevices and seem to be good climbers. Glenurus Hagen Glenurus UdigQn 1866. Stettin, ent. Ztg. 'll '312. Type-species: Formicaleo grata Say. Descripti'on: Antenna fossa separated from ocular rim by less than great- est diameter of pedicel; antenna long and slender; mesoscutellum shorter than metanotum; all legs about equal in length except hindlegs usually somewhat longer; tibial spurs longer than forecoxal length; pretarsal claws not capable of closing against distal tarsomere; forewing with anterior margin evenly curved toward apex, costal area at point of coalescing of subcostal and radial veins somewhat lower or higher than at middle of wing; forewing radial sector originates well beyond forking of CuA; posterior fork of forewing vein CuA at an oblique angle to hind margin; forewing vein 2 A widely separated from normal 3 A before strong angle toward posterior margin; hindwing vein CuA extends nearly to or somewhat beyond forking of MP2; male ectoproct simple without postventral lobe; male paramere in form of rigid plate; male gonarcus not greatly expanded laterally; female ectoproct with prominent digging setae; posterior gonapophysis weakly produced as a swelling. Discussion: This is a very homogeneous genus easily recognized by the considerable brown suffusion of the wing apices. G. heteropteryx is the only known species that lacks this prominent suffusion in the forewing, although the hindwing conforms with the rest of the genus. Structurally Glenurus seems close to Eremoleon but differs from that genus as well as all other glenurine genera in having the posterior gonapophysis poorly developed. Banks (1922) has provided a key to the South American species and (1928) to the Nearctic species. Navasoleon Banks Navasoleon Banks 1943. Bol. ent. Venez. 2:168. Type-species: Gymnocnemia boliviana Banks, by original designation. Description: Antenna fossa separated from ocular rim by less than great- est diameter of pedicel; antenna long and slender; mesoscutellum shorter than metanotum; foreleg (at least females) much longer than midlegs and hindlegs; tibial spurs absent; pretarsal claws capable of closing against distal tarsomere; forewing with anterior margin evenly curved toward apex, costal area at point of coalescing of subcostal and radial veins lower than at middle of wings; fore- wing radial sector originates well beyond forking of CuA; posterior fork of forewing vein CuA at an oblique angle to hind margin; forewing vein 2 A widely separated from normal 3 A before strong angle toward posterior margin; 12 Contributions in Science No. 186 hindwing vein CuA extends somewhat beyond forking of MP2; female ecto- proct with small digging setae; posterior gonapophysis swollen thumb-like. Discussion: Navasoleon is a poorly known genus and male specimens are unavailable for study. The genus appears very distinct. The lack of tibial spurs is diagnostic and the specialized pretarsal claws is found only in Elachy- leon and Dimarella. In the type-species of the genus and also in an undescribed species the foreleg is unusually long. The species known from Argentina has shorter front legs but the specimen is damaged to such an extent that the sex is undeterminable. Pachyleon Stange, new genus Diagnosis: This genus is distinguished from other glenurine genera in the Western Hemisphere except Dimarella by the following characters: (1) forewing vein 2A is closely associated with vein 3A for its entire course (see Fig. 2) ; (2) radial and subcostal veins in both wings crowded together; (3) antennal fossa separated by more than greatest diameter of scape from ocular rim; (4) forewing veins CuP + lA and posterior branch of CuA run parallel for a long distance. Pachyleon is closely allied to Dimarella but differs from that genus in two important characters: (1) pretarsal claws simple in Pachy- leon, specialized in Dimarella (capable of closing against tarsomere); (2) hindwing broadly triangular in Pachyleon (greatest breadth about equal to forewing) whereas it is more slender in Dimarella (greatest breadth of hind- wing much less than that of forewing). The structural gap between Pachyleon and Dimarella would be even greater if it were not for a recently discovered (apparently undescribed) species of Dimarella which has the condition of fore- wing vein 2A (closely associated with 3A) very similar to that in Pachyleon. Pachyleon and Dimarella share so many characters that are not found in almost all other Glenurini that they form a group apart. However Psammoleon cautus (Walker) represents an intermediate type between Psammoleon and these two genera. The unusual shape of the hindwing of Pachyleon provides a key character for the genus. It remains to be seen whether or not the broad hindwing is of generic or specific importance since considerable variation exists in the relative broadening of the hindwing in the allied genus Dimarella. Description: Antennal fossa separated from ocular rim by more than greatest width of antennal scape; antenna short and thick; pronotum much broader than long; thorax depressed, mesoscutellum as long as metanotum; foreleg moderately short and thickened, midleg greatly swollen and short, hindleg relatively long, about twice as long as foreleg; tarsomeres of foreleg and midleg about equal, those of hindleg much longer; tibial spurs present, much shorter than forecoxal length; pretarsal claws not capable of closing against distal tarsomere; femoral sense hair of foreleg short, much less than one-half length of femur; banksian lines absent; subcostal and radial veins crowded together, nearly touching well before their apical fusion; forewing 1970 Catalog of the Western Hemisphere Glenurini 13 Figure 8. Photo of holotype (male terminalia removed). vein CuP + 1 A runs parallel with posterior branch of CuA for a long distance; forewing vein 2A widely separated from lA, closely associated and almost fused with vein 3A, not greatly angled; male ectoproct with elongate postven- tral lobe; male genitalia not rigid, paramere with hook; gonarcus expanded laterally as a large round plate; female ectoproct simple with strong digging setae ventrally; lateral gonapophyses separated, with longer digging setae than on ectoproct; posterior gonapophysis digitiform, longer than wide. The generic name is from the Greek words Pachys meaning wide and leon meaning lion, in reference to the broad wings, pronotum and head. The following new species is the type-species of the genus. Pachyleon alvarengai Stange, new species Diagnosis: As the only species in the genus, the generic characters serve to distinguish this species. The unusually short and broadened wings, espe- cially the rather triangular hindwing (see Fig. 8) is distinctive among all known species of ant-lions in the Western Hemisphere. This unusual hindwing shape is also unknown to me in Old World species but I do not have a perfect knowledge of the Old World fauna. The most similar species is a Dimarella (probably undescribed) in which the wings are similar in shape although not as extreme as in P. alvarengai and in which vein 2A is closely associated with vein 3A. 14 Contributions in Science No. 186 Description (Holotype male): Length of body about 19 mm, forewing 15 mm, hindwing 1 1 mm, greatest wing width (about equal in forewing and hindwing) 4.75 mm. Color: Face (Fig. 1) pale yellowish with A-shaped dark brown mark at middle below antennae and crescent-shaped dark brown mark above each antenna, sublateral spot on each side of clypeus, larger dark brown area laterad of antenna toward eye, extending ventrally and dorsally nearly to first row of vertex scars; vertex with grayish bloom, scar pattern about as in Fig. 8; antenna with scape and pedicel pale yellow with dark brown behind, flagello- meres dark brown basally and much paler distally; maxilla and labius nearly all pale yellow except distal palpomere and cardo with dark brown; postgena mostly dark brown; pronotum mostly dark brown with grayish bloom inter- rupted by irregular paler areas; notum II and III dark brown with grayish bloom, darker velvety areas as follows: submedial triangular area on each side of prescutum II; dark brown oval spot each side of middle and touching, sur- rounded by golden colored microtrichia on prescutum III; large curved dark brown mark sublaterally near anterior margin of scutum; postscutellum II and III nearly all black; cervical sclerites pale yellow; pleuron and sternum mostly dark brown with grayish bloom, paler brown areas mostly in front of midcoxa and hindcoxa, mesopleural wing process yellowish; forecoxa mostly pale brown, other coxae darker brown; forefemur pale brown with numerous dark brown spots, spots becoming confluent on most of exterior face and on distal one-third of closing face; foretibia pale brown with dark brown spots and small dark brown stripes at basal one-third of exterior face, larger dark brown areas at middle, mostly complete but smaller apical rings, dark brown areas inter- connected along closing face; midtibia and midfemur with large, mostly separated dark brown spots on exterior face, somewhat denser and forming short transverse dark brown bands on closing face of tibia, subapical inter- rupted dark brown ring on femur, solid dark brown one on tibia; hindfemur pale brown with dark brown spots at setal bases, infuscated between spots especially on exterior face, large subapical dark brown band; hindtibia pale brown with large, mostly separated dark brown spots at setal bases on closing face, small dark brown stripe near base, spots becoming more stripelike dis- tally, exterior face nearly all pale brown except dark brown apical ring and smaller dark brown area toward base; tarsi of all legs similarly patterned, pale brown with apical dark brown ring on tarsomeres; wings nearly without suffu- sion, pale white stigmal area preceded by small darkish brown area in fore- wing; wing veins mostly pale brown, especially hindwing, with dark brown streaks along longitudinal veins principally at crossvein junctions, many cross- veins dark brown especially apically in forewing; abdomen with tergites mostly pale brown with complicated pattern formed by dark brown areas (see Fig. 8), sternites mostly dark brown with a prominent although narrow median longi- tudinal pale brown line. 1970 Catalog of the Western Hemisphere Glenurini 15 Chaetotaxy: Head with few prominent setae except on most of clypeus, white ones on dark brown areas of frons, longer ones on labrum and many long white setae on mentum and stipes; vertex with few inconspicuous ap- pressed dark setae; pale with very sparse, short dark brown setae; pronotum with some short setae on disc, longer dark brown ones at lateral and posterior margin; row of five long white setae on each side of middle along anterior margin of scutum II, several smaller dark brown setae on prescutum II, else- where nearly absent or very short; pleuron and coxa with rather conspicuous long white hairlike setae; forefemur with dense, mostly appressed white setae on exterior surface, some black and some white erect setae distally, closing surface with fewer setae and with more black ones; foretibia with subbasal row of three black bristles, followed by another similar row, then one of two bristles and finally a subapical black bristle on exterior face, interspersed with shorter, mostly dark setae, closing surface with many stout setae, predomi- nately white; midfemur with row of white bristles on exterior surface, many other setae elsewhere but more hairlike and white; midtibia with mostly black setae of various sizes, larger black ones toward apex of lateral face, larger white ones near base of median face; hindfemur with many short, mostly appressed white setae; hindtibia with many setae along closing surface, white ones mostly concentrated in middle, exterior surface nearly without setae, only a few scattered minute ones; femoral sense hair of foreleg and midleg shorter than width of femur at that point; tarsi with rather dense, dark brown setae on closing surface flanked (typically) on the four basal tarsomeres by one white seta near apex, exterior surface with fewer setae, especially hindtarsus; ab- domen with only microsetae on tergites I-V; starting with tergite VI, dark brown setae mostly laterally increasing in abundance toward terminalia; ster- nites with numerous, mostly white hairlike setae; postventral lobe with rather prominent black bristles toward apex (Fig. 3). Structure-. Head with vertex not much raised above eyes; greatest ocular width about one-third interocular distance (measured just below antennal fossae); antenna short and rather flattened with 19 flagellomeres, flagellomere I longer than wide, II about 1.5 times wider than long (in flat view) with increasing width toward apex; distal palpomere of labius rather slender, not much swollen, sensory pit oval and close to base; pronotum about twice as wide as long; femur of all legs broadened beyond base, tibia much more slender except midtibia, which is subequal to midfemur; midleg much shorter than either foreleg or hindleg; wing venation and shape as in Fig. 8; abdomen shorter than forewing, segments larger basally, decreasing in diameter begin- ning with segment IV; tergites I-V and sternite II with numerous scalelike spicules; terminalia as in Fig. 3, genitalia as in Figs. 3 and 4. Types: Male holotype and 1 female paratype from Jacare (N. Xingu), Mato Grosso, Brazil, XI 1961, M. Alvarenga collector 52° 24' W; 12° 00' S. 16 Contributions in Science No. 186 Holotype deposted in the collections of the Los Angeles County Museum of Natural History. Paratype female retained in the author’s collection. Variation: The female paratype agrees well in color, chaetotaxy and structure with the male except that the scalelike spicules on the abdomen are lacking. Female terminalia as in Fig. 5. Discussion : This species is dedicated to Moacir Alvarenga, Brazilian coleopterist, for his efforts in securing critical ant-lion material from Brazil. The holotype male upon arrival at the depository was found damaged in transit. Most of the broken parts (pieces of wings, legs and antenna) are associated with the specimen and no difficulty in identification is foreseen although this was an unfortunate mishap. The photo (Fig. 8) shows the holotype before shipment. Psammoleon Banks Psammoleon Banks 1899. Canad. Ent. 31 :69. Type-species: Myrmeleon in- geniosus Walker, by original designation. Diazus Navas 1914. Broteria 12:220. Type-species: Diazus clavatus Navas, by original designation. Description: Antennal fossa separated from ocular rim by much less than greatest diameter of scape; antenna variable; pronotum wider than long; mesoscutellum smaller than metanotum; legs about equal in length, hindleg sometimes longer than foreleg; tibial spurs variable, usually shorter than fore- coxal length; pretarsal claws not capable of closing against distal tarsomere; forewing with anterior margin evenly curved toward apex, costal area at point of coalescing of subcostal and radial veins lower than at middle of wing; fore- wing radial sector originates well beyond forking of CuA; posterior fork of forewing vein CuA at an oblique angle to hind margin; forewing vein 2 A widely separated from normal 3A before strong angle toward posterior mar- gin; hindwing vein CuA extends nearly to forking of M?2 or somewhat beyond; male ectoproct simple without postventral lobe, except in P. cautus (Walker) ; male paramere in form of rigid plate (except cautus) ; male gon arcus not greatly expanded laterally (except cautus) ; female ectoproct without digging setae (sometimes long bristles present) , often produced ventrally as a lobe; posterior gonapophysis well developed, often somewhat inflated. Discussion: I studied the holotype male of the type-species of Diazus (D. clavatus Navas) in 1962 on loan from the Vienna Museum. The specimen is in good condition except that the hindlegs and abdominal terminalia are missing. My studies of this type indicate that D. clavatus is a typical Psammo- leon, rather similar to Psammoleon minor Banks. Navas (1914:220) gave few remarks concerning the diagnostic features of Diazus except to point to a relationship with Incamoleon and Glenopsis. Probably the swollen femora of D. clavatus prompted him to consider this species generically distinct. However it is clear that Diazus is a synonym of Psammoleon. 1970 Catalog of the Western Hemisphere Glenurini 17 The genus Psammoleon is the largest genus in the Western Hemisphere Glenurini, with twenty-seven described species. Also I have seen a number of undescribed species, mainly from Mexico. Most species are found in North America. The genus is unknown south of Mato Grosso, Brazil. The long fore- femoral sense hair (over one-half length of femur) is diagnostic. Also the rather swollen forefemur is characteristic except for Pachyleon and Dimarella. P. cautus (Walker) is intermediate between Psammoleon and Dimarella and perhaps should be placed in a distinct genus. However this situation suggests a relationship between these genera. The modified female ectoproct in many species of Psammoleon is also without known counterpart among Western Hemisphere Glenurini. Considerable variation is present between species in the development of the tarsus, forewing costal area and in chaetotaxy. Key to Genera of Western Hemisphere Glenurini 1. Pretarsal claws capable of closing against distal tarsomere 2 Pretarsal claws not capable of closing against distal tarsomere 4 2. Posterior fork of forewing vein CuA and vein CuP + 1 A parallel with each other and hind margin for a long distance; antennal fossa separated by more than greatest width of antennal pedicel from ocular rim; male ectoproct with elongate postventral lobe Dimarella Posterior fork of forewing vein CuA at an oblique angle to hind margin; antennal fossa separated by less than greatest width of antennal pedicel from ocular rim; male ectoproct without elongate postventral lobe 3 3. Tibial spurs present; foreleg about equal or shorter than \\md\Qg..Elachyleon Tibial spurs absent; foreleg (at least in female) usually longer than hindleg Navasoleon 4. Antennal fossa separated by more than greatest width of antennal pedicel from ocular rim; hindwing about twice as long as wide, much shorter than forewing (Fig. 8); forewing vein 2A closely associated with 3 A, not much angled (Fig. 2); subcostal and radial veins of both wings nearly touching, especially before their apical fusion; male ectoproct with elongate post- ventral lobe Pachyleon Antennal fossa separated by less than greatest width of antennal pedicel from ocular rim; hindwing over three times longer than wide, about equal in length to forewing; forewing vein 2A widely separated from normal 3A before strong angle toward hind margin; subcostal and radial veins widely separated until their apical fusion; male ectoproct without postventral lobe (except P. cautus) 5 5. Forewing with anterior margin abruptly angled and swollen near coalescing of subcostal and radial veins, costal area at this point nearly twice as wide as at middle of wing; hindwing vein CuA ends well before midpoint to fork- ing of MP2; forewing radial sector originates well basad to forking of CuA Araucaleon 18 Contributions in Science No. 186 Forewing with anterior margin evenly curved toward apex, costal area at point of coalescing of subcostal and radial veins usually somewhat narrower than at middle of wing; hindwing vein CuA extends nearly to forking of MPo or beyond; forewing radial sector originates near or well beyond forking of CuA 6 6. Forefemur swollen, at widest point much wider than interantennal distance, with abundant pubescence, especially on closing surface; femoral sense hair or foreleg elongate, over one-half length of femur; female ectoproct often with ventral projection Psammoleon Forefemur slender and elongate, at widest point about equal at most to in- terantennal distance, with sparser pubescence; femoral sense hair of foreleg less than one-third length of femur; female ectoproct not produced ven- trally 7 7. Apical one-fifth of hindwing and usually forewing with predominant dark brown suffusion in marked contrast to basal one-fifth; posterior gonapophy- sis of female weakly produced as a swelling Glenurus Apical one-fifth of wings predominantly transparent, sometimes many scattered dark brown spots present; posterior gonapophysis of female well developed, digitiform Eremoleon Synoptic Catalog of the Western Hemisphere Glenurini WITH Type Information Genus Araucaieon Banks Araucaleon Banks 1938b: 127. Type-species: Araucaieon inca Banks, orig. design. inca Banks 1938b: 128. Hoi. 5, Huachi, Bolivia, IX.1925 (USNM). withycombei (Esben-Petersen) 1927:347 (as Glenurus). Hoi. $, St. Augus- tine, Trinidad, VI. 1924 (BM). New Comb. Genus Dimarella Banks Dimarella Banks 191 3a: 229. Type-species: Eremoleon angustus Banks, orig. design. =Nobra Navas 1915b:6. Type-species: N o bra mar tinsi as, ong. design. =Furgus Navas 1921:51. Type-species: Nobra riparius Navas, orig. design. =Mystroleon Banks 1924:436. Type-species: Myrmeleon praedator^ a\kev . Taxonomy: Banks ( 1943: 166) ; Stange (1963:810-812; 1968:59). angusta (Banks) 1908:31 (as Eremoleon). Hoi. 9, Santa Elena, Ecuador (MCZ). =Nobra nevermanni Navas 1936:167. Syntypes, Costa Rica, Farm Ham- burg am Reventazon, V.2.1934 (Zaragoza). Taxonomy: Banks (1913a:230; 1943 : 168) ; Stange (1963:812). 1970 Catalog of the Western Hemisphere Glenurini 19 Distribution: Neotropical (Costa Rica, Colombia, Ecuador, Venezuela, Brazil, Peru). campestris (Navas) 1930:107 (as Furgus). Hoi. $ , Guayaquil, Ecuador 1930 (Barcelona). New Comb. garciai (Navas) 1932:12 (as Nobra). Hoi. 9, Vinces, Ecuador (Turin). New Comb. martinsi (Navas) 1915b:6 (as Nobra). Hoi. 9, Rio Purus, Brasil 1904 (Paris). Taxonomy: Stange (1963:812). menkei Stange 1963:814. Hoi. 5, Rio Cuchujachi, 10 miles S.E. Alamos, Sonora, Mexico, V.22.1962 (CAS). praedator (Walker) 1853:391 (as Myrmeleon) . Hoi. 9, Santarem, Brazil (BM). Taxonomy: Hagen (1860:364) ; Stange (1963:812). psammophila Stange 1963:814. Hoi. 9, Veracruz, Mexico, IV.29.1962 (CAS). riparius (Navas) 1918b: 16 (as Nobra). Lect. 5, Santa Fe, Argentina, 1.9. 1917 (Barcelona). —Dimarella pallida Navas 1933a:89. Hoi. 9, S. Miguel, Argentina, XI. 25. 1932 (Barcelona). Taxonomy: Stange (1968:59). Distribution: Neotropical (Uruguay, Argentina). silvaticus (Navas) 1918b:6 (as Nobra). Hoi. 9, Mato Grosso, Brasil (La Plata). N. Comb. tarsalis (Guilding) 1829:47 (as Formicaleo) . Type (?s), ?Demerara (not located) . ^Myrmeleon efferus Walker 1853:388. Hoi. 9, Para, Brazil (BM). Taxonomy: Hagen (1860:364) ; Banks ( 1913a:229; 1913b:86; 1943:168); Stange (1963:812). Genus Elachyleon Esben-Petersen Elachyleon Esben-Petersen 1927:348. Type-species: Elachyleon piinctipennis Esben-Petersen, orig. design. =Oroleon Navas 1927:49. Type-species: Oroleon serranus Navas, orig. de- sign. =Sericoleon Esben-Petersen 1932:114. Type-species: Sericoleon paessleri Esben-Petersen, orig. design. New Synonymy. Taxonomy: Banks ( 1943 : 166) ; Stange (1963:810; 1968:56). gentilis (Banks) 1943:171 (as Sosa). Hoi. $, Kartabo, British Guiana, XI.4 (MCZ). New Comb. paessleri (Esben-Petersen) 1932:115. (as Sericoleon) . Hoi. 9, Taltal (Anto- fagasta), Chile, 1.30.1906 (Copenhagen). New Comb, punctipennis Esben-Petersen 1927:348. Hoi. (sex undet.), St. Augustine, Trinidad, V.20. 1925 (BM). 20 Contributions in Science No. 186 =Elachyleon punctipennis pulchellus Esben-Petersen 1932:113. Syntypes: 9 , Las Mercedes, Costa Rica, IV.27.1922 (Hamburg, destroyed) ; sex un- det., Brazil (Berlin?). Taxonomy: Banks ( 1943: 168) ; Stange (1963:815; 1968:56). Distribution: Neotropical (Mexico to Argentina). serranus (Navas) 1927:50 (as Oroleon) . Hoi. 9, Alta Gracia, La Granja, Sierras de Cordoba, Argentina, 1.8.1928 (Buenos Aires). Taxonomy: Stange (1968:57). Genus iremoleon Banks Eremoleon Banks 1901:366. Type-species: Mynneleon macer Hagen, orig. design. =Incamoleon Banks 191 3a: 229. Type-species: Psammoleon punctipennis Banks, orig. design. New Synonymy. —Glenopsis Banks 191 3a: 229. Type-species: Myrmeleon anomalus Ram- bur, orig. design. —Sosa Navas 1914a:218. Type-species: Sosa conspicuus Navas, orig. de- sign. New Synonymy. —Segura Navas 1914c: 18. Type-species: Segura vitreus Navas, orig. de- sign. —Belen Navas 1921:119. Type-species: Belen cerverinus Navas, orig. de- sign. =Cortesius Navas 1924:107. Type-species: Cortesius genini Navas, orig. design. New Synonymy. =Novulga Navas 1925:189. Type-species: Novulga mexicanaNdiwks, orig. design. —Dobla Navas 1926:428. Type-species: Dobla arcuata Navas, orig. design. New Synonymy. =Joergenia Esben-Petersen 1932:118. Type-species: Joergenia pulchra Esben-Petersen, orig. design. New Synonymy. =Antilloleon Banks 1943:168. Type-species: Glenurus cerverai Navas, orig. design. New Synonymy. Taxonomy: Navas (1916:232; 1917:275); Banks (1928:69-71; 1942: 144); Adams (1957:85; 1958:6); Alayo (1968:62, 70); Stange (1968: 57). anomalus (Rambur) 1842:388 (as Myrmeleon) . Type (?s), Colombia (not located) . — Glenurus mollis Gerstaecker 1888:101. Hoi. (sex undet.) Colombia (Greifswald) . Taxonomy: Banks (1913a:229; 1943:171); Navas (1916:232); Stange (1968:57). Distribution: Neotropical (Colombia, Venezuela, Brazil, Argentina). 1970 Catalog of the Western Hemisphere Glenurini 21 capitatus (Navas) 1913:52 (as Formicaleo) Hoi. (sex undet.), Rio de Ja- neiro, Brasil, VI. 1911 (not located). New Comb, cerverai (Navas) 1921:118 (as Glenurus). Hoi. 9, Santiago (de las) Vegas, Habana, Cuba, VIII.23.1907 (MCZ). New Comb. Taxonomy: Banks ( 1943 :68) ; Alayo (1968:68). cerverinus (Navas) 1921:120 (as Belen) . Hoi. 9, Rio Almendares, Habana, Cuba, VIII.1.1915 (MCZ). Taxonomy: Adams ( 1957:6) ; Alayo (1968:70). conspicuus (Navas) 19 14a: 21 9 (as Sosa). Hoi. Brasil (Vienna). New Comb. femoralis (Banks) 1942:146 (as Psammoleon) . Hoi. (sex undet.), 20 miles N.W. La Paz, Lower California VII. 16. 1938 (CAS). New Comb, genini (Navas) 1924:108 (as Cortesius). Syntypes (sex undet.), Veracruz, Mexico, 1921 (Paris). New Comb. =DobIa arcuata Navas 1926:428. Hoi. 9, San Jose, Costa Rica (Paris). New Synonymy. gracile Adams 1957:90. Hoi. 9, Riverside, California, VIII.31.1939 (CAS). impluviatus (Gerstaecker) 1893:30 (as Glenurus). Hoi. 9, Locotal, Bolivia (Greifswald), New Comb. insipidus Adams 1957:88. Hoi. 5, 5 miles S. San Miguel, Lower California, VII.20.1938 (MCZ). Distribution: Nearctic (Mohave-Colorado Desert; Vizcaino Desert). longior Banks 1938a:225. Syntypes 9 9 , Yucatan, Mexico (MCZ). Taxonomy: Adams ( 1957:92) . macer (Hagen) 1861 :236 (as Myrmeleon) . Hoi. (sex undet.), Mexico (not located) . =Segura v it reus Navas 1914c: 18. Hoi. <$ , Cuernavaca, Mexique, 1871 (Paris). =Novulga mexicana Navas 1925:189. Hoi. $ , Veracruz, Mexique, 1921 (Paris). =Hesperoleon atomarius Navas 1933b: 105. Hoi. (sex undet.) , Cuernavaca, Mexico, 1929-30 (Hamburg, destroyed) . New Synonymy. Taxonomy: Banks (1928:70) ; Adams (1957:86). Distribution: Nearctic (Arizona); Neotropical (Mexico) . nigri basis Banks 1920:329. Syntypes $ , 9 , St. George, Utah, VI. 5, 6 (MCZ) =Eremeleon affine Banks 1942:144. Hoi. $ , Miraflores, Lower California, VII.8.1938 (CAS). Taxonomy: Banks (1928:71 ); Adams (1957:91). Distribution: Nearctic (deserts). ornatipennis (Alayo) 1968:69 (as Antilloleon) . Hoi. Soroa, Pinar del Rio, Cuba, IV. 1963 (Havana). New Comb, pallens Banks 1941c: 101. Syntypes, Picacho Peak, Arizona, VII. 23 (MCZ) . Taxonomy: Adams (1957:92). 22 Contributions in Science No. 186 peterseni (Banks) 1922:59 (as Glenopsis) . Lect. $, Chanchamayo, Peru, XI (MCZ). New Comb. Taxonomy: Stange (1961:674). psilocerus (Gerstaecker) 1893:32 (as Glenurus) . Hoi. (sexundet.), Merida, Venezuela (Greifswald) . New Comb. Taxonomy: Banks (1922:59; 1943:172). pulchra (Esben-Petersen) 1932:118 (as Joergenia). Hoi. 9, Paraguay, II. 1932 (Copenhagen). New Comb. punctipennis (Banks) 1910:147 (as Psammoleon). Hoi. 9, San Antonio, Colombia (MCZ). New Comb. —Formicaleo tetrastictus Navas 1913:51. Types(?), Venezuela (not located). —Formicaleo stictopterus Navas 1916:232. New name for punctipennis Banks 1910. (nomen nudum). Taxonomy: Banks (1913a:299; 1943 : 169) ; Stange (1968:58). Distribution: Neotropical (Colombia to Argentina). sectoralis Adams 1958:7. Hoi. 9, 5 miles S. San Miguel, Lower California, VII.20.1938 (CAS). triguttatus (Navas) 1914c: 19 (as Formicaleo) . Hoi. 9, San Pedro de Sula, Honduras (Vienna). New Comb. Genus Glenurus Hagen Glenurus Hagen 1866:372. Type-species: Formicaleo grata Say, design, by Banks (1928:67). =Lec/o5'c/w5’ Navas 1918:493. Type-species: Ledoscius penningtoni Navas, orig. design. SYNONYMY: Navas (1924:107). gratus (Say) 1839:45 {as Formicaleo) . =Myrmecoleon roseipennis Burmeister 1839:995. Hoi. (sex undet.), Nord- Amerika (Halle). Taxonomy: Walker (1853:392); Hagen (1861:225; 1866:405); Banks (1892:360; 1928:67). Distribution: Nearctic (East 100° Meridian). heteropteryx GtTsXaQokQY 1885:17. Hoi. 9, Chiriqui (Greifswald). ^Glenurus discorsNawas 1920:202. Hoi. 9, La Trinite, 1916 (Paris). Taxonomy: Banks (1922:58); Navas (1922b:187; 1935:362). Distribution: Neotropical (Panama, Venezuela, Trinidad, Ecuador). incalis Banks 1922:58. Syntypes, Chanchamayo, Peru (MCZ). /i/n/ger Gerstaecker 1893:125. Hoi. 9, Chiriqui (Greifswald). Taxonomy: Banks (1938:420). Distribution: Nearctic (Arizona); Neotropical (Mexico to Panama). peculiaris (Walker) 1859:194 (as Myrmeleon) . Hoi. 9, Brazil (BM). 1970 Catalog of the Western Hemisphere Glenurini 23 —Glenuriis brasiliensis Navas 1920:416. Lect. $, Sao Sebastiao, Brasil (not located). New Synonymy. Taxonomy: Hagen (1866:405); Banks (1922:58); Navas (1923:771). penningtoni (Navas) 1918:493 (as Ledoscius) . Hoi. La Rioja, Argentina (not located). —Glenurus croesus Banks 1 922 : 59. Syntypes, Sara, Bolivia, 450 m. ( MCZ) . Taxonomy: Navas (1922b: 187; 1924:107). Distribution: Neotropical (Bolivia, Argentina), pro/ Navas 1929:17. Hoi. $, Colima, Mexico (Paris). snowii Banks 1907:100. Hoi. $, Baboquivari Mts., Arizona (MCZ). Taxonomy: Banks (1928:67, 69). Genus Navasoleon Banks Navasoleon Banks 1943:168. Type-species: Gymnocnemia boliviana Banks, orig. design. Taxonomy: Stange (1963:810). boliviana (Banks) 1929:330 (as Gymnocnemia) Hoi. $, Rio Longo, Bolivia (MCZ). Taxonomy: Banks (1943:168). bosqiii (Navas) 1922:258 (as Gymnocnemia) . Hoi. 9, Santiago del Estero, Argentina (not located). Taxonomy: Stange (1968:58). leptocera (Navas) 1915:125 (as Gymnocnemia) . Hoi. (sex undet.) Cata- marca, Argentina, 1.21.1910 (Buenos Aires). Taxonomy: Stange (1968:58). Genus Pachyleon Stange Pachyleon Stange 1970: Type-species: Pachyleon alvarengai Stange, orig. design. alvarengai Stange. Hoi. $ , Jacare, Mato Grosso, Brazil, XI. 1961 (LACM) . Genus Psammoleon Banks Psammoleon Banks 1899:69. Type-species: Myrmeleon ingeniosus Walker, orig. design. =Diazus Navas 1914:220. Type-species: Diazus clavatus Navas, orig. de- sign. Taxonomy: Banks (1913:226; 1928:60-67; 1942:166; 1943:169); Alayo (1968:65). albovaria (Banks) 1942:146 (as Puren) . Hoi. (sex undet.) , Venancio, Lower California, VII.17.1938 (CAS). New Comb, arizonensis Banks 1935:53. Syntypes : 19, Tucson, Arizona (MCZ) ; 2 9 9 , Phoenix, Arizona (MCZ). 24 Contributions in Science No. 186 Distribution : Nearctic (Arizona-Sonora Desert; Mohave-Colorado Desert) . banksi Esben-Petersen 1932:115. Hoi. Amapala, Honduras, XL 12. 1907 (Hamburg, destroyed). bipunctatus (Navas) 1915:465 (as Formicaleo) . Hoi. $, Guyana Francesa (Paris) . Taxonomy: Banks (1943:170). bistictus (Hagen) 1861:235 (as Myrmeleon) . Hoi. (sex undet.), Cuba (Berlin?). Taxonomy: Navas (1921 : 1 17) ; Banks (1928:61; 1941 : 177) ; Alayo (1968:65). Distribution: Neotropical (so. Florida, West Indies, Yucatan, Mex. ) . cautus (Walker) 1853:349 (as Myrmeleon) . Hoi. $ , Brazil (BM) . =Feinerus nebulosus Navas 1922:21. Hoi. 2, Peru 1920 (Paris). New Synonymy. Taxonomy: Banks (1943: 170). clavatus (Navas) 1914:221 (as Diazus). Hoi. $ , Amazones 1860 (Vienna) . New Comb. connexus (Banks) 1920:329 (as Puren) . Hoi. S , San Jacinto Mts., California VI.25 (MCZ). Taxonomy: Banks (1928:61, 64) . cubensis Alayo 1968:67. Hoi. (sex undet.), Tortuguilla, Guantanamo, Or- iente, Cuba, VI. 1964 (Havana). debilis (Gerstaecker) 1893:44 (as Formicaleo) . Hoi. 2, Chiriqui (Greifs- wald) . Taxonomy: Banks (1943:170). Distribution: Neotropical (Panama; Colombia). decipiens Banks 1935:54. Syntypes: IS, Georgia (MCZ) ; sex undet., Shreveport, Louisiana (MCZ). giittipes Banks 1906:99. Hoi. $ , Tryon, North Carolina (MCZ). Taxonomy: Banks (1928:61, 63). im bell is (Banks) 1941 : 102 (as Puren) . Hoi. 2 , Port au Prince, Haiti, XI.20 (MCZ). New Comb. ingeniosus (Walker) 1853:337 (as Myrmeleon) . Syntypes, Brazil & no data (BM). Taxonomy: Hagen (1861:236; 1866:404); Banks (1899:69; 1906:99; 1943:169). Distribution: Neotropical (Colombia, Brazil) . iniquus (Navas) 1914:208 (as Formicaleo) new name for Formicaleo in- aequalis Navas 1913:51 (Preoccupied by F. inaequalis Navas 1912) : Hoi. (sex undet.) , Amapola, Honduras, XI (not located) . New Comb, inscriptus (Hagen) 1861 :230 (as Myrmeleon) . Hoi. S , Pecos River, Western Texas (MCZ) . Taxonomy: Hagen ( 1866:424) ; Banks ( 1904: 106; 1928:66). 1970 Catalog of the Western Hemisphere Glenurini 25 Distribution: Nearctic (Arizona-Sonora Desert). leachii (Guilding) 1829:49 (as Formicaleo) . Types (?), Jamaica (not lo- cated) . minor Banks 1928:62. Lect. $, Dry Tortugas, Loggerhead Key, Florida, VI.1917 (MCZ). Taxonomy: Banks (1941:176, 177); Stange (1961:674); Alayo (1968: 66). Distribution: Neotropical (So. Florida; West Indies). normalis Banks 1942:145. Hoi. $, Venancio, Lower California, VII.17.1938 (CAS). nubipennis (Navas) 1917:275 (as Formicaleo). Hoi. (sex undet.), Coachi, Colombia, VI. 19 15 (not located). New Comb, parallela Banks 1935:54. Syntypes, Honduras (MCZ). Banks 1920:331. Hoi. 5, Chapada, Brazil (MCZ). Taxonomy: Banks (1943:170). posticatus Banks 1941:3. Hoi. (sex undet.), Colombia (AMNH). Taxonomy: Banks (1943:169). reductus Banks 1941:177. Hoi. 5 , Stakes Bay, Cayman Brae, Cayman Islands (MCZ). serrei (Navas) 1920:201 Formicaleo) . F\o\. 5 , La Trinite, 1914 (Paris). New Comb. —Formicaleo chaperi Navas 1922:256. Hoi. S, Venezuela, 1885 (Paris). New Synonymy. Taxonomy: Banks (1943:170). sinuatus Currie 1903:275. Hoi. $ , Santa Rita Mts., Arizona, V.31 (USNM). =Psammoleon serpentinus Navas 1922:185. Hoi. (sex undet.), Jemez Springs, New Mexico, VII. 1916 (Zaragoza). Taxonomy: Banks (1928:5, 64). zay as i Alsiyo 1968:67. Hoi. 9 , Peninsula de Guanahacabibes, Pinar del Rio, Cuba, 1956 (Private coll. Zayas, Habana). Resumen Revision Generica y Catalogo de la Tribu Glenurini del Hemisferio Occidental con la Descripcion de un Nuevo Genero y Especies de Brasil. Estudios sistematicos de la Glenurini del hemisferio occidental han llevado a grandes cambios en su clasificacion. Ocho nombres genericos han sido reducidos a sinominos y un genero y especie nueva, Pachyleon alvarengai es descripto de Brasil. Estos sinonimos son: Elachyleon Esben-Petersen 1927 i=Sericoleon Esben-Petersen 1932); Eremoleon Banks 1901 {=Incamoleon Banks 1913, =Sosa Navas 1914, —Cortesius Navas 1924, =Dobla Navas 1926, —Joergenia Esben-Petersen 1932, = Antilloleon Banks 1943); Psam- moleon Banks 1899 {—Diazus Navas 1914). Se dan ahora ocho generos validos. Se incluyen descripciones y observaciones sobre cada genero ademas 26 Contributions in Science No. 186 de una clave para su identificacion. Se discute la posicion sistematica de la tribu Glenurini que se pasa a la subfamilia Myrmeleontinae, con la inclusion de los generos de la tribu Dimarellini. Se agrega un catalogo sinoptico de las especies y generos de la tribu con datos sobre los tipos y se indican muchas combinaciones nuevas ademas de cuatro sinonimos especificos nuevos. Estos son: Psammoleon cautus (Walker) 1853 {—Feinerus nebulosus Navas 1922); Psammoleon serrei (Navas) 1920 {—Formicaleo c/zapm Navas 1922); Glen- urus peculiaris (Walker) 1859 {—Glenurus brasiliensis Navas 1920) ; Eremo- leon macer (Hagen) 1861 {~Hesperoleon atomarius ^diwis 1933). Literature Cited Adams, P. A. 1957. New ant-lions from the southwestern United States (Neurop- tera: Myrmeleontidae). Psyche, 63:82-108, 46 figs. . 1958. Further notes on the genus Eremoleon Banks, with a new species (Neuroptera: Myrmeleontidae). Psyche, 64:6-8, figs. la-e. Alayo, D. P. 1968. Los Neuropteros de Cuba. Poeyana (B), 2:5-127, 31 figs, 10 pis. Banks, N. 1892. A synopsis, catalogue, and bibliography of the neuropteroid insects of temperate North America. Amer. Ent. Soc., Trans., 19:327-373. . 1899. A classification of the North American Myrmeleonidae. Canad. Ent., 31:67-71. . 1901. A list of neuropteroid insects from Mexico. Amer. Ent. Soc., Trans., 27:361-371, 15 figs. . 1904. Neuropteroid insects from New Mexico. Amer. Ent. Soc., Trans., 30: 97-110, figs. 1-14. . 1906. Three new species of Neuroptera. Psyche, 13:98-100. . 1907a. A new ant lion-fly from Arizona. Ent. News, 18:100-101. — . 1907b. Catalogue of the Neuropteroid insects (except Odonata) of the United States. Amer. Ent. Soc., Philadelphia. 53 p. . 1908. New tropical American Neuroptera. Ent. Soc. Wash., Proc., 9:30-34. . 1910. New South American neuropteroid insects. Ent. Soc. Wash., Proc., 12:146-160. . 1913a. Synopses and descriptions of exotic Neuroptera. Amer. Ent. Soc., Trans., 39:201-242, 42 figs. . 1913b. Neuropteroid insects from Brazil. Psyche, 20:83-89, 14 figs. . 1920. New neuropteroid insects. Harvard, Mus. Comp. Zool., Bull., 64:297- 362, no figs. . 1922. South American Glenurus and some other Myrmeleonidae. Canad. Ent., 54:58-60. . 1924. Descriptions of new neuropteroid insects. Harvard, Mus. Comp. Zool., Bull., 65:419-455,60 figs. . 1928. Revision of the Nearctic Myrmeleonidae. Harvard, Mus. Comp. Zool., Bull., 68:1-84, 101 figs. . 1935. A few new North American Neuroptera. Psyche, 42:53-57. -. 1938a. A new myrmeleonid from Yucatan. In Pearse, A. S., et al.. Fauna of the caves of Yucatan. Carnegie Inst. Wash. Publ., 491:225. . 1938b. Two new genera of Myrmeleonidae. Ent. Soc. Wash., Proc., 40:125- 129, 4 figs. . 1941a. Some new and interesting Neuroptera in the American Museum of Natural History. Amer. Mus. Novitates, 1 143: 1-5. 1970 Catalog of the Western Hemisphere Glenurini 27 . 1941b. Results of the Oxford University Cayman Islands biological expedi- tion, 1938 (Neuroptera). Entomol. Mon. Mag., 77:176-177. . 1941c. Three new species of Myrmeleonidae (Neuroptera). Psyche, 48:101- 104. . 1942. Contributions toward a knowledge of the insect fauna of Lower Cali- fornia. No. 4. Myrmeleonidae. Calif. Acad. Sci., Proc., (4) 24:133-152, 17 figs. . 1943. Neuroptera of northern South America. Part 2. Myrmeleonidae. Bol. ent. Venezolana, 2:161-173. Burmeister, H. 1839. Handbuch der Entomologie. Berlin. Gymnognatha. (2): 757- 1017. Currie, R. P. 1903. Myrmeleonidae from Arizona. Ent. Soc. Wash., Proc., 5:272- 284. Esben-Petersen, P. 1927. New and little-known species of Neuroptera in British collections. Part III. Ann. Mag. Hist., (9) 20:343-350, 8 figs. . 1932. New and little-known Neuroptera. Dansk naturh. Foren, Vidensk. Medd. 94:109-123, 13 figs. Gerstaecker, a. 1885. Vier Decaden von Neuropteren aus der Familie Megalop- tera, Burm. Naturw. Ver. Greifswald, Mitt., 16: 1-49. . 1888. Weitere Beitriige zur Artenkenntniss der Neuroptera Megaloptera. Naturw. Ver. Greifswald, Mitt., 19:89-130. . 1893. liber neue und weniger gekannte Neuropteren aus der Familie Megal- optera, Burm. Naturw. Ver. Greifswald, Mitt., 25:93-173. Guilding, L. 1829. The generic characters of Fonnicaleo\ with the description of two new species. Linn. Soc. London, Trans., 16:47-51. Hagen, H. 1860. Beitrag zur Kenntniss der Myrmeleon-Avitn. Stettin, ent. Ztg., 21: 359-369. . 1861. Synopsis of the Neuroptera of North America. With a list of the South American species. Smithson. Misc. Coll., 4: 1-347. . 1866. Hemerobidarum Synopsis Synonymica. Stettin, ent. Ztg., 27:369-462. Markl, W. 1954. Vergleichend-morphologische Studien zur Systematik und Klassi- fikation der Myrmeleoniden. Naturf. Ges. Basel, Verb., 65: 178-263, 85 figs. Navas, L. 1913. Neuropteros nuevos de America. Broteria, 11:45-53, 6 figs. . 1914a. Neuropteros sudamericanos. Primera Serie. Broteria, 12:45-57, 215- 235, 11 figs. . 1914b. Neuropteros nuevos o poco conocidos. Segunda & Tercera Serie. Acad. Cienc. Barcelona, Mem., (3) 11(8, 13): 105-1 19, 193-215, 14 figs. . 1914c. Neuropteres nouveaux de I’Amerique du Nord. II. Ent. Zeitschr., 28: 18-20, 4 figs. — . 1915a. Neuropteros nuevos o poco conocidos. Quinta Serie. Acad. Cienc. Barcelona, Mem., (3) 1 1(27) :455-480, 18 figs., pi. 1, 6 figs. — . 1915b. Neuropteros sudamericanos. Segunda Serie. Broteria, 13:5-13, 6 figs. . 1916. Neuropteros nuevos o poco conocidos. Acad. Cienc. Barcelona, Mem., (3) 12(13):219-243. — — . 1917. Neue Neuropteren. Ent. Mitteil., 6:274-282, 5 figs. . 1918a. Algunos insectos de la Republica Argentina. Acad. Cienc. Madrid, Rev., 16:491-504, 10 figs. . 1918b. Insecta nova. IV Serie. Apcad. Pont. Nuovi Lincei, Mem., (2) 4:1- 23, 5 figs. . 1919. Algunos insectos Neuropteros de la Republica Argentina. Serie Ter- cera. Acad. Cienc. Madrid, Rev., 17:287-305, 6 figs. — — — . 1920. Sur des Nevropteres nouveaux ou critiques. Deuxieme Serie. Soc. sci. Bruxelles, Ann., 39:189-203, 5 figs. 28 Contributions in Science No. 186 . 1921. Insectos americanos nuevos o criticos. Broteria, 19:113-124, 5 figs. — . 1922a. Insectos sudamericanos. Cuarta Serie. Acad. Cienc. Madrid, Rev., 19:255-267. -. 1922b. Estudis sobre Neuropters (Insectes). Inst. Cien. Barcelona, Arx., 7: 179-203, 2 figs. . 1922c. Algunos insectos del museo de Paris. Acad. Cien. Zaragoza, Rev., 7: 15-27, 6 figs. -. 1924. Algunos insectos del Museo de Paris. Broteria, 21:99-114, 5 figs. -. 1925. Insectos exoticos nuevos o poco conocidos. Acad. Cienc. Barcelona, Mem., (3) 19:181-200, 12 figs. — -. 1926. Neue Insekten (Neuropt.). Deutsch Ent. Zeitschr., 1926:428-431, 2 figs- . 1927. Veinticinco formas nuevas de insectos. Bol. Soc. iber. cien. nat., Zara- goza, 26:48-75, 9 figs. . 1929. Insectos neotropicos. 5a Serie. Rev. chil. Hist. Nat. 33: 17-24, 3 figs. . 1930. Neuropteros del Ecuador. Rev. chil. Hist. Nat. 34:107-109, 1 fig. — . 1932. Alcuni insetti del Museo di Zoologie della R. Universita di Torino. Turin, Univ., Mus. ZooL, Boll., (3) 42 (26) : 1-38, figs. — -. 1933a. Insectos de la Argentina. Rev. Acad. Cienc. Zaragoza, 16:87-120, 22 figs. — . 1933b. Decadas de insectos nuevos. Broteria (Cienc. Nat.), 2: 101-110. . 1935. Insectos sudamericanos. Decima Serie. Acad. Cienc. Madrid, Rev., 32:360-375. . 1936. Decadas de insectos nuevos. Broteria (Cienc. Nat.) 32:161-170. Rambur, J. P. 1842. Histoire naturelle des insectes. Nevropteres. Paris. 534 p., tab. col. i-xii. Say, T. 1839. Descriptions of some N. American Neuropterous Insects, and observa- tions on some already described. Acad. Nat. Sci. Philadelphia, Journ., 8:9-46. Smith, R. C. 1931. The Neuroptera of Haiti, West Indies. Ent. Soc. Amer., Ann., 24: 798-823, 21 figs. Stance, L. A. 1961. Lectotype designations in the New World Myrmeleontidae. Canad. Ent., 93:674-677. . 1963. The Dimarellini of Mexico with the descriptions of two new species of Dimarella. Ent. Soc. Amer., Ann., 56:810-816, 16 figs. . 1968. Catalogo de Neuroptera de Argentina y Uruguay. Acta Zool. Lill., 22 (1967):5-86. — . 1970. Revision of the Ant-lion Tribe Brachynemurini of North America. U.C. Publ. Ent. 55:1-192, 231 figs., 31 maps. Walker, F. 1853. List of the specimens of neuropterous insects in the collection of the British Museum. Part II (Sialidae — Nemopterides) . London: 193-476. . 1859. Characters of undescribed Neuroptera in the collection of W. W. Saunders. Ent. Soc. London, Trans., 5 : 176-199. Accepted for publication February 18, 1970 I LOS * ANGELES I COUNTY i MUSEUM I CONTRIBUTIONS IN SCIENCE MBER 187 C;i L ^ a May 4, 1970 FURTHER STUDIES ON AUDIBLE VOCALIZATIONS OF THE AMAZON FRESHWATER DOLPHIN, INIA GEOFFRENSIS By Melba C. Caldwell and David K. Caldwell Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 ( 1 IQ,: CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author oi divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor FURTHER STUDIES ON AUDIBLE VOCALIZATIONS OF THE AMAZON FRESHWATER DOLPHIN, INIA GEOFFRENSIS By Melba C. Caldwell^ and David K. Caldwell^ Abstract: Evidence is presented indicating the absence of a quasi-pme tone or whistle type of sound production by the Ama- zon freshwater dolphin. A discussion and sonagrams are given of the audible squeal or screech often attributed to Inia geoffrensis. Phonations of the toothed whales and dolphins fall roughly into three categories. These include two types which generically are related (Norris, 1969) and probably are produced by the same mechanism, as they both con- sist of discrete clicks. One type demonstrates a slower click repetition rate and is usually associated with environmental investigation (echolocation), whereas the other is made up of bursts of clicks of higher repetition rates of 150 per second or more (Caldwell, Caldwell and Evans, 1966b). The latter create tonal sounds variously labeled by such names as barks, squawks, screeches or squeaks. These are associated with behavioral situations wherein we find a positive correlation between the time of their emission and the use- fulness of a signal that could cue conspecifics to approach or withdraw (Cald- well and Caldwell, 1967). The third type of sound, most frequently termed a whistle, is a quasi-^mo, tone sound that cannot be resolved into individual clicks. These are the harmonic emissions of Tavolga ( 1965: 9) and Evans (1967). All odontocetes probably emit click sounds (Norris, 1968), but some do not emit the so called pure tone. No whistles have been reported in sound work on the primitive Amazon freshwater dolphin, Inia geoffrensis (see Schevill and Watkins, 1962; Schevill, 1964; Caldwell, Caldwell and Evans, 1966a, 1966b; Caldwell and Caldwell, 1967; Evans, 1967; Poulter, 1968; Caldwell and Caldwell, 1969b). All of these earlier works contain two deficiencies. First, on only one brief occasion was work done with more than one or two animals in a tank (Caldwell and Caldwell, 1967) . This lack of community tank sound studies constitutes a potentially serious deficit, as vocalizations of odontocete ceta- ceans of the marine species with which we have worked tend to diminish in relative number in captivity if other animals are not present. Second, several ^Research Associate in Cetology, Los Angeles County Museum of Natural History; Staff Research Associate, Communication Sciences Laboratory, University of Flor- ida, Gainesville. ^Research Associate in Cetology, Los Angeles County Museum of Natural History; Associate Professor, Communication Sciences Laboratory, and Staff Research Asso- ciate, Florida State Museum, University of Florida, Gainesville. 1 2 Contributions in Science No. 187 people handling Inia have reported a loud squeal or screech emitted by the animal which, from the observers’ descriptions, might have been interpreted as a whistle (Allen and Neill, 1957; Layne and Caldwell, 1964; Caldwell, Caldwell and Evans, 1966b; Caldwell and Caldwell, 1969a, 1969b). This sound also has been described to us in personal conversation with Earl S. Herald of the Steinhart Aquarium, San Francisco; Lawrence Curtis, formerly of the Fort Worth Zoo, Texas; Leo Baumer of Iquitos, Peru; and W. J. LeBlanc and William C. Raulerson, Marineland of Florida. Until recently we had heard the sound out of water on one occasion, but did not record it. Further studies on this sound were thus considered necessary to determine whether these sounds are broad-band pulsed sounds and not a narrow-band whistle. Correction of these two deficiencies seemed mandatory, not only to help solidify our basic knowledge of cetacean phonations, but also because we have come to regard the absence or presence of the pure tone whistle as one of the significant characters in the precariously understood familial relation- ships between members of the order Cetacea. Evidence to date suggests that the species of cetaceans considered to be more primitive also either lack the quasi-^xxxQ tone whistle or that when present it is less clear-cut than in those species considered to be more advanced. We recorded Inia phonations from a community tank over a period of two years for a total of some 25 hours. The dolphins were held in a large display tank at Marineland of Florida and at various times the community contained from four to eleven animals of mixed sizes and sexes. Sound record- ings were made with the animals at rest or swimming slowly, feeding, fighting, copulating, sick, dying, and semi-stranded. The community tank contained eleven animals during one seven hour recording session of a semi-stranding. The animals were recent captives which, in our experience, along with strand- ing increases the possibility of eliciting at least a few cetacean phonations. Additionally we have air recorded isolated individuals as they were being handled and medicated, as well as during venipuncture and force feeding. Within the range of our equipment (40 to 20,000 Hz), we have not detected quasi-^mc tone sounds, nor did Evans {in Herald, et al., 1969; per- sonal conversation, August, 1969) indicate the production of such sounds in his studies of Inia phonations with equipment sensitive to frequencies up to 100,000 Hz. The loud squeals that we recorded are pulsed as shown when they are subjected to detailed analysis, and the quasi-^mo, tone or whistle is still unknown in the vocal repertoire of Inia and we believe now that it is not to be expected. An adult female emitted squeals both singly and in series 22 and 19 times on two successive days when she was removed from her tank for medi- cation. On another occasion this same female, although not being handled, intermittently emitted loud squeals for about an hour when the water level in her tank was so low that she was partly stranded; but no recordings were 1970 Vocalizations of the Amazon Dolphin 3 0 0.2 0.4 TIME (SEC.) 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 TIME (SEC.) Figure 1. Phonations of Inia geoffrensis. Squeals emitted in air in daylight by isolated animals during application of medication. Upper: Adult female (MLF 165), Janu- ary 27, 1969. Lower: Juvenile male (MLF 202), March 10, 1969. Effective filter bandwidths 300 Hz. (Photographs courtesy Marineland of Florida). made. A juvenile male emitted seven squeals on one occasion when removed for medication and venipuncture. The four occasions are the only times of perhaps 30 to 40 in-air recording or listening sessions that the squeal was emitted by any individual of this species although all were subjected to the same amount of handling. Sound spectrograms (sonagrams) of the squeals of both animals (Fig. 1) show them to be the burst-pulse type with a high click repetition rate (Wat- kins, 1967), Although these are broad band clicks with some energy extend- ing above 12 kHz (Fig. 2), a strong fundamental frequency at one to two kHz is demonstrated on the sonagrams of the sounds of both animals (Fig. 1). These sonagrams show several characteristics in common with the type 2 4 Contributions in Science No. 187 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 |.0 TIME (SEC.) Figure 2. Phonations of Inia geoffrensis. First two of the same three squeals illus- trated in Fig. 1 (lower), but played at half recorded speed. Effective filter band- width 600 Hz. (Photograph courtesy Marineland of Florida). or complaint type of signal of the pilot whale, Globicephala melaena, (Busnel and Dziedzic, 1966: 615, fig. 11), the tin horn sound emitted when excited by a strong stimulus by the Pacific whitesided dolphin, Lagenorhynchus obliquidens, (Caldwell and Caldwell, 1967 : 889, figs. 5-7), and the squawk component of the whistle-squawk emitted in air by an excited Atlantic bottle- nosed dolphin, Tursiops truncatus, (Caldwell and Caldwell, 1967: 897, fig. 22). Fundamental frequencies, depicted harmonics, and durations are similar in the four species. However, to the human ear the sounds are not alike. Sound Equipment The recordings discussed in this paper were made with a Uher 4000 Report-L recorder operating at a tape speed giving it a frequency response of 40 to 20,000 Hz with a compatible Uher microphone. Sound spectrograms were prepared on a Kay Sona-Graph model 662A Sound Spectrograph Ana- lyzer calibrated in two sections from 85 to 12,000 Hz. When the recorded tape speed is reduced by half, and then fed into the analyzer, the response of the latter is doubled to 24,000 Hz. The effective filter bandwidths used for the illustrated analyses are indicated in the figure captions. Acknowledgments Financial support for certain phases of this work came from the National Science Foundation (GB-1189), the National Institute of Mental Health (MH-07509-01), the Office of Naval Research (N00014-67-C-0358, and modifications POOl and P002), and Marineland, Inc. The photographs are by William A. Huck. One of our associates, Nicholas R. Hall, made the air recordings that included the squeals. 1970 Vocalizations of the Amazon Dolphin 5 Literature Cited Allen, Ross, and Wilfred T. Neill. 1957. White whales of the Amazon. Nat. Hist. 66:324-329. Busnel, Rene-Guy, and Albin Dziedzic. 1966. Acoustic signals of the pilot whale Globicephala melaena and of the porpoises Delphinus delphis and Phocoena phocoena. In K. S. Norris, editor. Whales, dolphins, and porpoises. Berkeley: Univ. Calif. Press, pp. 607-646. Caldwell, Melba C., and David K. Caldwell. 1967. Intraspecific transfer of information via the pulsed sound in captive odontocete cetaceans. In R.-G. Busnel, editor, Les Systemes Sonars Animaux, Biologie et Bionique. Jouy-en- Josas, France: Laboratoire de Physiologie Acoustique. 2:879-936. — AND — . 1969a. The ugly dolphin. Sea Frontiers, 15(5) :308-314. AND . 1969b. More about the ugly dolphin. Sea Frontiers, 15(6) : 349-355. — , AND William E. Evans. 1966a. Preliminary results of studies on the sounds and associated behavior of captive Amazon freshwater dolphins, Inia geoffrensis. In Conf. on Biol. Sonar and Diving Mammals, Stanford Research Institute, California, Proc., 3:178-185. , AND . 1966b. Sounds and behavior of captive Amazon fresh- water dolphins, Inia geoffrensis. Los Angeles County Mus., Contr. Sci., No. 108:1-24. Evans, William E. 1967. Vocalization among marine mammals. In W. N. Tavolga, editor. Marine bio-acoustics. New York: Pergamon Press. 2:159-186. Herald, Earl S., Robert L. Brownell, Jr., Frederic L. Frye, Elkan J. Morris, William E. Evans, and Alan B. Scott. 1969. Blind river dolphin : first side- swimming cetacean. Science, 166:1408-1410. Layne, James N., and David K. Caldwell. 1964. Behavior of the Amazon dolphin, Inia geoffrensis (Blainville), in captivity. Zoologica, 49(2) :81-108, pis. 1-4. Norris, Kenneth S. 1968. The evolution of acoustic mechanisms in odontocete cetaceans. In Ellen T. Drake, editor, Evolution and environment. New Haven, Conn. : Yale Univ. Press, pp. 297-324. — . 1969. The echolocation of marine mammals. In H. T. Andersen, editor. The biology of marine mammals. New York: Academic Press, pp. 391-423. Poulter, Thomas C. 1968. Marine mammals. In T. A. Sebeok, editor. Animal communication; techniques of study and results of research. Bloomington : Indiana Univ. Press, pp. 405-465. ScHEViLL, William E. 1964. Underwater sounds of cetaceans. In W. N. Tavolga, editor. Marine bio-acoustics. New York: Pergamon Press. 1:307-316. — AND William A. Watkins. 1962. Whale and porpoise voices. Woods Hole, Mass.: Woods Hole Oceanogr. Inst. Phonograph record and 24-page booklet. Tavolga, William N. 1965. Review of marine bio-acoustics; state of the art: 1964. Port Washington, N. Y.: U. S. Naval Training Device Center, Tech. Rpt.: NAVTRADEVCEN 1212-1. 100 pp. Watkins, William A. 1967. The harmonic interval : fact or artifact in spectral analysis of pulse trains. In W. N. Tavolga, editor. Marine bio-acoustics. New York: Pergamon Press. 2:15-43. Accepted for publication February 3, 1970 LOS CONTRIBUTIONS IN SCIENCE i ANGELES i COUNTY MUSEUM JMBER 188 May 4, 1970 o C:;iL FOUR NEW TEREBRID GASTROPODS FROM EASTERN PACIFIC ISLANDS By Twila Bratcher and R. D. Burch Los Angeles County Museum OF Natural EiiSTORY • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM. — (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor FOUR NEW TEREBRID GASTROPODS FROM EASTERN PACIFIC ISLANDS By Twila Bratcher^ and R. D. Burch- Abstract: Four new terebrids are described from Eastern Pacific islands: Terebra hertleini, T. jacqiielinae, and T. purdyae from the Galapagos Islands, Ecuador, and T. stohleri from So- corro Island, Mexico, While examining the terebrid collections dredged by the R/V Valero III during the Allan Hancock Foundation Pacific Expeditions and by the ex- peditions of California Academy of Sciences to the Eastern Pacific islands, we discovered three news species of Terebra’, the fourth was brought to our atten- tion by Mrs. Jacqueline DeRoy, who dredged specimens at Academy Bay in the Galapagos Islands. The following abbreviations are used in the text: AHF, Allan Hancock Foundation (material on loan to LACM); AMNH, American Museum of Natural History; ANSP, Academy of Natural Sciences of Philadelphia; B&B, Bratcher and Burch collection; BM(NH), British Museum (Natural History) ; CAS, California Academy of Sciences; LACM, Los Angeles County Museum of Natural History; MCZ, Museum of Comparative Zoology, Harvard Uni- versity; SB, Santa Barbara Museum of Natural History; SD, San Diego Museum of Natural History; SU, Stanford University; USNM, United States National Museum; WAM, Western Australia Museum. Acknowledgments We wish to express our appreciation to Drs. James H. McLean of the Los Angeles County Museum of Natural History, Leo G. Hertlein of the California Academy of Sciences, and Harald Rehder and Joseph Rosewater of the U.S. National Museum for their cooperation and loan of materials. We thank Ben and Ruth Purdy of San Diego, California, and Laura Shy of West- minster, California, in whose collections additional specimens of some of the new species were found, for making their collections available to us. We are indebted to Mrs. Jacqueline DeRoy for sending material from the Galapagos Islands for study. We also wish to thank Maurice Giles of the California Academy for preparing several of the photographs and Mrs. Margaret Hanna for retouching them. Terebra hertleini, sp. nov. Figures 1-2 Diagnosis: A small species distinguished from other small terebrids by turreted whorls and large subsutural nodes. ’Research Assistant in Invertebrate Zoology, Los Angeles County Museum of Natural History. 8121 Mulholland Terrace, Los Angeles, California 90046. “Formerly Research Assistant in Invertebrate Zoology, Los Angeles County Museum of Natural History. (Deceased, February 22, 1970.) 1 1 Contributions in Science Description of holotype: Size small; color white, whorls flat, turreted, ten in number plus two glassy, convex nuclear whorls; first postnuclear whorl con- stricted; sculpture of three spiral cords per whorl and obsolete axial ribs be- ginning in large nodes anterior to suture (nine on penultimate whorl); sculp- ture on body whorl of three spiral cords crossing obsolete ribs that end in faint nodes at periphery, these nodes being less prominent than those at suture; anterior to periphery spiral cords cross minute axial striations; aperture semi- quadrate; outer lip thin, white within; columella straight with one weak plication; siphonal fasciole well developed with sharp keel; anterior canal of medium length, recurved. Length, 1 1.8 mm; diameter, 5.0 mm. Type locality ; Academy Bay, Santa Cruz (Indefatigable) Island, Gala- pagos Islands, Ecuador, 08° 46' 16" S, 90° 19' 38" W, CAS loc. 27536, 3. 5-5.5 fms, dredged, 45 specimens. Most of the specimens are very small and appear not to have been live taken as they have a chalky appearance. Type material: Holotype, CAS 13222. Paratypes: AMNH 157281 ; ANSP 316670; B&B 741; BM(NH); CAS 13223-13227; LACM-AHF 1288; SB 27147; SD 51962; SU 9996; USNM 680212. The LACM paratype is from AHF station 168-34, 15-25 fms, and was live collected. Academy Bay. Discussion: This species shows variation in the spiral cords being well developed in some specimens and almost obsolete in others. The nodes at the periphery of the body whorl are inconspicuous in some individuals. T erebra hertleini has a superficial resemblance to T. jacquelinae sp. nov., but is a much smaller species. The holotype of the former with ten whorls measures 1 1.8 mm in length while that of the latter with 13 whorls measures 33.2 mm. T erebra jacquelinae has a row of large nodes posterior to the suture which are absent in T. hertleini. T erebra hertleini is named for Dr. Leo G. Hertlein of the California Academy of Sciences in appreciation of his encouragement and assistance in our work on the eastern Pacific Terebridae. Terebra jacquelinae, sp. nov. Figures 3-4 Diagnosis: A species differing from other west American terebrids in having extremely concave whorls and axial ribs that fade at the center of the whorl and become large nodes at each end. Description of holotype: Size medium; color shiny cream; first four postnuclear whorls flat, remainder of whorls very concave, 1 3 in number plus 1.5 shiny, opaque, somewhat bulbous nuclear whorls; sculpture on first five Figures 1-8. 1. T erebra hertleini sp. nov. Holotype, CAS 13222. X 9. 2. T. hertleini, early whorls of holotype. 3. T. jacquelinae sp. nov. Holotype, CAS 13215. X 2.5. 4. T. jacquelinae, early whorls of holotype. 5. T. purdayae sp. nov. Holotype, LACM- AHF 1 182. X 5. 6. T. purdyae, early whorls of holotype. 7. T. st older i sp. nov. Holo- type, LACM-AHF 1 180. X 5.5. 8. T. stohleri, early whorls of holotype. Photos 1-5 by Maurice Giles; 6-8 by Twila Bratcher. 1970 Four New Terebrid Gastropods 3 4 Contributions in Science postnuclear whorls of weak axial ribs ending in small nodes at suture and of obsolete spiral grooves that cross the ribs; apical angle increases after sixth postnuclear whorl and ribs begin to fade at center of whorl while nodes at rib endings become more prominent, those anterior to suture being slightly more prominent than posterior ones; interspaces on later whorls contain minute striae, body whorl of medium length with 12 obsolete ribs ending in nodes at periphery; anterior to periphery weakly incised spiral grooves cross equally weak axial grooves; aperture elongate, white within; columella white, straight, with one rounded plication that continues to become keel of well developed siphonal fasciole; anterior canal short, recurved. Length, 33.2 mm; diameter, 8.4 mm. Type locality: Academy Bay, Santa Cruz (Indefatigable) Island, Gala- pagos Islands, Ecuador, 0° 46' 16" S, 90° 19' 38" W, CAS loc. 39585, about 10 fathoms, on hard packed coralline sand bottom, collected by Allyn G. Smith and Andre and Jacqueline DeRoy, February, 1964, holotype and four para- types. Type material: Holotype, CAS 13215; 4 paratypes, CAS 13216-13219. Additional paratypes are all from Academy Bay. CAS Templeton Crocker Expedition, May, 1932, loc. 27536, 3. 5-5. 5 fms., 11 specimens, distributed as follows: 8 paratypes, CAS 13220; 1 paratype, BM(NH) ; 1 paratype, SU 9997; 1 paratype, USNM 680213. Hancock Expeditions, AHF sta. 807-38, 10-25 fms., January, 1938, 28 specimens, distributed as follows: 26 paratypes, L ACM- AHF 1179; 1 paratype, MCZ 271946; 1 paratype, SD 51963. Thirteen additional paratypes were dredged by the DeRoys, 5-6 fms., 6 remain in the DeRoy collection, others distributed as follows: 1 paratype, AMNH 157282; 1 paratype, ANSP 316671; 1 paratype, SB 27148; 1 paratype, WAM; 3 para- types, B&B. Referred material: Numerous specimens of this species were among material taken by the Hancock and California Academy of Sciences expedi- tions and by the DeRoys at collecting stations in the vicinity of Santa Cruz (Indefatigable), San Salvador (James), and Baltra (Seymour) Islands in the Galapagos Islands at depths of 2-20 fathoms. Discussion: Although there is little variation among mature individuals of this species except that some are more slender than others, occasional im- mature specimens have the peripheral nodes forming a sharp keel that differs in appearance from that of adults. Most of the specimens examined are of a light cream color, but a number of specimens, most of which were collected at San Salvador Island, range in color from beige to light brown with cream colored nodes. The largest specimen examined measures 36.3 mm in length and 10.8 mm in width and is in the DeRoy collection. This species should not be confused with Terebra frigata Hinds, 1844, which has less concave whorls, pronounced spiral sculpture, a more slender profile, and ribs that are con- tinuous from suture to suture. The Gulf of Mexico species, T. concava (Say, 1827), is a smaller and much more slender species with definite spiral sculp- 1970 Four New Terebrid Gastropods 5 tore, having small sharp nodes at the subsutural band and periphery of the body whorl. Terebra jacqiielinae is named for Mrs. Jacqueline DeRoy of Isla Santa Cruz. Galapagos Islands, who first brought this species to our attention. Terebm purdyae, sp. nov. Figures 5-6 Diagnosis: A small slender species that differs from other west American species in having finely cancellate sculpture and a straight columella. Description of holotype: Size small, slender; color shiny pale cream with weak fulvous blotches; whorls almost flat, 1 1 in number plus four purple-beige convex nuclear whorls; suture fairly well defined; barely evident subsutural band of beading, slightly more prominent than beading on remainder of whorl; sculpture finely cancellate, remarkably consistent from second postnuclear whorl through body whorl, consisting of axial ribs (29 on penultimate whorl) crossed by cords (four on penultimate whorl) giving a beaded effect; axial ribs about equal to interspaces; body whorl of medium length with sculpture con- tinuing anterior to periphery but less well defined; aperture elongate and slender; columella straight with no plication; faint siphonal fasciole with posterior keel; anterior canal short, recurved. Length, 13.9 mm; diameter, 3.3 mm. Type locality: North of Santa Maria (Charles) Island, Galapagos Islands, Ecuador, 0° 59' S, 90° 25' W, 70-80 fms., sand and rock bottom, AHF station 195-34, January 29, 1934, 2 specimens. Type material: Holotype, LACM-AHF 1182; 1 paratype LACM-AHF 1183. An additional paratype, CAS 13221, is from Post Office Bay, of the same island, 8-10 fms., sand and rock and algae bottom. Referred material: LACM, Hill coll. 1363, Costa Rica, 2 specimens; AHF 201-34, Panama, 4 specimens; USNM 192963, Panama, 4 specimens. Discussion: The sculpture of this species resembles that of Terebra shyana Bratcher and Burch, 1970, but T. purdyae is a smaller species having a nucleus of four whorls and a straight columella with no plication. Terebra shyana has a nucleus of three whorls and a slightly curved columella with a faint plication. Terebra panamensis Dali, 1908, also has cancellate sculpture, but it is a heavier, broader shell with coarse sculpture. Terebra purdyae is named for Ruth Purdy of San Diego, California, in recognition of her generosity in sharing specimens, not only with us in our work on Terebridae, but with those studying other families of mollusks. Terebm stohleri, sp. nov. Figures 7-8 Diagnosis: A sturdy, medium-small, ivory colored species with little resemblance to other species of Terebra. 6 Contributions in Science Description of holotype: Size medium-small; color shiny ivory with slightly darker blotches; whorls convex, 14 in number plus remaining 2.5 (part missing) glassy convex nuclear whorls; suture deeply channeled, constricted; subsutural band inconspicuous; whorls slightly shouldered anterior to suture; sculpture of slightly curved axial ribs (28 on penultimate whorl), about equal to interspaces and of evenly spaced spiral grooves (three including subsutural groove), the two whorls posterior to body whorl having an additional groove immediately posterior to suture; body whorl of medium length; sculpture posterior to periphery remains constant; anterior to periphery axial ribs con- tinue and spiral grooves become more numerous and close set (seven between periphery and siphonal fasciole); aperture elongate; outer lip thin with sculp- ture pattern showing through; columella straight with no plication; siphonal fasciole striate; anterior canal short, recurved. Length, 21.4 mm; diameter, 5.1 mm. Type locality: Braithwaite Bay, Socorro Island, Mexico, 18° 42' 20" N, 110° 56' 15" W, sand and red mud bottom, March 17, 1939, AHF station 922-39, 1 specimen. Type material: Holotype, LACM-AHF 1180; 1 paratype, LACM-AHF 1181. The paratype is from Cape Rule, Socorro Island, AHF station 291-34, 4-10 fms. Referred material: LACM A. 5498, Galapagos Islands, 1 specimen; LACM, Hill coll. 1365, Chamela Bay, Mexico, 2 specimens; LACM A. 375, Tres Marias Islands, Mexico, 1 specimen; Shy collection, Manzanillo, Mexico, 1 specimen. Discussion: There is no other species with which this beautifully sculp- tured, shiny species can be easily confused. This species is named in honor of Dr. Rudolph Stohler who has given so much of his time, his personal finances, and himself to the publishing of The Veliger, for the advancement of conchology and malacology. Resumen Se describen cuatro nuevos terebridos procedentes de las Islas del Pacifico Este: T ere bra hertleini, T. jacquelinae y T. purdyae procedentes de las Islas Galapagos, Ecuador, y T. stohleri de la Isla Socorro, Mexico. Literature Cited Bratcher, Twila, and R. D. Burch. 1970. Five new species of Terebra from the Eastern Pacific. Veliger 12(3) :295-300, pi. 14. Dale, W. H. 1908. Reports on the dredging operations off the west coast of Mexico, and in the Gulf of California. . . . XIV. The Mollusca and the Brachiopoda. Mus. Comp. Zool. Harvard, Bull. 43(6): 205-487, pis. 1-22. Say, Thomas. 1827. Descriptions of marine shells recently discovered on the coast of the United States. Acad. Nat. Sci. Philad., Journ., ser. 1, 5:207-208. Accepted for publication December 30, 1969 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE JMBER 189 May 4, 1970 REINSTATEMENT OF THE TURRID GENUS BELLASPIRA CONRAD, 1868 (MOLLUSCA: GASTROPODA) WITH A REVIEW OF THE KNOWN SPECIES By James H. McLean and Leroy H. Poorman Los Angeles County Museum of Natural History • Los Angeles, California 90007 Exposition Park CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Alterations or changes in the manuscript after galley proof will be billed to the author. Unless specifically requested, page proof will not be sent to the author. One hundred copies of each paper will be given free to each author oi divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor REINSTATEMENT OF THE TURRID GENUS BELLASPIRA CONRAD, 1868 (MOLLUSC A: GASTROPODA) WITH A REVIEW OF THE KNOWN SPECIES By James H. McLean^ and Leroy H. Poorman- Abstract: The clavinid turrid genus Bellas pira Conrad is redefined and its type species, from the Virginia Miocene, is illustrated. Six known living species of the genus are treated. Two are from the Caribbean and four are from the eastern Pacific. Three new species are described: B. acclivicosta, from Guaymas in the Gulf of California; B. clarionensis, from Clarion Island, Mexico; and B. margaritensis from Venezuela. The genus Bellaspira Conrad, 1868, type species Mangelia virginiana Conrad, 1862, from the Virginia Miocene, had served a need for a small group of American turrids showing rounded axial ribs, fine spiral sculpture and a closely constricted anal canal, not deeply notched as in most other clavinid genera. Conrad’s original description of the genus was adequate and his illus- tration of the type species, although small, showed the diagnostic features in suflficient detail to allow Dali (1919) to associate with it a Recent west Ameri- can species. Unfortunately, a specimen of some other species was mislabeled as the type of B. virginiana, causing Bartsch and Rehder (1939) to interpret it incorrectly. As a result, three Recent species, all of which had been associated with the genus Bellaspira at some time, have been without a satisfactory generic allocation until now. Abbreviations for institutions cited in the text are as follows: AHF, Allan Hancock Foundation (on loan to LACM); AMNH, American Museum of Natural History, New York; ANSP, Academy of Natural Sciences, Philadel- phia; CAS, California Academy of Sciences, San Francisco; LACM, Los Angeles County Museum of Natural History; SDNHM, San Diego Natural History Museum; SU, Stanford University, California; USNM, United States National Museum of Natural History. Acknowledgments We are greatly indebted to Mrs. Virginia Orr Maes of the Academy of Natural Sciences, Philadelphia, for locating the type specimen of Bellaspira virginiana and providing us with photographs and helpful information about the western Atlantic species. Dr. Joseph Rosewater and Mr. Jack Byas of the U.S. National Museum kindly located and arranged the loan of specimens in ^Curator of Invertebrate Zoology, Los Angeles County Museum of Natural History. ^Research Assistant, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 189 that institution. Radula slides were prepared by Mrs. Ellen Brennan, Museum research assistant. Bellaspira Conrad, 1868 Bellaspira Conrad, 1868, p. 261; Dali, 1918, p. 323; Woodring, 1928, p. 162; Grant and Gale, 1931, p. 584. Type species, by monotypy, Mangelia virginiana Conrad, 1862, Miocene, Yorktown, Virginia. Non Bellaspira Conrad. Bartsch and Rehder, 1939, p. 133; Gardner, 1948, p. 271; Powell, 1966, p. 103. Conrad (1868) described the genus as follows: “Subfusiform, longi- tudinally ribbed; canal very short, labrum obsoletely and widely notched, curved outwards, simple; labium reflexed, entire; columella straight.” With further reference to the type species, he added, “It is a beautiful and graceful shell with distant ribs and very minute, close revolving lines, which become gradually more distinct on the body whorl. The labium is slightly callous near the upper extremity.” The shell figured as the type of Mangelia virginiana by Bartsch and Rehder (1939), copied by Gardner (1948), and Powell (1966), in no way agrees with Conrad’s description or illustration, and as Powell noted, the afiinity of the specimen is “with the mangelias or eucitharids,” in the subfamily Mangeliinae rather than the Clavinae. Mrs. Virginia Maes has found that both the specimen figured by Bartsch and Rehder, ANSP 1610, and another specimen, ANSP 30737, are labeled "‘Mangelia virginiana'' in Conrad’s handwriting. There is no doubt that the latter specimen is the one figured by Conrad in 1868. There may have been another original specimen in lot 1610 which was subsequently lost. In order to avoid further confusion, the only extant specimen, ANSP 30737, is here con- sidered the holotype. This specimen has seven postnuclear whorls, not five as stated in the original description. Grant and Gale (1931) included two other genera in the synonymy of Bellaspira, the European Atoma Bellardi, 1875, and Haedropleura Bucquoy, Dautzenberg, and Dollfus, 1883, but these taxa, as treated by Powell, 1966, lack the features of Bellaspira. Bellaspira is therefore strictly an American genus, with Recent species in the Caribbean, Panamic, and Californian provinces. Radulae of two west American species, B. melea and B. acclivicosta, new species, have been examined (Figs. 1,2). The radula is of the prototypic type described by Powell (1966, p. 70) as occurring in many of the strongly ribbed, glossy shelled genera of Clavininae. The radula consists of a vestigial rachi- dian, comblike laterals and long, narrow marginals. 1970 Reinstatement of the Genus BELLASPIRA 3 Figures 1-2. Radulae of Bellaspira, greatly enlarged. 1. B. acclivicosta sp. nov., Sola- dita Cove, Guaymas, Mexico. 2. B. melea Dali, San Carlos Bay, Guaymas, Mexico. A new generic diagnosis, taking the features of the living species into account, follows: Shell 7-19 mm in height, solid, spire elevated; outline concave below the suture, convex above; nuclear whorls two, smooth; axial sculpture of low rounded ribs, spiral sculpture of fine incisions or threads; anterior canal broad, only slightly notched, posterior notch constricted, mature specimens with a small deposition of callus at top of inner lip, immature specimens lacking a turrid sinus; lip of mature shell not greatly thickened, but penultimate axial rib thickened, producing a swelling as in other clavinid genera; color white to pink, some with darker spiral banding or markings on the axial ribs; oper- culum leaf-shaped, nucleus terminal; radula prototypic. Bellaspira virginiana (Conrad, 1862) Fig. 3 Mangelia virginiana Conrad, 1862, p. 286. Bellaspira virginiana, Conrad, 1868, p. 261, pi. 21, fig. 12. Non Bellaspira virginiana, Bartsch and Rehder, 1939, p. 133, pi. 17, fig. 6; Gardner, 1948, p. 271, pi. 37, fig. 14; Powell, 1966, p. 103, pi. 16, fig. 6. Description: Postnuclear whorls seven, axial ribs eight, continuous from whorl to whorl, spiral striae well spaced; height, 14 mm. Type locality: Miocene of Yorktown, Virginia. Type material: Holotype (here regarded as discussed above): ANSP 30737 (Fig. 3). Discussion : We have no other information on the occurrence of the type species. Additional material is not represented in the Philadelphia Academy. 4 Contributions in Science No. 189 Figures 3-5. 3. Bellaspira virginiana (Conrad). Holotype, ANSP 30737. X 4. 4. B. pentagonalis (Dali). Holotype, USNM 9300. X 8. 5. B. pentepleiira Schwengel ( = B. pentagonalis) . Holotype, ANSP 176450. X 5. Bellaspira pentagonalis (Dali, 1889) Figs. 4, 5 Drillia pentagonalis Dali, 1889, p. 90. Bellaspira (?) pentepleura Schwengel, 1940, p. 51, pi. 3, figs. 3, 3a; Powell, 1966, p. 104. Description: Postnuclear whorls six, axial ribs five, continuous from whorl to whorl, color white with a peripheral row of squarish chestnut spots on the axial ribs, spiral striae weakly developed; height, 8-10 mm. Type material and type localities: Holotype, Drillia pentagonalis, USNM 9300 (Fig. 4); 1 paratype, USNM 679297. Type locality: USFC sta. 2596, 17 miles east of Cape Hatteras, North Carolina, 49 fathoms. Holotype, Bellaspira pentapleura, ANSP 176450 (Fig. 5) ; 3 paratypes, ANSP 315786. Type local- ity: Off Palm Beach Florida, 45 fathoms, T. L. McGinty. Distribution: Cape Hatteras, North Carolina, to Key West and Saint Petersburg, Florida, 20-60 fathoms. Approximately 80 lots are represented in the National Museum collection. Discussion: Dali’s early name Drillia pentagonalis has been overlooked, evidently because the type was never illustrated. The holotype is worn and faded, measuring 6.9 mm in height. This species has only five axial ribs and thereby differs from all the other species having more numerous ribs. 1970 Reinstatement of the Genus BELLASPIRA 5 Bellaspira margaritensis sp. nov. Fig. 6 Diagnosis: A relatively large species, having more numerous axial ribs and more deeply incised spiral striae than B. pentagonalis. Description of holotype: Shell relatively large for the genus; nucleus of two smooth, rounded whorls; postnuclear whorls eight, axial ribs seven, con- tinuous from whorl to whorl; suture simple, slightly undulating; axial ribs evenly rounded, no subsutural channel; spiral sculpture of well defined, broad- ly spaced spiral incisions, more crowded in the subsutural area, approximately 12 on the penultimate whorl. Color whitish except for a broad pinkish area about the periphery, weakly demarked and not extending over the rib surfaces. Parietal callus well defined, terminating in a small node in front of the anal sulcus. Edge of outer lip thin, with a shallow stromboid notch near the base, a minor axial rib slightly behind the edge; the penultimate rib more strongly developed than the others. Dimensions (in mm) : height 15.0, diameter 5.5. Type locality: Four to seven miles north of Margarita Island, Venezuela, 11° 08' N, 64° 17' W, 17-21 fathoms on sand, R/V Velero III Station A42-39, 21 April 1939, four specimens, two live-taken. Type material: Holotype, LACM-AHF 1337 (Fig. 6); two paratypes, LACM-AHF 1338; one paratype, USNM 675964. Figures 6-8. 6. Bellaspira margaritensis sp. nov. Holotype, LACM-AHF 1337. X 4.2. 7. B. melea Dali. Holotype, USNM 15952. X 4.5. 8. B. melea Dali. Off San Carlos Bay, Guaymas, Mexico. X 5. 6 Contributions in Science No. 189 Discussion: The only other species with roundly arched axial ribs is B. pentagonalis, from which B. margaritensis differs in having two additional axial ribs, more prominent and broadly spaced incised spiral sculpture, and a color band not developed across the rib surface. Beilaspira melea Dali, 1919 Figs. 7, 8 Beilaspira melea Dali, 1919, p. 29, pi. 19, fig. 8; Grant and Gale, 1931, p. 584. Elaeocyma melea, Pilsbry and Lowe, 1932, p. 1 10. Clavus {ICymatosyrinx) melea. Keen, 1958, p. 448, fig. 736. Description: Postnuclear whorls 7-8, axial ribs 6-8, continuous from whorl to whorl, first rib behind the lip greatly thickened; color whitish to flesh pink, with traces of darker banding, rib surfaces lighter in color; height, 13-19 mm. Type locality: Panama, collected by Col. E. Jewett. Type material : Holotype, USNM 15952 (Fig. 7) . Distribution: Tepoca Bay, Sonora, Mexico, to Port Utria, Colombia, 10-40 fathoms. Twenty-two lots are represented in the LACM collection. Intermediate stations in Mexico include Concepcion Bay, Baja California; Guaymas, Sonora; Cleopha Island, Tres Marias Islands; Sihauntanejo, Guer- rero; Guatulco Bay, Oaxaca; and in Costa Rica, Port Culebra; indicating that the species is widely distributed throughout its range. Remarks: Dali (1919) correctly assigned this species to Beilaspira, not- ing that it is “typical Beilaspira” It so resembles B. virginiana that it would undoubtedly be considered conspecific but for the discrepancies of age and distribution. The type specimen is a worn shell not showing spiral sculpture, but Dali observed that spiral sculpture could have been worn away. In fresh specimens (Fig. 8) the spiral sculpture is extremely minute and somewhat undulating. Beilaspira melea varies considerably in color. Some specimens are white and others have pink or flesh colored banding, the darkest band located at the periphery. Beilaspira acclivicosta sp. nov. Figs. 9-11 Diagnosis: A relatively large species with a characteristic color pattern, having more numerous and more strongly crested axial ribs than B. melea. Description of holotype: Shell large for the genus, pinkish white to flesh colored; nucleus of two smooth, rounded, glassy pink whorls; postnuclear whorls nine, axial ribs 10, not continuous from whorl to whorl; suture simple, slightly undulating at the base of the ribs; ribs weak on the shoulder, strong and nodulous at the periphery, becoming obsolete on the base. Entire shell 1970 Reinstatement of the Genus BELLASPIRA 1 Figures 9-11. All Bellaspira acclivicosta sp. nov. 9. Holotype, LACM 1193. X 3. 10. Off Isla Candelero, Guaymas, Mexico, Poorman collection. X 3. 11. Pulmo Bay, Baja California, Mexico, LACM 66-19. X 5.5. covered by microscopic, raised spiral threads, most numerous above the periphery and strongest on the base, crossed by incremental growth threads to form a minutely reticulated surface. Peripheral crests of ribs white, surrounded by red-brown, strongest above and below; a second and smaller row of dots on the ribs between the periphery and the anterior tip. Columella straight, parallel to axis of shell, twisting slightly to the left at the base. The parietal callus is well defined and terminates in a domelike pad in front of the anal sulcus. Outer lip thin, smooth within, nearly parallel to the columella, showing a shallow stromboid notch near the base and thickened by the final rib, which is pre- ceded by an unusually wide valley and the penultimate rib that is stronger than all the others. Periostracum lacking; operculum with apical nucleus. Dimensions (in mm) : height 18.1, diameter 6.8. Type locality: One kilometer south of the east point at the entrance to Bahia San Carlos, Guaymas, Sonora, Mexico; 21° 56' N, 111° 03' W; 15-20 fathoms on rock and broken shell bottom, December, 1965, to April, 1967, Forrest and Roy Poorman, ten specimens. Type material: Holotype, LACM 1193 (Fig. 9); one paratype, ANSP 315788, one paratype, AMNH 147999; one paratype, CAS 13159; one para- type, SDNHM 50000; one paratype, SU 9964; one paratype; USNM 679183; three paratypes, Poorman collection. Referred material: Poorman collection, two specimens dredged at Isla Candelero, Guaymas (Fig. 10); LACM 68-27, four specimens, diving at 60 feet, Soladita Cove, Guaymas; LACM 66-19, one specimen, diving at 5-20 8 Contributions in Science No. 189 feet, Pulmo Bay, Baja California (Fig. 11); Shy collection (Westminster, California) , one specimen dredged at Manzanillo, Colima, Mexico. Discussion: Immature shells of B. acclivicosta do not show a turrid sinus and lack the callus deposition defining the anal sinus. Figure 10 shows a speci- men slightly larger and broader than the holotype that has an immature lip. This specimen also is more distinctly colored than the holotype. The specimen from Pulmo Bay (Fig. 1 1 ) has six postnuclear whorls and measures 9.1 mm in height. It is evidently a miniature specimen of B. acclivicosta having a ma- ture lip. It is similar to the type lot except for size. The new species differs from B. melea in having about ten discontinuous axial ribs rather than seven continuous ribs per whorl. The spiral sculpture consists of fine raised spiral threads on B. acclivicosta, whereas the weak grooves on B. melea appear to be scratched into the surface. The name is derived from the Latin nouns acclivis, steep, and costa, rib, suggestive of the steep, concave slope of the ribs below the suture. The specific name is intended as a noun in apposition. Bellaspira grippi (Dali, 1908) Figs. 12-15 Bela grippi Dali, 1908, p. 137. Lora grip pH, Dali, 1921, p. 74, pi. 3, fig. 8; Oldroyd, 1927, p. 100, pi. 7, fig. 6. Figures 12-14. All Bellaspira grippi (Dali). 12. Holotype, USNM 200653. X 5.7. 13. San Martin Island, Baja California, Mexico, LACM 67-50. X 6.4. 14. Asuncion Island, Baja California, LACM 67-66. X 7.4. 1970 Reinstatement of the Genus BELLASPIRA 9 Bellaspira grippi, Grant and Gale, 1931, pp. 514, 565; Burch, 1946, p. 19 rgrippiV’]. Description: Relatively small, postnuclear whorls six, subsutural area markedly concave; axial ribs 12-14, not continuous from whorl to whorl, ribs obsolete at the base and often over the final whorl; color white to flesh pink, some with dark and light banding; height, 7-1 1 mm. Type locality: Five miles south of the entrance to San Diego Bay, 15 fathoms, C. W. Gripp, collector. Type material: Holotype, USNM 200653 (Fig. 12), one paratype, USNM 697389. Distribution: Redondo Beach, California, to Asuncion Island, Baja Cali- fornia; Guadalupe Island, Mexico. Not uncommon on rocky bottoms under kelp and dredged on gravel bottoms to depths of 25 fathoms. Discussion : Grant and Gale (1931 ) were the first to associate this species with the genus Bellaspira, a relationship missed by Dali, who had correctly assigned his B. melea in 1919. A wide range of variation is shown in this species. The typical southern Californian form, represented by the holotype (Fig. 12) and a specimen from San Martin Island, Baja California (Fig. 13) , is whitish or flesh colored and usually shows a slightly darker band in the concave area of the shoulder. One specimen from Asuncion Island, outer coast of Baja California (Fig. 14), 6.9 mm in height, is dark rose colored, except on the siphonal fasciole, and the shoulder is more strongly tabulate, the axial ribs forming projecting nodes. Incised spiral sculpture is unevenly spaced on the body whorl. The largest of five specimens from Guadalupe Island (Fig. 15) is 7.5 mm in height. All are strongly tabulate and the spiral incisions are evenly spaced; two are dark flesh-colored and one is white. Bellaspira clarionensis sp. nov. Fig. 16 Diagnosis: A medium sized species larger than B. grippi and having more numerous spiral striae. Description of holotype: Shell of medium size for the genus, nuclear whorls two, rounded; postnuclear whorls seven, axial ribs 10, alternating in position on early whorls, but aligned from whorl to whorl on the final three whorls. Suture simple, slightly undulating; shoulder concave; axial ribs well defined across the shoulder, cresting at the periphery, extending across the body whorl and becoming obsolete on the base. Spiral sculpture of closely spaced spiral striae, more crowded upon the shoulder and overriding the axial ribs. Color whitish with a narrow band of flesh pink upon the shoulder and another uninterrupted narrow band across the base. Parietal callus projecting, anal sinus narrow. Lip edge thin, stromboid notch barely perceptible, the 10 Contributions in Science No. 189 Figures 15-16. 15. Bellaspira grippi (Dali). Guadalupe Island, Mexico, LACM 65-42. X 9.5. 16. B. clarionensis sp. nov. Holotype, LACM-AHF 1366. X 4.6. second and third ribs behind the aperture coalesced and thickened. Dimensions (in mm) : height 13.5, diameter 5.9 mm. Type locality: Off Sulphur Bay, Clarion Island, Revillagigedo Islands, Mexico, 18° 20' N, 114° 44' W, 28-45 fathoms on sand, R/V Velero III Station 917-39, 16 March 1939, one specimen. Two additional paratype specimens were dredged at stations 136-34 and 921-39, Sulphur Bay, Clarion Island, 30-35 fathoms. Type material: Holotype, LACM-AHF 1366 (Fig. 16), one paratype, LACM-AHF 1367; one paratype, USNM 675565. Referred material: LACM A. 3 75, five specimens, dredged by George Willett, 20-35 fathoms. Clarion Island, 24 March 1938. Discussion: Bellaspira clarionensis differs from B. grippi in reaching a larger size, having a narrower shoulder area, elongate rather than nodular axial ribs, and more numerous spiral striae. Resumen El genero clavinido turrido Bellaspira Conrad se redifine y la especie tipo, del Mioceno de Virginia, se ilustra. Seis especies vivientes conocidas en este genero son aqui tratadas. Dos de ellas son del Caribe y cuatro del Pacifico 1970 Reinstatement of the Genus BELLASPIRA 11 Este. Se describen tres nuevas especies: B. acclivicosta, de Guaymas en el Golfo de California; B. clarionensis, de la Isla Clarion, Mexico; y B. marga- ritensis de Venezuela. Literature Cited Bartsch, P., and H. a. Rehder. 1939. New turritid mollusks from Florida. U.S. Nat. Mus. Proc., 86:127-138. Burch, J. Q. (ed.). 1946. Distributional list of the West American Marine Mollusks from San Diego, California, to the Polar Sea. [Extracts from the] Minutes, Conch. Club So. Calif., 62: 1-52. Conrad, T. A. 1862. Descriptions of new genera, subgenera and species of Tertiary and Recent shells. Acad. Nat. Sci. Phila., Proc., 14: 284-291. . 1868. Descriptions of new genera and species of Miocene shells, with notes on other fossil and Recent species. Amer. Jour. Conchology, 3: 257-270. Dale, W. H. 1889. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877-78) and in the Caribbean Sea (1879-80), by the U.S. Coast Survey steamer “Blake.” XXIV. Report on the Mollusca. — Part II. Gastropoda and Scaphopoda. Mus. Comp. Zook, Bull., 18: 1-492. . 1908. Some new Californian shells. Nautilus, 21: 136-137. . 1918. Notes on the nomenclature of the mollusks of the family Turritidae. U.S. Nat. Mus., Proc., 54: 313-333. . 1919. Descriptions of new species of mollusks of the family Turritidae from the west coast of America and adjacent regions. U.S. Nat. Mus., Proc., 56: 1-86. . 1921. Summary of the marine shellbearing mollusks of the northwest coast of America, from San Diego, California, to the Polar Sea, mostly contained in the collection of the United States National Museum, with illustrations of hitherto unfigured species. U.S. Nat. Mus., Bull., 112: 1-217. Gardner, J. 1948. Mollusca from the Miocene and Lower Pliocene of Virginia and North Carolina, Part. 2. Scaphopoda and Gastropoda. U.S. Geol. Surv., Prof. Paper, 199B: 177-310. Grant, U. S., IV, and H. R. Gale. 1931. Catalogue of the marine Pliocene and Pleistocene Mollusca of California and adjacent regions. San Diego Soc. Nat. Hist., Mem., 1: 1-1036. Keen, A. M. 1958. Sea shells of tropical west America; marine mollusks from Lower California to Colombia. Stanford Univ. Press, Stanford, California, xi -f 624 p. Oldroyd, I. S. 1927. The marine shells of the West Coast of North America. 2, Gastropoda, Scaphopoda, and Amphineura. Stanford Univ. Publ. Univ. Ser., Geol. Sci., 2(1): 1-297. PiLSBRY, H. A., AND H. N. LowE. 1932. West Mexican and Central American mol- lusks collected by H. N. Lowe, 1929-31. Acad. Nat. Sci., Phila., Proc., 84: 33-144. Powell, A. W. B. 1966. The molluscan families Speightiidae and Turridae. Bull. Auckland Inst, and Mus. 5: 1-184. SCHWENGEL, J. S. 1940. New Mollusca from Florida. Nautilus, 54: 49-52. Woodring, W. P. 1928. Contributions to the geology and paleontology of the West Indies. Miocene mollusks from Bowden, Jamaica. Part II. Gastropods and discussion of the results. Carnegie Inst. Wash., Publ. no. 385: 1-564. Accepted for publication February 28, 1969 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS ZYL, IN SCIENCE UMBER 190 May 20, 1970 A NEW SPECIES OF FOSSIL KELLETIA (MOLLUSCA: GASTROPODA) FROM THE LOMITA MARL, LATE CENOZOIC OF SAN PEDRO, CALIFORNIA By Leo G. Hertlein Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Changes in the manuscript after galley proof will be billed to the author. Unless otherwise requested, page proof also will be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor A NEW SPECIES OF FOSSIL KELLETIA (MOLLUSCA: GASTROPODA) FROM THE LOMITA MARL, LATE CENOZOIC OF SAN PEDRO, CALIFORNIA By Leo G. Hertlein^ Abstract: Kelletia kanakoffi, a new species of gastropod belonging to the family Neptuneidae, is described from strata of Late Pliocene or Early Pleistocene age in San Pedro, California. Introduction Numerous species of fossil mollusks have been reported from Hilltop Quarry and nearby localities (see Hanna, 1923; Schenck, 1945: 513-514; Woodring et ai, 1946: 49-52; Valentine, 1961: 411-413) in San Pedro, Los Angeles County, California. Intensive collecting in this area by George P. Kanakoff and his associates has yielded many interesting fossils now preserved in the Los Angeles County Museum of Natural History. Recently Mr. Kanakoff sent the author two specimens of a large gastro- pod, referable to the genus Kelletia. Study of these fossils revealed that they represent an undescribed species here described as new. The strata in which they were found have been considered to be of Late Pliocene age by some authors but as Early Pleistocene age by others. An age of three million years given by Obradovich (1965), based upon Potassium- Argon dating of glauconite, represents the most recent maximum age attributed to these sediments. This date would certainly place the Lomita Marl, a mem- ber of the lower San Pedro formation, in the Late Pliocene or very Early Pleistocene. Several species of mollusks reported from the Lomita Marl occur elsewhere only in strata believed by some authors to be of Pliocene age. Acknowledgments I am grateful to Mr. George P. Kanakoff, Curator Emeritus of the Section of Invertebrate Paleontology, Los Angeles County Museum of Natural History, who lent the specimens of the new species described herein for study and description. Acknowledgment also is due Victor Zullo, Allyn G. Smith and Barry Roth, California Academy of Sciences, whose advice and criticism was particularly helpful during the preparation of the manuscript. Alan J. Galloway of the same institution, and Warren O. Addicott, United States Geological Survey, called my attention to certain literature concerning the geochronology of the Lomita Marl. The illustrations in the text are from photo- graphs made by Maurice Giles, Staff Photographer, California Academy of Sciences. ^Curator (Emer.) of Invertebrate Paleontology, California Academy of Sciences; and Research Associate, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 190 Class Gastropoda Family Neptuneidae Tryon Genus Kelletia Bayle, in Fischer Kelletia Bayle, in Fischer, 1884: 625. Sole species, “5. Kelleti, Forbes.” Wenz, 1938 (reprint, 1962): 1166 Type species: Siphonalia kelleti Forbes [=Fusus kelletii Forbes (1852: 274, pi. 9, fig. 10)] “taken on the Californian coast.” This genus (and its type species) was named for Captain Kellett of the British Royal Navy. Some authors have used the orthography Kellettia but most recent authors have followed the original spelling given by Bayle for the genus and kelletii for the species as given by Forbes. West American authors have placed Kelletia in either the family Nep- tuneidae or the Buccinidae. Powell ( 1929: 63) stated that the shell characters as well as the radula of this genus are very similar to Austrosipho Cossmann, and these genera he placed in the family Buccinulidae Finlay. Wenz ( 1962: 1151) placed both that family and the Neptuneidae as subfamilies under the Buccinidae. He considered Austrosipho Cossmann, 1906, and Verconella Iredale, 1914, to be synonyms of Penion Fischer, 1884. Ruth ( 1942), in a monograph of Siphonalia in the Cenozoic of western North America, considered Kelletia a subgenus which included four species that he described and illustrated. The geologic range of the subgenus indicated by the included species was from Paleocene (Martinez) to Recent. Sub- sequent to Ruth’s paper, one species, Kelletia vladimiri Kanakoff, was described from strata of Pliocene age in southern California. One species of Kelletia now lives in the region between Santa Barbara, California, and San Quintin, Baja California, from the intertidal zone to a depth of about 64 meters (35 fathoms). Another species, Kelletia lischkei Kuroda, lives in Japanese waters where it has been reported to range from 32° to 39° North Latitude on the Pacific coast and to 40° North on the Japan Sea coast. In their catalogue of the Tertiary mollusks of Japan, 1952, Hatai and Nisiyama did not allocate any species to Kelletia but species assigned to Kel- letia (considered to be a subgenus of Siphonalia) have been reported by others from strata of Late Tertiary age in Indonesia and in India. Two species of Pliocene age from Ecuador, tentatively assigned to Kel- letia by Olsson ( 1964: K. ecuadoriana, p. 165, pi. 24, figs. 2-2c and K. rugosa, p. 165, pi. 24, figs. 1, la), do not closely resemble species of that genus described from California. Anderson ( 1964: 249) recently proposed a subgenus Boreokelletia, based upon Kelletia (Boreokelletia) hosiusi (Beyrich, 1856), from strata of Miocene age in Germany. The type species earlier was referred to the genus Strepto- chetus Cossman, 1889, by Kautsky (1925: 1 17), and others. 1970 A New Species of Fossil Kelletia 3 Kelletia kanakoffi, new species Figures 1 , 2, and 3 Diagnosis: A new species of Kelletia which differs from other west Ameri- can species of this genus in its large size, thick shell, coarse spiral sculpture and large umbilicus. Description: Shell large, thick, fusiform, the spire elevated; whorls of spire five (the apex lacking), angulated near the middle, posteriorly slightly con- cave and extending over a little more than one half of the preceding whorl, the suture appressed; axial sculpture consisting of well developed, axially elon- gated nodes, seven on the uppermost, eight on the succeeding one, ten on the penultimate whorl and eight on the body whorl, followed by a low, carinate ridge at an angulation posterior to the middle, with vague vertical ridges out- lined by lines of growth; the entire shell is sculptured with spiral ridges sepa- rated by deep grooves, about twenty on the penultimate whorl and about twenty-three on the body whorl, the anterior ones much coarser than the posterior ones; some ridges bear one to three spiral grooves and where eroded these separate small riblets; aperture ovate, outer lip not thickened, the inte- rior lightly lirate corresponding to the exterior sculpture, pillar smooth, twisted, inner lip smooth; canal well developed, narrow, twisted and recurved, a large siphonal fasciole present, in front of which is a large, rather wide umbilicus. Dimensions: The holotype, length (apex lacking), 181 mm, maximum diameter, 90 mm, length of aperture and canal, 102 mm; paratype, length, 178 mm, maximum diameter, 91 mm, length of aperture and canal, 104 mm. Type locality: Loc. 435 LACMIP, Lomita Marl about 50 feet below the crossing of West Park Drive, Host Place and Coralmount Drive, San Pedro, California. Late Pliocene or Early Pleistocene. John Sutherland collector, July, 1966. Type specimens: Holotype Catalog Number 2456, and Paratype Cata- logue Number 2457, Los Angeles County Museum of Natural History, Inver- tebrate Paleontology Collection. Comments: This new species differs from the Recent Kelletia kelletii (Forbes, 1852: 274, pi. 9, fig. 10) in the larger size, coarser sculpture, very large umbilicus, the more angulated shoulder and slightly carinate body whorl, and in that the whorls on the spire overlap the preceding ones to a less extent posteriorly than do those of the well known living species (see Arnold, 1903: pi. 4, fig. 5; Ruth, 1942: pi. 48, fig. 5; Abbott: 1954, pi. 24, fig. W). A large specimen of Kelletia kelletii 161 mm long and 76 mm in maxi- mum diameter, in the collections of the California Academy of Sciences, from Locality 31642 (CAS), from 3.7 miles off Huntington Beach, California, in 37 to 38 meters (20 to 21 fathoms), has only a short, narrow, shallow, umbili- cal groove in comparison to the large umbilicus of K. kanakoffi, new species. The largest specimen of K. kelletii reported in the literature (Anon., 1959: 4 Contributions in Science No. 190 Figure 1. Kelletia kanakoffi, new species. Holotype, LACMIP 2456, from Lomita Marl, San Pedro, California; Late Pliocene or Early Pleistocene age. Approximately X 5/6; true length (apex lacking), 181 mm, maximum diameter, 90 mm. 1970 A New Species of Fossil Kelletia 5 Figure 2. Kelletia kanakoffi, new species. Dorsal view of holotype shown in figure 1 . 6 Contributions in Science No. 190 Figure 3. Kelletia kanakoffi, new species. Paratype, LACMIP 2457, from the same locality as the holotype shown in figure 1. Approximately X 5/6; true length, 178 mm, maximum diameter, 91 mm. Note the large umbilicus on this specimen. 1970 A New Species of Fossil Kelletia 7 19), from southern California, is 170 mm long and 87 mm in maximum diameter. The same shell characters which serve to separate Kelletia kanakoffi, new species, from K. kelletii also differentiate it from Kelletia vladimiri Kana- koff (1954: 114, pis. 29, 30, 31), a smaller, more slender form described from the Pico formation, of Pliocene age, in Los Angeles County, California. This species is named for Mr. George P. Kanakoff, Curator Emeritus, Los Angeles County Museum of Natural History, whose careful collecting has brought to light many interesting specimens. Literature Cited Anonymous 1959. “Lost Operculum Club.” Min. Conch. Club South. California (J. Q. Burch, Editor), No. 190: 17-23, July-August. Abbott, R. T. 1954. American Seashells. (D. Van Nostrand Co., New York), pp. I-XIV, 1-541, pis. 1-40, figs. 1-100 in text. Anderson, H. J. 1964. Faunen aus dem Miocan Nordwestdeutschlands. Die Miocane Reinbek-Stufe in Nord- und Westdeutschland und ihre Mollusken-Fauna. Fort- schritte in der Geologie von Rheinland und Westfalen. Geologisches Landes- amt Nordrhein-Westfalen, 14: 31-368, pis. 1-52, figs. 1-18 in text. Arnold, R. 1903. The Paleontology and Stratigraphy of the Marine Pliocene and Pleistocene of San Pedro, California. Mem. Calif. Acad. Sci., 3: 1-420, pis. 1-37, June 27. Fischer, P. 1884. Manuel de Conchyliologie et de Paleontologie Conchyliologique ou Histoire Naturelle des Mollusques Vivants et Fossiles. (Librairie F. Savy, Paris), Fasc. 7; 609-688, figs. 370-456 in text, June 30. Forbes, E. 1852. On the marine mollusca discovered during the voyages of the Herald and Pandora, by Capt. Kellett, R. N., and Lieut Wood, R. N. Proc. Zool. Soc. London for 1850: 270-274, pis. 9, 1 1, January 24. Hanna, G D. 1923. Results of preliminary examination of seven samples of sedi- ments from near Lomita. Bull. South. Calif. Acad. Sci., 22 (2) : 64-65, July. (“List of the fossil mollusks found in the Lomita beds, as identified by Mrs. Ida S. Oldryod. . . .”, p. 65). Hatai, K., and S. Nisiyama, 1952. Checklist of Japanese Tertiary marine mollusca. Sci. Repts. Tohoku Univ., Sendai, Japan, Second Ser. (Geology), Special vol. 3: 1-449; supplement, pp. 452-464. Kanakoff, G. P. 1954. A new Kelletia from the Pliocene of California. Bull. South. Calif. Acad. Sci., 53 (2): 114-117, pis. 29, 30, 31, May-August (issued August 1). Kautsky, F. 1925. Das Miocan von Hemmoor und Basbeck-Osten. Abh. d. Preuss. Geol. Landesanstalt, Neue Folge, Heft 97, pp. 1-255, pis. 1-12. 8 Contributions in Science No. 190 Obradovich, J. D. 1965. The Potential use of Glauconite for Late-Cenozoic Geo- chronology. International Association for Quaternary Research, 7th Congress (United States), Proc., 8: 267-279. Olsson, a. a. 1964. Neogene Mollusks from Northwestern Ecuador. Paleo. Research Inst., Ithaca, New York, pp. 1-256, pis. 1-38, October 28. Powell, A. W. B. 1929. The Recent and Tertiary species of the genus Buccinulum in New Zealand, with a review of related genera and families. Trans, and Proc. New Zealand Inst., 60 (Quarterly issue), Pt. 1 : 57-101, pis. 1-4, figs. 83-132 in text. May 31. Ruth, J. W. 1942. The molluscan genus Siphonalia of the Pacific coast Tertiary. Univ. Calif. Publ. Bull. Depart. Geol. Sci., 26(3) : 287-306, pis. 47, 48, March 24. SCHENCK, H. G. 1945. Geologic Application of Biometrical Analysis of Molluscan Assemblages. Jour. Paleo., 19(5) : 504-521, pis. 66, 67, September. Valentine, J. W. 1961. Paleoecologic Molluscan Geography of the Californian Pleistocene. Univ. Calif. Publ. Geol. Sci., 34(7) : 309-442, figs. 1-16 in text, May 17. ( See especially pp. 4 1 1 -4 1 3 ) . Wenz, W. 1938 (1962). Handbuch der Palaozoologie herausgegeben von O. H. Schindewolf. Gastropoda von W. Wenz. (Gebriider Borntraeger: Berlin. Foto- mechanischer Nachdruck, 1962), Bd. 6, Teil 1, Gastropoda, Teil 1, Lieferung 5: 961-1200, Abbildungen 2788-3416. Woodring, W. P., M. N. Bramlette, and W. S. W. Kew. 1946. Geology and Paleontology of Palos Verdes Hills, California. U.S. Geol. Surv., Prof. Paper 207: I-V, 1-145, pis. 1-37, figs. 1-16 in text. (See especially “Hilltop quarry and nearby localities,” pp. 49-52). Accepted for publication March 16, 1970 i LOS ; ANGELES 1 COUNTY ' MUSEUM CONTRIBUTIONS cllsfa IN SCIENCE llMBER 191 May 20, 1970 STUDIES ON THE FROGS OF THE GENUS LEPTODACTYLUS (AMPHIBIA: LEPTODACTYLIDAE). VI. BIOSYSTEMATICS OF THE MELANONOTUS GROUP By W. Ronald Heyer ii Los Angeles County Museum OF Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Changes in the manuscript after galley proof will be billed to the author. Unless otherwise requested, page proof also will be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor STUDIES ON THE FROGS OF THE GENUS LEPTODACTYLUS (AMPHIBIA: LEPTODACTYLIDAE) . VI. BIOSYSTEMATICS OF THE MELANONOTUS GROUP By W. Ronald Heyer^ Abstract: Six species are recognized in the Melanonotus group: dantasi, discodactyliis, melanonotus, podicipinus, pustu- latiis, and wagneri. A synonymy, diagnosis, summary of charac- teristics, and distributional summary are presented for each species. Series of Leptodactyliis were analyzed from broad geo- graphic and ecologic areas and situations for standard length, ventral pattern, ventrolateral gland development, posterior thigh pattern, and toe disk development. Correlations are evident between 1 ) larger size and more mesic habitats in wagneri, 2) darker ventral pattern and greater annual rainfall in mela- nonotus, 3) lighter ventral pattern and greater ventrolateral- gland extent in wagneri, and 4) greater ventrolateral-gland extent and greater annual rainfall in wagneri. The hypothesis is pre- sented that character displacement is involved in the sharp differences in size and pattern between populations of melanono- tus and wagneri as well as podicipinus and wagneri in sympatry. Leptodactyliis melanonotus is the most primitive of the species, while discodactyliis and piistiilatus are the most ad- vanced. Leptodactyliis melanonotus, podicipinus, and wagneri are associated with old land masses and are species adapted for xeric conditions. Leptodactyliis dantasi and discodactyliis are limited to the Tropical Rainforest of the western Amazonian Basin. Lep- todactyliis piistiilatus is distributed in the xeric regions of eastern Brasil. Introduction Since the frog genus Leptodactylus was proposed by Fitzinger in 1826, it has had an unstable systematic history. No one since Boulenger, in 1882, has treated the group as an entity. The present paper is the first of a projected series attempting to elucidate the biosystematics of each of the species groups within the genus. In the Melanonotus species group, characters of gross morphology and distribution are used to analyze the interspecific relationships, ecological distribution patterns, and geographical distribution patterns of the species. Methods All adult specimens of the Melanonotus species group were examined for size, sex, ventral color pattern, ventrolateral gland development, posterior 1 Research Associate in Herpetology, Los Angeles County Museum of Natural History; and Biology Department, Pacific Lutheran University, Tacoma, Washing- ton 98447. 1 2 Contributions in Science No. 191 thigh pattern, and degree of toe-tip expansion. Analysis of these characters is the basis for the taxonomic conclusions. For convenience, taxonomic con- clusions are presented first before discussing the variation of characters an- alyzed in detail. Ail available adult individuals of L. dantasi, discodactylus and pustulatus were studied for variation in 34 characters. The same 34 characters were ex- amined in a series of 15 males and 15 females for each of the other three species. Thirty specimens of melanonotus, podicipinus, and wagneri were chosen to represent the extremes in geographic origin and morphological vari- ation. Details of the methods of examining individuals are the same as used previously (Heyer, in press). Percentages of standard length (SL) are pre- sented as follows, 5-6. 1-7, where the first number is the minimum, the second the mean, and the third the maximum percentage. The terminology follows Peters (1964) and Elias and Shapiro (1957). A dissecting microscope was used to examine the pattern of melanophore distribution on the ventral surfaces since melanophores in a contracted state are difficult to distinguish with the naked eye. Prior to this study, five characters appeared adequate to define popula- tions of the Melanonotus species group. Male and female sizes were recorded, and standardized forms were drawn which encompassed the range of variation for the ventral pattern, ventrolateral gland development, posterior thigh pat- tern, and toe-disk development. If a selected character appeared intermediate in any single frog, it was arbitrarily placed in the category it more closely resembled. One to 100 individuals have been collected from each of 454 localities in Mexico and Middle America. As the status of the frogs in this geographic area was reasonably clear, samples for analysis were selected every 100 km along both coasts and in any inland area. In regions where faunal changes might be expected on the basis of other anuran distribution patterns more samples were analyzed. The samples, usually 10 frogs per locality, from 56 localities from Mexico through Panama and ail available South American specimens of the Melanonotus species group were analyzed. The frogs were examined to deter- mine how many distinctive morphotypes were present. Examples demonstrat- ing the range of variation for individuals from each population were then recorded. Analyses of only the largest specimens of each distinct morphotype from each locality were recorded. Thus in the following analysis, size refers to the maximum for males and females for a given locality. A series of mating call recordings were made in western Mexico in July 1967 on an Uher 4000 L portable tape recorder at 7.5 ips. The tapes were analyzed on a Kay Sonagraph 6061 B. Information on number of notes per call group, dominant frequency, harmonics, and frequency shifts was re- corded from the sonagrams. The tapes are deposited af the University of Southern California, Department of Biological Sciences. Localities, tapes 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 3 (one individual per tape), and specimens on deposit at the Los Angeles County Museum of Natural History, respectively, are: COLIMA, Colima, WRH 67-12, LACM 37037; JALISCO, 5 km W Acatlan, 1400 m, WRH 67-20, LACM 37038, WRH 67-21, WRH 67-22, WRH 67-23, LACM 37039, WRH 67-24, WRH 67-25, LACM 37040; 7.7 km E La Huerta, 340 m, WRH 67-26, LACM 37041; 37.5 km SW Tecalitlan, 910 m, WRH 67-15, WRH 67-16, WRH 67-17; 0.5 km NE Tonila, 1300 m, WRH 67-18, LACM 37428; MICHOACAN, 2.6 km N Capirio, 300 m, WRH 67-13, LACM 37427; NA- YARIT, 13.2 km NE San Bias, WRH 67-31; Santa Cruz, 15 m, WRH 67-30, LACM 37043; Tepic, 950 m, WRH 67-27, LACM 37042, WRH 67-28; SINALOA, 36 km S Los Mochis turnoff on Mexican Highway 15, 10 m, WRH 67-2, LACM 37426, WRH 67-3, LACM 37036; SONORA, 3.1 km E Hermo- sillo, WRH 67-32. Certain environmental parameters were chosen to evaluate as possible correlates with SL, gland development, ventral pattern, and posterior thigh pattern. Total annual rainfall, number of months with less than 100 mm of rain- fall (dry months), and elevation were recorded for each locality. The climatic data were taken from Espinal and Montenegro (1963) for Colombia; Hold- ridge (1964) for Middle America; Tosi (1960) for Peru; Vivo Escoto (1964) for Mexico; Walter and Lieth (1960-67) for South America and Lesser An- tillean Islands. Elevational data were taken from museum catalogues or from the Millionth Map series (American Geographical Society). The correlation coefficient used to compare categories was the product moment method for ungrouped data, as presented by Arkin and Colton (1966, pp. 80-82). Mary Nafpaktitis wrote a Fortran program for this formula suited to my data. Robert J. Lavenberg wrote a Basic program for a Ltest, as pre- sented by Fisher (1948, pp. 193-196). The correlation coefficients and /-tests were run on a General Electric Time Share Computer terminal, provided by the Los Angeles County Museum of Natural History. The t values were located on a table (Fisher, 1948, p. 174) to determine the level of statistical significance (F value) of the correlation coefficients. A P value of .05 or less was considered statistically significant. Acknowledgments This report is a modified section from my doctoral dissertation on file at the University of Southern California. My dissertation committee has been a constant source of encouragement — Jay M. Savage, Chairman, Robert M. Chew, John S. Garth, John L. Mohr, Basil G. Nafpaktitis, and John D. Soule. My fellow graduate students have aided me considerably in helping to define my ideas through stimulating discussions. Of particular help have been John R. Meyer, Roy W. McDiarmid, Norman J. Scott, Jr., Philip A. Silverstone and David B. Wake. 4 Contributions in Science No. 191 My studies of the genus Leptodactylus would have been lessened without the help of the following field companions: James R. Dixon, Miriam Heyer, Roy W. McDiarmid, Marco Tulio Pacheco, Norman J. Scott, Jr,, and Charles F. Walker. The curators of collections with which I worked were particularly gra- cious in providing information, lending large amounts of material and allowing dissections and skeletal preparations in certain cases ( Museum abbreviations as used in the text in parentheses) : James E. Bohlke, Academy of Natural Sciences, Philadelphia (ANSP) ; Werner C. A. Bokermann, Sao Paulo (WCAB) ; Antenor Leitao de Carvalho, Museu Nacional, Rio de Janeiro; Javier Castroviejo, Museo Nacional de Ciencias Naturales, Madrid; Doris M. Cochran and James A. Peters, United States National Museum (USNM and field series GOV and JAP) ; James R. Dixon, Texas A. & M. University (TCWC) ; William E. Duellman, Museum of Natural History, University of Kansas (KU) ; Josef Eiselt, N aturhistorisches Museum, Vienna; Alice G. C. Grandison, British Museum (Natural History) (BMNH) ; W. Hellmich, Zoologische Sammlung des Bayerischen Staates, Munich; B. Huben- dick, Naturhistoriska Museet, Goteborg; Robert F. Inger and Hymen Marx, Field Museum of Natural History (FMNH); Alan E. Leviton, California Academy of Sciences (CAS) ; Clarence J. McCoy, Carnegie Museum (CM) ; Gunther Peters, Institut fiir Spezielle Zoologie und Zoologisches Museum, Berlin; Hobart M. Smith, formerly of Museum of Natural History, University of Illinois (UIMNH) ; Robert C. Stebbins, Museum of Vertebrate Zoology, University of California at Berkeley (MVZ) ; Paulo E. Vanzolini, Departa- mento de Zoologia, Sao Paulo (DZ); Greta Vestergren, Naturhistoriska Riksmuseet, Stockholm; Charles F. Walker, Museum of Zoology, University of Michigan (UMMZ) ; Ernest E. Williams, Museum of Comparative Zoology, Harvard University (MCZ) ; John W. Wright, Los Angeles County Museum of Natural History (LACM) ; Richard G. Zweifel, American Museum of Natural History (AMNH) . Field series from the University of Southern California collections are designated CRE (Costa Rica), JRM (Honduras), ^ and PAS (Colombia).- Robert J. Lavenberg and Mary Nafpaktitis facilitated my use of the General Electric Time Share Computer service provided by the Los Angeles County Museum of Natural History. I acknowledge the support from two National Science Foundation Sum- mer Fellowships for Teaching Assistants (Summers of 1964 and 1965), an NDEA Dissertation Travel Fellowship administered by the University of Southern California for travel in the summer of 1967, and a National Science -To be deposited in the herpetological collection of the Los Angeles County Museum of Natural History. 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 5 Foundation Graduate Traineeship for full academic support for the past two years, 1966-1968. For her constant encouragement and secretarial help, I thank my wife, Miriam. Species Accounts Adult members of the Melanonotus species group may be confused only with those of the Ocellatus species group because members of both groups have extensive toe fringes as adults. The Ocellatus group members always have smooth dorsolateral folds, while members of the Melanonotus group usually do not have any indication of a dorsolateral fold, or, if such are present, they are weak and warty, never smooth. Only brief synonymies are presented, including synonyms and generic reallocations of the valid species, as Gorham (1966) has recently published a bibliographic synonymy of the genus Leptodactylus. Leptodactyhis dantasi Bokermann Leptodactylus dantasi Bokermann, 1959; 5-8, figs. 1-5 (Type locality, Brasil: Acre; Feijo. Holotype WCAB 1240, female) . Diagnostic characters: The only other species in the Melanonotus group with dark bellies with discrete light spots are podicipinus and pustulatus. L. dantasi has large light spots on the belly and lacks a metatarsal fold; podici- pinus has small light spots on the belly and has a well developed metatarsal fold. Leptodactylus dantasi lacks light spots on the posterior face of the thigh; pustulatus has large, well defined light spots on the posterior face of the thigh. Summary of characteristics: Snout rounded from above, rounded acute in profile; canthus rostralis indistinct; loreal slightly concave in cross section; tympanum distinct, horizontal diameter 0.5 eye diameter; vomerine teeth in slightly arched series, posterior to choanae; head length greater than width, 43 per cent SL; head width 39 per cent SL; interorbital distance 6 per cent SL; first finger much longer than second, first just shorter than third, second longer than fourth; no distinct ulnar ridge; back warty-pustulose, tibia with many, pronounced warts; supratympanic fold does not reach shoulder; ventrolateral glands present but not distinct; toe tips very slightly expanded, dorsal surfaces not grooved; toes with well developed lateral fringes; subarticular tubercles moderately well developed; metatarsal fold absent; tarsal fold distinct along distal two-thirds of tarsus, weakly continuous with toe fringe; tarsus and foot with many horny spicules, spicules dark on scattered warts; standard length of female holotype 68 mm; femur shorter than tibia, 37 per cent SL; tibia shorter than foot, 39 per cent SL; foot longer than femur, 48 per cent SL; upper lip barred; light outlined dark band from interorbit to sacrum; rest of back uniform; upper surfaces of limbs faintly barred; venter with large light spots on 6 Contributions in Science No. 191 Figure 1. Geographic distribution of Leptodactylus dantasi, discodactyliis, piistii- latus, and podicipinus. Dashed line indicates 2000 meter contour. Open symbols indicate sites of sympatry with wagneri. 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 1 a dark ground, spots not always regular, spots anastomosing on limbs; posterior thigh uniformly dark. Distribution: Known only from the type locality (Fig. 1 ), 249 m. Remarks: L. dantasi is at present known only from the holotype. Leptodactylus discodactylus Boulenger Leptodactylus discodactyliis Boulenger, 1883 : 637, pi. 58, fig. 3 (Type locality, Peru: Loreto; Yurimaguas. Holotype BMNH 84.2.18.44, female). Gorham, 1966: 128 (Literature list, synonymy) . Leptodactylus nigrescens Andersson, 1945: 57-58 (Type locality, Ecuador: Rio Pastaza and Rio Napo, Watershed. Type series, Naturhistoriska Riks- museet, Stockholm, unnumbered. Gorham, 1966: 132 (Literature list, synonymy) . Diagnostic characters: The only other species in the Melanonotus species group that has disk-like swellings of the toe tips is wagneri. The upper disk surfaces of discodactylus have longitudinal grooves; if disks are present in wagneri, the upper surfaces are never grooved. Summary of characteristics : Snout rounded, subovoid, or subelliptical from above, usually rounded or rounded-vertical in profile; canthus rostralis indistinct; loreal slightly concave in cross section; tympanum distinct, hori- zontal diameter 0.5-0.66 eye diameter; male vocal slits elongate, arise lateral to posterior 0.25-0.33 tongue and parallel jaw almost to the angle of the jaw; single internal vocal sac in males; vomerine teeth in transverse or very slightly arched series, posterior to choanae; head length greater than width, 37-39.1-43 per cent SL; head width 32-35.2-39 per cent SL; interorbital distance 8-9. 5-1 1 per cent SL; finger tips swollen or not; first finger longer than second, first shorter than third, second longer than or rarely equal to fourth; no spines on thumb of male; arm of male not hypertrophied; no distinct ulnar ridge; back rarely smooth to scattered with coni apicales; tibia especially scattered with coni apicales; supratympanic fold extends to shoulder; gland present at corner of mouth or not; post-tympanic gland present or absent; ventrolateral glands if present, not extensive; glands usually not present on posterior face of thigh; no chest spines on males; toe tips expanded into disks, dorsal disk surfaces grooved; toes with well developed lateral fringes; subarticular tubercles well developed; metatarsal fold present, weakly to strongly developed; tarsal fold distinct along distal 0.5-0.66 tarsus, not continuous with toe fringe; tarsus scattered with coni apicales; foot smooth, or outer sole scattered with coni apicales; standard length of males to 35 mm, females to 35 mm; femur shorter than tibia, 38-43.6-48 per cent SL; tibia shorter than foot, 43-45.8-49 per cent SL; foot longer than femur, 48-52.0-56 per cent SL; upper lip uniform or barred; interorbital region usually with a light stripe bordered behind by a dark 8 Contributions in Science No. 191 triangle, the dark triangle may extend to the sacral region, or uniformly dark; rest of back usually uniform, may have faint large splotches; upper surfaces of limbs barred to uniform; venter with a profusion of melanophores anteriorly only, to profuse over the entire venter; posterior thigh mottled. Distribution : Known elevational range : 150-1000 m. Known from a few localities of the upper Amazonian drainage in Ecuador and Peru (Fig. 1). Nomenclature: Boulenger described L. discodactylus from Peru in 1883. The subsequent use of the name in the literature has been only a citation of the original description. Dr. James A. Peters allowed me to send one of his specimens, JAP 6197, to the British Museum for comparison with the holo- type of L. discodactylus. Specimen 6197 is from the upper known elevational range of the species, and its characteristics agree with those of other high elevation populations discussed in the remarks section. Dr. Alice G. C. Grandi- son of the British Museum (Natural History) compared specimen 6197 with the holotype of L. discodactylus. She noted the departures from the holotype characters associated with elevation (see remarks section, below), but con- cluded that in all other aspects, including the diagnostic toe fringe, toe disks, and upper toe disk surface grooves, the specimens were identical. Andersson described Leptodactylus nigrescens on the basis of three speci- mens from east Ecuador in 1945. His description is based upon the largest specimen. I was able to examine the largest syntype of L. nigrescens at the United States National Museum where Dr. Peters’ private collection was available for comparison. Mr. Werner C. A. Bokermann had previously bor- rowed the two smaller specimens and allowed me to examine them at the National Museum. The type series is composite. The largest syntype of L. nigrescens possesses the diagnostic combination of toe fringe and toe disks with grooved upper surfaces of L. discodactylus. The other two specimens have well developed toe disks, no finger disks, no toe fringe, no tarsal fold, and two rows (1 pair) of dorsolaterally arranged coni apicales. I consider the smaller two syntypes of L. discodactylus to represent Eleutherodactylus nigrovittatus Andersson, described in the same paper as L. nigrescens. In a cursory literature review, I was unable to find a senior synonym for this distinct little frog, but one may exist. To avoid future confusion, I hereby designate the largest (35 mm) specimen (the specimen has no museum number) the lecto- type of L. nigrescens. Leptodactylus nigrescens Andersson is a junior synonym of L. discodactylus Boulenger. Remarks: The few individuals available have certain differences which correlate with elevation. Contrasting the highland samples from Ecuador (1000 m) with the low- land samples from Ecuador and Peru (150-300 m), one finds the following consistent differences (lowland population characteristics in parentheses) : 1 ) the finger tips are not swollen (finger tips swollen, forming small disks) ; 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 9 2) the tarsal fold is usually distinct along the distal 0.66 of the tarsus (tarsal fold usually distinct on distal 0.5 of tarsus); and 3) smaller adult size, males to 30 mm, females to 32 mm (both males and females to 35 mm). I assume that the differences noted are responses to different climatic parameters associated with differences in elevation. Leptodactylus discodactylus has been taken in sympatry with L. wagneri at two localities, Ecuador: Pastaza; Puyo, 1000 m and Peru: Loreto; Rio Tamaya, Sobral, 150 m. Leptodactylus melanonotus (Hallowell) Cystignathus melanonotus Hallowell, 1860: 485 (Type locality, Nicaragua. Type apparently lost). Cystignathus echinatus Brocchi, 1877: 181-2 (Type locality, Guatemala: Rio- madre Nieja. Syntypes Paris Museum 6322-3) . Cystignathus microtis Cope, 1879: 265 (Type locality, Mexico: Guanajuato; Guanajuato. Syntypes USNM 9906, 9908, 9909). Cystignathus perlaev is CopQ, 1879: 269-270 (Type locality, Mexico: Oaxaca; Japana. Holotype USNM 10041, female). Leptodactylus melanonotus, Brocchi, 1881: 20 (Cites Hallowell’s record). Gorham, 1966: 131 (Literature list, synonymy). Leptodactylus occidentalis Taylor, 1937: 349-52, pi. 1, figs. 1, 2, 7 (Type locality, Mexico: Nayarit; Tepic. Holotype FMNH 100015, female). Gorham, 1966: 133 (Literature list, synonymy). Diagnostic characters: There is no one character that immediately dis- tinguishes L. melanonotus from the other members of the Melanonotus species group. The toe disks of L. discodactylus distinguish it from L. melanonotus (toe tips not expanded into disks). Leptodactylus dantasi, L. podicipinus, and L. pustulatus have dark bellies with discrete light spots; L. melanonotus may have a dark belly, but it is mottled and never distinctly spotted. Certain few individuals of L. melanonotus are difficult to distinguish consistently from L. wagneri (L. melanonotus characters in parentheses) : L. wagneri may have the toe tips expanded into distinct disks (toe tips never disk-like); L. wagneri may have a light longitudinal stripe on the posterior face of the thigh (no distinct stripe) ; L. wagneri reaches a larger adult size, standard length of males to 63 mm, females to 81 mm (males to 46 mm, females to 50 mm). The only way to distinguish consistently L. melanonotus from L. wagneri is by geography. Leptodactylus melanonotus is distributed from Mexico through Middle America, and west of the Andes in South America. Leptodactylus wagneri is found east of the Andes in South America. 10 Contributions in Science No. 191 Summary of characteristics : Snout rounded-nearly semicircular, round- ed, rounded-subelliptical, or rarely subovoid from above, rounded to rounded- vertical in profile; canthus rostralis indistinct; loreal slightly concave in cross section; tympanum distinct, horizontal diameter 0.5 to 0.75 eye diameter; male vocal slits elongate, arise lateral to mid-point of tongue to almost the angle of the jaw, usually parallel to jaw, rarely slightly oblique to the jaw; single internal vocal sac in males; vomerine teeth usually in transverse series, rarely very slightly arched, always posterior to the choanae; head usually longer than wide, rarely equal, or rarely head length shorter than wide, head length SO- SO.3-41 per cent SL; head width 30-34.1-38 per cent SL; interorbital distance 5-7. 2-9 per cent SL; finger tips not noticeably swollen; first finger just longer than or about equal to second, first shorter than third, second longer than fourth; finger ridges present, especially on second and third fingers; 2 spines on male thumb; male arm not hypertrophied; ulnar ridge not developed; head smooth or scattered with coni apicales; upper eyelids warty, glandular or smooth, usually lacking coni apicales; rest of back with scattered coni apicales, the coni may be on warts or not; upper femur and tibia scattered with coni apicales; supratympanic fold extends to shoulder, rarely indications of several warty dorsolateral folds; brown to orange ventrolateral glands poorly to ex- tensively developed, same glandular material may or may not be on jaw angle, post-tympanic region, groin, posterior thigh, or inner tibia and along the tarsal fold; no chest spines on male; toe tips usually not expanded, sometimes slightly expanded, never disk-like, never with dorsal surfaces grooved; toes with well developed lateral fringes; subarticular tubercles well developed; metatarsal fold present, weakly to well developed; tarsal fold distinct along distal 2/3- 5/6 tarsus, not continuous with toe fringe; tarsus with scattered coni apicales; foot smooth, with scattered coni apicales, or coni apicales on outer sole only; standard length of males to 46 mm, females to 50 mm; femur longer than, equal to, or usually shorter than tibia, 36-40.0-45 per cent SL; tibia shorter than foot, 37-42.7-47 per cent SL; foot longer than femur, 43-49.3-53 per cent SL; upper lip barred to uniform; interorbit with dark, light outlined tri- angle, a light and dark bar, or a light triangle reaching tip of snout anteriorly, bordered posteriorly by a dark triangle; rest of back brown to gray with darker indistinct spots, blotches, bands, stripes, or without pattern; upper limb surfaces barred to uniform; venter with few melanophores, appearing light to many melanophores profused especially anteriorly to rarely heavily profused over entire venter; posterior thigh mottled. Distribution : Elevational range : sea level — 1440 m. The species is known from both coasts of Mexico — Hermosillo, Sonora and the Gomez Farias region, Tamaulipas in the north, extending southward throughout Middle America. The species is known only from the western lowlands of South America to mid-Ecuador. (Figs. 2-4.) Nomenclature: The holotype of Cystignathus melanonotus Hallowell is 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 11 Figure 2. Geographic distribution of Leptodactylus melanonotus in Mexico. apparently lost. The late E. R. Dunn noted (Dr. Dunn’s notes presently on deposit in the library of Dr. Jay M. Savage) that the type was originally deposited in the National Museum as specimen number 6264 and that it was missing. The specimen is not listed in the catalog of type specimens of reptiles and amphibians in the U.S. National Museum (Cochran, 1961). In recent visits to the U.S. National Museum, although aided by the late Dr. Cochran and by Dr. Peters, I was unable to locate the type. The evidence indicates that the original holotype is lost. Leptodactylus melanonotus is very closely related to both L. wagneri and L. podicipinus. L. melanonotus is found in Mexico and Middle America, where there is relatively little question as to what the species is. L. melanonotus is found west of the Andes and some individuals are very difficult to differentiate consistently from some individuals of L. wagneri, an Amazonian species. For nomenclature stability, the name L. melanonotus should be applied to a single specimen from the population found in Nicaragua, the type locality of the ori- ginal, now lost, holotype. I therefore designate specimen number 84848 in the collection at the University of Kansas, an adult male, from Nicaragua, Zelaya, Bonanza, as the neotype of Cystignathus melanonotus Hallowell. I have examined the holotype and two paratypes of Cystignathus microtis 12 Contributions in Science No. 191 Figure 3. Geographic distribution of Leptodactylus melanonotus in Middle America. Cope. Only two species of Leptodactylus (labialis and melanonotus) occur in Mexico; they are quite distinct from each other. Members of the type series of C. microtis agree with L. melanonotus in having fringed toes and mottled thighs; both characters differentiate L. melanonotus from L. labialis. There is no morphological reason for retaining C. microtis. The status of C. microtis has always been questioned, because the type locality was given as Mexico : Guanajuato; Guanajuato. Guanajuato is well above the elevational limits of L. melanonotus. No other specimens of L. melanonotus have since been taken from Guanajuato. I think the locality data are in error and probably represent the shipping point from which the specimens were sent to the U.S. National Museum rather than the collecting site. I have examined numerous paratypes of Leptodactylus occidentalis Taylor and have collected topotypic specimens which correspond closely with the type series of L. occidentalis. As is shown in the character analysis portion, L. occidentalis does not differ in any consistent way from L. melanonotus and is therefore considered a junior synonym of the latter. I have examined the holotype of Cystignathus perlaevis Cope and concur with the numerous workers who have placed it in synonymy with L. mela- nonotus. 1970 Studies on the Frogs of the Genus LEPTODA CTYL US 13 Figure 4. Geographic distribution of Leptodactylus melanonotus in South America, Dashed line indicates 2000 meter contour. Open symbol indicates site of sympatry with wagneri. 14 Contributions in Science No. 191 Remarks: A single specimen of L. melanonotus is recorded from Starr County, Texas, in the United States. I have examined the specimen (UIMNH 29817); it is a juvenile L. melanonotus. This single record is about 400 km north of the next recorded locality, from which several samples are known. The locality data for the Texas specimen is possibly in error, and for that reason I have not included the record in the range of L. melanonotus. Leptodactylus melanonotus has been taken in sympatry with L. wagneri at one locality in western Colombia (Narino, La Guayacana) . This instance of sympatry is not too surprising, as the sources of the rivers Cauca and Patia interdigitate at about 1600-1800 m within the range of L. wagneri. Lepto- dactylus wagneri is found throughout the Valle de Cauca. Apparently L. wagneri follows the Rio Patia into the west-coast lowlands of Colombia. Leptodactylus podicipinus (Cope) Cystignathus podicipinus Cope, 1862; 156 (Type locality, Paraguay. Holotype ANSP 14539). Leptodactylus podicipinus, Boulenger, 1882: 248 (Catalogue, synonymy). Gorham, 1966: 135-6 (Literature list, synonymy). Leptodactylus nattereri Lutz, 1926: 1011-2 (Type locality, Brasil: Sao Paulo; Itapura, Ilha Seca. Syntypes AL 1314-1315). Leptodactylus podicipinus podicipinus, Gans, 1960: 305-6 (List) . Gorham, 1966: 136 (Literature list, synonymy) . Diagnostic characters : The only other species in the Melanonotus species group with a dark belly with discrete light spots are L. dantasi and L. pustu- latus. L. dantasi has large light spots on the belly and lacks a metatarsal fold. L. podicipinus has small light spots on the belly and has a well developed metatarsal fold. L. pustulatus has large, light discrete spots not only on the belly, but on the entire ventral surface of the leg and on the posterior surface of the thigh. L. podicipinus has small spots which, if on the leg, are only on the ventral surface of the thigh. The males of L. podicipinus have thumb spines, the males of L. pustulatus lack thumb spines. Summary of characteristics : Snout rounded, subelliptical to subovoid from above, rounded in profile; canthus rostralis indistinct; loreal slightly con- cave in cross section; tympanum distinct, horizontal diameter eye diameter; male vocal slits elongate, arise lateral to middle of tongue, parallel jaw almost to angle of jaw; single internal vocal sac in males; vomerine teeth usually in transverse series, rarely in very slightly arched series, always posterior to choanae; head length greater than width, 34-37.0-41 per cent SL; head width 32-34.6-38 per cent SL; interorbital distance 7-8.2-10 per cent SL; finger tips not swollen; first finger usually longer than second, sometimes 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 15 equal, first finger much shorter than, shorter than, or about equal to third, second longer than fourth; finger ridges present, especially on second and third fingers; 2 spines on male thumb; male arm not hypertrophied; ulnar ridge usually not developed, sometimes a glandular ridge or 4 coni apicales in a row; dorsum smooth, glandular, or warty, scattered with coni apicales, es- pecially on posterior dorsum and upper leg surfaces; supratympanic fold extends to shoulder, may or may not be faint indication of several warty dorsolateral folds; ventrolateral glands orange-brown, small to large or absent, glands present or absent on angle of jaw, dorsolateral folds, posterior thigh, inner tibia, and along tarsal fold; no chest spines on male; toe tips not ex- panded; toes with well developed lateral fringes; subarticular tubercles well developed; metatarsal fold well developed; tarsal fold distinct along distal 2/ 3-5/ 6 tarsus, usually not continuous with toe fringe, rarely weakly con- tinuous with toe fringe; tarsus scattered with coni apicales; foot smooth or outer sole with scattered coni apicales; standard length of males to 38 mm, females to 44 mm; femur usually shorter but sometimes longer than tibia, 35-40.9-48 per cent SL; tibia shorter than foot, 38-42.9-48 per cent SL; foot longer than femur, 44-50.7-56 per cent SL; upper lip barred or uniform; interorbital region with light and dark stripe, or light outlined dark triangle extending to sacrum or not; rest of back uniform or with indistinct darker bands; upper limb surfaces uniform or with indistinct dark bars; throat, chest, belly dark with small light spots, rest of venter dark, but not with distinct light spots, posterior portion of belly occasionally with limb pattern rather than light- spotted; posterior thigh usually mottled, rarely with a moderately distinct light longitudinal line. Distribution: Elevational range: 18-550 m. Leptodactylus podicipinus is found south of the Amazonian drainage region in the following river drainages: Sao Francisco, Parana, Paraguay. The species is absent from the narrow coastal lowland strip along eastern Brazil from the States of Rio Grande do Norte to Rio de Janeiro. The species follows the Serras de San Jose, Santiago, and Sunsas in Bolivia, and the Serras dos Parecis and Pacaas Novos in western Brasil. (Fig. 1 ) . Nomenclature: In his description of Cystignathus podicipinus, Cope designated as type(s) “Mus. Smithsonian, (No. 5831) Philada. Acad.” Dr. James A. Peters informed me that in the catalog under 5831 is a later remark stating that the specimen had been sent to the Academy of Natural Sciences in Philadelphia. Apparently there was only one specimen upon which Cope based C. podicipinus. I have examined the holotype of C. podicipinus from the Academy of Natural Sciences. The specimens I have examined from Paraguay are morphologically similar to the holotype in all diagnostic features. Lutz described Leptodactylus nattereri on the basis of specimens collected in the state of Sao Paulo, Brasil, and included the specimens figured by Stein- dachner (1864) from the collection of Natterer. The figures in Steindachner 16 Contributions in Science No. 191 (Plate XI, figs. 1 a-d) show the two spines per thumb and dark belly with small light spots characteristic of L. podicipinus. Remarks: Leptodactylus podicipinus has been taken in sympatry with L. wagneri at two localities: Bolivia (Santa Cruz: Buenavista, 500 m) and Brasil (Rondonia: Forte Principe da Beira, 100-200 m). At both localities the species are very distinct, markedly differing in size, belly pattern, and posterior thigh pattern. There is apparently a narrow band of sympatry between L. podicipinus and L. wagneri, with character displacement occurring in the band of sympatry. Leptodactylus pustulatus (Peters) Entomoglossus pustulatus W. Peters, 1870: 647, PI. 2, fig. 1 (Type locality, Brasil: Ceara. Type apparently lost). Leptodactylus pustulatus, Boulenger, 1882: 239 (Catalogue, synonymy) ; Gorham, 1966: 137 (Literature list, synonymy). Diagnostic characters : The only other species in the Melanonotus species group with a dark belly with discrete light spots are L. dantasi and L. podici- pinus. Leptodactylus pustulatus has large, light discrete spots on the posterior surface of the thigh. Leptodactylus dantasi and L. podicipinus do not have spots on posterior surface of the thigh. The males of L. pustulatus lack thumb spines, the males of L. podicipinus have thumb spines. Summary of characteristics: Snout rounded to subovoid from above, rounded to rounded-obtuse in profile; canthus rostralis indistinct; loreal slightly concave in profile; tympanum distinct, horizontal diameter 2/3 eye diameter; male vocal slits elongate, parallel to jaw, extend from mid-tongue almost to angle of jaw; single internal vocal sac in male; vomerine teeth usually in transverse series, rarely very slightly arched, always posterior to choanae; head length greater than width, 35-36.4-38 per cent SL; head width 33-34.0-35 per cent SL; interorbital distance 5-6. 7-8 per cent SL; finger tips not expanded; first finger longer than second, first shorter than third, second greater than fourth; finger ridges present, especially on fingers two and three; no spines on male thumb; male arm not hypertrophied; ulnar ridge not developed; dorsal texture pustulose, with scattered coni apicales; supratympanic fold extends to shoulder, may or may not be indications of 6-8 warty dorsolateral folds; brown ventrolateral glands extend onto belly, or entire frog appears glandular; no chest spines on males; toe tips not expanded; toes with well developed lateral fringes; subarticular tubercles well developed; metatarsal fold present; tarsal fold distinct along distal Vi-Va- tarsus, usually continuous with toe fringe, occasionally weakly continuous; tarsus with scattered coni apicales; foot usually with scattered coni apicales, rarely smooth; standard length of males to 38 mm, females to 51 mm; femur shorter than tibia, 37-39.0-41 per cent SL; tibia shorter than foot, 39-41.8-44 per cent SL; foot longer than femur, 45- 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 17 51.3-55 per cent SL; dorsum uniform or patterned, if patterned, a light inter- orbital triangle originating at tip of snout bordered behind by a dark triangle, rest of back with darker longitudinal lines; upper limb surfaces faintly barred to uniform; venter dark with large light spots extending onto lower leg surfaces and groin; posterior thigh with light spots as on venter. Distribution: Known elevational range: 100-200 m. The few localities from which L. pustulatus have been collected are all low Amazonian. The major river drainages for which accurate locality data are available include the Paranaiba, Araguaia, and Tocantins. (Fig. 1). Nomenclature: Entomoglossus pustulatus was described by W. Peters on the basis of a single female from Ceara, Brasil, in the collection of the Royal Zoological Museum, Berlin. Dr. Peter Beurton of the Berlin Museum states that he was unable to locate the type in the collections there. Dr. Hellmich of the Munich Museum informed me that the type of Entomoglossus pustulatus is not and has never been in the Munich collection, the likely alternative depository for the type. This suggests that the type is lost. The identity of L. pustulatus has been very unclear because of the rarity of specimens in collections. Because of this confusion and because there appears to be no other Ceara (topotypic) material in collections, I have selected an individual repre- sentative of the population that I consider to be L. pustulatus. The nearest locality to the type locality is the Rio Poti in Piaui. The specimen from this locality is only in a fair state of preservation, but it still retains the pattern diagnostic of L. pustulatus. I designate this specimen, MCZ 373, from the collection of the Museum of Comparative Zoology, Harvard University, as the neotype of Entomoglossus pustulatus Peters. Remarks: L. pustulatus has been taken in sympatry with L. wagneri at one locality (Brasil: Mato Grosso: Sao Domingoes, Rio das Mortes, 200 m). Leptodactylus wagneri (Peters) Plectromantis wagneri W. Peters, 1862: 232-3 (Type locality, Ecuador, west side of the Andes. Type formerly at Munich Museum, destroyed in World War II). Platy mantis petersii Steindachner, 1864: 254-6, PI. 16, fig. 2, 2 a-c (Type locality, Brasil: Amazonas; Marabitanas. Type lost). Leptodactylus brevipes CopQ, 1887: 51-2 (Type locality, Brasil: Mato Grosso, Chupada, 30 mi. NE of Cuyata, near the headwaters of the Xingu. Holotype ANSP 11270, female). Leptodactylus validus Garman, 1887: 14 (Type locality. West Indies: St. Vincent; Kingston. Syntypes ANSP 26108, MCZ 2185). Leptodactylus wagneri Nieden, 1923: 479 (Literature list, synonymy). Gor- ham, 1966: 140 (Literature list, synonymy). 18 Contributions in Science No. 191 Leptodactylus pallidirostris A. Lutz, 1930: 25-6, PL 1, fig. 3 (Type locality, British Guiana: Essequibo; Kartabo). Leptodactylus natalensis A. Lutz, 1930: 26-7, PI. 1, figs. 7-7a, PI. 3, figs. 1-2 (Type locality, Brasil: Rio Grande do Norte; Rio Bahu. Syntype USNM 81130, male). Leptodactylus intermedius A. Lutz, 1930: 27-8, PI. 3, fig. 6 (Type locality, Brasil: Amazonas; Manacapuri near Manaos). Eleutherodactylus leptodactyloides Andersson, 1945: 43-4, fig. 15 (Type locality, Ecuador: Pastaza; Rio Pastaza. Holotype Stockholm Royal Museum, male). Leptodactylus podicipinus petersii, Rivero, 1961: 47 (Redescription, syn- onymy). Gorham, 1966: 136 (Literature list, synonymy). Diagnostic characters: Leptodactylus wagneri may have expanded toe tips, but the upper surface of the toe tip is never with longitudinal grooves, distinguishing wagneri from discodactylus, which has well developed toe disks with the upper disk surfaces grooved. Leptodactylus wagneri may have a dark belly, but it is mottled, and not with distinct light spots, distinguishing it from the dark-bellied, light-spotted dantasi, podicipinus, and pustulatus. Some in- dividuals of wagneri are difficult to distinguish consistently from melanonotus (melanonotus characters in parentheses); wagneri may have the toe tips ex- panded into distinct disks (toe tips never disk-like) ; wagneri may have a light longitudinal stripe on the posterior face of the thigh (no distinct stripe); wagneri reaches a larger adult size, standard length of males to 63 mm, females to 81 mm (males to 46 mm, females to 50 mm). The only consistent way to distinguish melanonotus from wagneri is by geography. Leptodactylus mela- nonotus is distributed from Mexico through Middle America, and west of the Andes in South America. Leptodactylus wagneri is found east of the Andes in South America. Summary of characteristics: Snout usually rounded, rarely subelliptical or subovoid from above, rounded to rounded-vertical in profile; canthus rostralis indistinct; loreal slightly concave in cross section; tympanum dis- tinct, horizontal diameter ’/2-% eye diameter; male vocal slits elongate, arise lateral to tongue, parallel jaw and extend almost to the angle of the jaw or slightly oblique to jaw; single internal vocal sac in male; vomerine teeth in transverse to arched series, posterior to choanae; head length greater than, equal to, or less than width; head length 33-36.8-42 per cent SL; head width 32-34.9-39 per cent SL; interorbital diameter 6-7. 8-9 per cent SL; finger tips not noticeably swollen; first finger longer than second, first shorter than or equal to third, second longer than or equal to fourth; fingers with lateral ridges, especially on second and third fingers; two spines on male thumb; male arm 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 19 slightly hypertrophied only in largest specimens; no ulnar ridge; dorsal texture smooth to glandular, posterior portion of back and legs with scattered coni apicales; supratympanic fold extends to shoulder, may or may not be indica- tion of 2-4 faint dorsolateral folds; orange-brown ventrolateral glands moder- ately developed or absent; gland at angle of jaw developed or not; post- tympanic gland developed or not; posterior thigh gland present or absent; no chest spines on male; toe tips not expanded to moderately expanded and disk-like, never with longitudinal grooves on dorsal surface; toes with well developed lateral fringes; subarticular tubercles well developed; metatarsal fold present; tarsal fold distinct along distal 4/7-3/4 tarsus, not continuous with toe fringe; tarsus with scattered coni apicales; foot smooth, with scat- tered coni apicales, or outer sole with scattered coni apicales; standard length of males to 63 mm, females to 81 mm; femur shorter than tibia, 35-42.4-48 per cent SL; tibia shorter than foot, 41-46.9-51 per cent SL; foot longer than femur, 49-52.4-57 per cent SL; upper lip usually barred, sometimes uniformly patterned; light and dark interorbital bar, dark bar may extend posteriorly as a triangle or band as far as sacrum; rest of back uniformly patterned or indis- tinctly spotted or striped, warts on sides may be darker than ground color; upper limb surfaces barred to uniform; ventral pattern of scattered melano- phores, melanophores may be profuse anteriorly only, or profuse over entire venter; posterior thigh with distinct light longitudinal stripe to mottled. Distribution: Elevational range: sea level — 1900 m. The species is widely distributed throughout northern and central South America east of the Andes. There is a single record west of the Andes in Colombia. The species occurs in the Cauca and Magdalena valleys of west- central Colombia, and throughout the greater Amazonian Basin; the species is also distributed along the narrow eastern coastal strip of Brasil from the States of Rio Grande do Norte to Rio de Janeiro. The species occurs on the islands of Bequia, Granada, St. Vincent, Tobago, and Trinidad. (Fig. 5) . Nomenclature: Wilhelm Peters (1862) described Plectromantis wagneri from Ecuador on the basis of a single male specimen. Peters listed the type as “von Dr. Moritz Wagner an den Westseite der Anden in Ecuador . . . im zoologischen Cabinet du Munchen.” Dr. Richard Etheridge kindly looked for the type when he was at the Munich museum. He writes (personal communi- cation), “. . . the old museum had one specimen. No. 1080/0, from ‘Pastassa’ collected by Wagner. This may or may not have been the type, there is no indication in the record that it was, but in any event the specimen was de- stroyed during World War II.” It is likely that specimen number 1080/0 was the unique type of Plectromantis wagneri. Pastassa is a geographically more reasonable locality than the west side of the Andes. Peters’ description is quite good, and the secondary sexual characteristics of spines on the thumb of the male leave no doubt as to the identity of the frog described by Peters. It is the population herein called L. wagneri. The only Leptodactylus west of the 20 Contributions in Science No. 191 Figure 5. Geographic distribution of Leptodactyliis wagneri. Dashed line indicates 2000 meter contour. Open symbols indicate sites of sympatry with other species of the Melanonotus group. 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 21 Andes in Ecuador with thumb spines on the male is L. melanonotus. The males of L. melanonotus reach 40 mm SL in Ecuador; the type of P. wagneri was 68 mm SL. I consider that Peters’ locality is probably incorrect, and the data as recorded for Munich specimen 1080/0 are correct. The name L. wagneri has never been used in association with a Leptodactylus population in the literature since the original description. The status of the northern South American population of the Melanonotus group has always been uncertain. To stabilize the nomenclature I hereby designate the holotype of Eleutherodactylus lepto- dacty hides Andersson in the Stockholm Royal Museum of Natural History (no number), an adult male, from the Rio Pastaza, as the neotype of Plectromantis wagneri Peters. Steindachner described Platymantis petersii on the basis of a single male specimen from Marabitanas, Amazonas, Brasil. Dr. Josef Eiselt of the Vienna Museum has been unable to find the unique type of Platymantis petersii. The type has been searched for several times and is most likely lost. Leptodactylus petersii has been used in various combinations for various populations of frogs of the Melanonotus species group. I consider P. petersii to be a junior synonym of L. wagneri, and that nomenclatural stability would be served by designating a neotype which conforms with my analysis. I have not examined any material from Marabitanas, nor do I know of any material from there. The closest record that I know of to Marabitanas is in Venezuela. This specimen, AMNH 23182, is in very good condition. The locality is Venezuela: Amazonas; Rio Pescado, approximately 150 m. The specimen is a female. I hereby designate AMNH 23182 as neotype of Platymantis petersii Steindachner. Cope described Leptodactylus brevipes on the basis of a single specimen from Chupada, Mato Grosso, Brasil. The name has been cited in literature lists based upon the original description but not upon the basis of recent material. I have examined the type, and find it the same as wagneri. The belly has very few melanophores, so that no confusion is possible with either podicipinus or pustulatus, the only other members of the Melanonotus species group with which it could possibly occur or be confused. Leptodactylus brevipes is re- garded as a junior synonym of L. wagneri. Leptodactylus validus was described by Garman from a series of speci- mens from the island of St. Vincent. I have found no differences between any of the island populations and the mainland population of wagneri, as shown below in the character analysis section. No holotype was designated by Gar- man. I hereby designate MCZ 71920, an adult male, from Kingston, St. Vin- cent, as the lectotype of Leptodactylus validus Garman. Lutz described three species of Leptodactylus of the Melanonotus group in the same paper in 1930. The first, pallidirostris, was described from British Guiana: Essequibo; Kartabo. I have not been able to examine the type, but have examined numerous examples of the Melanonotus group from the type locality. All specimens represent a single species and conform with the type 22 Contributions in Science No. 191 description of pallidirostris. The specimens from Kartabo are not distinct from wagneri. Thus, L. pallidirostris is considered a junior synonym of L. wagneri. Leptodactylus natalensis is the second species described by Lutz. I have examined one syntype of the type series and find it the same as other specimens examined in this study from the type locality (vicinity of Natal). The syntype has a lightly pigmented venter and marbled posterior thigh surfaces. I con- sider L. natalensis Lutz to be a junior synonym of L. wagneri. No holotype was designated. I designate the specimen I have examined, USNM 81130, a male, as the lectotype of Leptodactylus natalensis Lutz. The third species of the Melanonotus group described by Lutz is L. inter- medins. As in the case of pallidirostris, I have not seen the types, but have examined material from the type locality. The topotypic material I have ex- amined from Manaus agrees with the description of intermedins. I consider the material from Manaus to be wagneri', hence intermedins is a junior synonym of L. wagneri. Andersson described Eleutherodactylus leptodactyloides on the basis of a single specimen from Ecuador. I have been able to examine the type specimen. It is a Leptodactylus as suggested by the specific name, rather than an Eleu- therodactylus. I have designated this specimen, above, as the neotype of Plectromantis wagneri, thereby making Eleutherodactylus leptodactyloides a junior synonym of L. wagneri. Remarks'. Leptodactylus wagneri has been taken in sympatry with dis- codactylus, melanonotus, podicipinus, and pustulatus. Key to Adult Members of the Melanonotus Species Group la. Toe tips expanded into disks, dorsal surfaces with longitudinal grooves discodactylus lb. Toe tips usually not expanded into distinct disks, dorsal surfaces never grooved 2 2a. Ventral surfaces of legs and posterior surface of thigh with large, light, distinct spots pustulatus 2b. Ventral surfaces of legs light or dark, if large spots present, spots not distinct; posterior surface of thigh light or dark, never with large, light, distinct spots 3 3a. Belly dark with distinct, light spots, spots usually distinct over entire belly, but sometimes anastomosing on posterior belly 4 3b. Belly light or dark; if dark, no distinct light spots 5 4a. Belly with large light spots; no metatarsal fold dantasi 4b. Belly with small light spots; well developed metatarsal fold podicipinus 5a. Moderate sized frogs, males to 46 mm SL, females to 50 mm; posterior thigh never with light longitudinal stripe; toe tips never noticeably ex- panded; Mexico, Middle America, South America west of the Andes to Ecuador .melanonotus 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 23 5b. Moderate to large frogs, males to 63 mm SL, females to 81 mm; posterior thigh may have a light longitudinal stripe; toe tips often noticeably ex- panded; South America east of the Andes wagneri Character Analysis Standard length: Two broad geographic patterns are evident in male L. melanonotus. The coastal populations vary between 30 and 40 mm. Modal sized individuals (36 mm Hermosillo, Sonora, Mexico) or smaller specimens (34 mm Gomez Farias region, Tamaulipas, Mexico; 33 mm Ecuador) are associated with the extreme northern and southern geographic range. The largest specimens are associated with two lowland pass regions in the montane backbone of the Americas: the lowland pass at Tehuantepec, Mexico; and the Arenal pass in Costa Rica. The geographic pattern of female L. melanonotus size parallels the male geographic pattern, but the trends are not as pro- nounced. In addition to the Tehuantepec and Arenal passes harboring the largest females (47-50 mm), the Yucatan Peninsula and eastern lowlands of Costa Rica also have females of large size (46-48 mm). The range of adult female size is approximately 35-50 mm. Leptodactylus wagneri males attain greatest size along the eastern slopes of the Andes in Colombia, Ecuador, and Peru, ranging from about 40-60 mm. Along the northern coast of South America, on the islands, and in the Ama- zonian basin, the sizes range from 30-40 mm. The geographic pattern of L. wagneri female size exactly parallels the male adult size with the exception of the ranges. The east Andean slope populations most frequently range in size from 50-80 mm; the north coast, island, and Amazonian populations usually range from 40-60 mm. Adult male L. podicipinus range from about 28-38 mm in length, with a trend towards smaller size at the southern end of the geographic range. Adult female L. podicipinus range in size from 36-44 mm also with a trend towards smaller size at the southern end of the geographic range. The geographic pattern of size variation is basically the same in males and females, the only difference being that the females average larger than the males. The relationship between size and rainfall, number of dry months, and elevation are statistically significant only in the case of L. wagneri. In L. wag- neri males, a statistically significant correlation is present between greatest size of adults and mesic habitats (r = .28, P = <.01 ). This correlation appears to be biologically valid for both males and females, as a statistically significant correlation was found with smaller size in more xeric habitats as measured by number of dry months (males r = —.50, P — <.01, females r = — .27, P — <.01). A statistically significant correlation was found for greatest size and increasing elevation for both males and females in L. wagneri (males 24 Contributions in Science No. 191 Figure 6. Ventral color pattern standards. Note that pattern F is restricted to Leptodactylus piistulatiis. 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 25 T = .38, P = <.01, females r = .42, P = <.01 ) . The elevational effects pro- ducing size differences in these frogs are probably associated with rainfall differences also. Character displacement apparently occurs in the narrow bands of sym- patry between L. melanonotus, podicipinus, and wagneri. At La Guayacana in Colombia, a female melanonotus (mature) measures 38 mm, and a female wagneri measures 59 mm. Two instances of sympatry are known for wagneri and podicipinus. At Buenavista, Bolivia, a female podicipinus measures 40 mm, and a female wagneri 46 mm. At Forte Principe da Beira, Brasil, a female podicipinus measures 34 mm, a female wagneri 43 mm, and a male podicipinus 29 mm, and a male wagneri 38 mm. Ventral pattern : Five categories were used in scoring the ventral pattern. These ranged from A, in which the throat and belly do not have any melano- phores, to E, in which the throat and belly are heavily covered with mela- nophores, leaving light spots (Fig. 6). The distribution of ventral pattern is variable throughout most of the geographic range of L. melanonotus. In many localities, conditions ranging from B to D, encompassing the variation encountered in this species, are found. The populations at the northern and southern extremes of the geo- graphic range lack much of the variability as in other populations, and are lighter in ventral coloration. Northwestern Mexican and South American populations are average in pattern density. There is a weak trend of darker pattern in Atlantic lowland populations. No clear geographic trend in ventral pattern in L. wagneri is evident. The usual range for the species is B to D, with a few populations having an inter- mediate condition between D and E. Most adequate samples (10 or more individuals) show a variation through only two categories, either B-C or C-D. The occurrence of individuals with pattern B seems to be random with per- haps more B category individuals throughout the northern part of the species range, including the populations of islands off northern South America. There is no geographic trend of ventral color pattern in L. podicipinus. Almost all populations fall in the E category, with a few individuals in D or between D and E. A correlation between a darker ventral color pattern and greater annual rainfall is present in L. melanonotus (r = .17, P — .05). There is no correlation between ventral pattern and glandular extent, number of dry months, or ele- vation in melanonotus. No correlations between ventral color pattern and glandular extent, annual rainfall, number of dry months, or elevation are present in podicipinus. A correlation is found between a lighter ventral color pattern and greater ventrolateral gland extent in wagneri (r = —.14, P = .05). This is most certainly not a cause-effect relationship, but it may indicate that the genes responsible for these two characters are on the same chromosome or are co-functional in some other way. No correlations between ventral color 26 Contributions in Science No. 191 Figure 7. Ventrolateral gland standards. 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 27 pattern and annual rainfall, number of dry months, or elevation are found for wagneri. Ventral color pattern is a character that has apparently undergone charac- ter displacement in the area of sympatry between L. podicipinus and wagneri. Certain individuals of wagneri closely approach the pattern of podicipinus. In the two known instances of sympatry of podicipinus and wagneri, the ventral color patterns are clearly distinct; wagneri individuals have a light belly; podicipinus individuals have a dark, light spotted belly. Leptodactylus dis- codactylus and wagneri have basically the same ventral color patterns. Lepto- dactylus pustulatus is very distinct from wagneri with respect to ventral pattern. Leptodactylus pustulatus has a distribution allopatric to those of dantasi and podicipinus, the species it most closely resembles in ventral color pattern. Ventrolateral glands: Five categories of ventrolateral gland extent were established, ranging from no glands. A, to the entire belly, and most other ventral surfaces covered with glands, E (Fig. 7). In addition, five categories of gland color were noted in preserved animals: 1 — brown; 2 — gray; 3 — black; 4— orange-brown; and 5 — -yellow-orange. The geographic distribution of ventrolateral gland extent and coloration in L. melanonotus is a mosaic. The ventrolateral glands are very compact and brownish to orange-brown (B — 1,4) in populations from northwest Mexico and Panama. The greater degree of glandular extent is found in populations from the Isthmus of Tehuantepec, Yucatan Peninsula, Atlantic coastal British Honduras and Honduras, and the Arenal pass region in Costa Rica. These generalities are complicated in that individuals from a single locality 1 ) have or lack glands and 2) demonstrate a range of variation of glandular extent which encompasses the range of the entire species. Most populations of L. wagneri either have no ventrolateral glands or have orange-brown diffuse glands of moderate extent (C — 4). A few wagneri specimens from the southern part of their geographic range have more ex- tensive glandular extent. Where moderate samples (10 or more individuals) are available, the specimens at any single locality demonstrate categories A and C — -4. There is no geographic trend. Many individuals of L. podicipinus lack ventrolateral glands. When ventrolateral glands are present, they are of moderate extent and orange-brown in preservative (C — 4, D — 4). Individuals from a single locality usually en- compass the range of variation found within the species. No clear geographic trend is evident. There are no correlations between ventrolateral gland extent and annual rainfall, number of dry months, or altitude in L. melanonotus and podicipinus. A correlation is present between greater ventrolateral gland extent and greater annual rainfall in wagneri (r = .13, P = .05). Such a correlation is more sug- gestive of a causal relationship than the correlation between ventral color pattern and ventrolateral gland extent in wagneri. There are no correlations 28 Contributions in Science No. 191 Figure 8. Posterior surface of thigh color pattern standards. Note that pattern D is restricted to Leptodactylus pustulatus. 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 29 between ventrolateral gland extent and number of dry months or elevation in wagneri. Changes of extent of ventrolateral glands may occur seasonally, but there are no collections made at all seasons from a single locality. In the summer of 1967 when breeding was just beginning, Dr. James R. Dixon and I collected several series of L. melanonotus from western Mexico. We noticed that no glands were evident on living animals. After a few days in preservative, how- ever, the ventrolateral glands on the animals were dark brown in color and distinct. The function of ventrolateral glands is not known. Color pattern of posterior surface of thigh'. Four categories were dis- tinguished: A) mottled; C) with a light longitudinal stripe; B) intermediate between A and C; D) with distinct light spots on a dark background (Fig. 8). Almost all L. melanonotus have a mottled posterior thigh, although a few individuals from scattered localities show some indication of a light stripe (category B). In large series of L. wagneri from one locality, character states A, B, and C are represented. This is true throughout the geographic range of wagneri, with the exception of the Cauca Valley (Colombia) population. Few large samples from L. podicipinus demes were available for this study, but, on the basis of the series available, the distribution of the thigh pattern appears to parallel the situation in wagneri. That is, categories A, B, and C are represented within any deme. There is no correlation between the posterior thigh pattern and the climatic and elevational parameters used. It appears likely that the light stripe in L. wagneri and podicipinus is genetically controlled by very few genes, perhaps one. The scattered B categories in melanonotus populations indicate that the gene potential for a posterior thigh light stripe is present. The D category is found only in pustulatus. The light stripe on the posterior thigh is another feature that shows character displacement in the instances of sympatry between L. podicipinus and wagneri. At both sites of known sympatry, wagneri individuals have well developed stripes. The podicipinus at the sites of sympatry either have a mot- tled thigh (Buenavista, Bolivia), or only an indication of a light stripe (Forte Principe da Beira, Brasil). At this latter locality, the differences in thigh pat- tern are very striking. The light stripe in wagneri individuals is completely bordered with black and is very distinct from the thigh pattern of sympatric podicipinus. Toe-disk size: Three categories were used (Fig. 9): A) toe tips not expanded at all; C) Toe tips expanded into definite disks; B) intermediate condition. There is a gradient of toe-tip size in L. melanonotus. In the northern part of the range, Mexico to Costa Rica, most are in category A, only 3 demes among those examined within this area had individuals with B condition. Many 30 Contributions in Science No. 191 Figure 9. Toe tip standards. Pattern D is the upper disk surface of Leptodactylus discodactylus showing longitudinal grooves. individuals from Costa Rica have intermediate disks, B. In the South, from Panama through Ecuador, all populations have B disks. In no case is the toe tip expansion developed into a distinct disk as found in L. discodactylus. A distinct geographic gradient in toe tip expansion is also seen in L. wagneri. The Cauca Valley (Colombia) population is marked by non-ex- panded toe tips. The rest of the Andean slope and upper Amazonian popula- tions of Colombia, Ecuador, Peru, and Bolivia have categories A and B, with B predominant. The frogs associated with the rivers draining northward, those upon the islands, and probably those of the eastern Amazon region are either B or C; that is, many individuals from these areas have distinct toe disks such as are found in L. discodactylus. No distinct geographic gradient in toe tip expansion is evident in L. podi- cipinus. Most specimens lack expanded toe tips. A, but a few individuals from scattered demes have slightly expanded toe disks, B. The possible role of character displacement in toe-tip expansion is evident at the two sites of sympatry between L. podicipinus and wagneri. At both sites, the toe tips of podicipinus are not expanded. A, and the toe tips of wagneri are slightly, but noticeably, expanded, B. Mating call: Mating call samples were chosen to answer the specific status concerning L. occidentalis. Demes were sampled in lowlands throughout the latitudinal geographic range of occidentalis from Sonora through Nayarit. Lowland samples of melanonotus from Jalisco, Colima, and Michoacan were 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 31 Figure 10. Mating calls of Leptodactylus melanonotus in western Mexico. Vertical scale in kilocycles per second, horizontal scale in seconds. Arrows indicate dominant frequency band. A — Cali specimen from Sinaloa, 36 km S Los Mochis turnoff on Mexico Highway 15. B-— Call specimen from Jalisco, 0.5 km NE Tonila. C — Call specimen from Jalisco, 7.7 km E La Huerta. A 8—1 also taken. Demes along elevational transects were sampled in the states of Jalisco and Nayarit. All of the calls have a basic similarity when heard and when viewed on sonagrams. All calls consist of quite short notes, with a very wide frequency range. The call components are difficult to compare with calls of other species in which the components are very distinct. I consider the dominant frequency the darkest, widest portion of the sonagram of the call, as substantiated in sections, measuring the call duration at that point (Fig. lOA). Any other noticeably prominent frequency below the dominant is taken as the funda- mental frequency (Fig. lOB). Any periodic emphasized frequencies above the dominant are considered harmonics, even though they may not be sharply differentiated (Fig. IOC). The duration of the call is from .04 to .09 seconds, the range .05 to .08 seconds being common in the call of a single individual. Call duration does not correlate with temperature in the samples at hand. The most striking differences among the calls are: 1) one or two notes per call group; 2) the fundamental frequency is dominant or not; 3) the dominant frequency values; 4) whether harmonics above the dominant are distinct or 32 Contributions in Science No. 191 not; 5) whether the note is straight on the paper, or components are displaced to the right. Most of the calls are characterized by having a single note per call group. Only in the state of Nayarit did individuals have calls consisting of two notes per call group. One of the calls recorded at Tepic has a call consisting of both one and two notes, as does the individual recorded from near San Bias. The specimen recorded from Santa Cruz has a call consisting of two notes. The calls recorded from near Acatlan, Jalisco, Tepic, Nayarit, and Her- mosillo, Sonora, show the fundamental to be the same as the dominant. One call from near Los Mochis, Sinaloa, has a fundamental distinct from the dominant; another call has the fundamental equal to the dominant. All other calls have fundamentals distinct from the dominant. Where fundamental fre- quencies are present, they approximate one half the frequency of the dominant. The dominant frequencies usually range from 2000 to 2500 cycles per second. A range of 2000 to 2400 cycles per second is recorded in the deme from Tepic, Nayarit. The only dominant differing from the range is from a call recorded from Santa Cruz, Nayarit, which has a dominant of 3000 cycles per second. The higher frequency can not be accounted for by differences in temperature or elevation. Three categories are used to distinguish whether harmonics above the dominant are evident: 1) not present; 2) some indication of presence; 3) present. Calls from Colima, Michoacan, and near La Huerta and Tecalitlan, Jalisco, have harmonics. The call from near Tonila, Jalisco, has some indica- tion of harmonics. The calls from near Acatlan, Jalisco, show all three states. The calls from Tepic, Nayarit, either lack harmonics, or show some indica- tions of harmonics. The calls from Santa Cruz, Nayarit, show harmonics present or absent. All calls recorded north of Santa Cruz lack harmonics above the dominant frequency. Notes with certain components displaced to the right on the sonagram are found in calls from localities that also have straight notes on the sonagram. These localities demonstrating both types of notes are near Los Mochis, Sina- loa, Santa Cruz, Nayarit, and near Acatlan, Jalisco. Two of the call characteristics are gradually clinal in nature; the relation of the fundamental to the dominant, and the presence or absence of harmonics above the dominant. Two of the call characteristics are largely variable within demes, and not constant on any geographic basis: number of notes per call group; and whether the call is displaced to the right on the sonagram or not. A call from a single deme shows a higher dominant frequency than the others. Discussion : Physical size appears to be a major, pre-mating isolating mechanism between L. melanonotus and wagneri. Additional collecting around the northern end of the Andes in Colombia is necessary to determine the interactions or lack of interactions between these two species. Character displacement is most noticeable in sympatric demes of L. podi- 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 33 cipinus and wagneri, where respective adult sizes, ventral color patterns, posterior thigh color patterns, and toe tip expansions are very distinct. Ex- pression of certain of these characters are similar in other populations of podicipinus and wagneri. Ventral color pattern in particular shows the effects of character displacement. Certain demes along the slopes of the Andes in Bolivia and southern Peru have ventral color patterns that closely approach the pattern of some podicipinus. The ventral patterns of wagneri and podici- pinus are distinct at points of sympatry; the specimens do not exhibit gradation of ventral pattern D to E. The status of the population of frogs in northwest Mexico (L. occidentalis Taylor) has been particularly uncertain. The main characters used to dis- tinguish the population are adult size and extent and color of the ventrolateral glands. The size of the population fits into a larger pattern discussed above and is not peculiar to a series of demes in northwest Mexico. The ventrolateral gland condition (B) is characteristic of the population in northwest Mexico, but the same condition is found in certain demes in Panama. The mating call data indicate that the extreme northwestern demes are differentiated from the lowland demes in Colima and Michoacan, but that the differences are con- nected by a series of demes demonstrating intermediate call types. The low- land demes in Colima and Michoacan are separated geographically from the highland demes of Jalisco and Nayarit by terrain with virtually no standing water; water is either cascading or nonexistent. The population from north- west Mexico appears to be isolated at present from the coastal demes of Colima and southward. Morphological differentiation of the northwest Mexican pop- ulation is slight, and the same differentiation is repeated in other demes of L. melanonotiis throughout its geographic range. The island demes of L. wagneri are in no way distinguishable from main- land wagneri', in fact, the presence of expanded toe disks and the adult size indicate that the demes originated from the population found in northern Venezuela and Guyana. I follow an earlier statement on the value of the subspecies concept (Savage and Heyer, 1967). I believe the methods treating variation within a species, as here presented, reveal more of the populational trends than does the arbitrary practice of designating subspecies. It is nearly impossible to fragment the three widespread species on the basis of any combination of characters. It is appropriate to interject a comment on my evaluation of the tax- onomic decisions at this point. Most of the study was based on morphology; I did not have the benefit of any biological data for most of the specimens; I regard the taxonomic decisions as conservative. With the evidence available for this study, my best judgment is that the Melanonotus species group is composed of six species. Relationships'. Four characters shared by members of the Melanonotus 34 Contributions in Science No. 191 group are useful in determining the degree of evolutionary advancement of the species. More adequate justification for determining primitive and ad- vanced character states is presented elsewhere (Heyer, 1968). The presence of toe disks is considered an advanced condition, and the presence of grooves on the upper disk surfaces is considered a further advancement. Absence of thumb spines in males, a striped or distinctly spotted pattern on the posterior thigh surface, and a ventral pattern of light spots on a dark background are considered advanced conditions. Certain trends among the species are clear. Leptodactylus melanonotus is primitive in all four characters, and is probably most like the ancestral Melanonotus group stock which gave rise to the six extant species. Lepto- dactylus discodactylus and pustulatus are the most specialized. Leptodactylus discodactylus has the most advanced disks, and the males lack thumb spines. Leptodactylus pustulatus has an advanced ventral and thigh pattern, and the males also lack thumb spines. Leptodactylus dantasi, podicipinus, and wagneri are intermediate. Leptodactylus dantasi has the advanced ventral pattern. The male is unknown for this species. Leptodactylus podicipinus and wagneri have individuals with the advanced thigh stripe. Leptodactylus podicipinus has a spotted belly and wagneri has individuals with toe disks. Based on only these four characters, a graphic representation is very difficult as evolution appears to have produced a mosaic pattern. Leptodactylus melanonotus, podicipinus, and wagneri are the most closely related. I think dantasi and pustulatus are more closely related to these three species than they are to discodactylus. Leptodactylus dantasi, pustulatus, and podicipinus resemble each other in general appearance, and the distinctive patterns of dantasi and pustulatus could have been derived from a species very much like that of podicipinus. For the present, I regard the loss of thumb spines to be a case of parallel loss in discodactylus and pustulatus. The relationships are presented as a dendrogram (Fig. 11). Distributional Analysis Although some areas of Latin America have been studied moderately intensively, the geology, climate, soil types, and vegetation of many parts of Middle and South America are known poorly or not at all. Savage (1966) has recently given an excellent synthesis on the origins and history of the Central American herpetofauna, but no such synthesis has been attempted for South America. Only the broadest generalizations can be made with respect to the distribution of the components of the Latin American herpetofauna and the following remarks must be considered as provisional. To discuss the distribution of the Melanonotus group, it is necessary to review briefly the broad geographic, paleogeographic, climatic, and vegeta- tional patterns of Latin America. Geography: The great American Cordillera forms a distributional barrier DISCODACTYLUS 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 35 Figure 11. Proposed phylogeny of members of the Melanonotus group. MELANONOTUS 36 Contributions in Science No. 191 to frogs of the Melanonotus species group; the upper elevational limit for any species is about 1800 meters. Breaks in the montane chain where faunal inter- change occurs are: the Isthmus of Tehuantepec in Mexico; the region around Lake Managua in Nicaragua; the Arenal pass in Costa Rica; and an extensive area including most of Panama and northernmost Colombia. The Andes of South America pass along the western outline of the continent, abut against the sea, or leave a very narrow lowland coastal strip to the west. In South America, there are three major upland regions in addition to the Andes. These are the Guiana shield, the central Brasilian shield, and the coastal Brasilian shield. These upland regions do not form an elevational barrier to the distribution of frogs of the Melanonotus group. The Guiana shield is somewhat continuous from eastern Colombia, extending through central and southern Venezuela, to and including almost all of the Guianas and part of Brasil just below the Guianas. The central Brasilian shield is of irregular outline bordering Bolivia on the west and extending as a band eastward along the southern edge of the Amazon basin. The coastal Brasilian shield extends from the easternmost tip of South America in the state of Rio Grande do Norte along the coast southward to the state of Sao Paulo. Paleogeography : The Cordillera of Latin America is geologically young, its present form attained in the very vigorous orogenesis from Mid to Late Tertiary, especially in Miocene and Pliocene times. Orogenesis continues in some parts of the Cordillera. (For a more detailed account and further refer- ences, see Childs and Beebe, 1963; Harrington, 1962; and Savage, 1966). The Guiana, central Brasilian, and coastal Brasilian shields date from the Early Precambrian and have probably been land positive since then (Harring- ton, 1962) . The history of the present land areas amidst the shields is poorly known, particularly since Mid-Tertiary to present, comprising, probably, the whole of the history of the Melanonotus species group. The Amazon basin has had a recurrent history of continental sediment deposition and marine embay- ment. Continental deposition has definitely occurred in the Late Tertiary, but whether any marine embayment alternated with continental deposition during or following this time is not known (Harrington, 1962) . No precise informa- tion on the isolation or continuity of the shields with relation to each other exists, and it is impossible to comment on the isolation of the ancestral stocks of the frogs of the Melanonotus species group. Climate: Comments are limited to climates affecting the distribution of members of the Melanonotus group. The coasts of Mexico and Middle Amer- ica are characterized by seasonal rainfall patterns. A very pronounced hot and dry season is found along the western Mexican coast southward to the middle of Costa Rica. Similar conditions are found along the east coast of Mexico south to Nicaragua. The Atlantic coastal area of Costa Rica is wet the year around, but a seasonality in rainfall occurs. The only place in Middle America with a nearly continuously wet tropical rainforest is the Osa peninsula region 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 37 of Pacific Costa Rica. A wet climate with no cool season and basically no dry season occurs in four restricted areas in South America. The first such area is the Choco region along the western coast of Colombia extending into north- west coastal Ecuador. The second and largest region is amidst the upper tributaries of the Amazon in southern Colombia, eastern Ecuador, most of eastern Peru, and northwestern Brasil. The third extends from coastal Guianas to the Brasilian coast just south of the mouths of the Amazon. The fourth area extends along a sometimes very narrow coastal strip of eastern Brasil from the states of Bahia south to Rio de Janeiro. All the other regions in South America occupied by the Melanonotus group have a pronounced dry season from one to nine months in duration. The driest regions in South America, also four, appear to limit the distribution of the Melanonotus group. The first runs along the western lowlands of middle Ecuador southward the length of the western lowlands of South America. The second lies along the north coast of Colombia and Venezuela, penetrating inland various distances. The third is in northeastern Brasil including the states of Maranhao, Piaui, and Ceara. The fourth is the Gran Chaco of northern Argentina, southeastern Bolivia, and southwestern Brasil (James, 1959; see also Walter and Lieth, 1960-1967). Vegetation: The vegetation of Latin America is very complex and very inadequately studied. Many different systems have been used to define and describe the vegetation of Latin America. For the present I use the broad vegetational maps found in James (1959) and Wagner (1964) as the basis for discussion. True tropical rainforests are found in the Osa peninsula region of Costa Rica; along western coastal Colombia and northwestern coastal Ecuador; along upper drainages of the Amazon characterized by high, continuous rainfall; and along the narrow coastal strip of Brasil from the states of Rio Grande do Norte to Rio de Janeiro. Tropical and subtropical rainforests which have a definite seasonal rainfall distribution are found in a continuous belt along the east coast of Mexico through Panama. Coastal patches of deciduous forest are found within the tropical and subtropical rainforests along the Carribean coast of Mexico, the tip of the Yucatan peninsula, easternmost British Hon- duras, easternmost Honduras and northeastern Nicaragua. Seasonal tropical rainforest is continuous with the tropical rainforest of the upper Amazon and covers almost all of the Amazon basin not covered by the tropical rain- forest. Tropical deciduous forests are found along the west coast of Mexico from the state of Jalisco southeast to the middle of the Pacific coast of Costa Rica; along the periphery of the Guiana shield; in eastern Brasil from the north coast of the state of Rio Grande do Norte to middle Bahia, excluding the narrow eastern coastal strip; and in the Gran Chaco. Thorn forests are found along the northern coasts of northeast Colombia and Venezuela. Savannas cover the Guianan and central Brasilian shields, and most of the coasts ad- 38 Contributions in Science No. 191 jacent to the Brasilian shields. A palm forest is found in northeast Brasil in the states of Piaui and Maranhao. Distribution of the Melanonotus Species Group: Most of the species are known from scattered localities. Leptodactylus melanonotus has been most thoroughly sampled, and moderately accurate distributional data are available for at least Mexico and Costa Rica. Records for dantasi, discodactylus, and pustulatus are very spotty, and more distributional data are needed to define the geographic limits of the species. Leptodactylus podicipinus has been moder- ately well collected in Argentina, but inadequate distributional data are avail- able for the rest of its geographic range. Distributional data of wagneri are most adequate for Ecuador, parts of Colombia and Guyana. Throughout the rest of the geographic range of wagneri, records are so spotty that it is impossible to determine whether it has a broadly continuous range or if it is composed of many isolated populations. The limited information gives the following patterns of occurrence : Leptodactylus melanonotus, wagneri, and podicipinus form a series of closely related, basically allopatric species ranging throughout low and mod- erate elevations of tropical Latin America. All three species are found in areas having a definite dry season. Judged by collections, melanonotus is not common along the continuously wet lowlands of Colombia and Ecuador. Only wagneri is commonly associated with the tropical rainforests of eastern Colombia, Ecuador, and Peru. The foam nest formed by these species appears to be an adaptation to a definite dry season. In some more xeric portions of its range (southern Sonora and northern Sinaloa, Mexico), melanonotus is found only in association with permanent bodies of water, such as persisting springs or irrigation water. The northern- most locality for melanonotus on the west coast of Mexico is a dam seepage by Hermosillo, Sonora. In more mesic areas, it is often associated with tem- porary bodies of water, such as rain filled ditches along roadsides. Leptodactylus melanonotus, podicipinus, and wagneri are found in re- gions with old land masses in South America: melanonotus with an old land mass in western Colombia and Ecuador (land positive at least during Late Mesozoic and Tertiary time, Harrington, 1962:1780) ; podicipinus with the Brasilian shields; and wagneri with the Guiana shield. If one assumes that the three species had a common, widespread ancestor, isolation of the land masses in the Cenozoic could account for the fragmentation of the stock that gave rise to melanonotus, podicipinus, and wagneri. Individuals of melanonotus in the region from the Isthmus of Tehuante- pec to the Arenal Pass of Costa Rica are characterized by having a larger size, a more variable ventral pattern, and a greater ventrolateral gland extent than individuals to the north or south. Such character gradients suggest that the major dispersal point of melanonotus has been from this land bloc in Middle America. Secondary dispersal in Middle America by a species which had its 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 39 evolutionary origin in South America has also been hypothesized for the hylid frog Phrynohyas venulosa by McDiarmid (1968). Dispersal into Middle America by a species from western South America is consistent with Savage’s (1966) scheme of historical herpetofaunal dispersal. He suggested that an assemblage of species, termed the Western South American Herpetofaunal Complex, evolved during the separation of nuclear Middle America from South America from Eocene through Miocene. When Middle America and South America were joined again at the beginning of the Pliocene, members of the Western South American Complex invaded Middle America. Leptodactylus wagneri today is found in a variety of climates and associated vegetation types, but is found consistently in more mesic situations than either melanonotus or podicipinus. Throughout most of its range, wag- neri is found in tropical or subtropical rainforest with either a short or no dry season. An exception occurs in the northern part of the range, particularly along the extreme north coasts of Colombia and Venezuela. Leptodactylus wagneri is not recorded from the palm forest or tropical deciduous forest of eastern Brasil in the states of Maranhao through Rio Grande do Norte. Rec- ords are so scattered for eastern Brasil that it is hard to tell whether its absence in any region is a collecting artifact. The islands of Trinidad and Tobago are on the continental shelf and were probably connected to the mainland during glacial maxima. The presence of wagneri on these islands is to be expected. The islands of Grenada, Bequia, and St. Vincent are not on the continental shelf; they probably have been always isolated from the continent of South America. Grenada and Bequia are on the same bank. Currently, Bequia lies only 10 km from St. Vincent and possibly was separated by no more than 5 km of water during glacial maxima. The distance from the 100 fathom contour outside of Tobago to the 100 fathom line surrounding Grenada is at most 100 km, and these two points are directly connected by westerly surface ocean currents. All island populations are indistinguishable from mainland populations of northeast Venezuela, suggest- ing either that dispersal to the Lesser Antilles has been a rather recent event or that Lesser Antillean populations have been populated repeatedly from the mainland stock. A single dispersion of wagneri from South America (in- cluding Trinidad and Tobago) could account for all of the populations on the Lesser Antilles. Other members of the genus Leptodactylus are more success- ful island colonizers, especially mystaceus which has apparently hopped from island to island along the Lesser Antilles to as far as Haiti. As judged from the similarity of the island and mainland populations, it is reasoned that wagneri dispersed to the Lesser Antilles during the glacial maxima of the Pleistocene. Leptodactylus podicipinus is associated with the southern river drainages and the Sao Lrancisco drainage system. All of the regions, such as the Gran Chaco, where podicipinus has been found have a very distinct dry season. L. podicipinus is generally adapted to more xeric conditions than wagneri. 40 Contributions in Science No. 191 This may be seen where the two species occur close together, most notably with wagneri along the thin mesic coastal strip of east Brasil, and podicipinus close by in the dry parts of the Sao Francisco drainage system. Leptodactylus dantasi, discodactylus, and pustulatus are known from spotty distributional records, which suggest a distribution associated with climatic and vegetational factors. Leptodactylus dantasi has been taken only from the tropical rainforest in western Brasil. Leptodactylus discodactylus is apparently limited to the tropical rainforest of southeastern Colombia, eastern Ecuador, most of eastern Peru, and western Brasil. Leptodactylus pustulatus is apparently limited to the more xeric conditions in the palm forest, tropical deciduous forest, and savannas of eastern Brasil. The total distribution of members of the Melanonotus species group indicates that three species have broad ecologic tolerances and are distributed in extensive geographic regions and that three of the species have much narrower ecologic tolerances and have restricted geographic distributions. Specimens Examined Leptodactylus dantasi (N = 1): BRASIL. TERRITORIO DO ACRE: Feijo, 249 m — WCAB 1240 (Holotype). Leptodactylus discodactylus (N = 15): ECUADOR. NAPO: Payamino, 304 m— GOV 7238; PASTAZA: Puyo, 1000 m— GOV 2062, JAP 6197-8; 6333; Veracruz, 1000 m—JAP 6151, 6153-6, 6164. PERU. LORETO: Es- tiron, Rio Ampiacu, 200 m — CAS 93316, 7; Rio Itaya, 150 m — AMNH 43217; Rio Tamaya, Sobral, 150 m— AMNH 42680. Leptodactylus melanonotus (N = 520) : BRITISH HONDURAS. STANN CREEK: Mango Creek, 120 m— MCZ 37884; 4.8 km S Waha Leaf Creek, 120 m— MCZ 37875-6. COLOMBIA. CHOCO: Rio San Juan, Playa de Oro, 400 m — FMNH 54374; NARINO: La Guayacana, 230 m — PAS 44; Rio Mataje, 200 m — USNM 147393; Valle de Cauca: Rio Raposo Vir- ology Field Station, 20 m— USNM 151463. COSTA RICA. Alajuela: Cari- blanco, 800 m — CRE 2928 (10); 0.5 km S Cariblanco, 820 m— CRE 694; Los Chiles, 70 m— AMNH 54638, 54641-3, CRE 7217, 7219; Punta Cortez, 70 m— CRE 7218; CARTAGO: Cervantes, 1441 m— KU 28189; IICA, Tur- rialba, 600 m— CRE 575, 2808 (2), 7074 (2), 7079 (7); GUANACASTE: Arenal, 520 m— CRE 6251, 6258 (4), 6259, 6260 (2) ; 4.8 km NW Arenal— CRE 2901 (3) ; Bebedero, 6 m— CRE 7162; Canas, 88 m— CRE 7164; 9.6 km S La Cruz — CRE 8091 ; Rio Lagarto at Inter. Am. Hiway, 100 m—CRE 7122 (2) ; 2 km W Liberia — CRE 728 (4) ; 9 km N and 4 km E Liberia on Inter. Am. Hiway on Rio Colorado — CRE 714 (2); 14.8 km S Liberia, 90 m — CRE 8168; 45 km SW Liberia, Hwy #21— CRE 8216; Hacienda La Norma, 5 km N and 4.5 km W Liberia on Rio Colorado, 150 m—CRE 105 (2), 106 (6), 107, 251 (2); 3-11 km E Playa del Coco, 45 m—CRE 8012 (2); Finca San Bosco de Tilaran, 640 m — CRE 6273, 6278; 3.2 km W Santa Cruz on the 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 41 Playa del Tamarindo rd. — CRE 8233; Santa Cruz, 50 m — CRE 7128, 8232 (3); Silencio de Tilaran, 825-850 m— CRE 6217 (8), 6225 (2), 6228 (2); 0.5 km NW Tilaran, 530 m— CRE 520 (7); 3 km NE Tilaran— CRE 524; 5 km NE Tilaran, 600 m — CRE 8021; 5.5 km NE Tilaran, 560 m — CRE 8022; 6 km NE Tilaran, 550 m — CRE 523 (3), 8020 (9) ; Finca Taboga, 9.6 km S and 8.0 km W Canas, 4 m — CRE 2902; HEREDIA: Puerto Viejo, 100 m — KU 65683; 6.5 km W Puerto Viejo— KU 91798; 7.5 km W Puerto Viejo— KU 86293; LIMON: Batan, 15 m — KU 34146-51 ; 2.4 km E Los Diamantes, 260 m— CRE 8049; El Tigre, 680 m— CRE 290 (2); PUNTARENAS: 9.6 km ESE Golfito, 10 m — CRE 7105; junction of road and rail line (Base of Peninsula), 5 m— CRE 238 (2), 253-4. ECUADOR. ESMERALDAS: San Javier, 200 m— JAP 9077-82; Cacnaui, 20 m— JAP 2889-9; Rio Bogota, 200 m— GOV 6398; LOS RIOS: Vinces, 150 m— UMMZ 55587 (7). EL SAL- VADOR. SAN SALVADOR: San Salvador, 620 m— FMNH 65071-86. GUATEMALA. EL PETEN : N of La Libertad— UMMZ 75367 (12); 5.2 km S La Libertad — MCZ 21454; Toocog, 15 km SE La Libertad, 170-200 m — KU 58959; ESCUINTLA: Rio Guacalate, nr. Masagua, 100 m— USNM 125246-51, 125253, 125261-3; ZACAPA: 8 km ENE Rio Hondo, 175 m— KU 58960; 23 km NE Zacapa— ANSP 22165-6; 23 km W Zacapa,.240 m— TCWC 16403, 16406. HONDURAS. ATLANTIDA: Corozal, 15 km E La Ceiba, 10 m— JRM 2316, 2328-33, 2335, 2337-8, 2341, 2343; CHOLU- TECA: 1.5 km NW Choluteca, 170 m— KU 65676; 5 km S Choluteca, 80 m —MCZ 26460-2; CORTES: Agua Azul, Laguna Yogoa, 700 m— AMNH 54778; FRANCISCO-MORAZAN: 31 km S Guaimaca, Santa Clara, 1000 m— AMNH 54884-5, 54887; ISLAS DE LA BAHIA: Isla de Guanaja, SE shore opposite Guanaja, 10 m — JRM 1938, 1961-6; Isla de Roatan, .5-4 km NRoatan, 10-30 m— JRM 1828-9, 1896, 1916; OLANCHO: Escuela Nacional de Agricultura, 4.5 km SE Catacamas, 400 m — JRM 1660-2. MEXICO. CAM- PECHE: Chuina, 46 km S Champoton, 50 m— KU 75017-26; CHIAPAS: 4 km NW La Esperanza, 100-300 m— USNM 114372-6, 114381-2, 114385, 114388; Tonala, 55 m— USNM 114338-44; Tuxtla Gutierrez— UMMZ 115530; 9.8 km W Tuxtla Gutierrez, 520 m— UMMZ 119147 (2); COLIMA: 25.9 km N Manzanillo, 10 m— CAS 97111-8, 97155-7; GUERRERO: 12.9 km SW Tierra Colorada, 300 m— TCWC 8323-6, 8333-9; JALISCO: 5.2 km W Acatlan, 1400 m— LACM 37381-408; MICHOACAN: Apatzingan, 540 m— FMNH 38835, 38837, 38853-5, 38859-60, 38862, 38865-6, 38873; NA- YARIT: San Bias, 3 m— UMMZ 110892 (7), 115543 (3); Tepic, 1000 m— LACM 37042-3; OAXACA: Cacahuatepec at river, 150 m— UIMNH 52854- 8; 8.3 km N Pochutla, 100 m— UMMZ 123982 (8) ; 5.2 km N Matias Romero, 150 m— UMMZ 115522 (3); 7.8 km S Matias Romero, 150 m— UMMZ 113804 (4); Tehuantepec, 35 m— FMNH 123248-50, 123253, 123256, 123264, 123267-9; Tolosa, 50 m— AMNH 53610; QUINTANA ROO: 3 km E Caobas, 5-10 m— KU 75033; 4 km WSW Puerto Juarez, 5 m— KU 71117- 42 Contributions in Science No. 191 8; SAN LUIS POTOSf: 5.1 km NE Tamazunchale, 150 m— TCWC 7285-91; 5.1 km SW Tamazunchale, 150 m— UIMNH 16802-3; SINALOA; Culiacan, 45 m— AMNH 58382-91; 18.9 km SW Mazatlan, 48 m— KU 78259-61; 5.2 km W Villa Union— KU 29901-7; SONORA; Alamos, 300 m— AMNH 51356-65; Hermosillo, 28 m— AMNH 62737, MVZ 26066; TABASCO; Teapa, 40 m— UMMZ 119148 (2); 20.7 km N Teapa— UMMZ 119901 (2); 33.7 km N Teapa— UMMZ 119903 (6); 57 km N Villahermosa, 30 m— AMNH 69007-8; TAMAULIPAS; Pano Ayuctle, 12.9 km NE Gomez Farias, 100 m— UMMZ 98947 (2), 101185 (5), 102263, 102912; VERACRUZ; 5 km SW Boca del Rio— KU 23700; Cuatotolapam, 13 m— UMMZ 41640-3, 41646-7, 41649; 1 km E Mecayucan, 30 m— KU 24084, 24086, 24091-4; 25.9 km WNW Temapache, 50 m— KU 60225-8; YUCATAN; 31 km N Tizimin, 5 m — CM 40108-10; 16.2 km S Yaxcopoil, Aguada Xcamal, 5 m — - CM 45233-43; Yuncu, 25-50 m— UMMZ 73155, 73158 (2). NICARAGUA, MANAGUA; 15.5 km NW Managua, 55 m— KU 43030-39; RIVAS; San Jorge, Shore of Lake Nicaragua, 50 m — MCZ 28980-89; ZELAYA; Bonanza, 260 m— KU 84839-41, 84843-48; 10.4 km SW Bonanza, 300 m— KU 84842; 25.9 km above Recreo, Rio Mico, 30 m— UMMZ 79734, 79735 (4). PAN- AMA. CHIRIQUI; Las Lagunas, ± 6 km W El Volcan, ± 1200 m— KU 76522; Finca Palosanto, 6 km WNW El Volcan, 1230 m — KU 76520-1; COCLE; El Valle, 560 m— KU 76523-32; DARIEN; Rio Canclon, 60 m— UMMZ 125008, 125009 (10). UNITED STATES, TEXAS; Starr County, 5.1 km SE Rio Grande City— UIMNH 29817. Leptodactylus podicipinus (N = 154) ; ARGENTINA. CHACO; Resis- tencia, 60 m— KU 84732-4; CORRIENTES; 30 km W Itati, 60 m— CAS 100504; Manantiales, 50 m — MCZ 35589-90; FORMOSA; Esteros Laguna Oca, 65 m — MCZ 32775-6; SANTA FE; Banados Monteira, 15 km N Santa Fe, 20 m — CM 38011; Banados del Rincon, 20 m — CM 37981-3, 38003, 38005; Santa Fe, 18 m— CM 39181, 39194. BOLIVIA. BENI; Ivon, 150-200 m— UMMZ 64099 (7) ; Puerto Almacen, 260 m— AMNH 72245-53; SANTA CRUZ; Buenavista, 500 m— AMNH 39538; El Carmen, 212 m— CM 36234- 7; Naranjal, 5 km from Montero, 350 m — USNM 146551 ; El Porton, 550 m —CM 36248; Robore, 330 m— MCZ 30032-6; 3 km S Robore, 330 m— CM 36238-41; San Jose de Chiquitos, 350 m— CM 36164, MCZ 30028-9. BRA- SIL. BAHIA; Barreiras, 500 m — UMMZ 109986; Rio Grande at Sao Jose, 450 m— UMMZ 109989; MATO GROSSO; Agua Clara, 304 m— FMNH 67088; Conceicao, Paraguay River, 150 m — FMNH 9161; Corumba, 115 m —CM 2429, 36165, UMMZ 104225 (2); 40 km from Corumba, Urucum Mts., 200-300 m— CM 2430; Descalvados, 142 m— FMNH 9097, 9100-14; 1 day’s run below Descalvados, 150 m — USNM 132736-7; Maracaju, 385 m —USNM 107705-8, 107710-11; Porto Esperan9a, 120 m— UMMZ 104232 (11), USNM 133005-8; Salobra, 190 m— UMMZ 104227 (2); MINAS GERAIS; Januaria, 454 m— UMMZ 109984; Pirapora, 472 m— UMMZ 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 43 109982 (10), USNM 98535, 99801-3; PARANA: Andira, 480 m— MCZ 32725-6; RONDONIA: Forte Principe da Beira, 100-200 m— WCAB 8330, 8982; Porto Velho, 60 m— KU 92933-37; SAG PAULO: Anhembi, 500 m— WCAB 30880-1; Bauru, 499 m — DZ 53-4; Cachoeira de Emas, Pirassununga, 650 m — DZ 4636-42; Lins, 396 m — DZ 9024; Porto Marcondes, Parana- panema, 250 m— =DZ 19979, 19999 (10 total). PARAGUAY. CAAGUAZJU: Pastorea nr. Caaguazu, 250 m — MCZ 17902-11; CORDILLERA: Villeta, 110 m— AMNH 50654; PRESIDENTS HAYES: Rio Pilcomayo, 38.8 km W Rio Paraguay, 80 m--»MCZ 25827-33. Leptodactylus pustulatus (N = 7) : BRASIL. GOIAS: Aruana, 200 m— DZ 4992, 8109; MATO GROSSO: Sao Domingoes, Rio das Mortes, 200 m -^DZ 986-7, 996, 1069; PIAUf: Rio Poti^MCZ 373. Leptodactylus wagneri (N = 711): BOLIVIA. BENI: Cachuela Es- peranza, 150 m — UMMZ 64102 (2); nr. Guauara-mirim, Rio Marmori, 160 m — CM 2666; Swamp along Rio Guapore, San Antonio de Guapore, 150 m—USNM 115973; Huachi, 250 m"'— UMMZ 64103 (3), 64104; Puerto Almacen, 260 m— AMNH 72251, 72254, 72398-402; Lake Rogoagua, 200 m-^UMMZ 64100 (7); Rurrenabaque, 227 m— UMMZ 64097 (3); Villa Bella, 120 m~CM 2659; COCHABAMBA: Puerto Chipiriri, 300 m“”AMNH 72242-3; PANDO: in front of Placido de Castro, Acre, Brasil—DZ 6529-31; SANTA CRUZ: Buenavista, 500 m— CM 4331, 4353, UMMZ 64028-29, 64030 (2), USNM 118688-9, 146523-4. BRASIL. ALAGOAS: Fazenda Canoas, Rio Largo, 60 m™— DZ 9279; Mangabeiras, 50 m--DZ 11982-3; Sao Miguel, 20 m---DZ 9250; AMAZONAS: Manaca- puro, 50 m—USNM 103621-2; BAHIA: Salvador, 15 m— DZ 9131, 9550-1; GOIAS: km 47 Estrada Goias- Aruana, 350 m-^DZ 1518; MARANHAO: Carolina, 168 m— DZ 21664-5; MATO GROSSO: Dumba, 200 m--DZ 1448; Banks of center of Rio Guapore, 250 m — CM 2508-9; Confluence of Rio Araguaia and Rio Tapirape, 190 m- — AMNH 68093; Sao Domingoes, Rio das Mortes, 200 m— DZ 10891-2, 10894-900, USNM 148677-91; PARA: Belem, 20 m-~-FMNH 83264-5, MCZ 36010-1, USNM 154068; Cachimbo, 500-600 m — DZ 21656-61; Rio Mapuera at the Equator, 80 m — AMNH 46181; San- tarem, betwn Rio Tapajos and Rio Amazonas, 16 m — CM 2641; RIO DE JANEIRO: km 31 on Estrada Mage-Estrada Rio Petropolis, 100 m — DZ 12034-5; RIO GRANDE DO NORTE: Natal, 120 m— AMNH 36261, MCZ 15847, UMMZ 68791 ; TERRITORIO DO ACRE: Iquiri, 150 m— DZ 6588- 92; TERRITORIO DO RIO BRANCO: Rio Urariquera, 100-200 m— WCAB 18792; TERRITORIO DO RONDONIA: Forte Principe da Beira, 100-200 m~-^WCAB 10690, 10703. COLOMBIA. AMAZONAS: Falls of Engano— MCZ 28063; Leticia, 275 m— CM 36081-2, USNM 142179-86, 142187-94, 146252, 147031; ANTIOQUIA: Envigado, 1600 m— AMNH 39265, 39274, 39281; Medellin, 1500 m—AMNH 38785, 38808-15, 39465; Sonson, 2500 m-^AMNH 39631; BOYACA: Muzo, 1240 m— FMNH 69744, MCZ 24920; 44 Contributions in Science No. 191 CAUCA: Popayan, 1700 m— FMNH 54377-83; CUNDINAMARCA: Ano- laima, 1500 m— USNM 147054-5; Bogota, 2600 m~AMNH 13471; Fusa- gasuga, 1700 m— AMNH 71581; La Mesa, 1000-1300 m-— USNM 144892-3; META: Buenavista, 1200 m — MVZ 63067-73; Menegua, upper Rio Meta, 260 m— USNM 147273-4; 46 km S, 22 km W San Martin, 490 m— -MVZ 63076; Villavicencio, 500 m— ANSP 25765, FMNH 30571, 81788-9, MCZ 16277, MVZ 63074-5, 63738-9, UMMZ 74811, USNM 144848-66, 147056- 62; 7 km NE Villavicencio, 500-700 m— USNM 146382-4; NARINO: La Guayacana, 100 m— FMNH 61754; NORTE DE SANTANDER: Astillero, 100-200 m— USNM 147063; E of Cucuta 200-500 m— ANSP 25758; PUTU- MAYO: Rio Putumayo, 5 km N Puerto Asis, 260 m — PAS 47; Rumiyaco, 1000 m— FMNH 54376; SANTANDER: nr. Lebrija, 1086 m— USNM 144883-8, 146250; San Gil, 1095 m— CM 7938, UMMZ 74799; TOLIMA: Guindio Mtns.— MCZ 8217; VALLE DE CAUCA: Buenaventura— USNM 147076; 2 km S Cali, 950 m— USNM 148800-23; VICHADA: Puerto Car- reno— CM 17122. ECUADOR. NAPO: Lagarto Cocha, 200 m— GOV 9208; Loreto, 400 m — GOV 9205, WCAB 36541 ; Rio Napo, Avila, 300 m — UMMZ 92145-6; PASTAZA: Upper Bobonaza— JAP 8732; Canelos, 530 m —MCZ 17950; Cerros de Abitagua, 1100 m— FMNH 25789, 26899, UMMZ 92147; Chichirota, 300 m— GOV 9207; Upper Rio Curaray — GOV 9210; Upper Rio Oglan, 500 m— GOV 9206; Rio Pindo, 190 m— GOV 9195-7; Puyo, 940 m— JAP 1958-71, 1973-2005, 2036, 2038, 6182, 6186-7, 6191-5, 6289-99, 6303-5, 6329-32, 6334, GOV 8021-2; Shell Mira, 1064 m— GOV 8503, JAP 2177; Don Tomas, 340 m— GOV 9209; Rio Villano, 300-400 m— GOV 7731-6, JAP 3776; ZAMORA-CHINCHIPE: Copal, 910 m— JAP 6747-8; Limon, 1550 m— JAP 6937, 6964-5; Rio Llushin— GOV 9198-201, 9203, JAP 3751; Mendez, 580 m— JAP 2037, 6805, 6870, 6872-5, 6880, 6832-4, 6837, 6845-6, 6855-6, 6936; Plan de Milagro, 1700 m— JAP 7008, 7014-6, 7018-23, 7027, 7057; Plan Grande, 880 m— JAP 6903; NE of Rio- bamba— FMNH 23509 (8) ; Sucua— JAP 2186, 2225. FRENCH GUIANA. Sophie, on trail leading S from St. Elie Gold Pits, 200 m — MCZ 44564. GUYANA. DEMARARA: Atkinson, McKenzie Trail, 50 m— USNM 162872-3; Dunoon, Demerara Riv., 20 m — UMMZ 50183-9; Georgetown, 5 m— UMMZ 80497; Wismar, 50 m— AMNH 45750, UMMZ 77517, 80417; ESSEQUIBO: Isheraton, 250 m— AMNH 53435-7; Karanambo, 100 m— AMNH 53438; Kartabo, 10 m— AMNH 10377, 10379-83, 11659, 13518-9, 25233, CM 5443-4, USNM 118059-62; Kuyuwini Landing, 250 m— AMNH 46281, 49352, 49355 (5); Lethem, 100 m— MCZ 50708; Marudi, 250 m— AMNH 46238 (2); Moku Moku, 100-200 m— USNM 146366-7; Oko Mts., 115 m — FMNH 26691 ; Santa Rosa Is., Monica Riv., 7 m — UMMZ 55833 (2), 80417; Shudikarwan, 200-300 m— AMNH 46274 (9). PERU. AYA- CUCHO: Candalosa, ± 1400 m— FMNH 39727-32, 39734-46; La Mar, ± 760 m— FMNH 39725; La Mar, Ayna ± 1900 m— FMNH 39726; HUAN- 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 45 UCO; Ganzo Azul, 150 m=-~FMNH 45144; JUNIN: Satipo, 500 m~^MCZ 24424-8, UMMZ 89476 (7); Tarma, Palmapata, 1200-1400 m-^FMNH 36827-30; Tarma, Valley of Vitoc River, 1200-1400 m—FMNH 36817-26; LORETO: Rio Maranon, mo. of Rio Pastaza, 150 m- — AMNH 42221-3, 42225-7; Pampa Hermosa, 176 m— AMNH 42142-4, 42661; Roaboya, 100 m----AMNH 42823-4; Rio Tamaya, Sobral, 150 m-— AMNH 43238; Rio Alto Tapiche, mo. R. Contaya, 200 m— AMNH 42988; Tibi Playa, 150 m— - AMNH 42783; Rio Ucayali, Pacapa, 150 m---FMNH 56322; Upper Uto- quina, 200 m — ^AMNH 42599; PASCO: Iscozazin Valley, Chontiila, 780 m — LACM 40660-5; Rio Pachitea, Monte Alegra, 150 m — AMNH 43013; SAN MARTIN: Tocache, 450-750 m-™AMNH 42627. SURINAM. MARO- WIJNE: Kaiserberg airstrip, Zuid R., 200-300 m— FMNH 128925, 128927- 8, 128933-7, 128939-40. VENEZUELA. AMAZONAS. Rio Pescado— AMNH 23182; ARAGUA: Colonia Tuvar, 1830 m — CM 7729; Cumboto, 150 m— UMMZ 113975 (2) ; Rancho Grande, nr. Maracay, 450 m — AMNH 70687; BOLIVAR: Arabopo, 1280 m-— UMMZ 85197 (5); Mt. Roraima, Paulo, 1216 m^AMNH 39753, 39758-9; DELTA AMACURO: 140 km. NE Barrancas, 5-10 m^LACM 31382-97; DISTRITO FEDERAL; Cotiza, Camino de Calipan, 900 m-— USNM 117526, 128837; El Limon, 577 m~ CM 7676-7, 7697, 7699, USNM 121146; FALCON: Cerro Cosme, 100-200 m— MCZ 26144-6; Lagoon, Palma Sola, 50 m — -UMMZ 55551; MERIDA: Merida, 1641 m—MCZ 2640, USNM 118176; Rio Albirregas nr. Merida, 1640 m-^AMNH 10517-9; MIRANDA: Petare, 885 m^CM 7769-71, 7791; USNM 121147; MONAGAS: Caripito, dz 50 m-™AMNH 70669-86, USNM 117088-9; SUCRE: Cumanocoa, 300 m-^CM 9055-76; Latal, Hacienda Mirasol, 975 m— --CM 9099, 9101; San Rafael, nr. Cumanocoa, 200-400 m-- CM 8972; El Yaque, 800 m— CM 9108-9; TaCHIRA: La Fria, Rt. Fork Rio Oropito, 110 m— -UMMZ 55552-3; ZULIA: Chama, 10 m — -AMNH 10685-8. WEST INDIES: BEQUIA: north end~-USNM 103976-8; GRE- NADA: Grand Etang-=-MCZ 2963-71, 2974-6; no specific locality— -AMNH 18961, 18970-2, 18992, USNM 67183-4, 67186-94; ST. VINCENT: Botani- cal Garden-^USNM 144246; Brighton^USNM 79068-75; Mt. St. Andrews “—USNM 79076-7; TOBAGO: Bacolet River“-AMNH 55873; Buccoo Bay ^AMNH 55863-5; nr. Speyside, 180 m-=-=MCZ 27788-9; TRINIDAD: Nariva: Brickfield^FMNH 49656-60; Upper Tucker Valley-— USNM 119055-60; St. Andrew: Mount Harris—FMNH 49654-5; Sangre Grande— MCZ 3295-8; San Rafael—FMNH 49664-5; St. George: St. Augustine-— -CM 5466. Resumen En el grupo Melanonotus se reconocen las siguentes seis especies: dantasi, discodactylus, melanonotus, podicipinus, pustuiatus, y wagneri. Para cada especie se ofrecen: sinonimia, diagnostico, sumario de carac- 46 Contributions in Science No. 191 teristicas y distribucion. Series de Leptodactylus procedentes de amplias areas geograficas y ecologicas fueron analizadas para estudiar largo normal, apa- riencia ventral, desarrollo de las glandulas ventrolaterales, apariencia de las extremidades posteriores, y desarrollo de los discos digitales. Se encontraron correlaciones evidentes entre: tamano grande y habitates humedos en wagneri, 2) apariencia ventral mas oscura y una cantidad mayor de Iluvia anual en melanonotus, 3) apariencia ventral palida y una mayor extension de las glandulas ventrolaterales en wagneri, y 4) mayor extension de las glandulas ventrolaterales y mayor cantidad anual de Iluvia en wagneri. Se postula la hipotesis de que existe displazamiento de caracteres en la estricta diferencia en tamano y apariencia tanto entre poblaciones de mela- nonotus y wagneri como entre podocipinus y wagneri las cuales viven sim- patricamente. Leptodactylus melanonotus es la especie mas primitiva, mientras que discodactylus y pustulatus son mas avanzadas. Leptodactylus melanonotus, podicipinus, y wagneri estan asociadas con regiones antiguas y son especies adaptadas a condiciones xericas. Leptodacty- lus dantasi y discodactylus estan limitados al bosque tropical humedo del valle al oeste del Amazonas. Leptodactylus pustulatus se encuentra en las regiones orientales xericas de Brasil. Literature Cited Andersson, L. G. 1945. Batrachians from East Ecuador collected 1937, 1938 by Wm. Clarke-Macintyre and Rolf Biomberg. Arkiv f. Zook, Stockholm 37(A): 1-88. Arkin, H., and R. R. Colton. 1966. Statistical methods. Coll. Outline Ser. Barnes and Noble, New York. 273 p. Bokermann, W. C. a. 1959. Una nueva especie de Leptodactylus da la region Amazonica (Amphibia, Salientia, Leptodactylidae) . Neotropica 5(16): 5-8. Boulenger, G. a. 1882. Catalogue of the Batrachia Salientia s. Ecuadata in the collection of the British Museum. 2nd ed. British Museum, London. 503 p. . 1883. On a collection of frogs from Yurimaguas, Huallaga River, Northern Peru. Proc. Zook Soc., London, 1883:635-638. Brocchi, M. P. 1877. Sur quelques Batraciens Raniformes et Bufoniformes de I’Amerique Centrale. Bull. Soc. Philom. Paris. Ser. 7(1): 175-197. . 1881. Etude des Batraciens de I’Amerique Centrale. Miss. Sci. Mex. et Centr. Amer. Part 3, Sect. 2: 1-122. Childs, O. E., and B. W. Beebe, ed. 1963. Backbone of the Americas. Tectonic history from pole to pole. Amer. Ass. Petr. Geok, Mem. 2. 320 p. Cochran, D. M. 1961. Type specimens of reptiles and amphibians in the U.S. Na- tional Museum. U.S. Nat. Mus. Bulk 220: 1-291. Cope, E. D. 1862. On some new and little known American Anura. Proc. Acad. Nat. Sci. Phila., 1862:151-159. . 1879. Eleventh contribution to the herpetology of tropical America. Proc. Amer. Phil. Soc. 18:261-277. . 1887. Synopsis of the Batrachia and Reptilia obtained by H. H. Smith in the Province of Mato Grosso, Brazil. Proc. Amer. Phil. Soc. 24:44-60. 1970 Studies on the Frogs of the Genus LEPTODACTYLUS 47 Elias, H., and J, Shapiro. 1957. Histology of the skin of some toads and frogs. Amer. Mus. Novitates 1819:1-27. Espinal T., L. S. and E. Montenegro M. 1963. Formaciones Vegetales de Colom- bia. Inst. Geogr. “Agustin Codazzi,” Dep. Agrologico. 201 p. Fisher, R. A. 1948. Statistical methods for research workers. Biol. Monogr. and Manuals. Oliver and Boyd, London. 354 p. Fitzinger, L. 1826. Neue Classification der Reptilien nach ihren natiirlichen Ver- vandtschaften nebst einer Verwandtschafts — tafel und einem Verzeichnisse der Reptilien. Sammiung des k. k. Zool. Mus. zu Wien., 1-66. Cans, C. 1960. Notes on a herpetological collecting trip through the southeastern lowlands of Bolivia. Ann. Carnegie Mus. 35:283-314. Garman, S. 1887. On West-Indian Reptiles and Batrachians in the Museum of Com- parative Zoology at Cambridge, Mass. Bull. Essex Inst. 19:1-53. Gorham, S. W. 1966. Liste der rezenten Amphibien und Reptilien. Ascaphidae, Leiopelmatidae, Pipidae, Discoglossidae, Pelobatidae, Leptodactylidae, Rhino- phrynidae. Das Tierreich 85:1-222. Hallowell, E. 1860. Report upon the Reptilia of the North Pacific Exploring Ex- pedition, under command of Capt. John Rogers, U. S. N. Proc. Acad. Nat. Sci. Phila., 1860: 480-510. Harrington, H. J. 1962. Paleogeographic development of South America. Bull. Amer. Ass. Petr. Geol. 46( 10) : 1773-1814. Heyer, W. R. 1968. Biosystematic studies on the frog genus Leptodactylus. Ph.D. Dissertation, Univ. Southern Calif. 234 p. . in press. Studies on the frogs of the genus Leptodactylus (Amphibia, Lepto- dactylidae). II. Diagnosis and distribution of the Leptodactylus of Costa Rica. Rev. de Biol. Trop. Holdridge, L. R. 1964. Life zone ecology. Trop. Sci. Cent., San Jose. 124 p. James, P. E. 1959. Latin America. 3rd ed. The Odyssey Press, New York. 942 p. Lutz, A. 1926. Sur deux especes nouvelles de Batraciens bresiliens. Compt. Rend, soc. Biol. Paris 95(29) : 101 1-1012. -. 1930. Second paper on brasilian and some closely related species of the genus Leptodactylus. Mem. do Inst. Oswaldo Cruz 23(l):21-34. McDiarmid, R. W. 1968. Populational variation in the frog genus Phrynohyas Fitzinger in Middle America. Los Angeles County Mus. Nat. Hist., Contrib. in Sci. 134:1-25. Nieden, F. 1923. Anura I. Subordo Aglossa und Phaneroglossa, Sectio I Arcifera. Das Tierreich 46: 1-584. Peters, J. A. 1964. Dictionary of herpetology. Hafner Publishing Co., New York. 392 p. Peters, W. 1862. Fine neue Gattung von Laubfroschen, Plectromantis, aus Ecuador vor. Monats. k. p. Akad. Wiss. Berlin, 1862:232-233. . 1870. Mittheilung fiber neue Amphibien des Koniglich zoologischen Mu- seums. Monats. k. p. Akad. Wiss. Berlin, 1870:641-652. Rivero, J. A. 1961. Salientia of Venezuela. Harvard Bull. Mus. Comp. Zool., 126 (1): 1-207. Savage, J. M. 1966. The origins and history of the Central American herpetofauna. Copeia 1966(4) :719-766. Savage, J. M., and W. R. Heyer. 1967. Variation and distribution in the tree-frog genus Phyllomedusa in Costa Rica, Central America. Beitr. zur neotropischen Fauna 5(2) : 1 1 1-131. Steindachner, F. 1864. Batrachologische Mittheilungen. Verb. Zool. Ges., Wien, 239-552. 48 Contributions in Science No. 191 Taylor, E. H. 1937. New species of Amphibia from Mexico. Trans. Kans. Acad. Sci. 39:349-359. Tosi, J. A., Jr. 1960. Zonas de vida natural en el Peru. Inst. Interamer. de Cienc. Agr. de la OEA Zona Andina. Bol. Tecnico 5. 271 p. Vivo Escoto. 1964. Weather and climate of Mexico and Central America, in West, R. C., editor. Handbook of Middle American Indians. I. Natural environment and early cultures. Univ. Texas Press, Austin, 570 p. Wagner, P. L. 1964. Natural vegetation of Middle America, in West, R. C., editor. Handbook of Middle American Indians. I. Natural environment and early cultures. Univ. Texas Press, Austin, 570 p. Walter, H., and H. Lieth. 1960-1967. Klimadiagramm-Weltatlas. Gustav Fischer Verlag, Jena. Accepted for publication February 17, 1970 LOS ANGELES ' COUNTY MUSEUM CONTRIBUTIONS c:lIs in science UMBER 192 June 26, 1970 SYSTEMATICS OF THE LIZARDS OF THE GEKKONID GENUS PHYLLODACTYLUS OF MAINLAND SOUTH AMERICA By James R. Dixon and Raymond B. Huey Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy, (2) Double space entire manu- script. (3 ) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Changes in the manuscript after galley proof will be billed to the author. Unless otherwise requested, page proof also will be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor SYSTEMATICS OF THE LIZARDS OF THE GEKKONID GENUS PHYLLODACTYLUS OF MAINLAND SOUTH AMERICA By James R. Dixon^ and Raymond B. Huey“ Abstract: A brief description of the habitat of species of Phyliodactyliis of mainland South America is presented. Micro- habitat preferences are given for each species where possible. Little or no microhabitat information is available for P. heter- urus, sentosus, and cUnatus. Four species, P. angiistidigitus, ger- rhopygiis, microphyUus, and sentosiis are primarily terrestrial and allopatric in distribution. Nine species, P. clinatus, dixoni, inaequalis, interandinus, johnwrighti, kofordi, lepidopygus, piimi- lus, reissi, and ventralis are scansorial and only johnwrighti, ventralis, and dixoni are not found sympatric with one or more of the other scansorial species. Where three or more species are sympatric, there is little overlap in daytime retreats and foraging areas. One egg is laid per clutch by most species of Phyliodactyliis of mainland South America except reissi, which lays two eggs. Information on clutch size is not available for F. heteriirus, sen- tosus, and clinatus. Communal egg laying appears to be common among most species, including such sympatric species as F. inter- andinus and reissi, kofordi and reissi. Species of Phyllodactylus appear to have specific adapta- tions to their microhabitat. Enlarged terminal digital lamellae, dorsoventrally compressed head and body, and lateral nostrils characterize species of Phyllodactylus from a boulder micro- habitat; small terminal digital lamellae, round body and head, dorsolateral valvular nostrils for species from a sand microhabi- tat; small terminal digital lamellae with enlarged claw, slim body and head, dorsolateral valvular nostrils for species from shell beaches. Depending upon the degree of scansorial or terrestrial behavior some species of Phyllodactylus are intermediate in the above adaptations. Distribution maps and photographs are given for each spe- cies, except F. heterurus, and a gazetteer of important localities is presented. Of the holotypes and syntypes for species described from the mainland of South America, ten (dixoni, inaequilis, lepido- pygus, magister, microphyUus, nigrofasciatus, phacophorus, pu- milus, reissi, and ventralis) are extant and eight have been exam- ined. Seven new species (angustidigitus, clinatus, interandinus, johnwrighti, kofordi, pumilus, sentosus) are described herein and nine older names (abrupteseriatus, baessleri, guayaquilensis, lobensis, magister, nigrofasciatus, phacophorus, underwoodi, variegatus) are relegated to the synonomy of other species. ^Department of Wildlife Science, Texas A&M University, College Station, Texas. ^Department of Zoology, University of Texas, Austin, Texas. 1 2 Contributions in Science No. 192 Introduction The leaf-toed geckos, genus Phyllodactylus, form two distinct geographic units in South America, a northeastern group in Colombia, Venezuela and the Caribbean Islands, and a western group in Ecuador (including the Galapagos Islands), Peru, and Chile. A study of the northeastern representatives was completed by Dixon ( 1962, 1964a). The present work treats the systematics and biology of the mainland South American Phyllodactylus. Although the Galapagos Islands species are evolutionarily close relatives to the mainland species of western South America, they are a special problem and will be reported on at a later date. The earliest description of a species for the mainland of South America was by Wiegmann (1835) for P. gerrhopygus and the most recent by Rivero- Bianco and Lancini (1968) for P. dixoni. A total of 19 species have been described, ranging from the coast and mountain areas of Ecuador to Chile, including the upper Amazon Basin of northwestern Peru in western South America; from the lowlands and foothills of eastern Colombia to central Venezuela in eastern South America. A common Mexican and Central American species, P. tuberculosus, described by Wiegmann ( 1835), has also been reported in the literature from Ecuador, Peru, the Galapagos Islands, and the Greater Antilles. The distribu- tion of this species is Mexican and Central American (Dixon, 1960), and specimens on the mainland and islands of South America to which this name has been misapplied represent populations of P. reissi, P. darwini, and P. wirshingi. Individual species were studied and described by Wiegmann (1835), Tschudi (1846), W. Peters (1862), Cope (1876, 1877), Werner (1901, 1907, 1910, 1912), and Noble (1924). Short notes on distribution, habitat and taxonomy of mainland South American species were given by Parker (1935), Taylor (1942), Codoceo ( 1950), Titschack (1951), Yanez (1951), Dixon (1962, 1964a), J. Peters (1967), Donoso-Barros ( 1966), and Rivero-Bianco and Lancini ( 1968). As a group, the mainland South American species were studied by Bou- lenger (1885) and Werner (1912), both of whom published a key to the arid and tropical American species known at the time. Burt and Burt 0933) pub- lished a preliminary checklist of lizards of South America. Wermuth (1965) published a world checklist of the genera, species, and subspecies of the fam- ily Gekkonidae, and Vanzolini ( 1968a) published a paper on the geographic distribution of South American Gekkonidae. Vanzolini (1968b) also pub- lished a key and list of Gekkonidae known to occur in Brazil, and included a list of the species of Phyllodactylus known to occur in South America. Mechler (1968) published a key, descriptions, and maps for all gekkonids in Colombia. Some of the latter studies contain errors in species determination, through the uncritical acceptance of publications of earlier workers who did not have 1970 Lizards of the Genus Phyllodactylus 3 adequate knowledge of the distribution and ecology of the species. With some of these problems in mind, a study of the genus was initiated by one of us (Dixon) in 1957. Recent collections of Phyllodactylus by Huey in 1967 and 1968, and by Dixon and Wright in 1968, from the coast and upper Amazon Basin of Peru and an examination of extant syntypes and holotypes of South American species has provided the necessary material to complete this study. A detailed study of the sympatric relations of P. kofordi, P. microphyllus, P. reissi, and F. clinatus will be presented by one of us (Huey) at a later date. Type Material Of the 19 available names for mainland South American species of Phyllodactylus, only nine have extant type material that has been examined by us: lepidopygus, phacophorus, reissi, inaequalis, microphyllus, nigrofascia- tus, magister, ventralis, and dixoni. The remaining 10 typical series are either definitely destroyed or lost. The museums in Hamburg and Dresden were seriously damaged during the fire bombings of the second world war. Among type material destroyed in the Dresden Museum were the nine syntypes of F. variegatus and the holotype of F. baessleri. The Hamburg Museum lost two syntypes each of F. abrup- seriatus and F, guayaqidlensis, and type material (if any existed) of F. lobensis (nomen nudum). The syntypes of F. gerrhopygus, reported to be housed in the Berlin Museum, are lost (Wermuth, pers. comm.). The holotype of F. heterurus, housed in the School of Medicine, University of Chile, was destroyed by a fire in the early 1950’s (Capurro, pers. comm.). Fortunately, Donoso-Barros (1947) was able to examine the type of F. heterurus in detail prior to its destruction by fire. However, some of the early type descriptions by Werner (1912) were brief and failed to give a detailed description that could be used to identify his species with certainty. There- fore, our analyses of his original descriptions and our subsequent synonomies based on these descriptions may be in error. A problem related to the loss of types is knowledge of the exact type locality for some species. Werner (1912) gives the type locality for F. abrup- terseriatiis as “probably Brazil.” Members of this genus have never been found in Brazil and we assume, as have earlier workers (e.g. Burt and Burt, 1931), that the type series probably came from Ecuador or northern Peru. The type description clearly describes individuals of Phyllodactylus reissi Peters in these areas. Tschudi’s (1846) original descriptions of F. lepidopygus and F. phacophorus gave only the type locality for the former species (Chorillos, Peru). The only statement made on the type locality of the latter species was “a wooded region.” Werner’s (1901) description of F. baessleri gave the town of Chanchamayo, Peru, as the type locality. Several towns of that name were located in west central and southern Peru. Most of these towns are sufficiently low in elevation to sustain populations of Phyllodactylus, but only two species are known to occur in the above areas, F. gerrhopygus and F. lepidopygus. 4 Contributions in Science No. 192 The only known populations fitting Werner’s description are found 1600 km NW of the nearest town named Chanchamayo. Therefore, the type locality of P. baessleri is presumed to be in error. These and other problems have been responsible for the misapplication of species names in a number of publications on Phyllodactylus over the past 100 years. Consequently, we present the following table of the chronological synonymy of sixty-one specific names proposed for North, Central, and South American species. Table 1 Chronological synonymy of sixty-one specific names proposed for the genus Phyllodactylus from North, Central and South America Date, original name and author present name 1830 pulcher Gray pulcher 1835 tuberculosiis Wiegmann tuberculosus 1835 gerrhopygus (Wiegmann) gerrhopygus 1836 gymnopygiis Dumeril & Bibron gerrhopygus 1846 lepidopygus (Tschudi) lepidopygus 1846 phacophorus (Tschudi) lepidopygus 1862 reissi Peters reissi 1862 spatiilatus Cope pulcher 1862 julieni Cope julieni 1863 iinctus Cope line t us 1863 xanti Cope xanti 1869 galapagoensis Peters galapagoensis 1875 ventralis O’Shaughnessy ventralis 1876 inaeqiialis Cope inaequalis 1876 microphyllus Cope microphyllus 1877 nigrofasciatus Cope lepidopygus 1887 martini Van Lidth de Jeude martini 1889 leei Cope leei 1892 bauri Garman bauri 1900 variegatus Werner lepidopygus 1900 baessleri Werner reissi 1903 gilberti Heller gilberti 1907 he ter unis Werner heterurus 1910 mentalis Werner xanti 1910 guayaquilensis Werner reissi 1910 lobensis Werner microphyllus 1912 abrupteseriatus Werner reissi 1912 barringtonensis Van Denburgh barringtonensis 1924 magister Noble reissi 1935 mulleri Parker ventralis 1935 homolepidurus Smith homolepidurus 1935 lanei Smith lanei 1936 delcampi Mosauer delcampi 1940 magnatuberciilatiis Taylor lanei 1970 Lizards of the Genus Phyllodactylus 5 1940 muralis Taylor muralis 1940 rutteni Hummelinck rutteni 1942 darwini Taylor darwini 1942 bordai Taylor bordai 1942 magnus Taylor tiibercidosiis 1952 eduardofischeri Mertens tubercidosus 1955 wirshingi Kerster & Smith wirshingi 1960 duellmam Dixon duellmam 1960 paucitiiberculatiis Dixon paucitiiberculatiis 1960 insidaris Dixon insidaris 1962 iinderwoodi Dixon ventral is 1964 davisi Dixon davisi 1966 angelensis Dixon angelensis 1966 partidus Dixon partidus 1966 tinklei Dixon tinklei 1966 bugastrolepsis Dixon bugastrolepis 1966 santacruzensis Dixon santacruzensis 1 966 apricus Dixon apricus 1968 palmeiis Dixon pal me us 1968 dixoni Rivero-Bianco & Lancini dixoni 1970 kofordi new species kofordi 1970 interandinus new species interandinus 1970 johnwrighti new species johnwrighti 1970 angustidigitus new species angustidigitus 1970 clinatus new species clinatus 1970 sentosus new species sentosus 1970 pumilus new species pumilus Materials and Methods Approximately 1,800 specimens of Phyllodactylus were examined from mainland South America. The material is housed in the American Museum of Natural History (AMNH); Academy of Natural Sciences, Philadelphia (ANSP); Berlin Museum (BM); British Museum (Natural History) (BMNH); California Academy of Science (CAS); Escuela Politecnica Nacional, Quayaquil (EPN); Field Museum of Natural History (FMNH); Hamburg Museum (HM); Los Angeles County Museum of Natural History (LACM); Museo de Ciencias Naturales, Caracas (MCNC); Museo de Bio- logia de la Universidad Central de Venezuela (MBUCV); Museum of Com- parative Zoology (MCZ); Museo “Javier Prado,” Peru (MJP); Museum of Vertebrate Zoology (MVZ); Neuchatel Museum, Switzerland (NM); San Diego Society of Natural History (SDSNH); Senckenberg Museum, Frank- furt am Main (SMFM); Texas Cooperative Wildlife Collection (TCWC); University of Illinois Museum of Natural History (UIMNH); University of Michigan Museum of Zoology (UMMZ); United States National Museum (USNM); private collections of Charles M. Fugler (CMF), and Raymond B. Huey (RBH). 6 Contributions in Science No. 192 Descriptions of habitat, behavior, ecology, and color notes of living specimens were obtained in the field by Huey during July, August, and Sep- tember, 1967 and 1968 and by Dixon and Wright during November and December, 1968. Climatological data were obtained from Robinson ( 1964) and details of the vegetation were obtained from Macbride ( 1936). Measure- ments and analysis of squamation were made by methods described by Dixon (1964b). Sex was determined by the presence of cloacal bones and enlarged postanal tubercles in males, or by dissection. When we present a description of a species, it will describe all individuals except those covered in the sections on variation. General Description of the Habitat of Western South America The Andes of western South America divide the coastal countries of Ecuador and Peru into three relatively distinct geographic and biological units: the narrow Pacific coast, the Andes and associated valleys, and the tropical forests of the east. Species of Phyllodactylus and other lizard genera such as Tropidurus, Dicrodon, and Callopistes are restricted primarily to the arid coast of western South America and to the northwestern interandean basins of Peru. The coastal zone, buttressed by the moist Chocoan forests in northern Ecuador (Peters J., 1960), passes south from the thorn scrub and mangrove forests of Ecuador to the barren deserts of Peru and northern Chile. With the exception of a few low passes in northern Peru the Andes to the east form a high barrier generally more than 4,250 meters in areas relevant to discussion. The deciduous thorn scrub forests extend from the Pacific coast and Andean foothills of Ecuador to the extreme northwestern corner of Peru in the Department of Tumbes, east to the base and foothills of the Andes in the Department of Piura, and then south along the foothills of the Andes into the Department of Lambayeque. Some of the dominant plants of this region include species of Loxopterygium, Caesalpinia, Bursura, Bombax, Cap pans, Mimosa, and Cereus (Weberbauer, 1936) . The junction of the Ecuadorian thorn scrub and the Peruvian deserts in Tumbes is very narrow and has been attributed to the climatic changes associated with the boundary between the cold waters of the Humboldt Cur- rent and the warmer Equatorial Counter Current (Dawson, 1963). The coast of Peru below Tumbes to northern Chile is a barren, arid desert. This desert is not induced by excessively high temperature regimes; rather, it is apparently the result of the aridity associated with the cold Humboldt Current and a rainshadow produced by the Andes (Dawson, 1963; Kendrew, 1961; Murphy, 1948) . Rainfall is very low and sporadic throughout much of the area. Most of the water available to plants comes from rivers penetrating the coast from the western slopes of the Andes, high water tables in the Quebradas, and fault zones of northern Peru and fog. 1970 Lizards of the Genus Phyllodactylus 7 Weberbauer (1936) recognizes two major plant zones in the Peru and Chile deserts. The first runs from Tumbes to a latitude of about 8° south. This area is essentially rainless, except for stormy El Nino years, and extremely barren. Encompassed are such areas as the Sechura Desert in the Department of Piura. Plants tend to be widely scattered and grow as shrubs rather than trees. The substratum is sandy throughout much of the region with a few rocky cerros and the rocky, hard soils of the Quebrada zones. Characteristic features are shifting barchan dunes intermingled with large hummock-like dunes stabilized by Cap parts scabrida (Koford, 1968) . Major plants inhabitat- ing this region include species of Cryptocarpus, Prosopis, Capparis, and A cacia (Weberbauer, 1936) . The second zone follows the narrow coast to northern Chile and contains the famous Lomas and Tillandsia plant formations. The Lomas are chiefly seasonal carpets of herbs and occasional low trees in areas where the coastal foothills of the Andes rise to heights sufficient to intercept the winter and spring clouds. The increased elevation and lowered temperatures produce con- siderable condensation in the form of mists making the existence of vegetation possible in an otherwise arid climate. The average annual precipitation recorded in the open is 201 mm for 29 years at Lomas de Lachay in the Department of Ancash (elevation 450 m) compared with only 46 mm for 18 years at nearby Lima (Holdridge, 1964) . Major Lomas plants include species of Aniantum, Bowlesia, Astrephia, Sicyos, and abundant lichens, mosses and ferns (Weberbauer, 1936). Unlike the Lomas, the Tillandsia association is not dependent on sea- sonal changes and appears to survive upon atmospheric moisture. Tillandsia is most often found on the leeward side and at a lower elevation of hills covered with Lomas vegetation, in drier valleys inland from the coast, and along the coastal plain where no other plants survive. There are large expanses of the deserts where even Tillandsia is not found and one is confronted only with barren sand reaching from the Pacific tides to several thousand feet up the Andean foothills. The deserts of northern Chile extend south to a latitude of about 27°. Extremely barren and arid, much of these deserts are colonized only by Til- landsia, South of 21° , the desert is bordered by brush and savanna (Mann 1960). To the east the height of the Andes has apparently prevented invasion by species of Phyllodactylus into the Larrea deserts of western Argentina. The northern interandean basins, lying between latitudes 4° to 6° south and longitudes 78° to 80° west (Weberbauer, 1936), are loosely connected with the Peruvian coast by the relatively low passes in the Department of Piura (e.g. Porculla Pass at 2,160 m). These low passes have served as an invasion route for species of Phyllodactylus and other species of coastal lizards into a portion of the Amazon drainage. Surrounded by the higher Andean passes to the north and south and by the Amazonian forests to the east, the interandean basins are limited in extent 8 Contributions in Science No. 192 and similar in appearance to the coastal foothills of northwestern Peru. The general aridity of these basins, less pronounced than the coast, apparently results from a rain-shadow produced by the eastern chain of the Andes cutting the basins off from the Amazonian moisture. The interandean basins vary considerably in species composition and physiognomy but are somewhat uniform in vegetative cover with cacti and deciduous plants covering the basin floors except along water courses. The higher slopes are generally matted with dense growths of woody plants (Weber- bauer, 1936). Further details of the habitat of Ecuador, Peru, and Chile can be obtained from Weberbauer (1936), MacBride (1936), Tosi (1960), Koepcke, M. (1954), Koepcke, H. (1961), Dawson (1963), Murphy (1948), Mann (1960), Robinson (1964), Kendrew (1961), Donoso-Barros (1966), and Verdoorn (1945). Details of the arid and semiarid habitats of Colombia and Venezuela may be obtained from Ruthven (1922) and Shelford (1926). Key The following key is based upon approximately 2,000 specimens of 15 species of Phyllodactylus from mainland South America. The size, shape, and number of scales in the median row beneath the tail is variable within some species but it is generally consistent if a careful examination of the entire, original tail is made. Regenerated tails do not have tubercular scales even if they were present on the original tail. In some instances, the width of the scales of the median row beneath the regenerated tail may be about the same size as those of the original tail. Key to the Species of PHYLLODACTYLUS of Mainland South America la. Preanal scale single, greatly enlarged, 10-20 times larger than belly scales ................................................... 2 lb. Preanal scales small, equal in size to other belly scales ............ 4 2a. Tail smooth, without enlarged rows of lateral tubercles ........... 3 2b. Tail keeled, with four rows of lateral tubercles (northern Chile) heterurus 3a. Digits short, fourth toe lamellae 12-15 (13.1), terminal lamellae large (southern Peru to northern Chile) .................. . gerrhopygus 3b. Digits long, fourth toe lamellae 15-20 (16.8), terminal lamellae very small (coastal south central Peru) . . angustidigitus 4a. Dorsal tubercles small, rounded, sometimes slightly keeled, not arranged in regular rows; ventral surface of tail with all scales of equal size. . . 5 4b. Dorsal tubercles large, trihedral, arranged in regular rows; ventral sur- face of tail with or without enlarged scales in median row ......... 6 5a. Terminal lamellae of digits much enlarged, most of claw hidden when viewed from above; nostril not swollen (extreme northwestern Peru and probably southwestern Ecuador) . inaequalis 1970 Lizards of the Genus Phyllodactylus 9 5b. Terminal lamellae of digits very small, most of claw exposed when viewed from above; nostril greatly swollen (central to extreme north- western Peru) microphyllus 6a. Dorsal surface of tibia without enlarged tubercles 7 6b. Dorsal surface of tibia with enlarged tubercles 9 7a. Dorsal surface of tibia with very small granular scales; ventral surface of tail with all scales of median row enlarged, rectangular (northwestern Peru) clinatus 7b. Dorsal surface of tibia with round, elevated scales; ventral surface of tail with or without some enlarged, rounded scales in median row. . . 8 8a. Tubercles in paravertebral row from head to base of tail 44-53; no scales on ventral surface of tail enlarged (coastal foothills of central Peru) lepidopygus 8b. Tubercles in paravertebral row from head to base of tail 65-81; some scales in median row of ventral surface of tail enlarged, rounded (Mara- non Basin, northwestern Peru) interandinus 9a. Tubercles present on dorsal surface of forearm 10 9b. Tubercles absent on dorsal surface of forearm 13 10a. Terminal lamellae of digits much enlarged; claw mostly hidden when viewed from above; ventral surface of tail with all scales of median row enlarged, rectangular 11 10b. Terminal lamellae of digits small, narrow; claw mostly exposed when viewed from above; ventral surface of tail with a few scales of median row enlarged, rounded (Lima, Peru) sentosus 11a. Large species, average snout-vent length of 64 mm; 19-26 paravertebral tubercles between axilla and groin 12 1 lb. Small species, average snout-vent length of 38 mm; 16-21 paravertebral tubercles between axilla and groin; (northwestern Peru and probably southwestern Ecuador) kofordi 12a. Distal one-third of tail with at least two or more rows of tubercles; scales across venter 18-24; (northern Venezuela and Colombia) . . . . ventralis 12b. Distal one-third of tail without tubercles; scales across venter 26-28; (central Venezuela) dixoni 13a. Scales of supraocular region largest of interorbital scales; enlarged tuber- cles usually present on dorsal surface of thigh; maximum snout-vent length 51 mm 14 13b. Scales of supraocular region smallest of interorbital scales; enlarged tubercles usually absent on dorsal surface of thigh; maximum snout-vent length 75 mm (northwestern Peru and west central Ecuador) . . . . reissi 14a. Scales on proximal one-fourth of tail homogeneous; internasals usually separated along median edges (Huancabamba Basin, northwestern Peru) johnwrighti 14b. Scales on proximal one-fourth of tail heterogeneous; internasals usually in contact along median edges (west central Ecuador) pumilus 10 Contributions in Science No. 192 Species Accounts Phyllodactylus heterurus Werner Phyllodactylus heterurus Werner 1907. Ann. Univ. Chile, 121 : 149. Holotype. Adult male, (lost in a fire), collected by C. Reiche 1902. Type- locality. Oasis de Pica (Desierto de Tarapaca), Department of Tarapaca, Chile. Definition and Diagnosis. This species is known only from the holotype that was destroyed by a fire. Phyllodactylus heterurus is a small gecko with a maximum snout-vent length of 38 mm; dorsal surfaces of head and body smooth, without tubercles; tail with four rows of tubercles, two on each lateral surface; preanal plate present; terminal lamellae of digits moderately large. This species is distinguished from all other species of Phyllodactylus except gerrhopygus and angustidigitus, by the presence of an enlarged preanal plate; from gerrhopygus and angustidigitus by the presence of four rows of tail tubercles, with anteriorly projecting tips. Description of holotype. Since the holotype is no longer extant, we sub- mit a translated version of Werner’s (1907) original description: Head ovoid, very much longer than wide; snout longer than the distance from the posterior corner of eye to ear and two times the length of eye; frontal region not con- cave; ear opening small, horizontal and elliptical; body decidedly thin; inferior aspect of digits similar on anterior and posterior limbs, with wide plates whose distal ends are visibly wider than their proximal portion; dorsal aspect of body with small scales that are ovoid, smooth, somewhat convex; rostral scales pentagonal, somewhat concave posteriorly, with a groove beginning on pos- terior border; nasal opening surrounded by rostral, supralabial and three nasals; nine supralabials and seven infralabials; nasal (internasals) scales separated on median line by an azygous scale; mental two times longer than wide with posterior border rounded, lateral sides parallel; chin without enlarged scales (postmentals absent) ; scales in gular region very small; ventral scales small, smooth, circular and overlapping; tail cylindrical, with transverse series of robust, conical tubercles whose points are directed anteriorly. Color and pattern on dorsum lead gray with black irregular spots and lines; lines unite on dorsal side to form transverse bands with obscure out- lines; tubercles on tail brick red. Total length 71 mm, head 12 mm in length by 7.5 mm in width; tail 30 mm in length. Comments. Donoso-Barros ( 1947) redefined the holotype prior to its destruction and further distinguished the species from gerrhopygus by its large, round eye, ovoid head with sharply pointed snout, and gave preserva- tion colors that differed from those given by Werner ( 1907) . As far as we can determine, no additional specimens have been taken since 1902, therefore data are not available on its habits, habitat, or natural history. Distribution. Known only from the type locality (fig. 5). 1970 Lizards of the Genus Phyllodactylus 11 Phyllodactylus gerrhopygus (Wiegmann) Diplodactylus gerrhopygus Wiegmann 1835. Nova Acta Leop. -Carol., 14:242. Phyllodactylus gymnopygus Dumeril and Bibron 1836. Erp. Gen., 3:394. Phyllodactylus gerrhopygus, Boulenger 1885. Cat. Liz. Brit. Mus., 1:95. Phyllodactylus inaequalis, (part) Burt & Burt, 1931. Bull. Amer. Mus. Nat. Hist., 61(7) :250. Holotype. Lost, formerly in Berlin Museum, collector and date of col- lection unknown. Type-locality. Chile. Definition. An average size gecko with a maximum known snout-vent length of 56 mm; large preanal plate present; dorsal tubercles absent; tibia and femur tubercles absent; ear denticulate on anterior and posterior margins; median row of scales beneath tail equal in size to other ventral scales; tail about 50 percent of total length; mental much longer than wide and post- mentals absent; terminal lamellae of digits large, longer than wide; claws of digits large, usually extending slightly beyond tip of terminal lamellae; dorsal markings present or absent, if present, consisting of broad bands or blotches; tail usually banded (fig. 8a). Diagnosis. This species may be distinguished from all other species, except angustidigitus and heterurus, by the presence of an enlarged preanal plate; from heterurus by the absence of enlarged lateral tubercles on the tail; from angustidigitus by the larger toe pads and shorter digits. Description. Rostral about as high as wide, its dorsal edge “M” shaped, with a short median vertical groove; two internasals, somewhat triangular in shape, their median edges not in contact but separated by three azygous scales (snout granules) ; internasals bordered posteriorly and medially by small granules and postnasal of each side; nostril surrounded by rostral, labial, inter- nasal, and two postnasals; nostril area somewhat swollen by elevation of postnasals and internasal; first supralabial in broad contact with ventral edge of nostril; moderate depression between internasals, slight depression in frontal region; scales in posterior loreal region about twice the size of midorbital scales; eye large, its diameter contained in snout length about 1.5 times; eyelid with two rows of granules and one larger outer row of scales, the last three to six are pointed; diameter of ear contained in eye diameter about 2 times; ear opening denticulate on anterior and posterior margins, rear of head granu- lar; usually 8 supralabials and 5 or 6 infralabials to point below center of eye; mental straight-edged, rounded posteriorly, about 2.5 times longer than wide; postmentals absent, mental bordered by small chin scales. Dorsum without enlarged tubercular rows, scales subequal in size, some- what flattened, about half size of midventral scales; latero-ventral scales smaller than dorsal and ventral scales; very large preanal plate (scaleless), surrounded by about 25 ventral scales; preanal plate about 20 to 30 times larger than adjoining ventral scales; enlarged postanal scales on each side, somewhat flattened. 12 Contributions in Science No. 192 Figure 1. Lateral, ventral and dorsal view of the distal end of the fourth toe of (A) P. gerrhopygus and (B) P. angustidigitus. 1970 Lizards of the Genus Phyllodactylus 13 Dorsal surface of upper arm with flattened scales, forearm with larger, slightly elevated scales; dorsal surface of thigh and lower leg with flattened, but slightly elevated scales; claw moderately exposed when viewed from below; terminal lamellae twice as long as wide, somewhat reduced in size with claw extending beyond tip of terminal lamellae; tail with all scales of ventral surface of equal size, scales of dorsal surface smaller than those of ventral surface. Color in alcohol: ground color tan, with light orange tint; dorsum with four broad light brown bands with undulating edges bands composed of suf- fusion of brown and tan scales intermixed (freckled) , with tan interspaces slightly wider than dark bands; upper and lower arms and legs with faint brownish bands; tail with nine bands similar in color to dorsal bands, with yellowish tan interspaces; posterior part of head and nape with reticulating brown lines on ground color; top of head spotted with brown; brownish stripe from nostril to eye; brownish reticulating lines between eye and ear; labials faintly flecked with dark brown; ventral surfaces of head, body, and tail immaculate white. Variation. Snout-vent length of young adult and adult males varies from 32 to 56 mm (43.9), females from 32 to 55 mm (42.8), juveniles from 25 to 31 mm (28.9), and males and females together average 42.7 mm; postmentals absent in 96.4 percent of sample; three of 98 specimens have two postmentals, one specimen had one postmental on one side, none on the other side; mental much longer than wide, extending posteriorly beyond the tips of adjoining labials; number of transverse chin scales immediately following and touching mental vary from 3 to 8 (4.6) ; scales across midorbital region vary from 13 to 20 (15.9), across the snout at level of third labial from 12 to 18 (14.7); num- ber of scales bordering internasals vary from 6 to 10 (7.7), internasals always separated along median line by 1 to 3 granules; transverse row of scales across venter vary from 20 to 30 (24.7), longitudinally 64 to 76 (69.1); preanal shield large, without scales, generally 20 to 30 times larger than adjoining scales; scales between eye and nostril vary from 8 to 12 (10.4); number of lamellae beneath fourth toe vary from 12 to 15 (13.1); all scales of dorsum, limbs, and tail subequal in size, relatively smooth, and slightly elevated to some extent; supralabials vary from 7 to 9 to a point below center of eye, 8 being most common; dorsal bands (if present) number 4 to 6, tail bands 6 to 9. Color pattern consists of body bands or blotches that tend to appear freckled with tan and brown scales intermixed, but occasionally some speci- mens are uniform tan or light brown; tail usually banded; belly usually immacu- late white; parietal region and lateral area of head behind ear usually with reticulated brown lines on ground color; limbs usually faintly banded with brown. Habitat and natural history. Of 98 individuals examined, only five were collected by one of us. One of these was collected during the day from beneath 14 Contributions in Science No. 192 a piece of tin in an open sand dune area where some vegetative cover was present. Four individuals were found active after dark along the base of sand dunes approximately 150 m from a sand and rock rubble beach, and approxi- mately 50 m from a ravine where some plant cover was present. The dunes were approximately 50 to 75 m in height and without vegetation. Silverfish, beetles, and spiders appear to be the only food items available in the area where the five specimens were taken. The absence of individuals along the beach might be explained by the presence of hundreds of ghost crabs occupying the intertidal zone. One specimen in the University of California Museum of Vertebrate Zoology with field data was taken from beneath a stone in a sandy wash. Carman (1875) mentions that his specimens were found beneath stones in a sandy plain near Arequipa, Peru. Through the courtesy of Stanley W. Taft, a Peace Corps Volunteer from New Mexico State University, we were able to obtain some habitat data on P. gerrhopygus from the San Juan de Marcona area of Peru. The San Juan Marcona area is approximately 157 airline km SSE of lea, at an elevation of 0 to 150 m. The general terrain is rolling sand hills with sandy beaches and large outcroppings of a red limestone-like formation. Taft (personal com- munication) indicated that large hemitite deposits were present in the area which probably give the red color to the sandy areas. The only vegetation of the immediate area consisted of a few scattered Capparis sp. shrubs. A series of P. gerrhopygus collected by Taft were taken from rock walled terraces, outside walls of houses, and from sidewalks. Taft indicated that the geckos were occupying small holes made by the absence of plaster or mortar between the rocks and cement blocks of terrace walls and houses. All of his specimens were collected at night, and two were taken from a terrace wall near a sodium vapor light. Taft stated that several were seen along the sandy beaches at night but were not collected. From an examination of living and preserved females, there appears to be only one egg laid per clutch. Nests of this species have not been found under natural conditions. Distribution. This gecko apparently occupies a variety of habitats along the desert coast, foothills, and mountains from central Peru to northern Chile between the elevations of sea level and 2,750 m (fig. 2). We have examined 98 specimens from the following localities: CHILE. Tarapaca: Arica UMMZ 1 1576; 3 mi S Arica AMNH 65343-59. 50 km from Iquique MCZ 56247. Pica FMNH 6327-29, 9984. Pintados FMNH 6326. Putre FMNH 6330, 6331(3), 6332-41. PERU. Arequipa: Arequipa AMNH 22238, 36525-26, FMNH 3577(4), 43769. 12 mi SE Camana CAS 84769. Chucarapi FMNH 34236-37. 16 km S. La Joya MVZ 84636. Mollendo CAS 84760-68, FMNH 34248(4). Tambo FMNH 4005-06. Vitor FMNH 34179. Yura FMNH 34047(9), 34205(5), 34182-83, 34190. Ica: 1 km NNW Jaguay LACM 48465-66, TCWC 28064-65: Marcona TCWC 27874-78. 6.3 km ESE Pozo 1970 Lizards of the Genus Phyllodactylus 15 Santo TCWC 28063. San Juan Port (Salamanquejo) SDSNH 31091-92. Lima: Isla San Lorenzo USNM 17542 (probably in error). Figure 2. Distribution of P. gerrhopygus, P. microphyllus and P. angiistidigitus in western South America. 16 Contributions in Science No. 192 Phyllodactyius angustidigitus sp. nov. Phyllodactylus gerrhopygus, (part) Burt and Burt 1930; Proc. U. S. Nat. Mus., 78(2849) :4. Holotype. Adult male, TCWC 27915, collected by James R. Dixon and John W. Wright, 1 1 November 1968. Type locality. One km S new port of Pisco, northeast side of Paracas Peninsula, 19 km by road south and west of Paracas, Department of Ica, Peru, sea level. Definition. A moderate-sized gecko, maximum known snout-vent length of 57 mm; large preanal plate present; scales of head, body, tail, and limbs uniform in size; tubercles absent; no enlarged scales beneath tail; tail 48 to 52 percent of total length; mental much longer than wide, without enlarged postmentals; terminal lamellae very small, claw greatly extended beyond tip of terminal lamellae; fourth toe lamellae numerous ( 15 to 20) (fig. 8a, 8b). Diagnosis. Phyllodactylus angustidigitus differs from all other species of Phyllodactylus except gerrhopygus and heterurus, by having an enlarged preanal plate; from heterurus by absence of tail tubercles and small terminal lamellae; from gerrhopygus by small terminal lamellae and high number of fourth toe lamellae, 15 to 20, rather than 12 to 15. Description of holotype. Rostral about twice as wide as high, its dorsal edge with a broad “M” shape, with dorsal vertical groove one-half depth of rostral; two internasals, somewhat rounded, their median edges separated by two granules, bordered posteriorly by two granules and postnasal of each side; nostril surrounded by rostral, labial, internasal, and two postnasals; first labial in narrow contact with ventral edge of nostril; shallow depression between internasals, no depression in frontal region; 10 scales on line between nostril and eye; scales in posterior loreal region about twice size of midorbital scales; 15 scales across snout at level of third labial, 14 scales at anterior edge of orbits, 18 midorbital scales; eye large, its diameter contained in snout length 1.7 times; eyelid (brille) with one row of granules and one larger outer row of scales, last three to four are pointed; diameter of ear contained in diameter of eye 2.3 times; ear opening denticulate on posterior margin; all dorsal scales small, smooth, slightly convex; seven supralabials and six infralabials to point below center of eye; mental with parallel sides, about 2.5 times longer than wide, bordered posteriorly by six chin scales; enlarged postmentals absent; dorsum with 75 granules and scales around body; venter with 29 scales across belly, from throat to vent 79; postanal tubercles number four on one side, three on opposite side; dorsal surfaces of limbs similar to dorsum; lamellae formula for hand 7-11-12-13-11, foot 7-12-16-18-16; claw completely visible when viewed from below; terminal lamellae very small, not much wider than width of fifth lamella counted from base of digit. Measurements in mm; snout-vent length 53.5, axilla-groin length 25.8, length of leg 24.0, length of arm 18.0, length of tail 55.0, length of head 13.3, depth of head 6.2, width of head 10.5, length of snout 5.0, diameter of eye 2.9, diameter of ear 1.2, distance from eye to ear 3.8. 1970 Lizards of the Genus Phyllodactylus 17 Color in life: ground color yellowish tan; dorsum with fine network of thin, dark brown, reticulating lines, broken at midbody by median stripe of ground color, bordered on each side by dark brown line; paravertebral lines begin at shoulder and terminate at level of groin; arm with four light brown bands with interspaces of ground color; lower part of leg with three indistinct light brown bands; upper part of thigh with faint indication of bands with irregular edges; tail with 1 1 broad, brownish black bands that contain light brown diffuse spots within their borders, interspaces ground color, about one-third width of dark bands; head with fine network of light brown, reticulat- ing lines; diffuse brownish black stripe from nostril to eye; iris of eye blue gray with black and silver flecks; ventral surfaces dirty white, with grayish wash across chest and throat. Variation. Snout-vent lengths of young adult and adult males vary from 41 to 57 mm (50.3), females from 37 to 54 mm (47.8), juveniles from 30 to 36 mm (32.8), and males plus females average 48.7 mm; postmentals absent in 99.3 per cent of sample, two of 254 individuals had three enlarged post- mentals; number of chin scales immediately following and touching mental vary from 3 to 8 (5.8); scales across snout at level of third labial vary from 13 to 17 (15.1), across midorbital region 16 to 20 (18.2); internasals sepa- rated by small granules in 99.5 per cent of sample; scales bordering inter- nasals vary from 7 to 10 (8.7) ; scales across venter vary from 23 to 28 (25.6) , from throat to vent 64 to 70 (67.5); enlarged preanal shield without scale covering, 20 to 25 times larger than surrounded scales; scales between eye and nostril vary from 10 to 12 (11.0); number of fourth toe lamellae vary from 15 to 20 ( 16.8) . Ventral color generally uniform dirty white with some grayish wash on chest and throat; dorsal coloration highly variable, from almost uniform wash of yellowish tan to broad, dark brown bands; majority of individuals with some form of dark dorsal reticulations or bands, usually heavily suffused with ground color; tail generally banded, but width of dark bands and light inter- spaces highly variable; regenerated tails either near black with light areas or ground color with dark flecks; ground color variable, from light tan to very dark grayish brown. Comment. Since populations of angustidigitus appear to be closely related to gerrhopygus, the following statement of our ideas on the evolu- tionary trends in this population emphasizes its rate of speciation. The Paracas Peninsula was apparently an island, probably as late as the Pleistocene, and within recent times has become connected to the mainland. The attributes of angustidigitus populations on Islas Sangallan and Viejas are identical with those of the Paracas Peninsula, lending some support to this idea. The original invasion of the islands by prototypes of a angustidigitus-Yiko. population may have occurred by rafting, or by land invasion while the land masses were con- nected early in the formation of the continental shelf. The islands and peninsula populations are presently restricted in distribu- 18 Contributions in Science No. 192 tion to the narrow shell and rubble beaches probably due to cover and food availability. The adaptations for living in the shell and rubble resulted in the reduction of the size of the terminal lamellae and increase in the exposure of the claw as a means of walking and digging in this environment. The longer toe length affords better traction on a shell and rubble substratum. The large terminal lamellae adapted to sand or a scansorial habit are unsuited for life on a shell or rubble beach. The presence of a few angustidigitus living along a mud beach towards the southeast part of the peninsula where it connects with the mainland and also two individuals found in beach dunes on the mainland a few kilometers away may indicate a recent invasion of the mainland by this species. This may be due, in part, to the great density of the population now occupying the shell and rubble beaches. Density dependent factors related to population pressure may be forcing individuals farther away from their preferred habitat. Hence, a few hardy individuals are surviving in a marginal habitat which is generally unfit for the population as a whole. An individual of P. gerrhopygus was found 25 kilometers from the easternmost record of P. angustidigitus, with no indication of intergradation or hybridization between the two species. Thus some period of genetic isola- tion of the peninsular population has allowed angustidigitus to evolve through natural selection. Habitat and natural history. This species is apparently restricted to the narrow shell and rubble beaches of the Paracas Peninsula and its associated islands. Of 241 specimens taken from the east side of the peninsula only three were found more than 10 m from the storm tide zone. One was found beneath a stone at the edge of a bluff approximately 20 m from the ocean and two were found about the same distance away on a small granite hill covered with a thin layer of sand. There are no macroscopic plants on the peninsula and no surface water. Away from the beach the only physical cover consists of a few outcrops of granite and small amounts of condensed salt forming a hard surface crust. The only food items available appear to be silverfish, amphipods, and spiders that live in beach debris of the storm and intertidal zones. The paucity of food and cover elsewhere on the peninsula may account for the concentration of individuals along the storm and intertidal zones. A total of 157 individuals were taken along a 100 m stretch of shell beach in one eve- ning between the hours of 8:00 and 9 : 30 p.m. Individuals of this species emerge from cover soon after dark, but they seem to be most abundant two to three hours following darkness. Several individuals were observed emerging from beneath dried seaweed and shell debris about 9 p.m. During the day individuals were found beneath the car- casses of dead birds, dried seaweed, and assorted trash cover. Seven specimens were found along a narrow mud beach where the only retreat consisted of cracks in a vertical mudstone cliff. All specimens were found in cracks one 1970 Lizards of the Genus Phyllodactylus 19 to two meters above the mud beach. The only food items observed were a few amphipods under damp seaweed on the beach and pseudoscorpions in cracks along the mudstone cliff. Wherever the beach was composed principally of sand, only a few individuals of this species could be found. These were taken along the base of cliffs bordering the beach even though there was considerable storm debris near the beach. The near absence of geckos along these stretches of beach may be due to the great abundance of ghost crabs occupying the storm and inter- tidal zones of these areas. Crabs were not observed feeding on geckos but their nocturnal activity corresponded with that of the geckos, and the crabs may have been possible predators. When disturbed, the majority of geckos attempted to escape by running towards the cliffs above the beach or hiding beneath shell debris. However, on two occasions individuals ran into the ocean. One immediately turned and made its way back to shore while the other swam a few meters out to sea. The latter individual, apparently tired, spread its limbs horizontal to the body and allowed wave action to return it to shore. A few individuals were observed climbing boulders and cliffs during their nocturnal activities. Their actions were slow and deliberate and when dis- turbed, frequently fell off the surface in an attempt to escape. They seem to have difficulty in clinging to vertical surfaces and any rapid movement on their part dislodges their hold on the surface. The absence of large terminal lamellae normally found in scansorial species is a probable factor in their inability to move rapidly on vertical surfaces. Body temperatures were taken from several individuals during their nocturnal activities and found to be closely associated with the substratum temperature, varying approximately 0,5 °C above or below the latter tem- perature. This species appears to lay only one egg at a time. All gravid females examined contained one large developing egg and those confined in captivity laid only one egg. Distribution. This species is apparently confined to the narrow shell and rubble beaches of the Paracas Peninsula and associated islands of Peru although a few individuals may be found on the periphery of the former area (fig. 2). Specimens (246) have been examined from: Ica: Islas Sangallan USNM 38570; Has Viejas SDSNH 31093-94; Lagunillas (Gallinazo) Bay FMNH 34337-40; Paracas Peninsula AMNH 36505-07, 3651042; 7.2 km SW Para- cas LACM 48705, TCWC 27983; 1 km S Puerto Pisco (new) LACM 48593- 627, TCWC 27915, 27944-79; 3 km S Puerto Pisco (new) LACM 48632-704, 48706, TCWC 27984-8062; 7 km S Puerto Pisco (new) LACM 48628-31, TCWC 27980-82. 20 Contributions in Science No. 192 Phyllodactylus inaequaiis Cope Phyllodactylus inaequaiis Cope 1876. Jour. Acad. Nat. Sci., Phila., ser. 2, 8:174. Holotype. Adult female, ANSP 11365, collected by James Orton in 1867-1868. Type-locality, from Pacasmayo, Department of Lambayeque, Peru. Definition. A small gecko with a maximum known snout-vent length of 42 mm; tibia and femur tubercles absent; ear denticulation absent; tubercles on rear of head absent; median row of scales below tail equal in size to adjoin- ing scales (occasionally widened in some specimens) ; enlarged preanal plate absent; enlarged dorsal rows of flattened tubercles usually absent at midbody (occasionally two rows present) ; tail about 45 per cent of total length; tail tubercles absent; terminal lamellae of digits distinctly widened and truncate; ground color light gray with dark gray bands usually present; tail usually banded with dark gray on light ground color; belly brilliant white to light yellow (fig. 12b). Diagnosis. Phyllodactylus inaequaiis may be distinguished from ger- rhopygus, angustidigitus, and heterurus by the absence of an enlarged preanal plate; from clinatus, dixoni, internadinus, pumilus, johnwrighti, kofordi, lepidopygus, sentosus, reissi, and ventralis by having less than 10 dorsal rows of enlarged, flattened tubercles at midbody; from microphyllus by its smaller maximum snout-vent length (42 mm rather than 56 mm), and by having the terminal lamellae of the digits distinctly enlarged and truncate with claw mostly hidden rather than very small terminal lamellae with claw mostly exposed. Description. Rostral twice as wide as high, its dorsal edge almost straight with a short median vertical groove; internasals somewhat rounded, their median edges in narrow contact, bordered posteriorly by small granules and postnasal of each side; nostril surrounded by rostral, labial, internasal and two postnasals; first supralabial in narrow contact with ventral edge of nostril; shallow depression between internasals, slight depression in frontal region; scales in posterior loreal region about two to three times larger than scales in midorbital region; eye large, its diameter contained in snout length about 1.4 times; eyelid with two rows of granules and one large outer row of scales, last three to five are pointed; diameter of ear contained in eye diameter four times; ear opening small, scales on anterior edge rounded, subequal, those on posterior margin rounded but with elevated scales lying within the ear open- ing; rear of head granular, without larger intermixed tubercles; six to seven supralabials and five to six infralabials to a point below center of eye; mental bell-shaped, longer than wide, bordered posteriorly by two postmentals; postmentals wider than long, their median edges in narrow contact, followed by transverse row of granular scales. Distal one-third of dorsum with none to 10 indistinct rows of slightly 1970 Lizards of the Genus Phyllodactylus 21 enlarged, flattened tubercles, paravertebral row of each side reach to shoulder in some individuals; paravertebral rows usually separated from each other by four rows of granules; each tubercle of enlarged dorsal series separated from preceding tubercle by two to four granules; postanal tubercles number four on each side, distinct and elevated. Dorsal surface of upper arm with flattened, rounded scales, forearm with slightly smaller scales that are rounded and slightly elevated; dorsal sur- face of thigh and lower leg with rounded, slightly elevated scales; claw short but visible when viewed from below; terminal pad large, longer than wide with somewhat rounded edges; about every other median scale on ventral surface of tail slightly widened; cloacal sacs with external plicate lips. Color in alcohol: ground color light gray; dorsum with dark gray trans- verse bands, somewhat broken along midline of back; width of bands equal to ground color interspaces; dorsal bands somewhat diffuse with white granules intermixed with few black granules; lateral area of body with suffusion of intermixed white, gray, and black granules; arms and legs with suffuse spot- ting of gray, brownish gray, and black granules and scales, without definite pattern; tail with five dark gray bands, twice as wide as whitish interspaces; top of head with suffusion of dark gray and ground color, no definite pattern; dark gray stripe from nostril through eye, over ear, to just posterior of arm insertion, bordered above and below by pale whitish line from internasal to just posterior of ear; belly brilliant white. Variation. Snout-vent length of subadult and adult males varies from 33 to 40 mm (37.0), females from 30 to 42 mm (36.8), juveniles from 19 to 28 mm (23.0), males and females together average 36.9 mm; two post- mentals in all specimens examined; postmentals contacting first labial in only 54.6 per cent of the sample, first and second labials in 22.7 per cent, and no labials contact in 22.7 per cent; number of transverse scales immediately fol- lowing the postmentals vary from 7 to 10 (8.3) ; scales across the midorbital region range from 20 to 22 (20.8), across the snout at level of third labial 20 to 24 (21.3) ; number of scales bordering internasals vary from 7 to 8 (7.3) with an azygous scale separating the internasals in 3 to 59 specimens; transverse row of scales across belly vary from 26 to 32 (29.4), longitudinally 55 to 62 (58.4) ; scales on a line between nostril and eye vary from 1 1 to 12 (11.6); lamellae beneath fourth toe vary from 10 to 12 (10.7); enlarged rows of dorsal tubercles scarcely evident at midbody, none present in seven speci- mens, two rows present in four specimens; dorsal tubercular rows more evident across rump and base of tail; rows of tubercles across base of tail vary from four to eight, six being most common. Color pattern fairly consistant, with shades of gray ground color and dark gray bands somewhat variable; bands may be straight or slightly undulat- ing; all have dark grayish brown spot in center of parietal area; dark head stripe from nostril to arm in seven specimens, to anterior third of body in 22 Contributions in Science No. 192 three, to ear in one; body bands (if present) number 4 to 6, five being most common; tail bands (if present) number 6 to 7 in about equal numbers; tail bands generally fade towards distal fourth of tail. Comments. (See comments under Phyllodactylus microphyllus.) Habitat and natural history. Phyllodactylus inaequalis principally inhab- its the desert foothills of northwestern Peru. Specimens have been taken in a variety of foothill localities where the vegetation may be classified as very sparse to dense. The Cerro Amotape area rises out of the northern end of the Sechura Desert and is contiguous to the northeast foothills of the western chain of the Andes. This area is moderately barren with exfoliating granite boulders and abundant cacti, but the vegetation is less dense than other foothill local- ities to the east. Three specimens were obtained in a dry wash on the south slope of the cerro where scattered small shrubs were present. The Las Lomas area lies at the foot of the Andes and contrasts with the Cerro Amotape region in having more dense vegetation, principally thorn scrub with scattered larger trees. One lizard was found foraging on the ground beneath small leafless shrubs. The Abra de Naupe area lies at the foot of the Andes south and east of the Las Lomas region. This region consists of low hills composed of weather- ing granite (nonexfoliating) with scattered cacti, mesquite, acacia, and small amounts of grass. One specimen was found during the day beneath a rotting cactus. The Pacasmayo area (type locality) lies near the base of the Andes where the southern end of the Sechura Desert meets the Jequetepeque River. The vegetation is very sparse, consisting of mesquite groves along the river and a few cacti on the foothill slopes. A few individual plants of Cap par is sp. occur in sandy flats at the base of the foothills. Seven individuals were taken at dusk from a flat area of hardpan soils at the base of the foothills on the north side of the Jequetepeque River. All seven specimens were found beneath such trash items as cardboard, cement sacks, and tarpaper. The majority of specimens (47) collected by us were taken in the Cerro de la Vieja region. Although geographically a foothill locality, this cerro is separated from the main foothills lying several kilometers to the north and east. The cerro is approximately 150 m in height, with large exfoliating granite boulders and moderate plant cover. The larger plants of the area are mesquite, palo verde, cacti, and Capparis sp. (tree form). During the day P. inaequalis were found beneath a bridge under a stone and under small exfoliating flakes of granite on large boulders. Individuals were found actively foraging on the ground or on small boulders at night. Most individuals were found on the ground within one-half meter of the base of boulders. One lizard was taken on open soil approximately five meters from the nearest cover, and one indi- vidual was observed climbing a small shrub. In all regions mentioned above except for the Pacasmayo area, P. inae- 1970 Lizards of the Genus Phyllodactylus 23 qualis is sympatric with reissi and kofordi. There are however, some micro- habitat preferences for daytime retreats unique to each species. Large exfoliat- ing granite flakes were utilized by P. reissi, small flakes by inaequalis, and small burrows under boulders or cracks in the soil by kofordi. After dark, P. reissi generally preferred large vertical faces of boulders on which to forage, P. inaequalis generally on the ground at the base of boulders, and P. kofordi on boulders and on the ground. The nocturnal body temperatures of P. inaequalis usually ranged between the substratum and air temperatures but generally closer to the substratum than to the air temperature. This was found to be true for other species as well. This species lays only one egg at a time. Captive females laid only one egg and preserved gravid females contained a single egg, either in the right or left oviduct. Distribution. This species occurs in widely scattered localities in north- western Peru. It appears to be restricted to the desert foothills bordering the Sechura Desert (fig. 4). Specimens (59) have been examined from the fol- lowing localities: Lambayeque: 1 km S Motupe LACM 48495, 48500-18, TCWC 28308-27, MVZ 82174, 85329-33, 7.2 km W Naupe TCWC 28328. Libertad: Pacasmayo ANSP 1 1365; 16.7 km ENE Pacasmayo LACM 48496- 99, TCWC 28305-07. Piura: Cerro Amotape MVZ 82170, 82172, 82283; 5 km E Las Lomas MVZ 82171. Phyllodactylus microphyllus Cope Phyllodactylus microphyllus Cope 1876. Jour. Acad. Nat. Sci. Phila., ser. 2, 8:175 Phyllodactylus inaequalis, (part) Noble 1924. Occ. Paps. Boston Soc. Nat. Hist., 5:111. Phyllodactylus Qerrhopycus, (part) Burt and Burt 1931; Bull. Amer. Mus. Nat. Hist., 61(7) :250. Holotype. Adult male, ANSP 11364, collected by James Orton in 1868- 69. Type locality. Valley of Jequetepeque, Peru (Department of Lambayeque) . Definition. An average-sized gecko with a known maximum snout-vent length of 58 mm; very small terminal lamellae on all digits with fingers and toes short; dorsal tubercular rows indistinct, composed of small flat, oval tubercles that seldom form regular rows; enlarged tubercles absent from head, limbs, and tail; scales of ventral surface of tail uniform in size; scales sur- rounding nostril greatly swollen; ventral color of body, limbs, and tail bril- liant white (fig. 9b). Diagnosis. This species is readily distinguished from P. gerrhopygus, heterurus, and angustidigitus by the absence of an enlarged preanal plate; from clinatus, dixoni, lepidopygus, interandinus, kofordi, sentosiis, john- w right i, reissi, ventralis and pumilus by the absence of enlarged, regular rows of trihedral tubercles on dorsum; from inaequalis by the presence of a distinct swollen region about the nostril, larger size, and very small terminal lamellae. 24 Contributions in Science No. 192 Description. Rostral about twice as wide as high, its outer posterior edges elevated along inner edge of nostril; internasals small, rounded and swollen, their median edges separated in most specimens by three rows of granules; median and posterior borders of internasals contact small granules and post- nasal of each side; nostril surrounded by rostral, internasal, and two postnasals, all greatly swollen about nostril; first supralabial not in contact with ventral edge of nostril; shallow depression between internasals and in frontal region; scales in posterior loreal region about twice size of midorbital scales; eye large, its diameter contained in snout length about 1.8 times; eyelid with 2 rows of granules and one larger outer row of scales, last 8 to 12 are pointed; diameter of ear contained in diameter of eye about 2.8 times; ear opening small, not denticulate, scales on anterior and posterior margins rounded; rear of head granular; usually six supralabials and five infralabials to point below center of eye; mental bell-shaped, slightly longer than wide, its distal tip extends beyond posterior border of first pair of labials; postmentals much longer than wide, their median edges usually in narrow contact, nearly sep- arated by posterior projection of mental; postmentals bordered across chin by very small granules. Dorsum with irregular rows of small, oval tubercles that tend to be flat, posterior dorsal edge with faint keel; dorsal tubercles not regular enough to count with accuracy; postanal tubercles usually number four on each side; limbs with somewhat small, flattened scales; main portion of claw visible when viewed from below; terminal lamellae very small. Color pattern in life: ground color yellowish tan; dorsum with profusion of dark brown, brown, light tan, and white granules and larger scales (tuber- cles) that tend to give entire dorsal color a sandy appearance; arms and legs similar to dorsum; dorsal surface of tail faintly banded with light gray bands on yellowish tan ground color; dorsal surface of head similar to dorsum; loreal region with larger diffuse spots of brown and tan granules; ventral sur- faces brilliant white. Variation. Snout-vent lengths of subadult and adult males vary from 33 to 56 mm (46.7), females from 32 to 58 mm (46.5), juveniles 23 to 31 mm (27.1 ), males plus females average 46.6 mm; tail length varies from 42 to 53 per cent of total length; postmentals number from two to four, two occur- ring in 90.6 per cent three in 8.0 per cent and four in 1.4 per cent of sample; postmentals contacting adjacent labials vary from none to two on each side, none in 4 per cent, one on one side and none on opposite side in 4 per cent, one on each side in 89.2 per cent one on one side and two on opposite side in 1.4 per cent and two on each side in 1.4 per cent of sample; number of scales across chin immediately following postmentals varies from 6 to 9 (7.4); scales across midorbital region range from 18 to 25 (20.8), across snout at level of third labial 18 to 26 (21.8); median and posterior scales bordering internasals vary from 7 to 12 (8.7), with median edges of internasals sep- arated by granules in 99.2 per cent of sample; scales from nostril to eye range 1970 Lizards of the Genus Phyllodactylus 25 from 10 to 15 (12.4); scales across venter vary from 21 to 38 (27.3), main- land populations from 26 to 38 (31.0), island populations 21 to 33 (25.8); scales from throat to vent range from 62 to 72 (66.3) in all populations; dor- sal tubercular rows range from 10 to 14 (11.9) when arranged in somewhat regular rows, but majority of individuals have irregular rows; lamellae beneath fourth toe vary from 12 to 15 (13.3); terminal lamellae generally very small, but variable in size dependent upon soil type from which specimens are obtained; tail generally shorter than snout-vent length and always devoid of tubercular rows, ventral surface with all scales of equal size. Dorsal color pattern variable, from light brown, brown, and white freckled appearance on yellowish tan ground color to dark brown to black reticulating lines on gray, tan or light brown ground color; broad, but indis- tinct brown bands are occasionally present in adults, more frequently in young and juveniles; a brownish line usually present from nostril to eye, occasionally extends to ear; ventral surfaces brilliant white. Comments. There has been considerable confusion in the application of the name microphyllus to the original population Cope described in 1876. Cope described microphyllus and inaequalis in the same paper, but failed to make clear the type of digital expansion found in inaequalis that would sepa- rate it from microphyllus. Noble (1924) used the name inaequalis (— micro- phyllus) for populations inhabiting the sandy deserts of the coast, apparently being unable to distinguish between Cope’s two species. Cope (1877) later described another species, nigrofasciatus, and in the same paper explained the differences in digital expansion between microphyllus and inaequalis, appar- ently overlooked by Noble and later workers. Burt and Burt (1930), apparently confused by Noble’s use of the name inaequalis for the sandy desert form, listed microphyllus, inaequalis, and gerrhopygus from Peru, but had misidentified a specimen of gerrhopygus as being inaequalis. They later (1931) pertpetuated this mistake by listing inae- qualis (= gerrhopygus) from Arequipa, Peru, and in part, confused speci- mens of microphyllus with inaequalis in northwestern Peru. This mistake was repeated by Donoso-Barros ( 1966). Habitat and natural history. Phyllodactylus microphyllus is a widespread species, but generally restricted to the western sandy areas of the Sechura Desert, sandy and sand rubble beaches of the Peruvian coast and offshore islands. The absence of vegetation does not seem to restrict its distribution in the drier parts of Peru. We have found specimens in areas where vegetation was absent, but these individuals were found within one kilometer of the beach. The storm and intertidal zones of the beach are probably the principal sources of food. The absence of food elsewhere is a possible factor limiting the distribution of this species. In northwestern Peru where vegetation is more abundant this species has been taken 60 kilometers (airline) from the nearest beach. During the day individuals have been found under rocks, cow chips. 26 Contributions in Science No. 192 boards, bones, paper, tin, and other objects strewn on sandy surfaces. In one instance individuals were found in open Indian graves where remnants of burial cloth were exposed. Some of these daytime retreats are often fully exposed to sunlight and the body temperatures of individuals found under these conditions are about 10°C higher than their nocturnal temperatures. In the vicinity of Culebras, this species was observed emerging just prior to com- plete darkness and actively foraging for about four hours. In the Cerro Illescas area of northwestern Peru this species was found foraging in leaf litter and on the branches and trunks of shrubs, mainly Cap pans avicennifolia, during winter months. We observed this species foraging on guano mounds on Cerro Aziil and it was reported from guano mounds on Isla Lobos de Afuera by Burt and Myers ( 1942). The behavior of this species is similar to that of members of the genus Coleonyx. When disturbed, the tail is raised, back is arched, and it feigns an attack accompanied by noise and walks diagonally away with full side view of body exposed to a potential predator. The movements of this lizard are deliberate, somewhat slow, and when disturbed, it does not seem able to run rapidly. One individual was observed utilizing a small rock to scrape shedding skin from its body. This species apparently lays a single egg per clutch. Preserved females contained a single egg in either the right or left oviduct. One specimen was removed from the stomach of a snake, Liophis sp., but actual predation was not observed. Potential predators consist of ghost crabs, snakes, and foxes. As in other species of Phyllodactylus, the nocturnal body temperatures of this species closely approximates that of the substratum. Distribution. This species is known only from Peru. It ranges from the northwestern coastal village of Los Organos, southward to the coastal village of Cerro Azul, an airline distance of 1,150 km. The western part of the Sechura Desert of northwestern Peru is the only area where its range extends inland from the coast (fig. 2). Specimens (277) have been examined from the following localities : Ancash: 2 km N Culebras LACM 48589-92, TCWC 28151-55. 12 km N, 1 km W Pativilca LACM 48570-72, TCWC 28148-50. 24.7 km N Pativilca MVZ 82145, 82231. Lambayeque: 1 1 km S Chiclayo MVZ 82181. Eten MCZ 18138, 17692-94, AMNH 28517-18. 10 km S Mocupe TCWC 28162. 3 km SE Morrope MVZ 82146. Libertad: Jequetepeque Valley ANSP 11364. 4 km N Pacanguilla LACM 48574, TCWC 28156-57. Lima: Ancon AMNH 36513-19. Cerro Azul LACM 48567-69, TCWC 28145-47. 7 km SSE Chilca MVZ 85430-34, 85436-38, 85440-41. Chorillos MJP 3 (no numbers). Cruz de Hueso USNM 75390. Isla San Lorenzo USNM 17541. Lima MJP 5 (no numbers), 3 km N San Antonio TCWC 28174. 5.5 km NE San Bartolo MVZ 85421-29. Piura: 1 km S Bayovar, Cerro Illescas MVZ 85353-61. 5 km SSE Bayovar MVZ 85342-52. 12 km S, 8 km E Bayovar MVZ 82157-61, 1970 Lizards of the Genus Phyllodactylus 27 82270. 20 km SE Bayovar MVZ 85362-63. Isla Lobos de Afuera SDSNH 31123-53, USNM 38571-74. Isla Lobos de Tierra FMNH 34226(7), 34234 (5), 40113-14, AMNH 38753-55, UMMZ 83099 (2). 3.7 km ENE Los Organos LACM 48588, TCWC 28173. Negritos FMNH 5726-27, 8363, 8453-54. Paita AMNH 28461, HM 1499 (2). mouth of Parinas Valley FMNH 41543-45. 13 mi N Piura CAS 85163. 10 km E Piura MVZ 82162. Punta Aguja MVZ 82148-52, 82228, 82248. Reventazon MVZ 82153-56. 30 km SSE Sechura MVZ 82147. Talara FMNH 53891, 53894, 57457. 2 km E Talara LACM 48575-78, TCWC 28158-61. 2 km N Talara TCWC 28163. 3 km S Talara LACM 48579-86, TCWC 28164-71. 5 km S Talara LACM 48587, TCWC 28172. Trie Trac MVZ 82163-68, 85334-41. Phyllodactylus clinatus sp. nov. Holotype. Adult female, MVZ 82227, collected by Raymond B. Huey, 19 July 1967. Type locality. Punta Aguja, 37 km SW Sechura, Department of Piura, Peru. Definition: A small gecko with a known snout-vent length of 46 mm; dorsum with 10 to 12 rows of enlarged, trihedral tubercles; tubercles absent on tail, thigh, tibia, and forearm; granules absent between internasals; ear very large, 56.7 per cent of diameter of eye; ear strongly denticulated on anterior and posterior margins; terminal lamellae of digits large, truncate; all scales of median row beneath tail widened, rectangular (fig. lib). Diagnosis. This species may be distinguished from P. heterurus, angusti- digitus, and gerrhopygus by the absence of an enlarged preanal plate; from inaequalis and microphyllus by much larger ear, dorsal rows of enlarged, trihedral tubercles arranged in regular series, and by a widened row of scales beneath tail; from dixoni, pumilus, kofordi, sentosus, johnwrighti, reissi, and ventralis by the absence of tubercles on tibia and much larger ear; from lepidopygus and interandinus by presence of a row of widened, rectangular scales beneath tail, larger ear, and small, flattened granular scales on tibia. Description of holotype. Rostral twice as wide as high with short median groove about one-third depth of rostral; two internasals, somewhat rectangu- lar, their median edges in broad contact, and bordered posteriorly by three granules and postnasal of each side; nostril surrounded by rostral, labial, internasal, and two postnasals; first labial in broad contact with ventral edge of nostril; deep depression between internasals, moderate depression in frontal region; scales in posterior loreal region about 2.5 times larger than midorbital scales; 22 scales across snout at level of third labial, 14 scales across anterior edge of orbits, 20 midorbital scales; eye large, its diameter contained in snout length 1.5 times; eyelid with two rows of granules and one larger outer row of scales, last 4 to 5 are pointed; ear diameter contained in eye diameter 1.8 times; ear opening large, denticulate; rear of head granular with larger inter- mixed tubercles; six supralabials and five infralabials to point below center of eye; mental bell-shaped, as wide as long; bordered posteriorly by two post- 28 Contributions in Science No. 192 mentals; postmental as wide as long, their median edges in broad contact, immediately followed by transverse row of 9 scales, followed by second row of 13 smaller scales; postmentals contact first labial only of each side. Dorsum with 10 rows of enlarged, trihedral tubercles that are somewhat flattened; paravertebral row with 45 tubercles from head to base of tail, 24 between axilla and groin; paravertebral rows separated from each other by 4 to 6 rows of granules; 4 rows of dorsal tubercles reach to nape, 6 to base of tail; each tubercle of enlarged dorsal series separated from preceding tubercle by 1 to 3 granules; postanal tubercles number two on each side, indistinct; 26 scales across venter; 60 from throat to vent; dorsal surfaces of arm with granules and larger flattened scales; dorsal surface of thigh with granules and larger flattened scales, tibia with small flat granules; lamellae formula for hand 6-8-10-13-9, foot 7-9-11-14-10; claw hidden when viewed from below; terminal lamellae large, longer than wide, truncate; digits slender. Measurements in mm: snout-vent length 46.0, axilla-groin length 19.0, length of leg 16.5, length of arm 14.4, length of tail 46.0, length of head 14.0, depth of head 4.9, width of head 8.7, length of snout 4.9, diameter of eye 2.3, diameter of ear 1.8, distance from eye to ear 3.6. Color in life: ground color sand brown; dorsum with six narrow, choco- late brown bars interrupted along median line, ground color interspaces four to five times width of dark bars; sides of body with light tan spots between dark bars, tending to form ocellated pattern with network of fine brown lines intermixed; arms with chocolate brown reticulating lines, tending to form bands on upper arm; hind limb with five chocolate brown bands, one-third as wide as ground color interspaces; tail with 1 1 dark brown bands, ground color interspaces twice width of dark bands; head with dark brown parietal spot, rear of head with thin, dark brown reticulating lines: dashed and partly wavy line from rear of eye to hind limb, connecting with lower edge of dark body bars; supralabials with upper half heavily pigmented with dark brown, lower half and all of infralabials with suffusion of minute black dots; dark brown stripe from nostril to eye, tending to be black from eye to side of body; ventral surfaces with faint brownish dots scattered along sides, more dense on tail; general ventral color pale yellow. Variation. Snout-vent length of juvenile male 28 mm, females vary from 34 to 46 mm (40.3); tail length about 45 to 50 per cent of total length; post- mentals number 2 in all, contacting first labial only in all; number of scales across chin immediately following postmentals range from 6 to 9 (7.3); scales across snout at level of third labial range from 21 to 23 (21.8), midorbital scales from 19 to 22 (19.5); scales bordering posterior edge of internasals number 5 in all; scales from nostril to eye range from 13 to 15 (13.8); scales across venter vary from 26 to 32 (28.8), from throat to vent 60 to 64 (61.8); dorsal tubercular rows range from 10 to 12 (10.5); paravertebral tubercles from head to base of tail range from 42 to 46 (44.5), from axilla to groin 24 to 29 (27.5); lamellae beneath fourth toe range from 13 to 15 (14.0); 1970 Lizards of the Genus Phyllodactylus 29 terminal lamellae of digits large and truncate in all; ratio of ear diameter to that of eye ranges from .500 to .653 (.567); ratio of ear diameter to distance between eye and ear ranges from .419 to .487 (.444) . Dorsal color pattern varies from distinct bands, broken along median line, to series of dorsolateral spots connected longitudinally by thin dark brown line; bands or spots (one side) vary from 6 to 8 (6.5) ; tail bands num- ber 11 in only specimens with complete tail; limbs usually banded but bands may be obscured by dark ground color; ventral surfaces pale yellow in all specimens. Habitat and natural history. Having taken only four individuals of this species, very little natural history information of the species could be obtained. All four specimens were taken from the vicinity of Cerro Illescas where Cap- paris avicennifolia, C. scabrada and mesquite are the dominant plants. Three specimens were obtained in July and August of 1967. One was taken from a snap trap at the base of a small boulder near the edge of a cliff. The substratum was sandy, although it was only one-third meter from a frag- mented shale area. One was captured on the trunk of a mesquite about one m above the ground. The third specimen was taken one afternoon under a small rock on sandy substratum, near rock outcrops and mesquite. In spite of intensive collecting in this general area in 1968, only one additional specimen was obtained. It was found 2.5 m above the ground on a mesquite branch. Distribution. This species is known only from the Cerro Illescas area of northwestern Peru (fig. 4). Four specimens have been examined: Piura: 4 km SSE Bayovar MVZ 82271; 5 km SSE Bayovar TCWC 28332; 10 km SE Bayovar MVZ 82269; Punta Aguja MVZ 82227. Phyllodactylus lepidopygus (Tschudi) Diplodactylus lepidopygus Tschudi 1844-46, Untersuchungen uber die Fauna Peruana. Herp.: 38. Discodactylus phacophorus Tschudi 1844-46, Untersuchungen uber die Fauna Peruana. Herp.: 38-39. Phyllodactylus nigrofasciatus Cope 1877. Proc. Amer. Philos. Soc. Phila., 17:36. Phyllodactylus variegatus Werner 1901. Abh. Mus. Dresden, 9(2): 2. Holotype. NM 43, adult male, obtained by Tschudi prior to 1844. Type locality, Chorillos, Peru. Definition. A moderate-sized gecko with known maximum snout-vent length of 55 mm; tibia and femur tubercles absent; ear denticulate on anterior margin; slightly enlarged tubercles among granules on rear of head; median row of scales on ventral surface of tail not enlarged; enlarged preanal scale absent; 10 to 16 (13.1) rows of enlarged trihedral tubercles on dorsum; tail length about 45 to 55 percent of total length, tail without enlarged tubercles; 30 Contributions in Science No. 192 terminal lamellae of digits moderately enlarged; body bands (may be broken on midline) 6 to 9 (7.2), tail bands 9 to 13 ( 1 1.0) (fig. 13a). Diagnosis. This species is easily distinguished from P. heterurus, angusti- digitus, and gerrhopygus by the absence of an enlarged preanal scale; from kofordi, reissi, cUnatus, dixoni, johnwrighti, pumilus, ventralis, and inter- andinus by having all scales beneath the tail of equal size, rather than median row distinctly widened or enlarged; from inaequalis and microphyllus by hav- ing distinct dorsal rows of enlarged trihedral tubercles, rather than indistinct dorsal rows of small flattened tubercles or rows occasionally absent; from sentosus by the absence of tail, tibia, and thigh tubercles, rather than very large tubercles present. Description. Rostral more than twice as wide as high; two internasals and two postnasals; nostril bordered by rostral, internasal, labial, and two postnasals; second supralabial separated from nostril by two granules and postnasal; posterio-dorsal loreal scales about four times larger than inter- orbital scales; internasals slightly wider than long, bordered posteriorly by small scales; auricular opening very small, slightly denticulate, contained in snout length about eight times; eye large, contained in snout length about one and one-third times; rear of head granular with few larger, rounded tubercles; 7 to 9 supralabials and 5 to 6 infralabials to point below center of eye; mental slightly longer than wide, bordered posteriorly by two to three well defined postmentals; postmentals followed by irregular row of granular scales; eyelid with two rows of granules and larger palebral row of scales; last 2 to 5 scales in larger supercilliary series bearing long spines; regular rows of longitudinal, enlarged keeled tubercles on dorsum, 8 rows reach to rear of head and 8 to base of tail; all rows equally spaced apart, no wide granular area separating paravertebral rows; four well defined postnasal tubercles on either side of anus in males; venter scales moderately differentiated from lateral body scales; no enlarged preanal shield; arm covered with flat imbricated scales, no enlarged tubercles intermixed with flattened scales; femur with flat, imbri- cated scales on dorsal and ventral surfaces, posterior-ventral surface with granular scales; lower leg similar to femur; terminal pads slightly longer than wide; claw short, hidden between terminal pads. Variation. Specimens of this species with a minimum snout-vent length of 32 mm are considered young adults. In a sample of 36 specimens, 8 are less than 32 mm snout-vent length, 13 are males with snout-vent from 32 to 50 mm (40.8), 15 are females with snout-vent from 36 to 55 mm (46.8) . The average snout-vent length for males and females is 44.0 mm. The number of postmentals varies from two to four, two occurring 58.4 per cent of the time, three 38.9 per cent and four 2.7 per cent; number of scales bordering the postmentals range from 5 to 8 (6.9) ; number of scales across the midorbital region 15 to 21 (18.2); number of scales across the snout at the level of the third supralabial 17 to 22 (20.1); scales bordering the internasal shields posteriorly number 5 to 1 1 (6.7), with one or two small 1970 Lizards of the Genus Phyllodactylus 31 azygous scales separating the internasals in 7 of 36 specimens; transverse venter scales number 26 to 34 (29.4), longitudinally 65 to 74 (69.6); scales from posterior edge of nostril to eye number 10 to 14 (11.4); fourth toe lamellae number 10 to 14 (11.5); dorsal tubercular rows number 10 to 16 (13.1); tubercles in paravertebral row from rear of head to base of tail num- ber 42 to 53 (47.8), axilla to groin 24 to 34 (28.9); body bands or blotches number 6 to 9 (7.2), tail bands 9 to 13 ( 1 1.0). The general ground color in life varies from a greenish brown to a deep blackish brown; dorsal body markings vary from black spots to brown cross- bands or light tan blotches; general appearance consists of light ground color with darker spots irregularly arranged; head with blackish line from nostril through eye to above arm, occasionally continuing along side of body to hind limb: top of head from prefrontal area to posterior tip of parietal region con- sistently darker than remainder of body in all specimens but one; one speci- men has reversal pattern of dark ground color with light blotches; tail gen- erally banded with alternating black or brown bands with light tan to whitish interspaces; interspaces usually about half the width of dark bands but occa- sionally as wide or wider than dark bands; regenerated tails usually reticulated with dark lines on tan ground color; juvenile pattern similar to that of adults; venter yellowish tan to yellowish brown with dark brown spotting along lateral edges or occasionally to center of venter; chin and throat creamy yellow in young and subadults, becoming progressively darker with suffusion of brownish dots in old adults; ventral surface of tail generally darker than venter. Comments. We have examined the holotype of P. lepidopygus, three syntypes of P. phacophorus, holotype of P. nigrofasciatus and find that they are identical in general shape, size of head and body. Their scale characters are within the range of variation of a relatively large sample of P. lepidopygus from a single locality and we therefore consider P. phacophorus and P. nigro- fasciatus as subjective synonyms of P. lepidopygus. The nine syntypes of P. variegatus were destroyed during the fire bombing of Dresden, Germany, during World War II and Werner’s (1901) original description is the only evidence we have for a comparison of P. lepidopygus and P. variegatus. Werner’s description agrees with the characters attributed to P. lepidopygus. His description agrees to some extent with the salient features of P. reissi, a species found further to the north. The major exception is the presence of tibia tubercles in P. reissi, absent in variegatus, and to a minor degree, the short outer row of dorsal tubercles in P. reissi, not so in variegatus. Werner gave the type locality as Lima (9 specimens) and Chanchamayo ( 1 specimen), both areas some 1,500 km south of the nearest locality where P. reissi has been taken. However, P. lepidopygus occurs in the vicinity of Lima and P. variegatus is considered a synonym of P. lepidopygus. Habitat and natural history. This species has been found associated with the lomas vegetation of the Lima region, the drier Tillandsia association of the desert coast, and the coastal foothills to an elevation of 1,400 m. The 32 Contributions in Science No. 192 Figure 3. Distribution of P. pumilus, P. kofordi, P. lepidopygus, and P. interandinus in western South America. 1970 Lizards of the Genus Phyllodactylus 33 microhabitat appears to be one of vertical surfaces of buildings, rock walls, large and small granite boulders, and crevices therein. A series of 34 specimens taken by Huey, Dixon, and Wright from Lomas de Lachay were found exclusively on vertical surfaces of granite boulders, rock walls, or walls of buildings. No specimens were observed on rocks with abundant mossy growth, but lichen covered surfaces were utilized. Occasion- ally P. lepidopygus were seen in crevices or under smaller rocks resting on boulders during the day. This species became active as soon as it became dark, even on cold misty nights. The air temperature on 4 July 1967 was 13.2°C. approximately an hour and half after nightfall and seven geckos were taken subsequent to that time. Potential food items (e.g., Porcellio, spiders, moths) in the Lomas de Lachay region were plentiful on the rock surfaces at night. One specimen taken 24.7 km north of Pativilca was collected from a burro femur lying in a rolling sand hill region with few exposed surface rocks and Tillandsia as the only plant. The only gecko observed in the Lomas de Lachay region was P. lepido- pygus, but the rolling sand hill region 123 km north of Lima also contained P. microphyllus. Of four specimens taken from the Asia area south of Lima, three were found actively foraging at night on a rock corral wall, and one was found on a small cerro on a rotting granite outcrop. All four were found on surfaces relatively free of heavy lichen growth. Two individuals were found east of Chosica at an elevation of 1,700 m. Both specimens were taken from a wall of an abandoned adobe building that had been plastered. The coloration of both specimens was atypical, being almost translucent, without pattern, and inconspicuous against the light colored plaster. The two individuals had a more normal color pattern following a 24-hour period in a cloth sack. An examination of the oviducts of preserved females reveals that this species lays a single egg per clutch. Natural nests were not obesrved but we suspect that eggs are deposited in cracks of large boulders in the Lomas de Lachey area. Several shells of Phyllodactylus lepidopygus eggs were found 40 km NNW Pativilca, 600 m, in dead cactus plants in a dry arroyo strewn with small boulders. No geckos were observed at this locality, but the location it within the known distribution of lepidopygus. Distribution. The known range of this species is approximately 300 km north and 300 km south of Lima, and appears to be restricted to the rocky foothills from near sealevel to an elevation of 1,700 m (fig. 3). Specimens (63) have been examined from the following localities: PERU, no specific locality BM 25440. Ancasli: Chimbote Valley ANSP 1 1366. 24.7 km N Pativilca MVZ 82175. Lima: 2 km ESE Asia LACM 48473-74, TCWC 28142-43. 8 mi SE Chilca MVZ 85435. Chorillos NM 43. vicinity of Chorillos NM 41 (3). Chosica FMNH 34175. 26 km ENE Chosica LACM 48475, TCWC 28144. Lima MJP (3) unnumbered. Lomas de Lachey CAS 34 Contributions in Science No. 192 85143, LACM 48467-72, MVZ 82182-91, 82222-25, 82252-65, TCWC 28135-41. 20 km E Lurin MVZ 85411; Verrugas Canyon MCZ 26674. lea: lea FMNH 39367. Phyllodactylus interandinus sp. nov. Phyllodactylus phacophorus, (part) Noble 1924, Occ. Paps. Bost. Soc. Nat. Hist., 5: 111. Holotype. Adult female, TCWC 27914, collected by John W. Wright and James R. Dixon, 26 November 1968. Type locality, 4.8 km S (by road) Bel- lavista, 520 m. Department of Cajamarca, Peru. Definition. A small gecko with a known snout-vent length of 47 mm; numerous rows of enlarged dorsal tubercles and tubercles in paravertebral row from head to tail and axilla to groin; terminal lamellae moderately large, digits short; distinctly banded, black, white, and yellowish green tail; tubercles absent on thigh, tibia, and forearm; distinct dorsal color pattern of three light and four dark longitudinal stripes (fig. 10a). Diagnosis. This species is distinguished from P. heterurus, angusti digitus and gerrhopygus by the absence of an enlarged preanal shield; from inaequalis and microphyllus by having some scales beneath tail enlarged, and by numer- ous rows of enlarged trihedral tubercles on dorsum; from clinatus, dixoni, ventralis, lepidopygus, kofordi, sentosus, johnwrighti, pumilus, and reissi by presence of 65 or more tubercles in paravertebral row from head to base of tail, and distinct dorsal color pattern of light and dark stripes. Description of holotype. Rostral twice as wide as high, its dorsal edge with vertical groove one-half depth of rostral; two internasals, somewhat rounded, their median edges in broad contact, bordered posteriorly by six granules and postnasal of each side; nostril surrounded by rostral, labial, inter- nasal, and two postnasals; first supralabial in narrow contact with ventral edge of nostril; shallow depression between internasals and in frontal region; 12 scales on line between nostril and eye; scales of posterior loreal region about two times larger than midorbital scales; 22 scales across snout at level of third labial, 14 scales across anterior edge of orbits, 16 midorbital scales; eye large, its diameter contained in snout length 1.6 times; eyelid with two rows of granules and one larger outer row of scales, last three are pointed; ear con- tained in eye diameter 2.6 times; ear opening denticulate, but denticulate scales low and somewhat rounded; rear of head with subequal scales; five supralabials and four infralabials to point below center of eye; mental bell- shaped, slightly longer than wide, bordered posteriorly by three postmentals; postmentals rounded, their median edges in broad contact, immediately fol- lowed by transverse row of 8 scales, followed by second row of 12 smaller scales; postmentals contact first labial only on each side. Dorsum with 20 longitudinal rows of enlarged trihedral tubercles, some- what flattened, paravertebral row with 7 1 tubercles from head to base of tail, 44 between axilla and groin; paravertebral rows separated from each other by 1970 Lizards of the Genus Phyllodactylus 35 Figure 4. Distribution of P. johnwrighti, P. inaequalis and P. clinatiis in western South America. 36 Contributions in Science No. 192 5 rows of granules, each tubercle of paravertebral row separated from preced- ing tubercle by 0 to 1 granule; 14 rows of tubercles extend across nape, 8 at base of tail; postanal tubercles number three on each side; 34 scales across venter, 58 from throat to vent; dorsal surface of arm with small flattened scales; dorsal surface of thigh without tubercles; tibia with subequal rounded, somewhat elevated scales; lamellae formula for hand 6-7-8-9-7, foot 6-9-10- 11-10; claw barely visible when viewed from below; terminal lamellae moder- ately large, slightly longer than wide, rounded. Measurements in mm: snout-vent length 41.0, axilla-groin length 18.6, length of leg 15.2, length of arm 12.1, length of tail 43.0, length of head 11.3, depth of head 5.2, width of head 8.1, length of snout 4.2, diameter of eye 2.6, diameter of ear 1.0, distance from eye to ear 3.3. Color in life: ground color dark gray; dorsum with one median and two dorsolateral grayish white stripes, two dorsal and two lateral dark brown stripes; median light stripe from rear of head to base of tail, anterior part of median stripe bifurcates and extends anteriorly to above each eye; dorsal dark stripes tend to form series of loosely connected spots because of small clusters of grayish white spots scattered in dark fields; lateral dark stripe extends from snout to groin; limbs heavily mottled with grayish white and grayish black spots; tail with 1 1 black bands, posterior edge of each black band with white tipped scales; interspaces three times width of black bands, greenish yellow; dorsal surface of head with diffuse dark brown lines; ventral surfaces dusky, chin generally whitish, each scale with black dots. Variation. Snout-vent lengths of subadult and adult males range from 32 to 45 mm (39.2), females 33 to 47 mm (41.1), young and juveniles 22 to 30 mm (29.0), males plus females average 40.1 mm; tail lengths vary from 48 to 55 per cent of total length; postmentals range from two to four, two occurring in 58.4 per cent, three in 38.9 per cent, and four in 2.7 per cent of sample; chin scales bordering postmentals range from 5 to 8 (6.9) ; postmentals contact first labial only of each side in 99.6 per cent, first labial on one side, none on opposite side in 0.4 per cent of sample; scales bordering internasals number 5 to 7 (6.2); scales from nostril to eye vary from 9 to 12 (10.8); scales across snout at level of third labial range from 22 to 27 (24.3), mid- orbital scales 18 to 22 (20.3); scales across venter vary from 26 to 33 (29.6), from throat to vent 56 to 65 (60.8); fourth toe lamellae range from 10 to 12 (11.6); longitudinal rows of dorsal tubercles vary from 14 to 22 (17.4); tubercles in paravertebral row from head to base of tail range from 65 to 85 (70.1 ), between axilla and groin 35 to 60 (40.7). Color pattern consists of one median and two dorsolateral grayish white to yellowish tan stripes, separated by two dorsal and two lateral grayish brown to grayish black stripes; dark stripes usually bifurcates on nape, each arm extending anteriorly to posterior edge of eye or to tip of snout; broad, dark brown stripe usually present from snout to rear of head, lying between dor- solateral light stripes; ventral surfaces usually dusky with black dots on each 1970 Lizards of the Genus Phyllodactylus 37 Figure 5. Distribution of P. reissi, P. heterurus and P. sentosus in western South America. 38 Contributions in Science No. 192 scale, chin and throat generally lighter in color than remainder of body; dorsal surface of tail with 11 to 16 narrow black bands edged with white posteriorly; broad interspaces bright reddish orange in juveniles, yellowish orange in subadults, yellowish green in adults. Comments. '^oh\Q (1924) labeled a relatively large series of specimens from northwestern Peru as P. phacophorus, but the majority of the specimens are referable to P. interandinus. A critical examination of the entire series reveals four species were present in the collection, interandinus, johnwrighti, kofordi, and reissi. Habitat a?td natural history. Noble (1924) recorded taking phacophorus (—interandinus) and magister (^reissi) from houses and deserted huts in Bel- lavista. We were unable to obtain interandinus in Bellavista but found reissi on the walls of houses. Phyllodactylus interandinus was abundant on a thorn scrub hillside a few kilometers south of Bellavista. Standing, dead Cereus sp. are the principal daytime retreats of this species with as many as seven indi- viduals occupying one plant. Some individuals were taken as high as two meters above the ground, but the majority of specimens were found within one meter of the ground, usually beneath the bark, in the hollow trunks or branches. Phyllodactylus reissi and interandinus were taken from the same cactus plant, but the former species was not abundant. The distribution of interandinus is spotty, for similar areas of vegetation on other hillsides yielded only reissi. Hillsides and valleys were examined in the Department of Ama- zonas, east of the Rio Marahon, but only reissi were found in abundance. Only six individuals of interandinus were found east of the river, taken beneath rotting agave stumps and dead cactus litter. A total of 79 reissi were taken east of the river in a wide variety of habitats, including agave. There seems to be potential competition between these two species for space but not for food. Phyllodactylus reissi is primarily scansorial while interandinus forages on the ground. Eggs of both species were found in a variety of rotting debris, but reissi eggs were far more numerous than those of interandinus. An examination of preserved females of both species indicates that P. interandinus lays one egg per clutch and reissi two eggs per clutch. There are distinct behavioral differences between P. interandinus and reissi when suddenly exposed to sunlight while in their daytime retreats. Phyl- lodactylus interandinus remains motionless for a few seconds, then seeks cover under debris within a few centimeters of where it was uncovered. Phyl- lodactylus reissi immediately runs up or down the plant upon which it was exposed, or jumps to the ground and runs to the nearest cover to hide. The uneven distribution of P. interandinus in the Maranon Valley sug- gests that its niche requisites are more restricted than those of reissi. Distribution. This species is an inhabitant of the Chinchipe, Maranon, and Utcubamba river valleys of the departments of Amazonas and Cajamarca, Peru. The locality of a series of specimens in the Museum of Zoology, Uni- 1970 Lizards of the Genus Phyllodactylus 39 versity of Michigan, is given as Sullana, Peru. This locality is approximately 70 km west of the Andes, and lies at the north end of the Sechura Desert. The specimens are in a similar state of color and preservation as those collected by Noble in 1916. We believe this locality to be in error, and represent some of the specimens taken by Noble from the Chinchipe or Maranon river valley (fig. 3). Specimens (149) have been examined from the following localities: PERU: Amazonas: 17 km S Bagua Chica LACM 48538, TCWC 28087. 10 km WSW Bagua Grande MVZ 82178-80, 82251. Cajamarca: Bellavista AMNH 39338, 39341-42, MCZ 18139, 18140(80), SMFM 8295, TCWC 24924-27. 4.8 km S Bellavista LACM 48519-37, 48539, TCWC 27914, 28066-68, 28070-86. Perico AMNH 28428-36. Pima: Sullana UMMZ 59034(4) (in error). Phyllodactylus kofordi sp. nov. Holotype. Adult male, TCWC 27912, collected by James R. Dixon and John W. Wright, 19-20 Nov. 1968. Type locality. 7 km S Motupe, Cerro de la Vieja, Department of Lambayeque, Peru, 150 m elev. Definition. A small gecko with a maximum known snout-vent length of 46 mm; tibia and femur tubercles present; ear denticulate on anterior and posterior margins; moderately enlarged tubercles on rear of head intermixed among granules; median row of scales below tail distinctly widened; enlarged preanal scale absent; 12 to 14 (12.1) rows of enlarged trihedral tubercles on dorsum; tail 50 to 55 per cent of total length; tail with an 8-6-4-2 or 6-6-4-2 reduction of enlarged tubercles per whorl from base to distal half of tail; terminal lamellae of digits moderately enlarged, truncate; body markings of three types, with spots, dots, or uniform coloration; tail markings of two types, uniform or banded; trihedral tubercles of tibia greatly enlarged (fig. 11a). Diagnosis. This species may be distinguished from P. heterurus, angiisti- digitus and gerrhopygus by the absence of an enlarged preanal plate; from johnwrighti, pumilus, interandiniis, clinatus, inaequalis, lepidopygus, and microphyllus by the presence of greatly enlarged trihedral tubercles on the dorsum, thigh, tibia, and forearm and by whorls of tubercles on the entire length of dorsal surface of the tail; from sentosus by having much larger terminal lamellae, smaller dorsal tubercles, and smaller snout-vent length; from re is si by the presence of tail tubercles, tubefcles on the thigh, smaller body size (maximum 46 mm rather than 75 mm), fewer paravertebral tuber- cles from rear of head to base of tail (31 to 36 rather than 46 to 60), and fewer paravertebral tubercles between axilla and groin (16 to 21 rather than 28 to 38) ; from dixoni and ventralis by the latter two characters and smaller maximum snout-vent length. Description of holotype. Rostral twice as wide as high; depression between internasals continues posteriorly through frontal and interorbital regions; internasals bordered posteriorly by 5 granular scales; 18 scales across 40 Contributions in Science No. 192 snout between third labials, 13 across snout between second labials; 14 between middle of orbits; 1 1 scales between eye and nostril; posterodorsal loreals slightly more than three times larger than interorbital scales; two postnasals; auricular opening strongly denticulate, two or three auricular scales bearing long spines, projecting posteriorly from anterior margin; eye diameter slightly less than two times the snout length; ear diameter contained in eye diameter three times, in snout length 5 times; second supralabial separated from nostril by two loreal scales and postnasal; anterio-dorsal border of first supralabial forms projecting edge bordering nostril below; 6 supralabial and 5 infralabial scales to center of eye; mental slightly longer than wide, bordered posteriorly by two postmentals which contact only the first labial on one side and none on the opposite side; postmentals followed by irregular transverse row of 6 scales, followed by second row of 10 smaller scales; Dorsum with 12 longitudinal rows of enlarged, keeled tubercles, 8 reach to rear of head, 6 reach base of tail; 28 scales across venter; 56 from gular region to vent; venter scales abruptly differentiated from lateral and dorsal scales, about 5 to 6 times larger; 6 rows of enlarged tubercles on anterior six whorls of tail, separated longitudinally by two annular rows of flat scales; distal half of tail with rapid reduction of tubercular rows, 6-4-2-0; ventral surface of tail with longitudinal row of 43 widened scales, about twice to three times larger than adjoining scales; three postanal tubercles on either side of vent; dorsal surface of femur and tibia with large trihedral tubercles inter- mixed with granular scales; posterio-ventral surface of femur granular; upper arm with large, flat scales; forearm with enlarged tubercles intermixed with smaller flattened scales; terminal pads slightly longer than wide; claw not extending beyond pad tips nor visible from below; lamellae formula for hand 7-8-9-11-8, foot 7-9-10-12-11. Measurements in mm: Snout-vent length 44.0; head width 9.0; head length 13.1; head depth 6.4; axilla-groin length 21.2; tail length 47.0; leg 18.6; arm 12.5; snout length 5.0; eye diameter 2.7; ear diameter 1.2; distance from eye to ear 4. 1 . Color in alcohol : Narrow broken, diagonal and transverse black lines across dorsum, 13 complete bands on tail; dark brownish stripe from nostril through eye to arm insertion; each enlarged dorsal tubercle flecked with black; venter immaculate, pale white; labials stippled with brown; limbs faintly spotted with brown. Variation. Snout-vent length of subadult and adult males ranges from 30 to 45 mm (38.0), females from 30 to 46 mm (38.6), juveniles from 18 to 29 mm (25.4), males and females together average 38.3 mm; postmentals vary from two to four, two occurring 104 times, three 5 times, and 4 one time; postmentals contact the first labial only in 89.8 per cent of the sample, first and second labial in 1 1.2 per cent; number of scales immediately fol- lowing postmentals across the chin vary from 5 to 8 (6.7) ; scales across the midorbital region range from 13 to 19 ( 15.8), across the snout at the level 1970 Lizards of the Genus Phyllodactylus 41 of the third labial 16 to 20 (17.3); number of scales bordering the internasals vary from 4 to 8 (5.8) with an azygous scale separating the internasals in 14 of 1 10 specimens; transverse rows of scales across the venter number 22 to 30 (25.0), longitudinally 47 to 56 (50.6); dorsal tubercular rows number 12 to 14 (12.1); paravertebral tubercles from head to tail number 31 to 36 (34.0), from axilla to groin 16 to 21 ( 18.0); scales on a line between nostril and eye number 10 to 13 (11.5); lamellae beneath fourth toe number 11 to 13 (11.5); body markings (if present) number 5 to 8 (7.1); tail bands (if present) 8 to 11 (9.9). Tubercles on the dorsum, tibia, and thigh usually much enlarged, strongly keeled and trihedral; tail tubercles large, somewhat flattened, keeled and ele- vated posteriorly. Color pattern consists of three distinct types; (1) presence of distinct black spots, lines or blotches, (2) distinct but very small black dots, (3) uni- form coloration; ground color generally pale brown in all specimens; in those with dorsal markings, the head is usually spotted with profusion of dark brown on ground color; dark brown to black stripe present from nostril through eye to above arm or slightly beyond; tail generally banded with dark brown to black on proximal three-fourths of tail, black and white on distal one-fourth; those with uniform pale brown dorsal color occasionally have dark lateral head stripe and faint tail bands; venter of all specimens pale white to white; chin, throat, and chest with suffusion of light brown flecks on each scale; ventral surface of tail usually heavily suffused with dark brown to black flecks, much darker than venter color. Comment. This species is named in honor of Carl Koford, who has added much to our knowledge of the mammalian and reptilian fauna of Peru in the past twenty years. Habitat and natural history. This small gecko is widespread and abundant in northwestern Peru. It has been found in every type of plant community from sea level to 650 m, except for the main dune areas of the Sechura Desert. This species is primarily a ground forager although it has been taken occa- sionally on boulders and plants. The daytime retreats of this lizard are usually holes beneath rocks, cracks in the ground, holes in sand banks, leaf litter at the base of plants, under or in hollow limbs of cacti, under bark of shrubs or stumps of rotting trees, in beach debris or trash dumped along the roadside. Occasionally a few individuals were found beneath exfoliating flakes of granite. Eggs of this species were found under rotting cacti (mainly Cereus) in the Piura and Tumbes areas. One dead Cereus sp. plant contained 15 eggs of P. kofordi and 4 eggs of reissi. Eggs were not found in other types of cactus litter, nor in debris of acacia and mesquite. It appears that dead Cereus sp. affords the necessary protection for the survival of the eggs and this plant is sought out by females for egg deposition. This species lays one egg per clutch. Captive females laid a single egg and preserved females contained only one egg. Gravid females and small juveniles were found in July, August, Novem- 42 Contributions in Science No. 192 ber, and December, indicating at least summer and winter breeding periods. Diurnal body temperatures were 8°C above the lizard’s nocturnal tempera- tures during foraging periods. This species is sympatric with P. reissi throughout most of its range, with reissi and inaequalis along the desert foothills, and with reissi, clinatus, and microphyllus in the Cerro Illescas area near the coast. Somewhere between Los Organos and Cancas (on the northwest coastline) the northern distribu- tion of microphyllus terminates, and the habitat (beach dunes community) left vacant is effectively taken over by kofordi from Cancas to the Ecuadorian border. Both were found in the same general microhabitat near Talara (beach dunes) but microphyllus was much more abundant than kofordi. Distribution. Phyllodactylus kofordi is a Peruvian species, found on the periphery of the Sechura Desert, western foothills of the Andes, and along the coast from Eten north to the Ecuadorian border. It probably occurs in the southwestern part of Lojas and the southern part of El Oro Departments, Ecuador (fig. 3). Specimens (167) have been examined from the following localities: Lambayeque: Chiclayo MVZ 82221. Eten BMNH 1900.2.26.1. 5 km SSW Motupe MVZ 82209. 7 km S Motupe MVZ 82249-50, LACM 48540, 48559- 66, TCWC 27912, 28107-14. 7.2 km W Naupe LACM 48543-46, TCWC 28117-20. 22 km E Olmos LACM 48541-42, TCWC 28115-16. Piura: Bayo- var FMNH 9803. 1 km S Bayovar, Cerro Illescas MVZ 85268, 85290-91, 85293-304. 4 km SSE Bayovar MVZ 82203-05, 82230. 5 km SSE Bayovar MVZ 85287-89. 12 km S, 8 km E Bayovar MVZ 82176. Cerro Amotape MVZ 82193-94, 82210-16, 82233-37, 82239-40, 82242, 82281, 82284. Chongollapi AMNH 28465. 8 km S, 8.4 km SSE Chulucanas LACM 48547, TCWC 28121. Fondo MVZ 82274. 2 km NE Las Lomas MVZ 85259, 85278- 84. 3 km ENE Las Lomas LACM 48556-58, TCWC 28131-34. 5 km E Las Lomas MVZ 82243-47, 82296. 30 mi S Lobitos BMNH 1932.9.5.1-2. 2 mi S Lobitos BMNH 1926.3.24.5. Negritos FMNH 8362. Paita AMNH 28460, 28462-64. Punta Aguja MVZ 82192, 82226, 82266-69. Reventazon MVZ 82229. 40 km WNW Sullana LACM 48548, TCWC 28122. Talara FMNH 57455. 2 km N Talara TCWC 28123. Trie Trac MVZ 82206, 85252-58, 85260-67, 85269-77, 85285, 85292, 85305-06. Tumbes: 1.2 km S Cancas LACM 48549-50, TCWC 28125-27. 1.2 km S, 1 km E Puerto Pizzaro LACM 48551-55, TCWC 28128-30. Quebrada Seca FMNH 9810, TCWC 28124. Phyllodactylus ventralis O’Shaughnessy Phyllodactylus ventralis O’Shaughnessy 1875. Ann. & Mag. Nat. Hist., (4)16:263. Phyllodactylus mulleri Parker 1935. Ann. & Mag. Nat. Hist., ( 10) 15:483. Phyllodactylus underwoodi Dixon 1962. Southwest. Nat., 7:218. 1970 Lizards of the Genus Phyllodactylus 43 75 70 65 R DIXON I 3 P. V ENTR ALIS 0 kilometers Figure 6. Distribution of P. dixoni and P. ventralis in northeastern South America. Holotype. Adult male, BMNH 53.6.1.33, collector unknown, date of collection unknown. Type locality. Island of Jamaica (in error). Definition. A large gecko with a maximum known snout-vent length of 75 mm; tibia and thigh tubercles present; ear denticulation usually present; enlarged preanal scale absent; terminal lamallae of digits distinctly widened and usually truncate at tips; median row of scales beneath tail distinctly wid- ened; dorsal rows of enlarged tubercles close together, trihedral, and strongly carinate; six to eight rows of tubercles across base of tail; rear of head with rounded tubercles dispersed among granular scales (fig. 14a). Diagnosis. Phyllodactylus ventralis is distinguished from heterurus, ger- rhopygus, and angustidigitus by the absence of an enlarged preanal scale; from inaequalis and microphyllus by the presence of regular rows of large trihedral, strongly keeled dorsal tubercles; from clinatus, lepidopygus, and interandinus by the presence of large tibia tubercles; from johnwrighti, pumi- lus and reissi by the presence of large tubercles on forearm; from sentosus by having terminal digital lamellae distinctly widened and enlarged; from 44 Contributions in Science No. 192 kofordi by greater maximum snout-vent length of 75 mm, rather than 46 mm; and 19 or more paravertebral tubercles between axilla and groin; from dixoni by having 24 or less scales across venter, rather than 26 or more, and dorsal tubercles four or more times the size of those in dixoni. Description. Rostral about one and one half times higher than wide; two internasals, bordered posteriorly by small granular scales and postnasal of each side; nostril surrounded by rostral, labial, two postnasals and internasal; ventral edge of nostril scale in broad contact with first labial; deep depression between internasals and in frontal region; scales of posterior loreal region six to seven times larger than midorbital scales; eye large, contained in snout length about two times; eye brille with one to two rows of granules and one larger outer row of scales, posterior half are pointed; ear diameter contained in eye diameter about three times; ear opening slightly denticulate anteriorly, none posteriorly; rear of head with very large tubercles dispersed among granular scales; supralabials usually six to seven and infralabials five to six to a point below center of eye; mental slightly longer than wide, somewhat triangular in shape. Dorsum with regular longitudinal rows of very large, trihedral tubercles, 10 to 14 rows reach nape, 6 to 8 rows reach base of tail; paravertebral rows of tubercles separated along median line by four to five rows of granules; postanal tubercles usually number three to four and very distinct in adult males. Forearm with large trihedral tubercles dispersed among granular scales; upper arm with large flat scales; dorsal surface of thigh and tibia with 10 to 20 very large tubercles dispersed among granular scales; terminal digital lamellae large, either rounded or truncate; claw short, hidden when viewed from below. Variation. Snout-vent lengths of adult males from 50 to 74 mm (62.4), adult females from 52 to 75 mm (62.3), subadults 40 to 45 (42.2), juveniles 24 and 29 mm; postmentals normally two, three occurring once in 22 speci- mens; postmentals contact first labial only in 25 per cent; first and second labials in 75 per cent of sample; scales bordering internasals vary from 4 to 6 (5.1); scales from nostril to eye vary from 9 to 12 ( 10.3); scales across snout at level of third labials vary from 17 to 23 (19.4); midorbital scales 12 to 17 (14.4); dorsum with 14 to 19 (16.4) longitudinal rows of enlarged trihedral tubercles; paravertebral tubercles from head to base of tail vary from 34 to 49 (41.2), axilla to groin 19 to 25 (22.1); venter scales from gular region to vent vary from 46 to 54 (52.0), across venter from 18 to 24 (20.6); fourth toe lamellae vary from 12 to 14 (13.6). Color pattern varies from seven to 10 narrow, undulating, dark brown cross bands on a light brown to tan ground color, to a linear series of 10 to 20 dark brown spots on a yellowish brown ground color; dorsal surface of head generally brown with or without dark brown spots or reticulating lines; 1970 Lizards of the Genus Phyllodactylus 45 side of head usually with a dark brown line from nostril through eye to arm insertion, bordered above and below by whitish lines or ground color; labials usually spotted or mottled with brown and dirty white; venter dirty white to yellowish brown. Comments. In 1962 one of us (Dixon) described a new species, Phyllo- dactyliis underwoodi, from the island of Grenada, Windward Islands. The holotype is the only available specimen and is unique in several features of scalation. Miss Alice G. C. Grandison (personal communication) informed us that there was an error in the labeling of this specimen, and it should have been labeled as Colombia. This specimen was a part of a relatively small collection made by F. A. Simons, purchased from a Mr. Janson by the British Museum of Natural History, and bore the locality New Grenada. Dixon (1962) noted the similarity of P. underwoodi to ventralis, but it differed in three major scale characters. It is now apparent that P. underwoodi is likely an aberrant ventralis, having three postmentals, 14 rows of enlarged dorsal tubercles, and six rows of tubercles across the base of the tail. No specimens of ventralis from Colombia or Venezuela have this combination of scale counts. However, only 22 specimens of P. ventralis have been examined, and additional material may reveal a wider range of variation. Habitat and natural history. This species has not been collected by either of us, therefore we rely upon information from the literature and from per- sonal observations of others. Ruthven (1922) records P. ventralis from houses, buildings, under stones and bark of trees. Two specimens were found in a dry thorn forest and one in a thorny scrub and cacti association under a stone. Mechler (1968) found three specimens in debris of an abandoned house in Bonda, Colombia. Janis A. Roze (personal communication) states that P. ventralis occurs in dry, semixerophytic areas, but occasionally pene- trates more moist habitats where cactus and thorny vegetation is replaced by a rich variety of shrubs and trees. Roze ( 1964) states that P. ventralis is often found in occupied houses, in abandoned buildings, and trees near buildings. Roze indicated that competition between Thecadactylus rapicauda and P. ventralis for human dwellings is being won by T. rapiciiada. Distribution. This species is found in the dry thorn forests and savannahs of northern Venezuela and Colombia (fig. 6). Specimens (22) have been examined from the following localities: COLOMBIA. Magdalena: Bolivar UMMZ 54737. “Grenada” BMNH 1880.2.26.6. Santa Marta UMMZ 48177. Santa Marta Mountains MCZ 24881, UMMZ 45462-63. VENEZUELA. Anzodtegui: Barcelona MBUCV 3144. A ragua: Maracay MCZ 39980. Carabobo: Caizuire MBUCV 3151. Isla de Burro MBUCV 3143. Falcon: Coro MBUCV 3148. Gudrico: San Juan de los Morros MBUCV 3149. Sosa MBUCV 3145. Portuguesa: Acarigua MCZ (unnumbered). Aparicion MBUCV 3146. Sucre: Isla Patos BMNH 1935.10.1. 10-11, 1940.2.15.1, 1945.11.1.41-42. Puerto de Hierro MCZ 50744-45. 46 Contributions in Science No. 192 Phyllodactylus dixoni Rivero-Bianco and Lancini Phyllodactylus dixoni Rivero-Bianco and Lancini 1968. Mem. Soc. Cienc. Nat. La Salle, 27:168. Holotype. Adult female, MBUCV III-8.450, collected by Juhani Ojasti in a cave on 5 October 1964. Type locality, Desembocadura del Rio Parguaza, afluente del Orinoco Medio en el Estado Bolivar, Venezuela. Definition. A large gecko with a maximum known snout-vent length of 76 mm; tibia, thigh, and forearm tubercles present; ear with little or no den- ticulation; enlarged preanal scale absent; terminal lamellae of digits distinctly widened and truncate; median row of scales beneath tail distinctly widened; dorsal rows of enlarged tubercles somewhat small, low and slightly trihedral; each dorsal row of tubercles separated from the other by three to five rows of granules; six to eight rows of dorsal tubercles across base of tail; rear of head with rounded tubercles dispersed among granular scales, almost subequal in size (fig. 14b). Diagnosis. Phyllodactylus dixoni may be distinguished from all other mainland species except ventralis, by having no enlarged preanal scale; regu- lar dorsal rows of keeled tubercles; presence of tubercles on forearm, thigh and tibia; distinctly enlarged and widened terminal digital lamellae; large maximum snout-vent length of 76 mm; 23 or more paravertebral tubercles between axilla and groin; from ventralis by having dorsal tubercles about one fourth the size of those of ventralis; 26 or more scales across the venter rather than 24 or less. Description. Rostral twice as wide as high, its dorsal edge with median vertical groove about one half depth of rostral; two internasals, bordered posteriorly by small granules and postnasal of each side; nostril surrounded by rostral, labial, internasal and two postnasals; first labial in broad contact with ventral edge of nostril scale; shallow depression between internasals, moderate depression in frontal region; scales in posterior area of loreal region about three to four times larger than scales in midorbital region; eye large, its diameter contained in snout length about 1.5 times; eye brille with two rows of granules and one larger outer row of scales, last five to eight pointed; ear diameter contained in eye diameter about two times; ear opening without denticulation, scales on posterior and anterior borders rounded; rear of head with granular scales with slightly larger rounded and elevated scales inter- mixed; supralabials six to seven and infralabials five to six to a point below center of eye; mental bell shaped, about as wide as long; two postmentals, slightly longer than wide and in contact on median line; postmentals contact only first labial. Dorsum with regular rows of enlarged, keeled tubercles that are some- what oval in outline; paravertebral rows separated from each other by four to five rows of granules; 10 rows of dorsal tubercles reach nape, 6 to 8 reach 1970 Lizards of the Genus Phyllodactylus 47 base of tail; postanal tubercles number three to four on each side in males; venter scales abruptly differentiated from lateral body granules. Dorsal surface of upper arm with flattened scales; posterodorsal surface of forearm with tubercles interspersed among small scales; dorsal surface of thigh with 15 to 25 tubercles dispersed among smaller flattened scales; tibia with large tubercles dispersed among granular scales; terminal lamellae large, somewhat truncate, about twice as long as wide; claw long, visible when viewed from below. Variation. The known snout- vent lengths of adults range from 56 to 76 mm (67.8), subadults 39 and 46 mm, juvenile lengths unknown; scales bordering postmentals vary from 5 to 8 (6.1); scales from nostril to eye vary from 11 to 12 (11.3); scales across snout at level of third labials vary from 17 to 20 (18.6), midorbital scales 15 to 19 (17.5); scales bordering inter- nasals vary from 5 to 6 (5.1); longitudinal rows of enlarged dorsal tubercles vary from 13 to 16 (14.9); number of paravertebral tubercles from head to base of tail vary from 41 to 50 (46.7), axilla to groin 23 to 26 (24.7); venter scales from gular region to vent vary from 59 to 66 (61.7), across venter 26 to 28 (27.1); fourth toe lamellae vary from 11 to 13 (11.3). Dorsal color pattern usually consists of five to eight undulating, dark brown cross bands on a brownish gray ground color from occiput to base of tail; tail banded with dark brown and light gray in juveniles, sometimes irregu- lar bands or spots in adults; head usually brownish gray with a suffusion of dark brown spots; labials spotted with dark brown; venter usually brownish cream; lower sides of body brownish yellow; ventral surfaces of toes dark brown. Habitat and natural history. Dr. Janis Roze (personal communication) states that the type series of P. dixoni was taken from beneath flakes of black basaltic rocks. The flakes were associated with immense, grayish black rock bluffs some 500 meters from the mouth of the Rio Parguaza. The immediate vicinity was essentially void of vegetation except for a few scattered shrubs and bromeliads. During the day, the temperature of the rocks reached 60°+C and the lizards took refuge beneath flakes or within cracks of boulders. The majority of lizards were taken by breaking off the rock flakes or by beating upon the flakes until the lizards ran out. Some lizards were observed running about in the late afternoon in the shade of scant vegetation or in the shade of rock boulders (Roze, personal communication) . Distribution. Phyllodactylus dixoni is known only from the type locality. The type locality is approximately 35 km NE of Puerto Carreno, Colombia, and 385 km (airline) south of San Juan de los Morros, Venezuela, the nearest locality of P. ventralis (fig. 6) . Seven specimens have been examined from the following locality: VENEZUELA. Bolivar: mouth of Rio Parguaza, on the Rio Orinoco MCNC 5479-80, MBUCV III 8451-54, TCWC 23827. 48 Contributions in Science No. 192 Phyllodactylus sentosus sp. nov. Phyllodactylus phacophorus, (part) Boulenger 1885, Cat. Lizards in the British Museum (Natural History) 2:84. Holotype. Adult female, TCWC 27913. Collected by Fortunato Blancas, November 1949; Type locality, Lima, Department of Lima, Peru. Definition. An average-sized gecko with maximum snout-vent length of 56 mm; large preanal plate absent; very large trihedral tubercles present on dorsum, forearm, thigh, tibia, and tail; ear denticulate on anterior and pos- terior margins; median row of scales beneath tail with few scales slightly enlarged, rounded; tail 45 to 50 per cent of total length; terminal lamellae of digits separated along median line, small, claw visible between and beyond tip of terminal lamellae when viewed from below (fig. 12a). Diagnosis. This species may be distinguished from all other western South American species except P. kofordi, by the presence of very large tubercles on dorsum, thigh, tibia, forearm, and tail. Phyllodactylus sentosus differs from kofordi by having very small terminal lamellae on the digits and generally uniform sized scales beneath ventral surface of tail, rather than terminal lamellae large and all scales of median row beneath tail enlarged, rectangular. Description of holotype. Rostral completely divided by vertical groove, twice as wide as high, its dorsal and lateral edges straight; two internasals, some- what rounded, their median edges in narrow contact, bordered posteriorly by four granules and supranasal of each side; nostril surrounded by rostral, labial, supranasal, and one postnasal; first supralabial in narrow contact with ventral edge of nostril; no depression between internasals, slight depression in frontal region; 9 scales between eye and nostril, scales in posterior loreal region about three to four times larger than scales in midorbital region; 18 scales across snout at level of third labials, 12 scales across head between anterior edge of eyes, 14 midorbital scales; eye large, its diameter contained in snout length 1.7 times; eyelid with one to two rows of granules and one larger outer row of scales, last four to five are pointed; ear diameter contained in eye diameter 2.3 times; ear opening denticulate, scales on anterior and posterior margins pointed; rear of head granular with many larger intermixed tubercles; six supralabials and five infralabials to a point below center of eye; mental bell- shaped, slightly longer than wide, bordered posteriorly by three postmentals; postmentals rounded, slightly longer than wide, their median edges in broad contact, followed by transverse row of eight small scales, followed by second row of 1 1 smaller scales; postmentals contact first labial. Dorsum with 16 longitudinal rows of enlarged, trihedral tubercles, paravertebral row with 26 tubercles from rear of head to base of tail, 13 between axilla and groin; paravertebral rows separated from each other by three rows of granules; 10 rows of tubercles reach to rear of head, 8 to base of tail; each tubercle of enlarged dorsal series separated from preceding tubercle by one to two granules; postanal tubercles number three on each side; 1970 Lizards of the Genus Phyllodactylus 49 venter with 18 longitudinal and 48 transverse rows of scales; tail with 8 rows of tubercles in proximal eight whorls, distal one-third of tail regenerated. Dorsal surface of upper arm with flattened scales, forearm with larger tubercles interspersed among smaller flattened scales; dorsal surface of thigh with 8 to 9 tubercles interspersed among granular scales, tibia with 13 to 15 tubercles; lamellae formula for hand 6-8-12-11-8, foot 7-10-11-12-10; claw largely visible between and beyond tip of terminal lamellae when viewed from below; terminal lamellae small, about three times as long as wide. Measurements in mm: Snout-vent length 50.0; axilla-groin length 25.0; leg length 19.3; arm length 16.0; tail length 46.3; head length 9.6; head height 7.8; head width 9.6; snout length 5.1; distance from eye to ear 4.6; eye diam- eter 3.0; ear diameter 1.3. Color in alcohol: ground color fawn; dorsum with nine undulating cross- bands of reddish brown, first two crossbands complete, remainder broken along midline of dorsum by cream middorsal stripe that begins about level of shoulder; each dorsal tubercle generally flecked with dark brown; limbs with brownish spots, forming faint reticulations on ground color; tail with six brown crossbands, about one-half width of ground color interspaces; distal one-third of tail regenerated; head brownish with wash of light brown; dark brown stripe from nostril to posterior edge of ear, bordered above by cream line on snout, above and below by cream line between eye and ear; ventral surfaces pale white, scales below tail with flecks of brown. Variation. Snout-vent length of two males 49 and 56 mm (52.5); two females 39 and 50 mm (44.5); postmentals number two in three specimens, three in one; postmentals contact first labial only in all; number of scales across chin immediately following postmentals varies from 6 to 8 (7.3); number of scales bordering posterior edge of internasals vary from 5 to 7 (6.3), with internasals completely separated by an azygous scale in two of four specimens; scales across snout at level of third labial varies from 16 to 18 (16.5) midorbital scales vary from 13 to 14 (13.3); scales across venter vary from 18 to 21 ( 19.3) ; longitudinally 47 to 55 (50.3) ; number of enlarged rows of dorsal tubercles vary from 14 to 16 (15.0); number of paravertebral tubercles from rear of head to base of tail vary from 26 to 31 (28.8); from axilla to groin 13 to 16 ( 15.0) ; scales along line from nostril to eye vary from 9 to 10 (9.3); number of fourth toe lamellae vary from 12 to 13 (12.5); terminal lamellae of all digits very small, twice as long as wide; claw visible beyond tip of pad when viewed from below; number of enlarged rows of tubercles at base of tail number 8 in three specimens, 6 in one; 8 tubercles occur in each of 8 to 1 1 whorls on proximal two-thirds of tail, reducing rapidly in sequence to 6-4-2-0 on distal third of tail; median row of scales beneath tail generally uniform in size to other scales but occasionally a larger, rounded scale may be present. Number of dorsal body bands varies from 8 to 10 (9.0), with majority of bands broken along midline of body in three specimens, bands loosely con- 50 Contributions in Science No. 192 nect and form reticulated pattern in one; tail bands number 13 and 14 in two specimens with complete tails; venter pale white to dirty cream; dark lateral face stripe continues to shoulder region as dashed line in two specimens, reaches to ear in two. Comments. Phyllodactylus sentosus closely resembles kofordi (NW Peru), darwini (Galapagos Islands), wirshingi (Puerto Rico), and martini (Dutch Leeward Islands) in general appearance. The latter four species have enlarged trihedral tubercles on the dorsum, tail, thigh, tibia, and forearm. However, the four species have the median row of scales beneath the tail much enlarged and widened, and have moderately large, well defined ter- minal lamellae on all digits. In addition, martini and wirshingi have higher numbers of enlarged rows of dorsal tubercles ( 18 to 20) and higher num- bers of paravertebral tubercles from head to base of tail (32 to 40). The num- ber of enlarged dorsal rows of tubercles in darwini are identical with those of sentosus, but darwini has higher numbers of paravertebral tubercles from head to base of tail (33 to 38) and differs in several other features of squa- mation. Natural history. The only known habitat of this species is the grounds surrounding the University of San Marcos in Lima. Dr. Gambini ( personal communication) indicated that she found one specimen running across a side- walk on the campus just after dark. Distribution. Known from the University of San Marcos campus, Lima, Peru, and from the vicinity of the city (fig. 5). Six specimens have been examined: Lima: Lima BMNH 63.2.3.18-19, TCWC 27913, 28329-31. Phyllodactylus reissi Peters Phyllodactylus reissi Peters, W. 1862. Monatsb. Konigl. Akad. Wiss. Berlin, Nov., 1862, pp. 626-27. Phyllodactylus baessleri Werner 1901. Abh. u. Ber. d.k. Zool. u. Anthr.- Ethn. Mus. zu. Dresden, 9(2) : 2. Phyllodactylus guayaquilensis Werner 1910. Mitt. Naturh. Mus. Ham- burg, 27(2): 4. Phyllodactylus abrupteseriatus Werner 1912. Mitt. Naturh. Mus. Ham- burg, 30:4. Phyllodactylus magister Noble 1924. Occ. Pap. Boston Soc. Nat. Hist., 5:110. Syntypes. Berlin Mus. 3734 (4) , 4567 (2) ; BM 4567, ad. female, snout- vent 64 mm, herein selected as the Lectotype; Carl Reiss, collector; date of collection unknown. Type locality: Guayaquil, Ecuador. Definition. A large gecko, maximum snout-vent length of 75 mm; femur tubercles absent, tibia tubercles present; ear strongly denticulate on anterior and posterior margins; enlarged tubercles on rear of head present; median row of scales beneath tail distinctly widened; enlarged preanal plate absent; 12 to 18 rows of enlarged trihedral tubercles on dorsum; tail about 48 to 58 1970 Lizards of the Genus Phyllodactylus 51 per cent of total length, without whorls of enlarged tubercles; terminal lamellae of digits moderately enlarged, truncate; dorsal color pattern variable, uniform in color to bold bands; venter usually yellowish (fig. 1 3b) . Diagnosis. This species may be distinguished from P. heterurus, angusti- digitus, and gerrhopygiis by the absence of enlarged preanal plate; from lepidopygus, microphyllus, clinatus, and inaequalis by the presence of tibia tubercles and larger size; from interandinus by its larger size (average s-v of 58.3 mm rather than 40.1 mm), fewer number of head to tail tubercles in paravertebral row, 47 to 60 rather than 64 to 8 1 ; from kofordi by its larger size (as above) rather than average of 44.0 mm, and by having more head to tail tubercles in paravertebral row (as above) rather than 3 1 to 36; from johnwrighti and pumilus by larger size, a maximum snout-vent length of 75 mm rather than 5 1 mm; thigh tubercles absent, rather than present; from sentosus by absence of tail and forearm tubercles, larger scales in median row beneath tail, larger terminal lamellae, and color pattern. Description. Rostral about twice as wide as high; internasals rounded, their median edges in broad contact, bordered posteriorly by small granules and postnasal of each side; nostril surrounded by rostral, labial, internasal, and two postnasals; first labial in broad contact with ventral edge of nostril; shallow depression between internasals, deep depression in frontal region; scales in posterior loreal region about 3 to 4 times larger than midorbital scales; eye large, its diameter contained in snout length about 1.8 times; eyelid with 2 rows of granules and 1 larger outer row of scales, last 3 to 6 of which are pointed; diameter of ear contained in eye diameter about 2.1 times; ear opening strongly denticulate with pointed scales on the anterior and posterior margins; rear of head granular with larger intermixed tubercles; 6 to 7 supra- labials and 5 to 6 infralabials to a point below center of eye; mental bell shaped, as wide as long, bordered posteriorly by two to four postmentals; postmentals slightly longer than wide, their median edges in broad contact. Dorsum with longitudinal rows of enlarged trihedral tubercles that are somewhat elevated; paravertebral rows separated from each other by 4 to 5 rows of granules; 6 rows of tubercles reach to nape, 6 to base of tail; each tubercle of enlarged dorsal series separated from preceding tubercle by 0 to 2 granules; postanal tubercles number 4 on each side, distinct and slightly elevated posteriorly in males; dorsal surface of upperarm with flattened scales, forearm with slightly larger flattened scales; dorsal surface of thigh without tubercles intermixed among smaller flattened scales; tibia with large tubercles scattered among granular scales; claw slightly visible when viewed from below; terminal pad large, slightly longer than wide, somewhat truncate at tip. Variation. Snout-vent length of subadult and adult males range from 42 to 75 mm (59.4), females from 37 to 73 mm (57.5), juveniles from 22 to 35 mm (28.9), and males plus females average 58.3 mm; 48 per cent of total sample (753 specimens) were 60 mm or more in snout-vent length; post- mentals vary from two to four, two occurring 97.4 per cent of the time, three 52 Contributions in Science No. 192 1.9 per cent and four 0.7 per cent; postmentals contact the first labial only in 96.8 per cent of sample, first and second labials 3.2 per cent; number of scales immediately following postmentals across chin vary from 5 to 8 (6.3); scales across midorbital region vary from 16 to 24 (20.2) ; scales across snout at level of third labial vary from 18 to 27 (23.7); scales bordering internasals posteriorly vary from 5 to 9 (6.7) with an azygous scale separating the inter- nasals in 23.1 per cent of sample; transverse rows of scales across venter vary from 24 to 37 (29.7), longitudinally 54 to 77 (60.0); dorsal tubercular rows vary from 12 to 18 (13.2); paravertebral tubercles from head to tail vary from 47 to 60 (53.0); between axilla and groin 25 to 38 (31.3); scales along a line between nostril and eye vary from 10 to 16 (12.5); lamellae beneath fourth toe vary from 11 to 17 (13.5). Dorsal color pattern highly variable, ranging from uniform yellowish tan without darker markings to light gray ground color with bold black undu- lating crossbands half the width of ground color interspaces; those with dark dorsal markings on ground color may have spots, crossbands, blotches, or lines that are either bold black, dark brown, brown, to indistinct grayish brown; in a few specimens the lines coalesce, forming a variegated pattern; when body markings are distinct enough to be counted, they generally vary from 5 to 7 pairs or single bands. Coloration of dorsal surface of tail somewhat similar to that of dorsum; when bands are present they vary from 8 to 11 with 9 and 10 being most common; interspaces usually lighter than those of dorsum, tending to be more in contrast with darker bands; dorsal surface of head is either uniform in color, spotted or reticulated with dark brown to black lines or spots on ground color; side of head with or without distinct dark line from nostril through eye to arm; limbs similar to dorsum in color and pattern, with fore- arm and lower leg generally banded with dark on ground color in most specimens. Ventral surfaces generally grayish white in alcohol (uniform pale yellow to tan with yellowish cast in life), with dark color of sides encroaching on venter in large individuals; ventral surface of tail usually darker than venter. An examination of three populations of P. reissi from the coastal and coastal foothills of Ecuador, desert foothills surrounding the Sechura Desert of Peru, and the Maranon-Chinchipe Valley of northern Peru reveal very little variation in the majority of characters. The Ecuadoran population has an azygous scale separating the internasals in 43.1 per cent of the sample, 32.3 per cent in the Sechura Desert sample, and 14.3 per cent in the Maranon- Chinchipe sample. The number of enlarged dorsal tubercular rows vary from 12 to 16 (14 the common number), 12 to 16 (12 common), and 14 to 18 (16 common) respectively, in the three samples indicated above. Comments. The types of P. baessleri, abrupteseriatus, and quayaquilensis were destroyed during World War II, but the characters mentioned in the 1970 Lizards of the Genus Phyllodactylus 53 original descriptions of the latter two species fall within the variation present in reissi. An analysis of the characters of a large sample of magister from the Marafion and Chinchipe Valleys of Peru indicates only minor differences between it and populations of reissi from Ecuador and northern Peru. We, therefore, consider abrupteseriotus, guayaquilensis, and magister as subjective synonyms of reissi. Werner’s (1901) original description of P. baessleri is the only evidence we have for comparison of his species with those known to occur in Peru. His description does not correspond in its entirety to any population of Phyl- lodactylus we currently recognize in Peru. A translation of Werner’s (1901, p. 2) original description follows: “Closely related to P. tuberculosus, snout one and a half times as long as eye, somewhat longer than distance between eye and ear; frontal region with slight concave; ear opening diagonal, small diameter of ear one half that of eye; snout with large flattened, rounded scales, those of the rear of the head smaller, here they seem to differ by being mixed with small rounded tubercles; ear opening with three pointed scales on the anterior margin; rostral four-sided, the posterior margin with a short median notch; nostril surrounded by first labial and three nasals, of which the upper one contacts the upper nasal of the other side; 7-8 upper labials, 6th below eye; 7 infralabials; mental large, bell-shaped, followed by two small polygonal postmentals in contact with each other; followed by successfully smaller scales in size; upper side of rump with 12 rows of strongly keeled tubercles, tubercles of each row close together; the two outer rows near the venter do not reach the forelimb; venter scales flat, overlapping, cycloid-hexagonal; tibia with enlarged keeled tubercles on the upper side; finger and toe ending with strongly enlarged lamellae, similar to P. tuberculosus; 4th toe with 16 lamellae, of which the distal one and mostly proximal ones are divided. Color above, light grayish brown with black spots on head, tail, base of tail and legs; four long, black, parallel lines, from rear of eye to base of fore- leg, two from back of head to middle of back, between the limbs; underside whitish. Total length 91 mm, most of tail regenerated (39 mm); The single example on hand (D 1780 M. Dr.) came from Chanchamayo. The species is named in honor of Prof. A. Baessler.” From Werner’s description (1901), we may eliminate the Peruvian species without strongly keeled tubercles on the dorsum, i.e. P. microphyllus, inaequalis, gerrhopygus, angustidigitus, and heterurus. His description men- tions the presence of tibial tubercles, therefore we may eliminate clinatus, interandinus, and lepidopygus. The remaining species, reissi, johnwrighti, sentosus, and kofordi agree in part with his original description. The latter four species have denticulate scales on the anterior and poste- rior margins of the ear, tibia tubercles, and strongly keeled dorsal tubercles. 54 Contributions in Science No. 192 Of these four species, only P. reissi has a fourth toe lamellae count ( 12 to 17) that falls within the range given by Werner for baessleri. Occasionally, P. interandinus has a few small tubercles on the tibia. The color description given by Werner ( 1901 ) does not agree with any species of Phyllodactylus in Peru, while all other features mentioned in his original description agree with most of the species. Werner’s type locality, Chanchamayo, is also questionable. Chanchamayo is a well known, tropical lowland locality at the headwaters of the Rio Perene on the east side of the Andes. There are no known species of Phyllodactylus inhabitating the selva of eastern Peru. We provisionally place P. baessleri as a subjective synonym of reissi, based upon the destruction of the holotype, a possible error in the type locality, and discrepancies in the original description when compared with known species in Peru. Habitat and natural history. Phyllodactylus reissi is the most abundant scansorial gecko of northern Peru. Daytime retreats consist of any object that affords protection from predators and adverse temperatures. Individuals were found beneath bark of living, dead, standing, or prone trees, fence posts, fallen banana plants, standing and prone cacti, cracks in adobe walls, crevices in boulders, knot holes in trees, beneath bridges, rotting tree stumps, beneath palm leaves, construction materials and fruit crates. Their nocturnal activities were essentially on vertical surfaces, i.e., trunks of trees, walls of buildings, boulders, cacti, and fence posts, but occasionally were found actively foraging on the ground. Several were found foraging on open ground two to three meters from large boulders and in leaf litter in cacao groves. This species is always found in association with some type of plant cover. When exposed to daylight, P. reissi actively seeks a dark place to hide. When exposed in its daytime retreat, it hesitates only a few seconds before rapidly moving to another dark retreat. Its movements are usually vertical, but on horizontal limbs or tree trunks reissi usually runs to the ventral surfaces. Two eggs are laid and usually deposited in rotting debris at the base of trees, beneath leaf litter in forests, in rotting agave stumps, hollow branches of dead cacti filled with debris, or in any rotting debris associated with its general habitat. Phyllodactylus reissi has been found occupying the same cactus plant with interandinus in the upper Amazon Basin, and with kofordi in north- western Peru. It has been found on the same tree with Pseudogonatodes bar- bouri and Gonatodes atricucullaris, and in the same general habitat with inaequalis, clinatus, and microphyllus. As in all of the species of Phyllodactylus found in Peru, the nocturnal body temperatures of reissi closely approximated that of the substratum upon which it was active. Distribution. This species is found from sea level to about 2,000 m in southern Ecuador and northwestern Peru. Its principal habitat seems to be 1970 Lizards of the Genus Phyllodactylus 55 arid tropical scrub west of the Andes, and cacti dominated areas of the north- ern interandian basins of Peru (fig. 5) . Specimens (772) have been examined from the following localities: ECUADOR, 4 km E San Francisco USNM 167183 (23). El Oro: Machacha USNM 167182. Santa Rosa AMNH 22068. Guayas: Ancon Camp BMNH 1931.10.21.1-2. 6 km SW Colonche USNM 167189. Daule River MCZ 4740 (2). Guayaquil AMNH 21845. ANSP 7526. BM 3734 (4), 4567 (2). CMF 16504, 16540-46, 16658-61, 16676, 16695, 16700, 16850-51, 16889-900. EPN 4618 (18). FMNH 61210. HM 3386. Playas MVZ 77217- 20, 77268. USNM 167181. Poza Huancauilca CMF 16512, 16515. Puente Chimbo AMNH 24343. Puna CAS 84770. 1 km NW Punta Cornero USNM 167190 (6). 0.5 km N Punta Cornero USNM 167191 (10). Salinas CMF 16531. Loja: BMNH 1930.10.12.13. Rio Casanga AMNH 18304-05. Manahi: Manta CMF 16522; FMNH 53896-98. 1 km S Manta USNM 167186 (5). 6 km E Manta, 5 km W Montecristi USNM 167188 (8) . No specific locality in Ecuador AMNH 64569 (4); ANSP 7530; USNM 14052-53. PERU. A mazonas: 14 km S Bagua Chica TCWC 28259. 17 km S Bagua Chica LACM 48766-72, TCWC 28260-64. 19 km S Bagua Chica LACM 48739-58, 48786; TCWC 28228-49. 10 km WSW Bagua Grande MVZ 82301. 8 km WSW Bagua Grande MVZ 82208, 82302-06. 15.5 km WSW Bagua Grande LACM 48759-64. TCWC 28250-55. 3.5 km SE Bagua Grande LACM 48765. TCWC 28256-58. Cajamarca: Bellavista AMNH 28525-27, 39339-40. FMNH 73379. LACM 48734-38. MCZ 42233 (3), 18126-29, 18141-42 (+ 208 unnumbered). TCWC 24923, 28210-21. UIMNH 41515, 40747. UMMZ 59029 (4). 4.8 km S Bellavista LACM 48734-38. TCWC 28222-27. Perico FMNH 100182-83. MCZ 17974, 18145-50. Lamhayeque: 1 1 km N Jayanca MVZ 82197-201. 5 km SSW Motupe MVZ 82299. 7 km S Motupe LACM 48707-21, 48811-29. MVZ 82300; TCWC 28175-208. 15 km S Naupe MVZ 82202. 7.2 km W Naupe LACM 48773-74. TCWC 28265. 22 km E Olmos TCWC 28209. Liber tad: Chiclin FMNH 34266. Pacasmayo HM (3 unnum- bered); SMFM 8285, 8289. USNM 167180. Piura: Bayovar FMNH 9798- 802, 9804. 1 km S Bayovar (Cerro Illescas) MVZ 85308-09, 85316-28. 1.1 km E Bocana LACM 48810. TCWC 28278-79. 1 km N, 28 km E Cabo Blanco MVZ 82177, 82232, 82280. Capullana Point FMNH 41551. Catalinas FMNH 9816-17. Cerro Amotape MVZ 82169, 82173, 82195, 82217-18, 82238, 82241, 82282, 82285-95. 8 km S, 8.4 km ESE Chulucanas LACM 48775-81. TCWC 28266-71. 2.4 km SW Chulucanas LACM 48782, TCWC 28272-73. El Alto FMNH 41548, 41550. Fondo MVZ 82272-73. La Breita BMNH 1926.3.24.6. 5 km E Las Lomas MVZ 82219-20, 82297-98. 2 km NE Las Lomas MVZ 85310-15. 3 km ENE Las Lomas LACM 48802-09. TCWC 28297-304. 5 km E, 4.2 km NNE Lobitos LACM 48798-801. TCWC 28291- 96. Monte Grande MVZ 82279. Negritos FMNH 8364-65, 8384. Parinas MVZ 82207, 82276-77. Punta Sal SMFM 41913. Sechura Desert CAS 92346. 40 km WNW Sullana LACM 48783-84; TCWC 28274-75. 4 mi W Suyo 56 Contributions in Science No. 192 TCWC 24072-91. Talara AMNH 66600-01; FMNH 41547, 41549, 53895, 53892-93, 57456. 1 km SW Talara MVZ 82275. 2 km E Talara LACM 48785; TCWC 28276-77. Tamboa AMNH 28426-27. Trie Trac MVZ 85307. between Verdun Alto and Negritos FMNH 41546. Tumbes: Las Vocas USNM 38569. 1.2 km S, 1 km E Puerto Pizzaro LACM 48792-97; TCWC 28285-90. Que- brada Seca FMNH 9809, 9811-12; LACM 48787-91; TCWC 28280-84. no specific locality in Peru ANSP 13121; FMNH 41552-53; SMFM 8290. LEP1D0PY6US KOFORDl SENTOSUS MICROPHYLLUS J0HNWRI6HTI INTERANDINUS ANGUSTIDIGITUS GERRHOPYGUS REISSI CLINATUS INAEQUALIS Figure 7. Ventral view of the fourth toe of all species of Phyllodactyliis (except hetemms and pumilis) found on the mainland of western South America. Phyllodactylus johnwrighti sp. nov. Phyllodactyliis phacophorus, (part) Noble 1924, Occ. Paps. Boston Soc. Nat. Hist., 5:111. Holotype. Adult male, TCWC 27916, collected by James R. Dixon and John W. Wright, 28 November 1968. Type locality. 13 km SSE (by road) Hacienda Molino Viejo, 1190 m, Department of Cajamarca, Peru. 1970 Lizards of the Genus Phyllodactylus 57 Definition. A small gecko with a known maximum snout-vent length of 50 mm; enlarged preanal shield absent; dorsum and tibia with tubercles, absent from thigh, forearm, and tail; some scales of median row beneath tail widened and rounded; tail length about 49 per cent of total length; terminal lamellae moderately enlarged, rectangular; dark stripe from nostril to arm insertion always present (fig. 10b). Diagnosis. This species differs from P. angustidigitus, gerrhopygus and heterurus by the absence of enlarged preanal shield; from interandinus by absence of dorsal stripes, presence of tubercles on tibia, and maximum num- ber of 49 tubercles in paravertebral row from head to tail, rather than mini- mum of 65 in interandinus; from kofordi and sentosus by absence of forearm, tail, and thigh tubercles; from reissi by smaller maximum size, 50 mm rather than 75 mm, and absence of wide, rectangular row of scales beneath tail; from microphyllus and inaequalis by presence of regular longitudinal rows of enlarged trihedral tubercles on dorsum, from lepidopygus and clinatus by presence of tibia tubercles; from pumilus by absence of tail tubercles and smaller supraorbital scales. Description of holotype. Rostral rectangular, twice as wide as high, its dorsal edge with vertical groove one-half depth of rostral; two internasals, subrectangular, their median edges separated by two granules; internasals bordered by rostral, nostril, six granules and postnasal of each side; nostril surrounded by rostral, labial, internasal, and two postnasals; first labial in broad contact with ventral edge of nostril; slight depression between inter- nasals and in frontal region; 10 scales between eye and nostril; scales in posterior loreal region about 2.5 times larger than midorbital scales; 22 scales across snout at level of third labial, 14 scales across anterior edge of orbits, 20 midorbital scales; eye large, its diameter contained in snout length 1.6 times; eyelid with two rows of granules and one larger outer row of scales, last three to four are pointed; ear diameter contained in eye diameter 2.4 times; ear opening slightly denticulate, scales on anterior and posterior mar- gins small, rounded and subequal; rear of head granular with intermixed tubercles; six supralabials and five infralabials to point below center of eye; mental bell-shaped, as wide as long, bordered posteriorly by two postmentals; postmentals rounded, slightly larger than other chin scales, their median edges in broad contact; postmentals immediately followed by transverse row of 7 scales, followed by second row of 10 smaller scales; postmentals contact first labial only of each side. Dorsum with 14 longitudinal rows of enlarged trihedral tubercles; 10 rows reach to nape, 6 to base of tail; paravertebral row of 45 tubercles from head to tail, 26 between axilla and groin; paravertebral rows of tubercles separated from each other by five rows of granules, each tubercle separated from preceding tubercle by one to two granules; postanal tubercles three on each side; 26 scales across venter, 55 from throat to vent; tail without tuber- cles, median row of scales beneath tail with every other scale distinctly 58 Contributions in Science No. 192 a Figure 8. (a) dorsolateral view of P. angustidigitus; (b) ventral view of preanal plate of P. angustidigitus. 1970 Lizards of the Genus Phyllodactylus 59 widened; arms and thigh with small flattened scales; thigh with five to six tubercles intermixed among scales; tibia with 17 to 19 tubercles intermixed among granules; lamellae formula for hand 6-8-9-10-8, foot 7-9-1 1-12-11; claw slightly visible when viewed from below; terminal lamellae large, rectan- gular, longer than wide. Measurements in mm: Snout-vent length 39.0, axilla-groin length 17.4, leg length 14.8, arm length 11.8, tail length 34.0, head length 11.1, head depth 4.5, head width 7.6, snout length 4.2, eye diameter 2.7, ear diameter 1.1, distance from eye to ear 3.3. Color in life: Ground color grayish tan; dorsum without distinct bands; grayish brown spots on anterior half of body, fading posteriorly; grayish yellow median stripe from nape to base of tail; posterior edge of scales on limbs stippled with dark brown, rest of scale ground color; head with diffuse spotting of grayish brown; grayish brown line from nostril to above arm inser- tion, bordered above by dirty cream line from nostril to ear; tail with 2 brownish black bands with yellowish brown interspaces; ventral surfaces dirty gray, with three to five black dots on each scale, more dense on tail. Variation. Snout-vent lengths of subadult and adult males range from 32 to 40 mm (37.9), females from 33 to 44 mm (39.0), juveniles 21 to 31 mm (26.4) , males plus females average 38.4 mm; tail length ranges from 46 to 51 per cent (48.8% ) of total length; postmentals vary from two to four, three occurring twice and four once in 40 specimens; postmentals always contact first labial; number of scales immediately following postmentals vary from 5 to 8 (6.2) ; midorbital scales vary from 17 to 21 ( 18.8), scales across snout at level of third labials 20 to 26 (21.7) ; scales bordering internasals vary from 6 to 9 (7.5), internasals separated along median line by granules in 97.7 per cent of sample; scales from nostril to eye vary from 10 to 12 (11.1); scales across venter range from 26 to 32 (27.4), from throat to vent 47 to 57 (51.8); dorsal tubercular rows vary from 12 to 15 (13.9); paravertebral tubercles from head to base of tail range from 40 to 49 (45.0), between axilla and groin 22 to 29 (26.4) ; lamellae beneath fourth toe range from 10 to 12 ( 1 1.4) ; tubercles absent from tail. Dorsal pattern generally uniform tan to pinkish tan, few individuals with double row of 9 to 12 small, grayish brown spots or transverse lines; tail usually banded with 1 1 to 16 black bands, with ground color interspaces of equal width; tail bands of juveniles usually bright orange and black, subadults yellowish orange and black, and adults yellowish brown and black; iris of eye pinkish gold, ventral surfaces generally dirty white in all specimens. Comments. A distinct population of this species occurs on the east slope of the Andes at an elevation of 2100 m. One specimen of this population was taken by Noble from near the headwaters of the Rio Huancabamba in 1916, and three individuals were taken by us from near the summit of the Andes on the Olmos to Bagua road in 1968. These four specimens are similar in squama- tion to our series from near Molino Viejo, but differ in color pattern and 60 Contributions in Science No. 192 have a more decidedly depressed head and body. One male measured 50 mm and a female 46 mm in snout-vent length. The dorsal pattern consists of a fine reticulation of dark lines on ground color with a light median stripe. The lateral dark stripe extends from the nostril to midbody. Until additional speci- mens of this population are available, we are not in a position to analyze this population further. This species is named in honor of Dr. John W. Wright, who discovered the original population along the Rio Huancabamba near Molino Viejo, and who provided us with an immeasurable amount of help in the field. Habitat and natural history. This species occupies two distinct vegeta- tion communities, the high, barren eastern slopes of the Andes where only a few cacti and shrubs are present, and a more extensive community of cacti, shrubs, mesquite, and broad leafed trees along the Rio Huancabamba and associated hillsides. The specimens from 2100 m were taken beneath dead Cereus sp. and a cholla-like cactus. Those along the Rio Huancabamba, taken during the day, were found beneath the bark of mesquite or under fallen trees. After dark, individuals were found emerging from beneath fallen Cereus sp. and from their branches as high as two meters above the ground. Later in the evening about 20 individuals were found foraging around the bases of small rocks, Opunta sp., weeds, and on open gravel soil. When disturbed, most individuals sought cover in small patches of weeds or among small rocks. Their nocturnal body temperatures averaged 23 °C and closely approximated that of the substratum. Eggs of this species were found in rotting stumps of dead trees. An examination of the oviducts of preserved females indicates that this species lays one egg per clutch. Distribution. This species has been found only on the arid slopes of the Rio Huancabamba Valley and higher barren slopes of the east side of the Andes in northwestern Peru (fig. 4). Specimens (41 ) have been examined from the following localities. PERU. Cajamarca: El Arenal MVZ 82196. 4.5 km W El Tambo TCWC 28088. 23.5 km WNW El Tambo TCWC 28104-05. 25.2 km WSW El Tambo TCWC 28106. 13 km SSE Molino Viejo LACM 48476, 48479-94, TCWC 27916, 28089-90, 28092-103. 11 km WNW Las Juntas LACM 48477-78, TCWC 28091. Piura: Huancabamba AMNH 28466. Phyllodactylus pumilus sp. nov. Holotype. Adult male, USNM 167227, collected by James A. Peters, October 1965. Type locality. 1 km S Manta, Department of Manabi, sea level, Ecuador. Definition. A small gecko with a known maximum snout-vent length of 51 mm; enlarged preanal shield absent; dorsum, tibia, and part of tail with tubercles; scales of median row beneath tail widened and rounded posteriorly; tail about 50 per cent of total length; scales in supraocular region largest of interorbital series; dark brown to black stripe from nostril to arm insertion 1970 Lizards of the Genus Phyllodactylus 61 always present; terminal digital lamellae moderately enlarged and somewhat rectangular in shape (fig. 14c). Diagnosis. This species differs from P. angiistidigitus, gerrhopygus, and heterurus by absence of enlarged preanal shield; from interandinus by absence of dorsal stripes and presence of tubercles on tail and tibia; from kofordi and sentosus by absence of forearm tubercles and tubercles on distal two-thirds of tail; from microphyllus and inaequalis by presence of regular rows of enlarged trihedral tubercles on dorsum; from lepidopygus and clinatus by presence of tibia turbercles; from reissi by smaller maximum snout-vent length of 51 mm rather than 75 mm, scales in supraocular region largest of inter- orbital series, and at least two rows of enlarged, flattened scales on proximal one-fourth of tail; from johnwrighti by presence of at least two rows of enlarged, flattened scales on proximal one-fourth of tail; more transverse Figure 9. (a) dorsolateral view of P. gerrhopygus', (b) dorsal view of P. micro- phyllus. 62 Contributions in Science No. 192 ventral scales from throat to vent, 53 to 62 (57.6) rather than 47 to 57 (46.4) ; fewer paravertebral tubercles from head to base of tail, 37 to 43 (39.4) rather than 40 to 49 (45.0). Description of holotype. Rostral twice as wide as high, its dorsal edge with short median, vertical groove; two internasals, somewhat rectangular and in contact along median edges; nostril surrounded by rostral, labial, inter- nasal, and three postnasals; first supralabial in broad contact with ventral edge of nostril; shallow depression between internasals and in frontal region; 1 1 scales between nostril and eye; scales in posterior loreal region about three times larger than interorbital scales; 19 scales across snout at level of third labials, 17 interorbital scales, 18 at anterior edge of orbits; eye large, its diam- eter contained in snout length 1.7 times; eyelid (brille) with one to two rows of granules and one larger outer row of scales; diameter of ear contained in eye diameter about three times; ear opening strongly denticulate on anterior and posterior margins; rear of head granular with a few larger tubercles inter- mixed; six supralabials and five infralabials to point below center of eye; mental bell-shaped, as wide as long, bordered posteriorly by two postmentals; postmental about as long as wide, their median edges in broad contact; post- mentals immediately followed by transverse row of six scales, followed by second row of 10 smaller scales; postmentals contact first labial only. Dorsum with 14 longitudinal rows of enlarged trihedral tubercles that are somewhat flat; paravertebral row with 42 tubercles from rear of head to base of tail, 25 between axilla and groin; paravertebral rows separated from each other by four rows of granules; five rows of tubercles reach nape, six to base of tail; each tubercle of dorsal series separated from preceding tubercle by none to two granules; postanal tubercles number four to each side; venter with 27 longitudinal and 61 transverse rows of scales. Dorsal surface of arm with large, flat scales; dorsal surface of thigh with three to five tubercles scattered among granular scales; lamellae formula for hand 7-S-9-9-7, foot 7-8-12-12-10; claw slightly visible when viewed from below; terminal lamellae slightly longer than wide, moderately enlarged; distal three-fourths of tail regenerated, proximal one-fourth with a 6-4-2 reduction in number of tubercles per whorl, decreasing in number towards distal portion of tail; median row of scales beneath tail widened, rounded posteriorly. Measurements in mm: snout-vent length 49.0, axilla-groin length 22.0, leg length 17.5, arm length 12.5, tail length 44.5, head length 13.0, head width 9.3, head depth 5.7, snout length 5.2, eye diameter 3.0, ear diameter 1.1, eye to ear distance 3.7. Color in alcohol : ground color grayish tan; dorsal surface of body with faint indication of six light brown bands, broken along median line of body by longitudinal stripe of ground color; limbs without pattern, ground color; top of head from occiput to snout uniform light brown; brownish stripe from nostril to arm insertion, passing over ear; supralabials spotted with dark brown. 1970 Lizards of the Genus Phyllodactylus 63 less so on infralabials; venter grayish white with a few minute black dots on each scale; ventral surfaces of hands and feet slightly darker than venter; tail with faint indication of banding. Variation. Snout-vent length of young adult and adult males range from 37 to 51 mm (44.0), females from 41 to 44 mm (42.0), juvenile 25 mm; males plus females average 43.2 mm; tail partly regenerated in all specimens except one and averages about 50 per cent of total length; two postmentals in all, contacting first labial only in 100 per cent of sample; number of scales immediately following postmentals vary from 4 to 8 (6.1 ) ; midorbital scales vary from 15 to 18 ( 16.9) , scales across snout at level of third labial 19 to 22 (20.1 ) ; scales bordering internasals vary from 5 to 8 (6.0) ; internasals in contact along their median edges in 90.1 per cent of sample; scales between nostril and eye vary from 10 to 13 ( 1 1.6) ; scales across venter range from 27 to 30 (28.5), from throat to vent 53 to 62 (57.5) ; dorsal tubercular rows vary from 12 to 14 ( 13.3) ; paravertebral tubercles from head to base of tail range from 37 to 43 (39.4), between axilla and groin 23 to 27 (24.7) ; lamel- lae beneath fourth toe vary from 1 1 to 13 ( 12.0) ; at least two rows on enlarged, flat scales (may be elevated and tubercular-like) on proximal one-fourth of tail. Dorsal color pattern highly variable, from an almost uniform light brown to definite paravertebral rows of dark brown spots or bands; head usually uniform light brown, but faintly mottled with light gray in some specimens; dark brown stripe from nostril to arm insertion in all but one specimen which has stripe continuing to groin region; median grayish stripe on dorsum in all specimens, but faintly interrupted by black reticulating lines in one; tail banded brown and gray in only specimen with complete, original tail; venter grayish white in all. Habitat and natural history. Dr. James A. Peters (personal communica- tion) states that the type locality is a vertical rocky cliff on the beach, just above the high tide zone, 1 km south of Manta. The cliff surface is somewhat rotten, crumbles easily, and contains numerous holes and crevices. Four speci- mens of P. pumilus were taken from beneath rocks or in crevices along the cliff face. Peters indicated that he had not seen other specimens of Phyllo- dactylus this close to the ocean at any other point on the Ecuadorian coast. Peters indicated that all specimens of this species were taken during day- light hours. Other specimens were found beneath loose bark on trees and stumps in a semi-desert area with low scrubby vegetation, or in a transition, dry to subdry forest. Phyllodactylus reissi were taken with pumilus from the stump of an old ceiba tree 6 km E of Manta, and at the type locality. Distribution. This species has been taken west of the Cordillera de Balzar in the department of Manabi of Ecuador, from sea level to approximately 200 m (fig. 3). Specimens (11) have been examined from the following localities: ECUADOR. Manabi: 8 km W Jipijapa USNM 167184. 1 km S Manta USNM Figure 10. (a) dorsal view of P. interandinus; (b) dorsolateral view of P. john- wrighti. 1970 Lizards of the Genus Phyllodactylus 65 Figure 11. (a) dorsal view of P. kofordi (holotype); (b) dorsal view of P. clinatus (holotype). 167227-30. 6 km E Manta, 5 km W Montecristi USNM 167233. 11 km W Portoviejo USNM 167185, 167226. 9 km N San Vicente USNM 167187, 167231-32. Adaptation to Habitat Species of Phyllodactylus are, in the more restricted sense, geckos of arid and semiarid regions. Only on the mainland of British Honduras and some islands of the Caribbean are they found in a more tropical environment. Adaptations of species of Phyllodactylus for living in arid deserts and 66 Contributions in Science No. 192 tropical deciduous forests are generally of two types, scansorial and terrestrial. Most of these lizards are found in areas where sufficient physical objects are available for scansorial habits, i.e. trees, rocks and cliffs. All species of Mexico, Central America, and eastern South America are scansorial. Most prefer a rock substratum, but are frequently found on trees and shrubs. Only on the mainland of western South America does one find a shift to the terrestrial habit, and even then the terrestrial species are occasionally scansorial. Underwood (1954) indicates that the most primitive condition in geckos is one of terrestrial habits with simple, clawed digits. Adaptations to a scan- sorial habit involved the development of pilose, friction lamellae beneath the digits, and in some cases, beneath the tip of the tail. He also states that there has been a number of independent reversions to terrestrial habits with a loss or reduction in the size of the lamellae. Of the 15 mainland species of western South America, four are ter- restrial : P. microphyllus, angustidigitus, gerrhopygus, and sentosus. Of these four species, microphyllus has become well adapted to an open, deep sand habitat, through reduction in size of the terminal lamellae, shorter and more muscular limbs, and swollen scales about the nostril with the nasal opening a very small slit and lying in an anterior, dorsolateral position. Phyllodactylus angustidigitus has become adapted to a shell and rubble beach habitat through reduction in the terminal lamellae, enlargement of the claw, elevated and swollen scales about the nostril, but unlike microphyllus, the limbs are very slender, long, and without much muscular development. Little is known about the habits of sentosus, but all known specimens were taken from the ground and they have relatively small terminal lamellae and well developed claws, but the region about the nostril is unswollen. We find it difficult to explain the presence of enlarged terminal lamellae in a terrestrial species, such as gerrhopygus. This species has very large, well developed ter- minal lamellae like those of scansorial species (fig. 7), but has the area about the nostril greatly swollen, and the limbs are intermediate in size and shape between angustidigitus and microphyllus. There appears to be differential development for a terrestrial habit for each of the four species, with evolutionary trends variable within each popu- lation. For example, P. microphyllus taken from hard surface sands or con- glomerate soils have slightly larger terminal lamellae than those from soft sands. However, in gerrhopygus, the size of the terminal lamellae varies slightly within each population and does not seem dependent upon soil type. The reduction in the size of the terminal lamellae coupled with an enlargement of the claw in angustidigitus, seems to be a special modification for living on shell and rubble beaches. Whether the reduction in the terminal lamellae and enlargement of claws are for walking on shell and rubble, or for turning over these objects in search of food is not known, but the claw is approximately four times the size of the claw of gerrhopygus, a species that is frequently found on sandy beaches. 1970 Lizards of the Genus Phyllodactylus 67 The terminal lamellae of the scansorial species (P. reissi, lepidopygus, and probably dinatus) are generally greatly enlarged. Phyllodactylus reissi has exceptionally large toe pads, possibly influenced by its large body size. Several geckos are intermediate in their microhabitat preferences, P. kofordi, inae- qualis, phnwrighti, and interandinus, usually spending the day under bark or exfoliations, and at night foraging about on the ground. As one might expect, the toe pads are intermediate in size, reflecting an adaptation to both terrestrial and scansorial habits. In addition, a toe pad of intermediate size may facilitate terrestrial locomotion over a broken substratum and surface debris typical of the habitats where these geckos are found. Seemingly, the more scansorial a species becomes, the more natural selection favors an enlargement of the ter- minal lamellae. However, this does not explain the enlarged terminal lamellae of gerrhopygus. This species has been taken from terrace walls and walls of houses indicating it is not completely terrestrial. There may have been a recent shift from a scansorial to a terrestrial habit, and natural selection for small terminal lamellae is presently acting upon the population. However, it is pos- sible that gerrhopygus is adapted for running upon soft surface sand and for- aging for surface food items. This would, in part, account for the small claws and enlarged terminal lamellae. Larger and more lengthened claws may be used for turning over surface litter or for digging under it in search of food (microphyllus, sentosus, angustidigitus). Large terminal lamellae would likely hinder the ability to dig or clasp objects in the search for food, wheras a surface feeder might retain the large lamellae, enabling it to move faster, farther, and with more ease, in search of surface food items. The swollen area about the nostril seems to be an adaptation for living in regions where deep sand is constantly being moved about by wind action. The scales about the nostril are elevated, rounded (swollen), and the position of the nostril has been shifted to a more anterior, dorsolateral position. The nasal opening is narrow and partially protected by an anterior extension of the upper postnasal scale into the nasal aperture. In effect, the elevated and rounded scales surrounding the nostril form a tubular opening. This type of nostril provides adequate air passage and prevents sand from clogging the aperture wLile the individual rests in daytime retreats or forages about on the sand at night. Phyllodactylus sentosus is the only representative of a terrestrial species that does not have a tubular condition about the nostril. The substratum where it lives is more of a conglomerate soil, and selection for a tubular nostril is not one of the critical adaptive features for survival. None of the scansorial species have tubular nostrils. Sympatric Relations of PHYLLODACTYLUS in Western South America An awareness of sympatry is essential in overcoming the subjectivity of morphological taxonomy. This section will briefly document known cases of 68 Contributions in Science No. 192 Figure 12. (a) dorsal view of P. sentosus (holotype) ; (b) dorsal view of P. inae- qualis. 1970 Lizards of the Genus Phyllodactylus 69 sympatry in western South America as well as describe certain aspects of ecological and evolutionary differences among the sympatric species. The geographic distribution of species of Phyllodactylus is summarized in figs. 2-5. Known cases of sympatry are noted in Table 2. When geckos occur sympatrically one consistently observes distinct dif- ferences in the degree of arboreality. For example, although both P. reissi and interandinus utilize arboreal perches for diurnal retreats, only interan- dinus moves to the ground for nocturnal foraging. Phyllodactylus reissi and kofordi are commonly sympatric but have very little overlap in perch height. Phyllodactylus reissi is usually a scansorial for- ager, while kofordi is primarily a ground forager. In some localities P. inaequalis is found with kofordi and reissi in sym- patry. Phyllodactylus inaequalis, like kofordi, is a terrestrial forager. The relative abundance of these species seems to vary from locality to locality based on collecting records. It is not clear whether this is due to competition and/or to subtle environmental differences favoring one species at a given locality. There is some evidence of microhabitat separation between these two species at Cerro de la Vieja where inaequalis forages near the bases of boulders and kofordi forages more on open ground. Another difference commonly found among sympatric geckos is in body size, presumably reflecting differences in prey size. In sympatric situations the only known exceptions to extreme body size differences are with P. kofordi and inaequalis as well as kofordi and cUnatus. The fact that all of these geckos are small and appear to forage in different microhabitats probably accounts for their coexistence. Some cases of sympatry are not clearly understood. The replacement of P. microphyllus, a sand gecko, by kofordi, normally a foothill gecko, along the sandy coast in extreme northwestern Peru is puzzling. Climatic, competi- tive, or physical environmental factors may be involved. Selander and Giller ( 1963) observed that broad sympatry in wood- peckers of the genus Centurus occurs between distantly related species. Con- versely, closely related species are allopatric or only narrowly sympatric. These workers concluded . . it is suggested that the evolution of reproductive isolating mechanisms tends to proceed more rapidly than does the evolution of ecologic isolation, with the result that competition and other interspecific interactions prevent extensive sympatry of closely related species . . .” Pending completion of our osteological and karyological studies, the intrageneric relationships of the western South America Phyllodactylus are tentative. Strictly on the basis of external morphology, we suggest the follow- ing species are closely related and group them: gerrhopygus, heterurus, an- gustidigitus; microphyllus, inaequalis; lepidopygus, clinatus; johnwrighti, pumilus, interandinus; sentosus, kofordi; and reissi. Comparison of these groups with Table 2 shows only one known case of sympatry within a group, that of gerrhopygus and heterurus. However, the known extent of sympatry 70 Contributions in Science No. 192 Table 2 Sympatric relationships of 15 species of Phyllodactylus from the mainland of South America. 5 3 3 5c Cc •5 3 5c 5c s; >3 clinatus dixoni CX o Oc heterurii *3 3 3^ 3 3 •2? 3 -3 O kofordi 3. 0 s. -3 3. 2 3 ’i pumilus reissi sentosus ventralis angiistidigitus cl i fiat us dixoni gerrhopygiis heteriirus inaequalis interandinus johnwrighti kofordi lepidopygus microphyllus pumilus reissi sentosus V entrails X XXX X X X X X X X X X X X X XX XX X X XX X XX X (X in sympatry) of these species is based on one locality, and the status of heterurus is in doubt. Where broad sympatry does occur (e.g., among inaequalis, kofordi, and reissi), it is between or among very distinct species. Hence, Selander and Ciller’s (1963) conclusions might well be applicable to species of Phyllodac- tylus in South America. 1970 Lizards of the Genus Phyllodactylus 71 % Figure 13. (a) dorsolateral view of P. lepidopygus; (b) dorsal view of an adult and young of P. reissi. 72 Contributions in Science No. 192 Figure 14. (a) dorsal view of P. ventralis; (b) dorsal view of P. dixoni; (c) dorsal view of P. pumilus (holotype). 1970 Lizards of the Genus Phyllodactylus 73 Gazetteer There are a large number of collecting localities in Ecuador, Peru, and Chile not shown in gazetteers, on maps, or generally unavailable to scientists working with material from western South America. We present the following list of locali- ties with coordinates, arranged by country and political division. CHILE Tarapaca Arica 18° 29'S o O r- 20'W Iquique 20 13S 70 low Pica 20 30S 69 20W Pintados 20 38S 69 38W Pisagua 19 36S 70 13W Putre 18 12S 69 35W ECUADOR Azuay San Erancisco 3 37S 79 30W El Oro Machacha 3 16S 79 57W Santa Rosa 3 27S 79 58W G nay as Ancon Camp 2 20S 80 52W Colonche 2 OIS 80 40W Daule River 1 53S 79 58W Guayaquil 2 lOS 79 52W Playas 2 38S 80 23W Puente de Chimbo 2 IIS 79 07W Puna 2 42S 79 55W Punta Cornero 2 18S 80 55W Salinas 2 13S 80 50W Loja Loja 4 OOS 79 13W Rio Casanga 4 08S 79 49W Manabi Jipijapa 1 20S 80 35W Manta 0 57S 80 45W Portoviejo 1 03S 80 27W San Vicente 0 36S 80 24W PERU Amazonas Bagua Chica 5 38S 78 35W Bagua Grande 5 47S 78 26W Marahon Valley 4 30S 78 27W Ancash Chimbote Valley 9 07S 78 38W Culebras 9 56S 78 14W Pativilca 10 40S 77 48W Areqiiipa Arequipa 16 24S 71 32W Camana 16 37S 72 42W Chucarapi 17 04S 71 44W La Joy a 16 45S 71 52W Mollendo 17 02S 72 OIW Tambo 17 06S 71 48W Vitor 16 26S 71 49W Yura 16 12S 71 42W Cajamarca Bellavista 5 37S 78 39W Chinchipe Valley 5 28S 78 32W El Arenal 5 56S 79 17W El Tambo 5 46S 79 24W Las Juntas 5 57S 79 14W Molino Viejo 5 49S 79 23W Perico 5 15S 78 45W Ica Ica 14 05S 75 44W Isla Sangallan 13 50S 76 28W Isla Viejas 14 17S 76 12W Jaguay 13 23S 76 12W Lagunillas Bay 13 54S 76 18W Marcona 15 20S 75 low Paraca Bay 13 50S 76 16W Paracas 13 48S 76 15W Paracas Peninsula 13 48S 76 24W Pozo Santo 13 54S 76 03W Puerto Pisco (new) 13 47S 76 17W San Juan Port 15 22S 75 12W Junin Chanchamayo 10 42S 75 08W 74 Contributions in Science No. 192 Lambayeque Chiclayo 6 46S 79 SIW Eten 6 54S 79 S3W Jayanca 6 24S 79 SIW Mocupe 7 oos 79 37W Morrope 6 32S 80 02W Motupe 6 09S 79 43W Naupe 5 37S 79 S4W Olmos Libertad 6 OOS 79 4SW Chiclin Jequetepeque 7 SOS 79 low Valley 7 21S 79 36W Pacanguilla 7 lOS 79 28W Pacasmayo Lima 7 24S 79 34W Ancon 11 46S 77 low Asia 12 47S 76 3SW Cerro Azul 13 03S 76 30W Chorillos 10 46S 77 46W Chosica Cruz de 11 54S 76 42W Hueso Isla San 12 23S 76 46W Lorenzo 12 OSS 77 ISW Lima Lomas de 12 03S 77 03W Lechay 11 27S 77 18W San Antonio Verrugas 12 37S 76 37W Canyon Pi lira 11 S3S 76 32W Bayovar 5 SOS 81 03W Bocana 4 41S 80 18W Cabo Blanco 4 ISS 81 14W Capullana Point 4 29S 81 17W Catalinas 4 ns 80 S8W Cerro Amotape 4 40S 80 S8W Chongollapi S 26S 79 37W Chulucanas S 06S 80 low El Alto 4 17S 81 12W Fondo 4 28S 81 09W Huancabamba S 14S 79 27W Isla Lobos de Afuera 6 S7S 80 42W Isla Lobos de Tierra 6 27S 80 S2W Las Lomas 4 40S 80 ISW Lobitos 4 26S 81 16W Los Organos 4 09S 81 07W Monte Grande 4 27S 81 02W Negritos 4 38S 81 19W Paita S 06S 81 07W Parinas Valley 4 30S 81 06W Piura S ns 80 38W Punta Aguja s 47S 81 04W Punta Sal s SIS 80 S6W Reventazon 6 lOS 81 oow Sechura s 34S 80 sow Sechura Desert s SOS 80 40W Sullana 4 S3S 80 41W Talara 4 34S 81 17W Trie Trac S SIS 81 OOW Tiimbes Cancas 4 OOS 81 oow Las Vocas 3 2SS 80 19W Puerto Pizzaro 3 39S 80 24W Quebrada Seca 4 23S 81 14W Acknowledgments We are endebted to V. Aellen, C. Rivero-Bianco, J. A. Bohlke, C. M. Fugler, E. Gambini, A. G. C. Grandison, R. Inger, K. Klemmer, W, Ladiges, A. E. Leviton, G. Orces, J, A. Peters, J. A. Roze, J. A. Sloan, H. M. Smith, R. C. Stebbins, D. W. Tinkle, H. Wermuth, E. E. Williams, J. W. Wright, and R. G. Zweifel for the loan of material in their care. We are especially grateful to Drs. Carl Koford, John W. Wright, F. McCollum, and R. B. Hamilton for their able assistance and companionship while pursuing our studies of geckos in Peru; to Drs. H. W. and M. Koepcke, Mr. and Mrs. W. Grindley, Ing. H. Ramirez, G. and Sr. S. Bertarelli, Drs. F. Blancas and E. Gambini for the many courtesies extended to us during our 1970 Lizards of the Genus Phyllodactylus 75 trips to Peru; to Dr. Janis A. Roze for field notes and a photograph of habitat in Venezuela; and to Dr. James A. Peters for field notes on Ecuadorian geckos. This study was supported through a grant to Dixon from the Texas A&M University Organized Research Fund (1968), a National Science Foundation Traineeship to Huey through the University of Texas (1968), and a expedi- tion to Peru ( 1967), sponsored by the Museum of Vertebrate Zoology, Uni- versity of California. Resumen Se hace una breve descripcion del habitat natural de las especies de Phyllo- dactylus encontradas en el continente de America del Sur. El micro-habitat na- tural de cada especie se describe hasta donde ha sido posible. Poca a ninguna informacion se aporta sobre el micro-habitat de P. heterurus, setosus y clinatus. Cuatro especies, P. angustidigitus , gerrhopygus, microphyllus y setosus son primariamente terrestres y alopatricas en distribucion. Diez especies, P. clina- tus, dixoni, inaequalis, interandinus , johnwrighti, kofordi, lepidopygus, pumi- lus, reissi y vent rails son aboricolas y unicamente johnwrighti, v entrails y dixoni no viven simpatricamente con una o mas de las otras especies arbori- colas. En donde tres o mas especies viven simpatricamente, las areas utilizadas para reposar durante el dia y para la obtencion de alimento se sobreponen escasamente. Con la excepcion de P. reissi, la cual produce dos huevos, todas las especies de Phyllodactylus encontradas en America del Sur continental depositan un solo huevo por nidada. La deposicion de huevos en forma comunal parece ocurrir con frecuencia entre el mayor numero de especies, incluyendo aquellas que viven simpatricamente : P. interandinus y reissi, y kofordi y reissi. Las especies del genero Phyllodactylus parecen tener adaptaciones especifi- cas a sus micro-habitates naturales. Aquellas especies que habitan areas roco- sas se caracterizan por la presencia de laminillas digitales terminales ensan- chadas, cabeza y cuerpo comprimidos dorsoventralmente y ventanas de la nariz localizadas lateralmente; en el caso de especies que habitan un micro- habitat arenoso las laminillas digitales terminales son pequenas, la cabeza y el cuerpo redondeados y las ventanas de la nariz son valvulares y estan localiza- das dorsolateralmente ; finalmente, las especies que viven en playas de concha tienen laminillas digitales con unas alargadas, cabeza y cuerpo delgado y ven- tanas de la nariz con valvulas situadas dorsolateralmente. Dependiendo del grado en que el comportamiento es arboricola o terrestre algunas especies de Phyllodactylus poseen las adaptaciones descritas en forma intermedia. Para cada especie, excepto P. heterurus, se incluye un mapa de distribucion junto con fotografias; ademas, se presenta una lista de localidades importantes. De los holotipos y sintipos de las especies descritas de America del Sur con- tinental, existen diez (dixoni, inaequalis, lepidopygus, magister, microphyllus, nigrofasciatus, phacophorus, pumilus, reissi, y v entrails), ocho de los cuales han sido examinados. 76 Contributions in Science No. 192 Se describen siete nuevas especies (angustidigitus, clinatus, interandinus, johnwrighti, kofordi, pumilus y setosus); nueve nombres antiguos se relegan al sinonimio de otras especies ( abrupteseriatus, baessleri, guayaquilensis, lo- bensis, rnagister, nigrofasciatus, phacophorus, underwoodi, variegatus). Literature Cited Burt, C. E., and M. D. Burt, 1930. The South American lizards in the collection of the United States National Museum. U.S. Natl. Mus., Proc., 78(2849) : 1-52. — , 1931. South American lizards in the collection of the American Museum of Natural History. Amer. Mus. Nat. Hist., Bull., 61 : 227-395. — , 1933. A preliminary check list of the lizards of South America. Acad. Sci. St. Louis, Trans., 28: 1-104. Burt, C. E., and G. S. Myers, 1942. Neotropical lizards in the collection of the Natural History Museum of Stanford University. Stanford Univ. Publ., Univ. Ser., Biol. Sci., 8(2): 1-52. Boulenger, G. a., 1885. Catalogue of the lizards in the British Museum, London, 1: 1-436. CODOCEO, M., 1950. Reptiles de Tarapaca. Inv. Zool. Chile, 1(1) : 15. Cope, E. D., 1876. Report on the reptiles brought by Professor James Orton from the middle and upper Amazon, and western Peru. Acad. Nat. Sci. Phila., Jour., (2)8: 159-183. , 1877. Synopsis of the cold blooded Vertebrata, procured by Prof. James Orton during his exploration of Peru in 1876-77. Amer. Philos. Soc., Proc., 17: 33-40. Dawson, E. Y., 1963. Ecological paradox of coastal Peru. Nat. Hist., 72(8) : 32-37. Dixon, J. R., 1960. The discovery of Phyllodactylus tiibercidosus (Reptilia: Sauria) in Central America, the resurrection of P. xanti, and description of a new gecko from British Honduras. Herpetologica, 16(1): 1-11. • — , 1962. The leaf-toed geckos, genus Phyllodactylus, of northeastern South America. Southwest. Nat., 7: 211-226. , 1964a. Further data on the geckos (Phyllodactylus) of islands of the extreme southern Caribbean. Southwest. Nat., 9: 203-206. -, 1964b. The systematics and distribution of the lizard genus Phyllodactylus of North and Central America. New Mex. St. Univ. Res. Center, Sci. Bull., 64(1): 1-139. Donoso-Barros, R., 1947. Breves notas sobre Reptiles Chilenos. Mus. Nac. Hist. Nat. Santiago, Bol. 23: 107-1 11. — , 1966. Reptiles de Chile. Ediciones de la Universidad de Chile, Santiago, Chile, pp. 1-458. Dumeril, a. M. C., and G. Bibron, 1836. Erpetologie Generale ou Histoire Natu- relle des Reptiles. Paris Impr. de Fain., 8: 394. G ARMAN, S. W., 1875. Exploration of Lake Titicaca. I. Fishes and Reptiles. Harvard, Mus. Comp. Zool., Bull., 3: 276-278. Holdridge, L. R., 1964. Life Zone Ecology. Trop. Sci. Center, San Jose, Costa Rica, pp. 1-124. Kendrew, W. G., 1961. The climates of the continents. Oxford Univ. Press, London, England, pp. 1-608. 1970 Lizards of the Genus Phyllodactylus 77 Koepcke, H. W., 1961, Synokologische Studien an der Westseite der Peruanischen Anden, Verlag Ferdinand Dummler, Bonn, pp. 1-320. Koepcke, M., 1954. Corte ecologico transversal en los Andes del Peru central con especial consideracion de las Aves. Parte 1: Costa, Vertientes Occidentales y Region altoandina. Lima, Univ., Mus. Hist. Nat. “Javier Prado,” Mem., 3: M19. Koford, C. B., 1968, Peruvian desert mice: water independence, competition, and breeding cycle near the equator. Science, 160: 552-553, MacBride, J. F., 1936. Flora of Peru. Field Mus. Nat. Hist., Chicago, Bot. Ser., 13(1): 1-320. Mann, F. G., 1960. Regiones biogeograficas de Chile. Inv, Zool. Chilenas, 6: 15-49. Mechler, B., 1968. Les Geckonides de la Colombie. Rev. Suisse Zool., 75: 305-371. Murphy, R. C., 1948. Oceanic birds of South America, Vol. I. Macmillan Co., New York, pp. 1-640. Noble, G. K., 1924. New lizards from northwestern Peru. Boston Soc. Nat. Hist., Occ. Pap., 5: 107-113. Parker, H. W., 1935. Some lizards from Venezuela and the Dutch Leeward Islands. Ann. Mag. Nat. Hist., 15: 480-484, Peters, J. A., 1960. The snakes of Ecuador, a check list and key. Harvard, Mus. Comp. Zool., Bull., 122: 491-541. Peters, W. C. H., 1862. Mittheilung uber einen neuen Phyllodactylus aus Guayaquil. Monatsb. Akad. Berlin, pp. 626-627. Rivero-Blanco, C., and A. R. Lancini, 1968. Phyllodactylus dixoni; Una nueva especie de lagarto (Sauria: Gekkonidae) de Venezuela. Soc. Cienc. Nat. La Salle, Memoria, 27: 168-175. Robinson, D. A., 1964. Peru in four dimensions. American Studies Press, S.A., Lima, Peru, pp. iii-424. Roze, J, A., 1964. La herpetologia de la isla de Margarita, Venezuela. Soc. Cienc. Natl. La Salle, Mem., 24(69) : 209-241. Ruthven, a. G., 1922. The amphibians and reptiles of the Sierra Nevada de Santa Marta, Colombia. Univ. Mich. Mus. Zool., Miscl. Publ., (8) : 1-69. Selander, R. K., and D. R. Gillier, 1963. Species limits in the woodpecker genus Centurus (Aves). Amer. Mus. Nat. Hist., Bull., 124: 217-273. Shelford, V. E., 1926. Naturalist’s guide to the Americas. Williams & Wilkins Co., Baltimore, pp. 1-761. Taylor, E. H., 1942. Some geckos of the genus Phyllodactylus. Univ. Kans. Sci. Bull., 28: 91-112. Titschack, E., 1951. Beitrage zur Fauna Perus. Bd. I Jena (Gustav Fischer), pp. iv-405. Tosi, J. A., 1960. Zonas de vida natural en el Peru, Memoria Explicativa sobre el mapa ecologico de Peru. Bol. Tec., (5), Inst. Interamer. Cien. Agr. OEA Zona Andina, pp. i-27 1 . Tschudi, j. j. Von, 1846. Untersuchungen uber die Fauna Peruana. Herpetologie. Zeiter und Zollikofer, pp. 37-39. Underwood, G., 1954. Categories of adaptation. Evolution, 8(4) : 365-377. Vanzolini, P. E., 1968a. Geography of the South American Gekkonidae (Sauria). Arq. Zool. S. Paulo, 17(2) : 85-1 12. 78 Contributions in Science No. 192 — , 1968b. Lagartos Brasileiros de familia Gekkonidae (Sauria). Arq. Zool. S. Paulo, 17(1): 1-84. Verdoorn, F., 1945. Plants and plant science in Latin America. Chronica Botanica Co., Waltham, Mass., pp. 1-381. Weberbauer, a., 1936. (in Macbride), Phytogeography of the Peruvian Andes. Field Mus. Nat. Hist., Bot. Sen, 13(1): 13-81. Wermuth, H., 1965. Liste der rezenten Amphibien und Reptilien, Gekkonidae, Pygopodidae, Xantusidae. Tierreich, 80: 1-246. Werner, F., 1901. Reptilien und Batrachier aus Peru und Bolivien. Abhand. und Berich. K. Zool. Anthrop. Ethn. Mus. Dresden, 9(2) : 1-14. — , 1907. (in Burger), Estudios sobre reptiles Chilenos. Anales Univ. de Chile, 121: 149-155. — — — , 1910. Uber neue oder seltene Reptilien des Naturhistorischen Museums in Hamburg. II. Eidechsen, Hamburg, Naturhist. Mus., Mitt., 27: 1-46. — — , 1912. Neue oder seltene Reptilien und Frosche des N aturhistorischen Museums in Hamburg. Hamburg, Naturhist. Mus., Mitt., 30: 1-51. Wiegmann, a. F. a., 1835. (in Meyen), Beitrag zur Zoologie, gesammelt auf einer Reise un die Erde. Nova Acta Acad. Caes. Leop.-Carol., 17(1) : 183-268. Yanez, a. P., 1951. Vertebrados Marinos Chilenos. III. Reptiles. Revista Biol. Marina, 3(1-2): 1-18. Accepted for publication February 10, 1970 I LOS 1^' ANGELES ' COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE ' ikuMBER 193 July 1, 1970 HADROSAURIAN DINOSAUR BILLS— MORPHOLOGY AND FUNCTION By William J. Morris *!! 1 Hi ii Ni 4 Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF.— Authors will be sent galley proof which should be corrected and returned promptly. Changes in the manuscript after galley proof will be billed to the author. Unless otherwise requested, page proof also will be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor HADROSAURIAN DINOSAUR BILLS— MORPHOLOGY AND FUNCTION By William J. Morris^ Abstract: Reconsideration of the morphology and function of the bill of hadrosaurian dinosaurs is based upon an excellently preserved bill mold of a recently discovered specimen (LACM 23502) assigned to the genus Anatosaunis. Morphology suggests a filtering function for the bill and adaptation for discharge of fluid and small particles following intake and mastication of plant or invertebrate food. Acknowledgments The excellent specimen of Anatosaurus cf. annectens (LACM 23502) upon which this study is based was discovered and collected by H. Garbani and skillfully prepared by M. Odano, both of the Los Angeles County Museum of Natural History. The skeleton was found in the Hell Creek Formation south of Ft. Peck Reservoir, NE Va, Sec. 36, T21N, R35E, Montana. The project, leading to this discovery as well as other important dinosaur finds, was generously supported by Mr. and Mrs. William T. Sesnon, Jr., patrons of the Los Angeles County Museum of Natural History. Constructive criticism of the manuscript by Dale Russell, Canadian National Museum, and Theodore Downs of the Los Angeles County Museum of Natural History is appreciated. Others on the staff of the Los Angeles County Museum whose help is acknowledged are: David Forstch and Alan Tabrum for their interest and discussion, and Joseph Cocke for his drawing of Figure 2. The remainder of the illustrations and photographs were done by the author. Introduction Dinosaurs of the family Hadrosauridae are among the most familiar Late Cretaceous fossils. Indeed, the first dinosaurs described from North America, belonged to these so called duckbilled forms. Prior to a brief description by Cope (1883) the presence of a horny bill in hadrosaurs was conjectural. Cope, in preparing a specimen of Anatosaurus annectens, discovered portions of a mold of the bill. Later two other specimens in which most of the bill was preserved were described for A. annectens by Versluys (1923) and for Corythosaurus excavatus by Sternberg ( 1935) . In all three cases we must rely upon the description of the bill as the original molds were either destroyed during preparation or subsequently lost. As a result, a recently discovered specimen of Anatosaurs cf. annectens (LACM 23502) from the Hell Creek beds of Montana containing a well preserved mold is most important in ^Research Associate in Vertebrate Paleontology, Los Angeles County Museum of Natural History; and Professor of Geology, Occidental College, Los Angeles, California 90041. 1 2 Contributions in Science No. 193 evaluating the morphology and function of the hadrosaurian bill. The letter designations are used to designate specimens in various museums: LACM, Los Angeles County Museum of Natural History; ROM, Royal Ontario Museum; NMC, National Museum of Natural Sciences, Ottawa. Bill Morphology of Anatosaurus cf. A. ANNECTENS (LACM 23502) The specimen is a mature individual approximately forty feet long. The skull is only slightly distorted mainly along a parasagittal section approximately midway between the orbits and narial openings. The mandibles, though offset to the right, are not distorted relative to the midline so that the bill mold is undistorted, (Fig. 1). Morphologically, the specimen is very similar to A. annectens but there are some differences between it and the holotype. Until the time when careful study has been made regarding individual variability of hadrosaurian dinosaurs, it is hazardous to make positive, specific references for specimens other than the type. This is especially true regarding the LACM specimen for it was found further north than the Lance Formation, Wyoming, localities from which the types of A. annectens were obtained. The bill mold is composed of gray, lithic siltstone, the surface is accentu- ated by a thin veneer of iron oxide. In the figures, the mold appears dark due to the application of preservative although it has the same composition as the matrix. The entire anterior surface (attached dorsally to the premaxillary region) is preserved and only a small portion of the ventral edge is lacking. Only the left side is preserved. Before discussing the form and function of the bill it is important to describe the geometry of the mold. Errors in interpretation by other authors seems largely to have resulted from misinterpretation of the geometry of the mold. When the specimen was discovered, anterior and posterior sides of the mold were preserved with a very thin surface of weakness separating them. The surface, shown in Figures 1, A, B, C and D, is the posterior side of the mold, the one closest to the premaxillary region of the skull. It is important to note that matrix on both sides of the preserved surface was carefully prepared and other structures were not found within the matrix except the surface of the mold. Hence the entire bill, as preserved, is represented by two surfaces separated by less than 0.01 mm, one of which is shown in the figures. The surface of the mold is vertically fluted, containing nine or possibly ten V-shaped channels. These are separated by dorsal-ventral pillars, rounded Figure 1. Partially prepared skull of Anatosaurus cf. annectens, LACM 23502. (A) Complete skull, left side. The bill mold is not preserved on this side but the edge of the mold can be seen extending ventrally from the premaxillary below the predentary of the lower mandible. Note that the free margin of the mold extends below the predentary and is separated from the predentary by siltstone matrix. (B) Anterior undulating surface of the bill mold showing the fluted channels preserved as an iron oxide veneer on the siltstone matrix. (C) Dorsal oblique aspect of the skull illustrating the very terminal position of the bill mold along the ventral margin of the premaxillary. (D) Ventral oblique view showing the bill mold and the matrix layer separating it from the predentary. 1970 Hadrosaurian Dinosaur Bills 3 4 Contributions in Science No. 193 1970 Hadrosaurian Dinosaur Bills 5 in cross section. Laterally, towards the maxillaries, the fluted surface is re- placed by a smooth triangular plate. The mold is approximately 110 mm long at the junction of the pre- maxillaries and it extends ventrally about 50 mm below the anterior margin of the predentary, decidedly overlapping the front of the lower jaw. Although the extreme ventral margin of the mold appears to be missing, the taper of the pillars and channels suggests that the entire bill was not much longer than shown in this specimen. The premaxillary of the specimen, dorsal to the mold on the left side, is not well preserved and the surface separating bill and upper jaw cannot be seen. There is, however, little doubt that the bill was united with the upper mandible. A complimentary structure that would have been attached to the lower mandible was not found even though conditions for its preservation must have been similar to those for the upper bill mold. Interpretive Morphology The surface of the mold forms an immediate extension of the pre- maxillary without break or offset. The actual horny bill must have been affixed anterior to the mold and, in order to be secure, must have extended up and onto the surface of the premaxillaries. Although there is no direct evidence indicating the posterior limit of the bill, it probably did not extend farther than the excavation surrounding the nares in the premaxillaries. The geometry suggested by the anterior position of the mold suggests that the mold reflects the interior surface of the bill. Probably some compression of the bill occurred during compaction of the silt now forming the matrix, and Figure 2. Restoration of the bill on the skull of Anatosaunis cf. annectens, LACM 23502. The undulating, fluted inner surface of the bill is shown. When the mandibles were closed and the predentary opposed to the bill, channels opening from the mouth were formed. 6 Contributions in Science No. 193 it is impossible to determine the original thickness of the specimen. The fluted surface was evidently confined to the interior of the bill as no comparable surface was found against the more dorsal surfaces of the premaxillary. Con- ceivably the original exterior fluting could have been destroyed during com- paction but then why was it not also destroyed on the more ventral portion? It seems best to reconstruct a bill with the fluted surface only on the internal side. The predentary is poorly preserved and furnishes no evidence for a bill on the lower jaw. Nor was evidence for the presence of a lower bill found in the matrix anterior to the predentary. A restoration of the probable appearance of the upper bill is shown in Figure 5. Discussion of Previously Described Bill Molds The specimen described by Cope was destroyed during preparation of the skull. Cope did not give a reason for destroying the mold, but one cannot help conjecturing that he thought his description vivid enough that future reference to the specimen would be unnecessary. The mold of the Cory- thosaurus excavatus specimen, except for a very small and unilluminating fragment, could not be located. In the third example Versluys apparently used the mold to cast what he took to be a plaster replica of the bill and, after obtaining the cast, the natural mold was not retained. Cope appears to have misinterpreted a structure very similar to that of LACM 23502. The full text follows (1883:106) : “Dermal or corneous struc- tures have left distinct traces in the soft matrix about the end of the beak-like muzzle. Laminae of brown remnants of organic structures were exposed in removing the matrix. One of these extends as a broad vertical band round the sides, indicating a vertical rim to the lower jaw, like that which surrounds some tea trays, and which probably represents the tomia of the horny sheath of a bird’s beak. At the front of the muzzle its face is sharply undulate, presenting the appearance of vertical columns with tooth-like apices. Corresponding tooth-like processes, of much smaller size, alternate with them from the upper jaw. These probably are the remains of a serration of the extremital part of the horny tomia, such as exists on the lateral portions in the lamellirostral birds.” Cope described an undulatory surface with tooth like terminations; how- ever, he did not make it clear whether he considered the object examined a mold or the actual bill. He implies that the structure was attached to the lower jaw and that only small protuberances were present on the upper. Cope prob- ably misinterpreted the area for attachment of the bill, judging from the LACM specimen. Versluys (1923) also believed that Cope’s analysis was in error, but Lull and Wright ( 1942) and Ostrom ( 1961 ) both accepted Cope’s (1883) description of a horny beak on the lower jaw. Probably the smaller, tooth like processes described by Cope were rugosities normally present distally on most hadrosaurian premaxillaries. The morphology of the hadrosaurian bill was further confused by Lull and Wright who state (1942:43), “Tn the vSenckenberg specimen [the one 1970 Hadrosaurian Dinosaur Bills 7 described by Versluys] the impression shows a beak on the upper jaw which projected with a free lower border for about 8 cm over the ventral edge of the premaxillae. The beak stood vertically, with a regular, undulating surface which became smoother towards the sides. Cope described the same wavy appearance corresponding and alternating with the tooth-like processes in the jaw itself. ... He [Cope] also described a similar horny beak on the lower jaw.” Cope of course had not described a wavy appearing structure occurring on both jaws, the upper being described as small tooth like processes. Ostrom (1961:152) apparently agreed with Lull and Wright stating, “Cope (1883), however, described beak impressions at the anterior extremities of both the upper and lower jaws in a specimen of A. copei {Diclonius mirabilis) Yet neither LACM 23502 nor the equally well preserved Senckenberg specimen give any indication of a lower bill. Sternberg figured a bill mold in describing a specimen of C. excavatus (NMC 8676) , but in neither the plate explanation nor in the body of the paper did he state that the bill was attached to the lower mandible (Fig. 3). Ostrom, however, (1961:152) states, “Further evidence of a lower beak has been found in a third specimen {Corythosaums excavatus) , (NMC 8676) , described by Charles M. Sternberg (1935), in the form of an impression of an incomplete horny beak in front of and along the left side of the predentary.” As the plate from Sternberg’s paper shows (Fig. 3) the mold is a fluted surface associated with the lower jaw but it is separated from the predentary by a considerable thickness of rock matrix. The presence of matrix between the surface and predentary is evidence that the mold is not the opposing surface of a predentary sheath. Upon examination of the plate and the small fragment in the collection of the Canadian National Museum, it appears as though the mold is that portion of the upper bill which originally projected over and in front of the predentary. Although the fluted surface has been observed on three reported species, there is a problem regarding it. Does the fluted surface belong on the external surface or the internal surface of the bill? Cope, (1883) having interpreted the mold as being attached to the lower jaw, apparently assigned the undulatory surface to the exterior of the lower bill. On the other hand, Versluys (1923) assigned the undulatory surface to the outer surface of the upper bill. Versluys presents excellent drawings show- ing that the bill was indeed attached to the upper mandible and that the feature is similar to the mold in LACM 23502. However, in reconstructing the bill he placed the undulatory surface on the exterior. Versluys removed the posterior portion of the bill mold and then poured a substance, probably plaster, against the anterior part of the mold, judging from the wording as translated from German. The resulting cast was retained but the rest of the mold was not. Versluys interpreted this cast as being the bill. He had apparently reproduced the posterior part of the original mold formerly occupied by matrix. Such a restoration would erroneously cause the undulatory surface to appear on the exterior rather than on the interior of the bill. Sternberg (1935) did not state whether he believed the undulatory surface 8 Contributions in Science No. 193 of the beak to be on the exterior or the interior. Ostrom, however, in discussing the specimen of Corythosaurus excavatus described by Sternberg, states, (1961:152), “This specimen indicates the presence of small, tooth-like pro- jections on the inner surface of the beak, which probably contributed to a firm union of this structure with the lower jaw.” Sternberg’s plate clearly shows a layer of matrix between the mold and predentary which is evidence against a firm union. In addition, the surface of the predentary, although rugose, is not appressed against the mold of the bill and the rugosities do not match the channels in the bill mold. The various interpretations of the relationship of the undulatory surface to the mandibles is illustrated in Figure 4. Function of the Hadrosaurian Bill Almost every conceivable mode of food gathering has been proposed at one time or another for the hadrosaurian dinosaurs. There are no existing Figure 3. Lower jaw of Corythosaurus excavatus, NMC 8676, showing the bill mold (H.B.), predentary (P.D.), and dentary (D). This figure is reproduced from Plate II, Sternberg (3). 1970 Hadrosaurian Dinosaur Bills 9 MORRIS (1969) COPE (1883) B-l NEGATIVE MOLD INTERIOR surface UPPER BEAK B-2 NEGATIVE MOLD EXTERIOR SURFACE LOWER BEAK B-2 NEGATIVE MOLD EXTERIOR SURFACE LOWER BEAK C- POSITIVE REPLICA UPPER BEAK b-2 NEGATIVE MOLD OUTER SURFACE UPPER BEAK Figure 4. Diagram illustrating the interpretations of the four reported hadrosaurian bill molds. All interpretations are based upon molds preserved with specimens of Anatosaiinis annectens except Sternberg (1935) which is based upon Corythosaurus excavatus. Blocks labeled represent the mold as discovered. Block 1 is the posterior part of the mold and 2 is the anterior. Blocks labeled B represent the parts of the mold preserved after preparation. Block C in the case of the interpretation by Versluys represents a plaster cast that was poured against B-2- Each group of blocks is followed by the interpretation placed upon them by the authors listed. Lastly, outline restorations are shown illustrating the interpretation by each other. 10 Contributions in Science No. 193 reptiles that have comparable feeding or masticating features. Cope suggested a diet of soft vegetable matter but this was based on the erroneous assumption that the teeth, in particular those of the dentary, were only loosely attached. Lull and Wright (1942:36) give the following picture, “The analogy now seems to be with the moose Alces whose diet is that of a browsing animal, that is the twigs and leaves of deciduous trees . . . and also certain conifers. In summer it turns to the tall, lush grass in swamps or near the margins of streams and lakes and, actually invading the water, feeds upon the leaves of the water lily or, with head entirely submerged, on the roots and stems of aquatic plants.” Versluys strongly rejected the idea of a ducklike feeding habit as reported by Abel (1912). Instead, he suggested that the structure was more beak than bill adapted for scraping bark and leaves from small scrubs and trees or even for the uprooting of plants. There is only one report of stomach contents associated with a hadro- saurian skeleton. Unidentified plant debris and recognizable conifer needles were reported (Krausel, 1922) from the stomach cavity of A. annectens. As pointed out by Ostrom ( 1961 ), this might indicate merely that these creatures could feed on land plants, and it does not eliminate the possibility that they fed on aquatic vegetation as well. Indeed, the harsher fibers of land plants might be more readily preserved than the softer materials of aquatic plants. In any analysis of their feeding habits, several morphological features must be taken into account and the assumption made that all of the features functioned together in an integrated well adapted system. The hadrosaurians were remarkably successful. Not only are their remains relatively common but genera have been recognized in North and South America and Eurasia. Despite considerable disparity, particularly in the shape of the skull, there are features related to feeding habits common to all genera. One is bulk. The probable weight of Corythosaurus and Anatosaurus was approximately 4 tons and yet the 30 to 40 foot hadrosaurians are not the largest recognized, being surpassed by Edmojitosauriis and Hypacrosaurus. Small hadrosaurian remains have been collected but it is not known if these represent a truly small taxon or are immature individuals of the larger species. The few apparently undistorted skulls indicate that the hadrosaurians had a constricted buccal passage formed by the dentaries and maxillaries, and leading to the throat. In large anatosaurs, for example, it is doubtful that particles having a cross-sectional diameter larger than three or four inches could have passed along this narrow channel. Hadrosaurian dental batteries are very similar, consisting of dozens of teeth arranged in closely spaced rows. The occlusional surface is a pavement formed of the diamond-shaped crowns of the opposing dental batteries. In addition, in all genera three to six teeth are arranged in vertical rows and in each, only the most dorsal are in active use. The others serve for a highly effective replace- ment mechanism. The bill as preserved in LACM 23502, seems to impose certain limiting parameters on hadrosaurian feeding habits. The structure seems best adapted for filtering. Although several workers have suggested that the bill closed 1970 Hadrosaurian Dinosaur Bills 11 against the predentary of the lower jaw forming a cropping device, three characteristics of the bill seem to negate this function: 1 ) The bill was attached to the anterior and dorsal segment of the premaxillaries and its free edge extended well below the anterior dorsal margin of the predentary; 2) The very thin space between the anterior and posterior mold suggests that the bill itself was a relatively thin structure certainly not well adapted for foraging leaves and branches of trees and scrubs; 3) The last feature is the fluted, undulating inner surface of the bill which would have little or no function for foraging or browsing, and yet seems ideally suited as a filtering device. Hadrosaurian dinosaurs, particularly such forms as Corythosauriis and Hypacrosaurus, are envisioned as mainly aquatic animals. A filtering device would be very important in assuring that these large animals could ingest large amounts of concentrated food relatively free of water in a manner similar to that of the dabbler ducks, such as Dendrocynga, Anas and Aix. These eury- phagus forms have laminations on the inner surface of the dorsal bill which, when closed against the ventral bill, form an efficient filtering device. Such an interpretation certainly makes the name “duckbill” even more fitting for the hadrosaurian dinosaurs than previously supposed. With their very efficient food gathering system, the hadrosaurs could take in all manner of food including mollusks and small crustaceans as well as plant material. In this sense they would, like many ducks, be relatively high protein as well as carbohydrate feeders. Plant material as well as more resistant invertebrates could be crushed between the broad occlusional surfaces of the impressive dental battery. By lowering the anterior part of the head, excess water as well as food particles smaller than approximately one cm would drain out of the mouth along the channels formed by the fluted bill and the opposing predentary. The resulting masticated material, highly concentrated, could then be made to flow down the very small passage between the dentaries to the stomach simply by raising the head. The presence of the fluted bill, multiple grinding surfaces, mechanism for continuous replacement of teeth, and a constricted passage formed from the bones of the posterior mouth region makes such a feeding habit feasible. Ostrom (1964) presents a very thorough compilation of prior hypotheses and information regarding the feeding habits of hadrosaurian dinosaurs. In his analysis of the paleoecology the conclusions reached vary only in detail from those expressed earlier (1961 ). In summary Ostrom states (1964:995), “Past interpretations of hadrosaurian ecology have pictured these animals as pre- dominantly aquatic in their habits, living and feeding in swamps, lakes, or rivers and making only occasional and limited excursions out on surrounding lands. Anatomic evidence, however, contradicts this interpretation, indicating instead that the hadrosaurs as highly adept bipeds were primarily terrestrial animals. Other anatomic and botanical evidence shows them to have been active terrestrial foragers adapted for browsing on the harsher, fibrous, or even woody tissues of trees and shrubs.” The bill, as reconstructed from the mold of LACM 23502, is very thin relative to its lateral and vertical extent, and suggests a structurally weak 12 Contributions in Science No. 193 feature, too weak to be of use in browsing on fibrous or woody plants. Ostrom is in agreement with the conclusions presented here that hadrosaurian dental batteries together with a high degree of tooth wear must have been used to masticate food of substantial resistance; however, he suggests that this is evidence for foraging upon more resistant terrestrial vegetation. Such a dental battery could be used just as effectively for mastication of invertebrates. Ostrom (1964:987-989) suggests correctly that in the hadrosaurian localities of the Edmonton, Belly River, Judith River, Kirkland-Fruitland, Ojo Alamo and Lance marginal and aquatic vegetation are relatively rare in the paleoeco- systems. This might, of course, be the result of fortuitous preservation, but even if truly representative it could be interpreted as evidence that aquatic and marginal vegetation was rare in the area where hadrosaurians were abundant, a position taken by Ostrom. On the other hand it could also be evidence for the needed addition of invertebrates in the hadrosaurian diet. One must keep in mind that the rarity of marginal and aquatic vegetation as computed by Ostrom is an absolute occurrence affected by all vagaries of preservation, but in the dynamics of the ecosystem it is the relative availability of the plants compared to the needs of the hadrosaurian that is significant. Hadrosaurians were efficient bipeds and as Ostrom indicates (1964:990- 993) the articular surfaces of the pes and hind limbs, the large fourth tro- chanter, ossified sacral tendons, and reduction of the carpus point strongly to this conclusion. These characteristics when treated by themselves may well suggest a terrestrial habitat but when the specialized nature of the bill is taken into consideration an alternative hypothesis seems more fitting. Therefore the conclusion that hadrosaurians were either aquatic or shallow water feeders seems appropriate. Under these circumstances the bipedalism as well as the specialized bill would be adaptively advantageous. In addition such a feeding habit would give credence to the more obvious adaptive significance of the laterally compressed tail and the probable presence of webbed hind feet, rather than to suggest that these characteristics were for protection allowing the otherwise defenseless hadrosaurian to retreat to the water when threatened (Ostrom, 1964:993-995) . Restoration of Hadrosaurian Head Figures 5, A, B, and C are sketches illustrating the appearance of billed hadrosaurian dinosaurs. Evidence of bills have been found in Anatosaurus annectens and Corythosaurus excavatus but not in C. intennedius as restored in Figure 5C. In the skull of C. intermedins, however, the distal part of the dentary is greatly deflected while the maxillaries are not. The paratype skull of C. intermedins (ROM 4671 ) shows this deflection while, in the same skull, the tooth rows are occluded in normal position. If the specimen did not have a bill on the upper mandible, then there would have been a wide, opened gape at the anterior end of the mouth. The restoration of C. intermedins was made as this form has the most exaggerated gape of any hadrosaurian, a condition which suggests the presence of a bill, although the magnitude of the gape may be partly due to accidents of preservation. 1970 Hadrosaurian Dinosaur Bills 13 Figure 5. (A) Restoration of the skull of Anatosaiirus cumectens. Figure 5. ( B) Restoration of the skull of Corythosaurus excavatiis. 14 Contributions in Science No. 193 Figure 5. (C) Restoration of the skull of Corythosauriis intermedins. Literature Cited Abel, O. 1912. Grunziibe der Palaeobiologie der Wirbectiere. Stuttgard, E. Schweizerbart’s Verlagsbuchhandlung, xvi-708. Cope, E. C. 1883. On the characters of the skull of the Hadrosauridae. Acad. Nat. Sci. Phila. Pr. 35:97-107. Krausel, R. 1922. Die Kaubewegung und Lebenweise von Edmontosaiirus spec, auf Grund der mechanischkonstruktiven Analyse, Palaeobiologica 5:409-422. Lull, R. S. and Wright, N. E. 1942. Hadrosaurian dinosaurs of North America. Special Paper Geol. Soc. Amer. 40:xi-242. OsTROM, J. H. 1961. Cranial morphology of the hadrosaurian dinosaurs of North America, Amer. Mus. Nat. Hist. Bull. 122(2) : 37-186. . 1964. A reconsideration of the paleoecology of hadrosaurian dinosaurs, Amer. Jour. Sci. 262:975-997. Sternberg, C. M. 1935. Hooded hadr’osaurs of the Belly River series of the Upper Cretaceous: A comparison with descriptions of new species, Canada, Dept, of Mines, Nat. Mus. Bull. 77(52) : 1-37- Versluys, J. 1923. Der Schadel des Skelettes von Trachodon annectens im Senken- berg — Museum, Abh. Senckenb. Naturf. Ges. 38: 1-19. Accepted for publication March 17, 1970 LOS ,1 ANGELES ' COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE Number 194 July 1, 1970 ANOTHER NEW NIGHT LIZARD {XANTUSIA) FROM DURANGO, MEXICO By Robert G. Webb Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Changes in the manuscript after galley proof will be billed to the author. Unless otherwise requested, page proof also will be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor ANOTHER NEW NIGHT LIZARD (XANTUSIA) FROM DURANGO, MEXICO By Robert G. Webb^ Abstract: Ten specimens of Xantiisia from a desert habitat in eastern Durango, Mexico are described as a new subspecies of the previously monotypic X. henshawi. The Durango popula- tion of X. henshawi is known only from the type locality; indi- viduals occur in cracks and crevices of igneous (andesite) outcrops. Xantiisia exlonis Webb is regarded as a subspecies of X. vigilis. In May 1969, Dr. Charles C. Carpenter sent me one xantusiid lizard from the area in eastern Durango from which I had previously described Xantusia extorris (Webb, 1965). The specimen sent by Carpenter was not X. extorris, but resembled X. henshawi. Field work in August, 1969, yielded nine addi- tional specimens from igneous outcrops and thereby revealed the presence of another disjunct population of Xantusia in eastern Durango. Its morphology suggests a subspecific relationship with X. henshawi. I am grateful to Dr. Carpenter for sending me the specimen obtained by his field party in April, 1969; to Robert M. Kinniburgh for field assistance, and to the University of Texas at El Paso University Research Institute for defraying expenses in the field in August, 1969; to Dr. John W. Wright, Los Angeles County Museum of Natural History, and Drs. Alan E. Leviton and Steven C. Anderson, California Academy of Sciences, for the loan of com- parative material; and to Dr. Jerry M. Hoffer, Department of Geology, Uni- versity of Texas at El Paso, for analysis of rock samples. Preserved specimens are in the collections of the following institutions, to which the abbreviations in parentheses refer in the text: The California Academy of Sciences (CAS), the Los Angeles County Museum of Natural History (LACM), and the University of Oklahoma (OU). The terminology for scales follows Savage (1963). Xantusia henshawi bolsonae, new subspecies Type material-. Holotype: LACM 55956 (Fig. 1), obtained 6-6.5 road miles NE Pedricena on Mexican Highway 40, at an elevation of 4400 feet, Durango, Mexico, by Robert G. Webb and Robert M. Kinniburgh (original field number, RGW 5365) on August 24-25, 1969. Paratypes: Eight specimens (LACM 55957-64) with same data as holotype. One specimen (OU 32848) ^ Department of Biology, University of Texas at El Paso, El Paso, Texas 79999. 1 2 Contributions in Science No. 194 from same locality as holotype, obtained by Jan Sassaman (member of field party headed by Charles C. Carpenter) on April 1, 1969. Diagnosis: A subspecies of X. henshawi differing from X. h. henshawi in having fewer longitudinal rows of dorsal scales (averaging 50 rather than 63), seven (instead of six) supralabials, fewer femoral pores (averaging 7 rather than 11), in mostly lacking microscopic dark peppering laterally on the belly, in having a narrower head in relation to body length (head width/ body length averaging 0.16 rather than 0.18), and probably in attaining a smaller (body length 57 rather than 68 mm) maximal size. For further detail, see compari- sons and relationships. Description of Holotype Female (on basis of relatively small size of femoral pores) with normal xantusiid body proportions; snout pointed, rounded at tip; nostrils partly visible in dorsal view; eye relatively large, slightly closer to nostril than to ear opening; ear opening ovoid and slightly diagonal; anterior pregular fold indicated only by smaller scales that form two transverse rows medially; posterior pregular fold, four scale rows anterior to gular fold, indicated only by short intercalary row of scales on either side; distinct gular fold with enlarged scales on edge abruptly differentiated from scales of fold immediately posterior; limbs rela- tively short, fingers and toes overlapping when limbs adpressed to body. Measurements (in mm from dial calipers) : Snout-vent or body length, 56.8; tail (seemingly regenerated), 61.9; head length, from tip of snout to upper anterior margin of ear opening, 12.0; maximal width of head, 8.7; depth of head, 4.4; length of large-scaled part of head, 11.5; diameter of eye, 2.4; distance from eye to tip of snout, 4.2; distance from tip of snout to gular fold, 9.1; length of leg, about 22.0; length of fourth toe, about 6.4; axilla-groin length, 28.3. Squamation: Rostral broader than high, followed in order by two nasals, a frontonasal, two prefrontals, a median, two frontals, a hexagonal inter- parietal that separates two parietals, and two large postparietals; interparietal with obscure parietal eyespot; postparietals separated posteriorly by small, wedge-shaped interpostparietal that touches nuchals; nasals, prefrontals, frontals, and postparietals in contact; labials, loreal region and dorsal head scales minutely pustulose (pustules confined to edges of posterior head scales) ; nostril pierced in suture between rostral, nasal, postnasal, and first supralabial; nostril followed on side of head by postnasal, anterior loreal, large posterior loreal, two loreolabials, uppermost smallest, and small preocular scales of ocular ring; postnasal touching nasal and frontonasal above and first supra- labial below; anterior loreal touching frontonasal and prefrontal above and first and second supralabials below; posterior loreal touching prefrontal and first supraocular above and second (barely) and third supralabials below; small, upper loreolabial touching first supraocular and posterior loreal; large, 1970 Another New Night Lizard 3 triangular, lower loreolabial touching posterior loreal and fourth supralabial; scales of ocular ring small, two postoculars largest; uppermost postocular touching fifth supraocular; five supraoculars, first touching upper loreolabial, posterior loreal, prefrontal, median, and frontal (first supraocular divided into two scales on right side); fifth supraocular touching uppermost postocular, pretemporal and first temporal; pretemporal, touching fourth and fifth supra- oculars, frontal, and parietal, followed by four (left) and five (right side) temporals that border parietal and postparietal; seven supralabials; four (left) and five (right side) infralabials; pretympanic scales variable in size, largest bordering temporal and auricular scales; five to seven small auricular scales; mental about as broad as long; three large pairs of postmentals, first pair touching medially; anteriormost pregular scale largest, between second pair of postmentals; 35 transverse rows (some intercalary from either side) of pregulars; 15 rows of pregulars between first pair of postmentals and anterior- most pregular fold; 17 rows between pregular folds, and four rows (curved from either side) of pregulars between posteriormost pregular fold and gular fold, excluding 13 enlarged scales on edge of gular fold; five rows of small scales in gular fold, anteriormost scales smallest. Small, roundish, granular scales on back, sides, and limbs; some radials and femorals enlarged; 14 longitudinal rows of squarish ventrals at middle of body, lateralmost scales with curved outer edges and slightly smaller than adjacent ventrals; transverse rows of ventrals alternating along midline in some places, so that number is 32 on right and 34 on left side; two pairs of enlarged preanal scales, posteriormost pair largest; about 49 longitudinal dorsal scale rows at midbody; about 1 18 transverse rows of middorsal scales (from post- parietals to level above posterior surface of thighs); 25 lamellae under fourth toe of left foot; seven femoral pores on each side; caudal scales rectangular, squarish proximally underneath tail. Coloration and pattern (recently preserved, in alcohol); Dorsal ground color of head, body, limbs, and tail pale yellow-buff; sides of body whitish; ventral surfaces pale yellow-white; ground colors showing uneven distribution of minute black peppering under magnification; black peppering absent on postorbital head stripes, and mostly lacking proximally underneath tail and on chest and belly. Dorsal surface having pattern of dark markings; dark brown spotlike markings on head, largest posteriorly; brown markings (paler than elsewhere) on limbs; dark brown-black markings on back, some markings lineate or of irregular shape; ground color of back less extensive than markings; dark brown-black markings on tail, tending to form ringed pattern; distinct, pale postorbital stripes extending posteriorly from upper margin of eye along juncture of temporals, and parietal and postparietal to small granular nuchal scales of neck; postorbital stripe with wide (as wide as pale stripe), continuous, brown-black lower border, but having border of two dark spots above; two. 4 Contributions in Science No. 194 pretympanic dark spots (right) or distinct, barlike mark (left side) between eye and ear opening; dark markings on side of neck and body paler and smaller than those dorsally; dark markings on side of head; labial region heavily suf- fused with dark punctations. Variation The general morphology of the nine paratypes closely approximates that of the holotype, except that the anterior pregular fold is indicated by a fold of skin in OU 32848. The limbs seem to overlap more than in the holotype with the fourth toe reaching the region of the elbow when the limbs are adpressed to the body. The measurements (in mm, means with ranges in parentheses) are based on the holotype and nine paratypes. The body length of the smallest specimen is 34.4. The body length of the largest female (holotype) is 56.8, and of the largest male is 50.8. Severed tails that have undergone complete regen- eration are often difficult to distinguish from tails that have never been broken. The body length/ tail length ratio of three specimens that appear to have original tails is 0.69 in one female with a body length of 51, and 0.74 and 0.75 in two males with body lengths of 50 and 51, respectively. The head length/ body length ratio is 0.22 (0.21-0.24); the head width/body length ratio is 0. 1 6 (0.1 5-0. 17); the head width/ head length ratio is 0.72 (0.68-0.79) ; and the head depth/head length ratio is 0.38 (0.35-0.42). The squamation of the paratypes resembles that of the holotype. The size of the small interpostparietal is often variable; the scale is either lacking or very small, or in LACM 55957, it is especially elongate. Four specimens have either one or two small intercalary scales at the rear of the interparietal; in LACM 55961 the intercalary scale touches the small interpostparietal, and in LACM 55964 the two scales nearly touch. In LACM 55963 the postparietals are completely separated by a large interpostparietal. The nasals, prefrontals, frontals, and postparietals touch medially in all specimens. An incomplete longitudinal suture partly divides the posterior loreal (right side) in LACM 55961. All specimens have minutely pustulose head scales. The supraoculars are five, in one case four (second and third supraoculars fused). There is often one (sometimes two) intercalary scale between the first and second supra- oculars and the adjacent median head scales; the intercalary scale seems to represent a fragment of the first supraocular. The temporals, variable in size, are usually five, in some cases four or six. The supralabials are usually seven, in one case six, and the infralabials are usually five, in one case four, and in another six. The largest postmentals are usually three, in one case four. The transverse rows of ventrals average 32.8 (31-34), excluding the two (usually) or three rows (each of two scales) of enlarged preanals, and small intercalary rows, usually in the chest region. The longitudinal rows of dorsal granular scales across the middle of the back average 49.6 (48-52), and the 1970 Another New Night Lizard 5 Figure 1. Comparison of dorsal pattern of Xantiisia h. holsonae and X. Ii. henslunvi\ four individuals of holsonae at left and two individuals of henslunvi at right; museum numbers of specimens from left to right: LACM 55956 (holotype), LACM 55960, LACM 55962, LACM 55964, CAS 57852, CAS 58089. femoral pores average 6.9 (6-8). The femoral pores in females are often not well developed being marked by shallow depressions. The average number of transverse rows of pregulars between the anterior pregiilar fold and gular fold is 20.8 ( 1 9-23 ) , and of the enlarged scales on the gular fold is 11.8 (10-13). The nine paratypes resemble the holotype in general features of pattern. All specimens have distinct pale postorbital stripes, which are accentuated by lacking microscopic black peppering and by having blackish borders that may be narrowly interrupted, especially the upper border. The black peppering is usually lacking proximally on the underside of the tail, is lacking or diffusely scattered on the underside of the head, and is mostly lacking on the chest and belly. The pattern on the neck and back of the largest specimens consists of small, irregularly shaped, dark blotches (generally intermeshed in jigsaw fashion) with the total area of blotching about equal to or slightly exceeding that of the ground color. However, ontogenetic variation is suggested since the corresponding pattern of the smallest specimen (LACM 55964, 34 mm body length) is obscure, consisting of small, mostly punctate, dark markings. See Figure 1 for variation in dorsal pattern. 6 Contributions in Science No. 194 Living individuals had pale yellow to buff ground color on the back, with the color somewhat more intense on the head. The tails were pale yellow, whereas the limbs and sides of the body were white. The dorsal markings on head, body, limbs and tail were dark brown-black. The ventral surfaces were white with the belly and tail tinged with yellow, especially ventrolaterally. The irises were orange with black reticulations concentrated anteriorly and posteriorly. Etymology. The name bolsonae refers to the geographic position of this race in a southern outlier of the Bolson de Mapimi. Comparisons and Relationships Based on gross morphological similarities there are two distinct species of Xantusia — vigilis and henshawi. Hitherto, X. vigilis contained six subspecies, vigilis, gilberti, wigginsi, utahensis (Savage, 1963:35), sierrae (Bezy, 1967a), and arizonae (Bezy, 1967b). Xantusia extorhs closely resembles X. vigilis, but because of its geographic isolation and because of the status of the mor- phologically similar arizonae as a distinct species at that time, extorris was considered a distinct species (Webb, 1965). The degree of morphological dif- ference between the subspecies of X. vigilis and between X. henshawi and X. vigilis indicates that relationships within the genus are best expressed by considering X. extorris as a subspecies of X. vigilis, and the isolated population described herein as bolsonae as a subspecies of the previously monotypic X. henshawi. The distinct species, X. henshawi, is most readily distinguished from the other species in the genus, X. vigilis, by 14 longitudinal rows of ventral scales (instead of 12) , and a dorsal pattern of relatively large, black blotches (instead of small dark dots) . The 108 specimens examined from throughout the range of X. h. hen- shawi do not seem to show any geographic variation, except in the number of femoral pores, and possibly in the number of temporals. In 54 specimens from the northernmost part of the range in the vicinity of the San Jacinto Mountains, Riverside County, California, the femoral pores average 1 1 .9 (8-16), whereas the femoral pores in 54 specimens from the other localities to the south average 9.7 (7-12); the combined number of femoral pores averages 10.7. The num- ber of temporals shows a slight increase from north to south. In populations to the north the number averages 5.2 (4-8), whereas the temporals in other populations to the south average 5.9 (4-7); the combined average number of temporals is 5.6. There seems to be no significant geographic difference in the number of dorsal granules across the back (62.8, 56-71), transverse rows of ventral scales (32.7, 30-36), enlarged scales on edge of gular fold (10.3, 7-14), transverse rows of pregulars between the anterior pregular and gular folds (22.6, 19-29), infralabials (5.1, 4-7), supralabials (6.2, 5-8), in the ratio, head width/ body length (0.18, 0.16-0.28), and in dorsal pattern. The average 1970 Another New Night Lizard 7 body length of the 10 largest males is 60.4 (59-62), and of the 10 largest females is 65.3 (63-68). Xantusia h. bolsonae has fewer longitudinal rows of dorsal granules (49.6, 48-52) than X. h. henshawi (62.8, 56-71). The average number of femoral pores is fewer in X. h. bolsonae (6.9, 6-8) than in X. h. henshawi (10.7, 7-16) . The number of supralabials is usually seven in X. h. bolsonae (95%) rather than six as in X. h. henshawi (72%; the frequency of seven supralabials is 25% ). In some specimens of X. h. henshawi it is difficult to determine which scale is the last supralabial. Xantusia h. bolsonae seems to have a slightly narrower head in relation to body length (head width/body length averaging 0.16, 0.15-0.17) than X. h. henshawi (0.18, 0.16-0.28). The two lateralmost longitudinal scale rows of the belly have a few scattered dots but are mostly devoid of microscopic black peppering in X. h. bolsonae. The ventral black peppering is extensive in X. h. henshawi but is diffuse or absent medially and most concentrated laterally; the two, and often the four, lateralmost scale rows are usually liberally black-peppered in X. h. henshawi. The postorbital stripes are distinct in all specimens of X. h. bolsonae\ when present in X. h. henshawi the stripes are often indistinct. A comparison of the dorsal patterns on the back is difficult to evaluate, but generally the pattern is of relatively smaller, more often irregular-shaped, dark markings in X. h. bolsonae than in X. h. henshawi (see comparison in Fig. 1 ). Xantusia h. bolsonae may be a smaller subspecies (maximum size, 57) than X. h. henshawi (maximum size, 68). Comparative material examined of X. h. henshawi (108 specimens): Baja California: 14 specimens — CAS 1 1932, 57294-96, 57455-58; LACM 36542-44, 36555-56, 36564. California. Imperial County: 20 specimens — CAS 22574-75, 57846-59; LACM 52668-71. Riverside County: 54 specimens —LACM 3769-72, 3774-76, 14513, 14519, 14524-26, 14528-29, 14531-49, 22370-73, 22376, 22381-84, 22389, 22391-92, 22395, 22397, 26800, 52672- 77. San Diego County: 20 specimens — CAS 22017-18, 22566, 22581-85, 58083, 58086, 58089, 58093-94, 64298; LACM 52678-83. Geographic Range and Habitat The ten known specimens of X. henshawi bolsonae are from within about a one-half mile radius, 6-6.5 road miles (Mexican Highway 40) NE Pedricena, Durango, Mexico. The nearest localities for X. henshawi, some 800 miles to the northwest, are in southern California and adjacent northern Baja California (see map 72 in Stebbins, 1966). The extent of geographic range of X. h. bol- sonae is not known. Apparent habitat was intensively examined only at the type locality. Perhaps X. h. bolsonae is restricted, at least in Durango, to the general desert habitat south of the Rio Nazas where X. vigilis extorris, Sceloporus maculosus, and Cnemidophorus inornatus paululus also have restricted dis- tributions in Durango. The general desert habitat in this part of the Chihuahuan Desert has been 8 Contributions in Science No. 194 Figure 2. Type locality of Xantusia lienshawi bolsonae, 6-6.5 road miles NE Pedricena, elevation 4400 ft., Durango. Photographs taken August 25, 1969. Top: General view of hills and low mountains showing isolated, small, andesitic out- crops. Bottom: Outcrop of large weathered boulders of andesite. 1970 Another New Night Lizard 9 described elsewhere (Webb, 1965). The general area is of relatively low elevation and physiographically is part of the extensive Bolson de Mapimi to the north; the topography is mostly irregular consisting of hills and low moun- tains. At the type locality there is a series of low hills with rock outcrops (Fig. 2). The principal vegetation covering the low hills includes lechuguilla (Agave lecheguilla) , a maguey (Agave sp.), a treelike yucca (Yucca sp.), ocotillo (Fouqideria splendens) , cholla (Opuntia sp.), small prickly pear (Opimtia sp.), and leatherplant (Jaimpha cuneata) . A shrubby legume (re- sembling Mimosa) and a small-leaved shrub (resembling Rhus) are often associated with the rock outcrops. Rock outcrops in the area are not extensive and generally form either a cluster of large boulders (Fig. 2) or a rim-rock escarpment of limited extent up to 10 feet in height. Xantiisia h. holsonae is associated with these igneous outcrops; the rock is best described as a porphyritic andesite (with phenocrysts of biotite and feldspar) and may have either a gray or pink hue. Most of the rock exposures in this desert are gray dolomitic limestone; these outcrops frequently are extensive and have rough corrugated surfaces. Although collecting activities were limited, X. henshawi was not found to be associated with the limestone. The andesite weathers mostly into large chunks. The lizards were found in both vertical and horizontal crevices. They occurred in some horizontal crevices that had small amounts of soil. All individuals occurred singly, and were observed only after moving large chunks of andesite by using a large crowbar. Temperatures at 2 pm on August 24 were: rock surface in sun, 48° C; air in sun, 37° C; and air in shade of rock crevice, 34° C. Three species of rock-dwelling Sceloporus — S. poinsetti, S. jarrovi, and S. maculosiis — are associated with both the andesite and limestone outcrops. One specimen of S. poinsetti and one of X. henshawi were found in the same rock crevice. Xantusia h. holsonae is sympatric with X. vigilis ext orris in eastern Dur- ango; one specimen of X. vigilis was found under a fallen yucca a few feet below an andesitic outcrop. Both populations are disjunct from the species in the southwestern United States and northwestern Mexico. The two Durangan populations occur east of the continental divide, and their ranges appear to be limited to the west by grasslands at higher elevations. The morphological differences between the two subspecies of X. henshawi seem to be greater than those between X. vigilis extorris and X. vigilis gilberti (the subspecies that extorris most closely resembles). Resumen Dies especimenes del genero Xantusia de la habitacion del desierto en el este de Durango, Mexico, estan descritos como una nueva subspecie del previo monotipico X. henshawi. La poblacion de Durango de X. henshawi es 10 Contributions in Science No. 194 conocida solamente por el tipo local; individuos ocurren en gfietas y hendiduras en las pretuverancias de andesita ignea. j ^ Xantiisia extorris es ahora considerada como una subespecie de X. vigilis. Literature Cited Bezy, Robert L. 1967a. A new night lizard {Xantiisia vigilis sierrae) from the southern Sierra Nevada in California. Jour. Ariz. Acad. Sci. 4(3) : 163-167. . 1967b. Variation, distribution, and taxonomic status of the Arizona night lizard {Xantiisia ari zonae) . 1967(3) :653-661. Savage, Jay M. 1963. Studies on the lizard family Xantusiidae IV. The genera. Los Angeles Co. Mus., Contrib. Sci. 71-1-38. Stebbins, Robert C. 1966. A field guide to western reptiles and amphibians. Hough- ton Mifflin Co., Boston, xiv-f-279 p. Webb, Robert G. 1965. A new night lizard (genus Xantiisia) from Durango, Mexico. Amer. Mus. Novitates 223 1 : 1-16. Accepted for publication January 28, 1970 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS cX,, IN SCIENCE Number 195 July 1, 1970 THE BIRDS OF THE KALINZU FOREST, SOUTHWESTERN ANKOLE, UGANDA By Herbert Friedmann and John G. Williams Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF. — Authors will be sent galley proof which should be corrected and returned promptly. Changes in the manuscript after galley proof will be billed to the author. Unless otherwise requested, page proof also will be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor THE BIRDS OF THE KALINZU FOREST, SOUTHWESTERN ANKOLE, UGANDA By Herbert Friedmann^ and John G. Williams- Abstract: The Kalinzu Forest, southwestern Ankole, Uganda, had never been explored zoologically before our field team worked there late in 1969. The forest is being cut down rapidly, and the present report on its avifauna may well be the final as well as the first account of its bird life. In all some 123 species are here recorded, and indicate that the Kalinzu Forest is (or was) a meeting place for some montane forest birds with the lowland forest species of western Uganda. The Kalinzu Forest, a medium altitude (5000 feet) moist, evergreen forest, in southwestern Ankole Province, western Uganda, lies a short distance to the south of Lake George and to the east of Lake Edward, at 0° 25' S, 30° 05' E. It is close to, but 1400 feet higher than, the larger Maramagambo Forest, which extends to the border between Ankole and Kigezi, the extreme southwestern province of Uganda. The Kalinzu Forest had never been explored zoologically before, and, indeed there seem to be no Kalinzu bird records in print. The survey on which the present report is based was carried out under sponsorship of National Science Foundation Grant GB-7787, by a team led by Robert M. Glen and Andrew Williams. Their stay in the Kalinzu covered just a month; their first specimens were collected on October 27, the last on November 26, 1969. The collection consists of 433 specimens of 121 species of birds, and also sight records of two more species. This comprises the first, and quite possibly, the last, collection to be made in that forest. In the original plan submitted to National Science Foundation in 1966 for the survey of the terrestrial vertebrates of the isolated forests of western Uganda it was pointed out that these areas have, or may be expected to have, faunas that are west African rather than east African in their main affiliations, and, as such, they form the easternmost outliers of the ranges of many Congo- lese species. It was also stressed that there is an unfortunate urgency in the survey as most of these forests are being cut down at an alarming rate to make room for agriculture, and that if the survey were postponed by even a few years, it might be too late, and our potential knowledge of the original, natural distributional limits of many of these sylvan species would be thereby re- stricted forever. 1 Director, Los Angeles County Museum of Natural History. (Currently, Research Associate in Ornithology, Los Angeles County Museum of Natural History.) -Research Associate in Zoology, Los Angeles County Museum of Natural History. 1 1 Contributions in Science No. 195 The Kalinzu Forest may be taken as an extreme instance of the urgency for collecting data before it is too late; it is an area in which there is reason to think the survey was accomplished literally at the last opportunity. In a letter from the field, dated November 17, 1969, Robert M. Glen wrote as follows: “. . . The Kalinzu is probably the poorest forest we have yet collected. Vast areas, almost the whole forest, have been worked for timber with the majority of the remaining non-timber trees poisoned by the forestry department . . .” He was informed by the area forest officer, “. . . who spent two days in our camp, that we are seeing the last of the forest in its present state, since it is his inten- tion, and that of the deparment, to utilize all standing timber as charcoal to supply a nearby tea factory which is at present being erected. Within a rela- tively short space of time ... all vegetation will be of a secondary nature only.” It must be emphasized that it is not our intention to deplore what is happening to original, natural environments, as these changes are probably 1970 Birds of the Kalinzu Forest 3 inevitable, and no criticism of the trend is here implied. We merely wish to stress the need for gathering information about such areas while it is still possible to do so. The following catalog of Kalinzu birds may well be smaller than it would have been if made five or ten years earlier, but we must be grateful that it was still possible to report so many kinds of birds. The fact that the forest lies at an altitude of 5000 feet makes it a meeting place of many lowland forest species with a smaller number ordinarily considered highland forest birds. Among the latter may be mentioned Pycnonotus masukensis kakamegae, Coracina caesia pura and Erithacus aequatorialis, all of which were collected, and Tauraco johnstoni and Nectarinia purpureiventris, which were seen, but not obtained. The name Kalinzu Forest has previously been used for the lowest level of the Impenetrable Forest, in southwestern Kigezi Province, as was pointed out recently by Keith et al. ( 1969: 9-10). While it is unfortunate thate this ambigu- ous application of the name exists, it must be clear that the Kalinzu Forest with which this paper deals is quite a different area, in Ankole Province, some 40 miles north-north-east of the nearest edge of the Impenetrable Forest, and at an altitude of 1500 feet higher than the lowest levels of the latter. In the following annotated list we have kept to a minimum discussions of different systematic treatment by different authors, as our real purpose is merely to record what birds are known to occur (possibly by the time this paper appears, what are known to have occurred) in the Kalinzu Forest. By and large, we follow White (1960-1965) in nomenclature, and in the recognition or rejection of subspecies, except where more recent revisions have been issued, principally in the volumes completing the Check-list of Birds of the World begun by the late J. L. Peters. The order of the passerine families follows that of Chapin (1953, 1954). Accipitridae Polyboroides typus Smith On November 26 an African harrier hawk flew into a tree in an area in the Kalinzu Forest where several mist nets had been placed. It landed among a troop of monkeys of several species, all of which made a tremendous racket, but which dispersed as the hawk was shot. It proved to be an adult male, testes not enlarged; bill black, the base and soft parts whitish; iris dark brown; feet yellow; weight 720 grams; stomach empty. Accipiter tachiro sparsimfasciatus (Reichenow) One specimen, male, testes not enlarged, November 19; bill black, cere and gape greenish yellow; basal ring gray; iris yellow; feet yellow; weight 227 grams; stomach empty. This goshawk was caught in a net in open under- growth in an area of remaining natural forest. 4 Contributions in Science No. 195 Hieraaetus africanus (Cassin) Cassin’s hawk-eagle is still generally considered a rare bird everywhere in its range, chiefly because it is rare in museum collections, yet three examples were obtained in eleven days, November 13 to 23, suggesting that in the Kalinzu Forest it is by no means uncommon. The three examples, two males and one female, were in non-breeding condition; bill black, the base gray, the soft parts, cere, and gape pale yellow; ring at base pale gray; iris medium brown to brownish cream; feet pale yellow; weight, female 1153 grams; males 938 and 1049 grams; all stomachs were empty. One of these birds was shot out of a tall tree, and while the collector’s assistants were searching for it in the undergrowth, another one, probably its mate, flew into the same tree and was also collected. The other example was caught in a trap baited with a small chicken. The Kalinzu Forest is the second area in Uganda from which this hawk is now known, the other being the Impenetrable Forest in the extreme southwestern part of the country. It may be worth noting that in the Im- penetrable Forest this hawk must also be common as our collectors in 1969 noted (but did not collect) the species on 21 occasions during two months in the field. Keith and Twomey (1968: 537-538) considered it significant that their single male specimen from the Impenetrable Forest had an unusually long tail, 248 mm, as compared with 225 and 234 mm in two topotypical male birds from Gabon. Our two Kalinzu males have tails 228 and 231 mm long, the female 266 mm. Our males therefore agree in this dimension with Gabon examples and negate any suspicion that the eastern population of the species may prove to be a race characterized by its tail length. Rallidae Sarothrura elegans (Smith) One buff-spotted crake, a female with the ovary much enlarged, was collected on October 30, when it flew into the camp at night, apparently attracted by one of the pressure lamps hanging in the mess tent. It had the bill blackish gray, the lower base whitish; iris dark brown; feet dark brown; eye skin gray; weight 44.5 grams; stomach contents grit and small beetle fragments. Sarothrura pulchra centralis Neumann The white-spotted crake was not obtained in the Kalinzu Forest, but on November 8 an adult male was taken in the nearby Maramagambo Forest, 3600 feet (as compared with 5000 feet at Kalinzu). It showed only slight gonadal swelling; bill black; iris dark brown; feet dark gray; weight 46 grams; stomach contents insect fragments. 1970 Birds of the Kalinzu Forest 5 COLUMBIDAE Cohimba unicincta Cassin The afep pigeon is represented by an unsexed adult taken on November 21; bill gray, paling to whitish gray at the tip; iris orange-red; feet pale gray; weight 379 grams; stomach contents seeds of small fruits. The bird was seen feeding in a fruiting tree in an open area of the forest. Streptopelia semitorquata (Riippell) Two adults with enlarged gonads, one male, one female, were collected on November 14, from a flock of several red-eyed doves feeding on the small berry-like fruits of a tall tree. Bill dull black, the gape pinkish in the male, dull maroon in the female; iris dark red; bare skin dull crimson to dull maroon; feet dull pale maroon; weight 162.5 grams (female), 192.5 grams (male); stomach contents seeds of small fruits. Tiirtiir tympanistria (Temminck) One male with enlarged testes, November 13; bill black; soft parts dark dull maroon; iris very dark brown; bare skin dark gray; feet dull maroon; weight 75 grams; stomach contents seeds and grit. Treron aiitralis gihherifrons (Madarasz) This green pigeon was met with twice, on Novemeber 12 and 16. On each day an adult male was taken, the second of which had large testes; bill pale gray; soft parts and cere dull orange; iris bright metallic sky-blue; feet orange- red; weight 225 and 226.5 grams; stomach contents small fruits. Musophagidae Taiiraco schutti emini (Reichenow) On October 29 one male of this black-billed lourie was collected in a patch of secondary forest which had recently been felled; testes not enlarged; bill black, the lower base red; iris dark brown; feet black; bare eye skin red; weight 267 grams; stomach contents fruit seeds. Tauraco johnstoni (Sharpe) Robert Glen saw, but was unable to collect, what “. . . was certainly a Ruwenzori turaco . . .”, which would establish a new minimal altitude for this montane bird. A specimen would not only have been welcome to corroborate the identification, but also to determine which of the two races occurs in the Kalinzu Forest. 6 Contributions in Science No. 195 CUCULIDAE Ciiculus clamosus gabonensis Lafresnaye An almost adult female in the pale-bellied ''mabirae'" type of plumage of this forest cuckoo was taken on November 25; maxilla black; mandible slightly grayish; iris dark brown; feet white with pale gray scales; weight 87 grams; stomach contents hairy caterpillars. This specimen has the chin, throat and upper breast pale rufescent barred fairly broadly with blackish; the lower breast and abdomen whitish cream with only a small number of incomplete fuscous cross bars, especially toward the sides of the breast and abdomen, the under tail coverts unmarked cream. Chrysococcyx cupreus cupreus (Shaw) The emerald cuckoo undoubtedly occurs in every forest in Uganda. In the Kalinzu an adult female with an enlarged ovary was taken on November 2; maxilla blackish gray; mandible below greenish; iris dark brown; eye ring apple-green; feet blue-gray; weight 38 grams; stomach contents caterpillars. A few of the abdominal feathers have a pronounced yellowish wash in their pale areas — a slight variation in the direction of the richly yellow abdomen of the male. Ceuthf nochares aereus aereus (Viellot) This cuckoo is represented in the collection by two males, one with large, the other with small, gonads, November 14 and 22; bill lemon-yellow; iris chestnut-brown to dark red; feet black; bare skin behind eye blue, in front of eye apple-green; weight 56 and 66 grams; stomach contents caterpillars, grass- hoppers and beetles. These birds were found feeding in creeper-covered bush in mixed, exploited forest. Strigidae Glaucidium tephronotum medje Chapin Three examples, two males and one female, of this owl were taken November 1, 22, and 26, respectively. It follows that the species must be fairly common in the Kalinzu Forest, and in the forests of western Uganda generally, as six more specimens from the Bwamba, Kibale, and Impenetrable Forests are now in the Los Angeles County Museum of Natural History. One of the present individuals was kept alive for some days and was heard to give a rather mournful sounding, whistled poop several times in rapid succession. This owl was caught in mist nets in the undergrowth in both primary forest and in areas where the trees had been heavily exploited. All three examples were in non- breeding state; bill pale greenish yellow to dull apple-green; cere yellow to greenish yellow; iris bright yellow (with gray flecks in one of the males); feet 1970 Birds of the Kalinzu Forest 7 yellow; eyelids olive; weight 83.5, 94, and 103 grams (the female the heaviest) ; stomach contents insect fragments and one green beetle. There is much individual variation among these owls, especially in the amount of rufescence on the underparts, and in the degree of difference between the dark grayish of the crown and occiput and the brown of the back, the distinction being more marked in some individuals than in others. As we pointed out (Friedmann and Williams, 1968: 15) some of the dorsal variability is probably due to wear as the brown feathers of the back have dark grayish tips when fresh, and as these are abraded, the brown color becomes correspondingly more prominent. Because of the scarcity of specimens of this owl in most museum collec- tions, the measurements of the nine examples now in Los Angeles form a useful addition to those discussed by Keith and Twomey (1968 : 540-541 ) . Our eight males have wing lengths of from 1 12 to 119 mm; tail lengths 80 to 88 mm; our one female, wing, 115 mm, tail, 85 mm. Apodidae A pus affinis affinis (Gray) This swift is fairly common in the Kalinzu area, where three males were taken November 1, 4, and 6. Two of them had enlarged gonads, the third showed only slight testicular swelling; bill black; iris dark brown; feet black; weight 25 to 28 grams; stomach contents flying insects. The collectors noted that this species was nesting under the eaves of the forestry department rest house at the edge of the forest. Chaetura sabini Gray Sabine’s spine-tailed swift appears to be commoner and more widely distributed in western Uganda than was formerly thought. Five females were taken in the Kalinzu area October 31 to November 10, and others have been reported or collected in the Bwamba, Budongo, and Kibale Forests as well. Of our present examples only one showed much gonadal enlargement, the others had small ovaries; bill black; iris dark brown; feet maroon to maroon-gray; toes blackish; weight 16.5 to 18 grams; stomach contents small flying insects. These specimens were shot from groups flying over the swampy forest together with other swifts and with black rough-winged swallows {Psalidoprocne) . In an earlier paper we (Friedmann and Williams, 1968 : 16) noted that the wing length of our sole Kibale specimen was only 1 10.9 mm, which seemed to uphold the validity of the supposedly short winged race ogowensis. Our present five examples have wing lengths varying from 1 14 to 126 mm, and they, in turn, make it more probable that ogowensis is not a recognizable subspecies. The use of a binomial for this swift thus appears to correctly reflect our knowledge of its variability. 8 Contributions in Science No. 195 Trogonidae Apaloderma narina brachyurum Chapin Two male trogons, taken on the same day, November 25, differ in that in one the feathers of the rump and the upper tail coverts are somewhat bluer, less greenish than in the other; one had enlarged testes, the other showed only slight gonadal swelling; bill greenish gray becoming yellow at the base; iris chestnut-brown; feet pale flesh-brown; bare skin around eye apple-green; weight 68 and 70 grams, respectively; stomach contents caterpillars, one butter- fly and other insect fragments. The birds were flying after insects about 20 feet from the ground, but the presence of caterpillars in their stomachs is evidence that not all of their food is caught on the wing. Alcedinidae Halcyon malimhica malimhica (Shaw) Two specimens of this kingfisher were taken on November 18 and 19, one of each sex; the female with large ovary, the male with slightly swollen testes; bill crimson-red, the mandible and the lower base blackish; iris dark brown; feet orange, scales brownish; weight 89 and 92.5 grams, respectively; stomach of the female held a blue-green tree frog while that of the male was empty. Meropidae M crops gularis australis (Reichenow) One male and two females of this bee-eater were taken October 28 to November 4, all in non-breeding state; bill black; iris red; feet black; weight 30 to 34 grams; stomach contents beetles and one wasp. On November 8 another male was taken in the nearby Maramagambo Forest, 3600 feet. CORACIIDAE Eurystonius gularis neglectus Neumann Two male rollers with small gonads and one female with enlarged ovary collected November 10 to 16. The males are in a changing state of plumage with a mixture of blue and maroon-brown on the abdomen; bill yellow; iris dark brown; feet yellowish brown, the toes darker; weight 99 (female) to 1 17.5 grams (males) ; stomach contents termites and flying black ants. These birds were seen catching small insects on the wing in the open, exploited forest. Phoeniculidae Phoeniciilus hollei jacksoni (Sharpe) Two wood hoopoes were taken October 30 and November 4, the former Linsexed, the latter a male; bill orange-red to deep pinkish crimson; iris dark brown; feet red; weight 48 and 55 grams; stomach contents beetle larvae and other insect remains. 1970 Birds of the Kalinzu Forest 9 Bucerotidae Bycanistes siihcylindriciis suhquadratiis Cabanis One male hornbill in breeding condition, taken November 14; bill dull scaly brown, with base and half of casque pale gray; iris rich reddish brown; bare skin dull black; feet scaly black; weight “over 1600 grams”; stomach contents fruits. The collectors noted many more single birds (males?) than pairs, which seemed to suggest that the females were already holed up in their nests. Capitonidae Lyhius hirsutus ansorgei (Shelley) The hairy-chested barbet is fairly common in the Kalinzu forest. Three examples were collected October 31 to November 22; all with slight gonadal enlargement; bill black; iris dark brown; feet black; weight 45 to 49 grams; stomach contents berries and fruit seeds. These specimens were collected in the undergrowth in an unspoiled area of natural forest. Gyninobucco honapartei cinereiceps Sharpe The brown barbet was numerous and four specimens were taken Novem- ber 13 to 19, three of them with large gonads, one with small ones; bill blackish horn; iris cream to yellowish white; feet black; weight 62 to 67 grams; stomach contents green fruits. A female, November 19, was shot at a nesting (?) hole in a dead tree. Buccanodon duchaillui duchaillui (Cassin) On November 10 three adults, two females, 1 unsexed, were collected; this was the only time the field team met with this barbet. The birds were in non-breeding state; bill black; iris dark brown; feet dark gray; weight 33 to 36.5 grams; stomach contents small fruit seeds. All were netted in the under- growth in a heavily exploited part of the forest. Pogoniidus scolopaceus aloysii (Salvadori) This little tinker bird was common in the Kalinzu. Specimens were obtained October 28, November 14, and November 20, the first and the last were in breeding condition; bill black; iris cream to grayish yellow; feet dark gray; weight 15.5 to 17 grams; stomach contents insect fragments and small fruit seeds. Found in the undergrowth in exploited forest areas. Pogoniulus hdineatus mfiiinbiri (Ogilvie-Grant) The commonest tinker bird in Kalinzu; seven examples, October 29 to November 19; non-breeding condition; bill black; iris dark brown; feet black- ish; weight 10.5 to 13.5 grams; stomach contents fruit seeds and insect fragments. 10 Contributions in Science No. 195 Trachyphonus purpuratus elgonensis Sharpe This brightly colored species was one of the commonest barbets in the Kalinzu. Large numbers were caught in the mist nets (and released) where they were catching flying termites as they escaped from the ground. Five specimens were taken October 30 to November 12; four of them with small, one with large, gonads; bill lemon-yellow; iris dark red; feet blackish gray; weight 72 to 85 grams; stomach contents fruit and fruit seeds. Indicatoridae Indicator exilis pachyrhynchus (Heuglin) The least honey guide is represented in the collection by one adult male, with enlarged testes, taken October 30; bill black, the lower base whitish; iris dark brown; eye skin dark gray; feet olive-gray; weight 21.5 grams; stomach entirely filled with beeswax. It was collected when it came to a spot where a number of sunbirds were excitedly mobbing a snake. Although this was the only example noted, there is every reason to assume that this honey guide is no less numerous in the Kalinzu than it is in the other western Uganda forests. The local abundance of small barbets, of at least two species of Pogoniulus, would assure an ample supply of hosts for this brood parasite as well as for the next species. Indicator willcocksi willcocksi Alexander That this forest honey guide occurs in the Kalinzu is of interest in round- ing out our concept of its range in western Uganda, where it is already known from the Impenetrable and the Kibale Forests. It does not occur in the Bwamba woodlands, which are at too low an altitude, as its range seems to be above 4200 feet. Three specimens, one male and two females, were taken in the Kalinzu on October 27, November 13, and November 23, respectively; all were in non-breeding state; bill black, the lower base pale pinkish gray; iris dark brown; feet olive-gray; weight 16 and 17 grams in the females, 20.5 grams in the case of the male; stomach contents entirely beeswax in two, beeswax and small insect fragments in the third. In an earlier paper we (Friedmann and Williams, 1968: 17-18) described seven specimens from the Impenetrable and Kibale Forests as being slightly paler and with more of a yellowish tone below and less heavily streaked on the crown than were the available Cameroon and Congo specimens, and we raised the possibility of an eastern subspecies of /. willcocksi. The Kalinzu birds merely add to the variability, or the inconstancy of coloration as noted in the western Uganda population, and thereby tend to negate the idea of an eastern race of this honey guide. Furthermore, as we pointed out in the earlier paper, it would be necessary to see specimens from localities all across the Congo before it would be possible to ascertain where, if anywhere, such a division might be made. 1970 Birds of the Kalinzu Forest 11 PiCIDAE Campethera nivosa herberti (Alexander) Inasmuch as the buff-spotted woodpecker is known from many forested areas throughout Uganda, it is no surprise to find it in the Kalinzu as well. On November 22, two males were collected as they were watched searching for insects on a dead branch sticking up out of the undergrowth in an undisturbed area of primary forest. Both birds showed only slight gonadal enlargement; bill horn black becoming brown on the basal part of the mandible; iris dark brown; bare eye skin olive-gray; feet olive; weight 35 and 39 grams; stomach contents insect fragments. Campethera tullbergi taeniolaema Reichenow and Neumann The fine-banded woodpecker is a very common bird in the Kalinzu Forest. Our collectors obtained ten specimens from October 30 to November 8, all in non-breeding condition; bill horn-black to dark gray, the basal part of the maxilla pale gray; iris bright crimson to deep chestnut; eye skin gray; feet olive; weight 48 to 55 grams; stomach contents ants in all cases. These woodpeckers were found feeding along long stems in the undergrowth in areas where the forest had been heavily exploited. Dendropicos fuscescens lepidus (Cabanis and Heine) The cardinal woodpecker was met with on November 23 when one adult male was collected as it was pecking on one of the upper branches of a dead forest tree that had been poisoned by the forestry department as part of its scheme to remove the remnants of the Kalinzu to make way for agriculture. It is not known if the methods used to poison non-timber trees affects the edibility of the insects still found on the latter, and, so, in turn, the birds that feed on them; but the question is a real one. The specimen obtained had the bill dark horn-gray above, paler toward the base on the mandible; iris red-brown; eye skin dark gray; feet olive; weight 25 grams; stomach contents beetles and beetle larvae. Mesopicos xantholophus (Hargitt) The yellow-crested woodpecker occurs in many Ugandan forests and its presence in the Kalinzu is therefore not surprising. One male and two unsexed specimens were collected, October 31 to November 25; bill dark horn-gray, paler on the mandible; iris red-brown to deep crimson; eye skin dark gray; feet olive-gray; weight 57.5 to 66 grams; stomach contents beetle larvae. The birds were noted feeding extensively on spurge stems (Neobutonia) , which appar- ently house a great many beetles. Robert Glen wrote that he had observed many species of beetles, including even the very large goliath beetle, frequent- ing these spurge stems. 12 Contributions in Science No. 195 Mesopicos elliotii elliotii (Cassin) Elliot’s woodpecker is generally considered to be fairly uncommon in the forests of Uganda, but it must be common in Kalinzu, as four specimens were obtained in less than a week, October 30 to November 4; all in non-breeding state; bill blackish horn, paling to gray on the mandible; iris dark brown to deep crimson; eye skin brown to dark gray; feet olive; weight 34.5 to 38 grams; stomach contents ants, caterpillars and beetle larvae. Eurylaemidae Smithornis capensis meinertzhageni van Someren The African broadbill is common in Kalinzu; five specimens, four males, one female, all in non-breeding state, were collected, November 8 to 22; maxilla black; mandible whitish gray to pale pink; iris dark brown; feet pale olive- yellow to olive; weight 23 to 27.5 grams; stomach contents insects, mainly small beetles. These birds were all netted in the dense undergrowth in natural forest areas. Hirundinidae Hirundo rustica rustica Linnaeus One male example of the European barn swallow, a winter visitor to Uganda, was collected on November 16; weight 20 grams; stomach contents small flies. Psalidoprocne pristoptera ruwenzori Chapin One male, with large testes, collected November 14, has a short tail (77.5 mm) well within the variational range of ruwenzori; bill black; iris dark brown; feet dark brown; weight 1 1 .5 grams; stomach contents insect fragments. Motacillidae Motacilla drier ea drier ea Tunstall One female gray wagtail was taken November 16; ovary small; bill blackish horn; iris dark brown; feet pale brown; weight 16.5 grams; stomach contents insect fragments. Motacilla capensis wellsi Ogilvie-Grant Common in the open areas about the Kalinzu Forest. From October 29 to November 1 3 four examples were secured, two males with large gonads, one female with a small ovary, and one of unrecorded sex; bill black; iris dark brown; feet blackish gray; weight 19 to 25.5 grams; stomach contents insect remains. 1970 Birds of the Kalinzu Forest 13 Timaliidae Trichastoma fulvescens ugandae (van Someren) This brown babbler would seem to be relatively uncommon in Kalinzu, judged by the fact that only a single example was obtained, as compared with four or five of T. rufipennis and T. alhipectiis. One male in non-breeding state was obtained November 4; maxilla black; mandible gray; iris brown; feet dark gray; stomach contents beetles and small snails. The bird was caught in a mist net in the dense undergrowth in an area of heavily exploited forest. Trichastoma pyrrhoptera pyrrhoptera (Reichenow and Neumann) One specimen, a non-breeding male, November 8; maxilla black; man- dible pale gray; iris chestnut; feet gray; weight 25 grams; stomach contents beetles and small snails. Trichastoma rufipennis rufipennis Sharpe Common in the Kalinzu Forest, where four specimens were caught in mist nets placed in the undergrowth of natural, undisturbed forest, November 1 3 and 14; all were in non-breeding state; maxilla black; mandible blue-gray; gape yellow; iris dark brown; feet dark gray, the toes paler; weight 21 to 26 grams; stomach contents insect fragments and small snails. Trichastoma albipectus (Reichenow) Like the previous species this one is also numerous in the Kalinzu Forest, where it occurs with the very similar T. rufipennis. Five specimens, four males, one of which had enlarged gonads, and one female, were taken November 1 to 23; maxilla black; mandible pale gray; gape yellow; iris brown to reddish brown; feet opaque pale gray in some cases with a pinkish or a pale violet tinge; weight 28 to 32 grams; stomach contents small beetle fragments and parts of one millipede. This species was caught in nets in both natural and exploited forest. Pycnonotidae Pycnonotus barbatiis tricolor (Hartlaub) This ubiquitous bulbul is represented in the collection by an adult male in non-breeding state, taken on November 19 outside the forest; bill black; iris dark brown; feet black; weight 45 grams; stomach contents fruits. Pycnonotus masukensis kakemegae (Sharpe) The Kalinzu Forest would seem to be at the lower level of the range of this bulbul, which is chiefly a bird of the highland forests. Between October 28 and November 21 four males and one female were collected; gonads ranging 14 Contributions in Science No. 195 from small to much enlarged; bill blackish above, dark gray below; iris red- brown; feet blue-gray; weight 25 to 27 grams; stomach contents insect frag- ments, and, in one case, small fruits. These birds were collected in natural or little exploited forest areas. The Kalinzu is the only area from western Uganda other than the Impenetrable Forest from which we have received examples of this species; it might have been expected in the Kibale Forest, but our collectors did not meet it there. Pycnonotus v ire ns holochlorus (van Someren) One female in non-breeding condition, November 4; bill brownish black; iris dark gray; feet yellow-brown; weight 21.5 grams; stomach empty; netted in undergrowth in heavily exploited forest. Pycnonotus gracilis ugandae (van Someren) Common in the Kalinzu; six specimens, four males and two females, all with small gonads, were taken October 30 to November 22; bill black; iris dark brown; feet olive-gray; weight 19 to 22 grams; stomach contents fruit seeds, insects and insect larvae. Pycnonotus curvirostris curvirostris (Cassin) Another common Kalinzu bird; seven examples October 29 to November 24; gonads varying from small to large; bill dark horn gray to blackish, more yellowish brown at tip; iris chestnut-brown to red-brown; feet olive; weight 23.5 to 26 grams; stomach contents fruit seeds and insect fragments. Pycnonotus latirostris eugenius (Reichenow) As might have been expected this bulbul proved to be numerous in the Kalinzu Forest. Seven specimens were obtained October 28 to November 14, three of them with enlarged gonads, four with small ones; bill black, tomium and gape yellow; iris dark gray-brown; feet brownish yellow; weight 25.5 to 29 grams; stomach contents fruit seeds and insect remains. The yellow- whiskered bulbul was found (netted) chiefly in undergrowth in heavily exploited areas of the forest. Pycnonotus gracilirostris congensis (Reichenow) Very common. Seven examples were taken in the Kalinzu Forest and two more in the nearby Maramagambo Forest, October 28 to November 19. A long series of this bulbul from many of the western Uganda forests suggests the use of the name congensis rather than the recognition of chagwensis. In this allocation we follow White (1962; 78) rather than Rand (1960: 256) . The birds were mostly in non-breeding state, but one female, November 2, had an enlarged ovary; bill black; iris brick-red to crimson; feet black; weight 29 to 37 1970 Birds of the Kalinzu Forest 15 grams; stomach contents insect fragments entirely. These slender-billed bulbuls were noted feeding in the tree tops in the ironwood climax forest. Baeopogon indicator indicator ( Verreaux) One male in non-breeding state, October 28; bill black; iris white; feet dark gray; weight 47 grams; stomach contents ants; collected in a heavily exploited area of the forest. Phyllastrephus hypochlorus (Jackson) Apparently fairly common in the Kalinzu Forest; three specimens, No- vember 8 to 18; all with small gonads; bill black above, grayish below with whitish sides at the tip; iris brown; feet gray; weight 21 to 28 grams; stomach contents insect fragments. Phyllastrephus fischeri sucosus Reichenow Very common in the Kalinzu; seven examples were collected, October 28 to November 22; gonads ranging from small to much enlarged; bill blackish horn to black, the tomium and mandible paler; iris gray to yellow-gray to pale brown; feet gray; weight 20 to 27 grams; stomach contents ants and other insect fragments. These birds were netted in the undergrowth in heavily exploited areas of the forest. Phyllastrephus xavieri xavieri (Oustalet) Xavier’s bulbul was found to be common in the Kalinzu, where four specimens were obtained October 31 to November 24; all with small gonads; bill black, tomium and gape yellow, mandible gray at base; iris gray to grayish cream; feet dark gray, the toes paler; weight 25.5 to 27 grams; stomach con- tents insect remains. These birds were caught in nets in the undergrowth in the heavily exploited areas of the forest. Bleda syndactyla woosnami Ogilvie-Grant Very common in the Kalinzu; seven specimens taken November 4 to 2 1 ; all but one with small gonads; bill black above, pale gray below; iris dark red to reddish brown; feet brownish gray to pinkish gray; weight 40 to 49 grams; stomach contents beetles and other insect fragments. Criniger calurus emini Chapin I'his bulbul was found to be very common and seven specimens were obtained November 4 to 22; gonads ranging from small to much enlarged; bill black above, blue-gray below and with the tomium whitish in some but not in all individuals; iris chestnut to crimson; feet gray to dark blue-gray; weight 23 to 31 grams; stomach contents beetles and other insect fragments. All these 16 Contributions in Science No. 195 specimens are definitely C. calurus and not C. ndussumensis, as all have broad bills and lack any grayish on the lores. Campephagidae Coracina caesia pura (Sharpe) The Kalinzu Forest (5000 feet) must be close to the lower level of the range of this cuckoo-shrike, which does not occur in the lower Bugoma, Bwamba or Kibale Forests. Two females, both in non-breeding condition, were taken November 20 and 22. Both were in late stages of molt and both were in nearly complete adult plumage; bill black; iris dark brown; feet black; weight 49 and 50 grams; stomach contents caterpillars and other insect remains. Campephaga phoenicea petiti Oustalet One adult male, testes much enlarged, taken on November 25; bill black; gape rich yellow; iris dark brown; feet black; weight 34 grams; stomach con- tents caterpillars. The yellow gape of the male is conspicuous in the field. Muscicapidae Muscicapa acliista pwnila (Reichenow) One male of this little flycatcher, testes not enlarged, was taken Novem- ber 7; bill black above, the basal half of mandible cream; iris dark brown; feet black; weight 7.5 grams; stomach contents beetles. Muscicapa griseigularis (Jackson) This flycatcher is fairly common in the Kalinzu Forest; three males and one female, all with small gonads, were collected November 4 to 18; maxilla black; mandible gray; iris dark brown; feet gray; wegiht 10.5 to 13 grams; stomach contents small beetle and other insect fragments. These birds were caught in mist nets in the undergrowth of both natural and exploited areas of the forest. Muscicapa comitata (Cassin) Another common bird, of which four specimens were taken November 2 to 14, two of each sex, all with small gonads; bill black; iris dark brown; feet black; weight 14 to 15 grams; stomach contents small beetle fragments. Artomyias fuliginosa J. and E. Verreaux This dusky brown flycatcher was found to be very numerous in the Kalinzu; six specimens were collected November 1 to 13; four with enlarged gonads, the other two unsexed because of damage; bill black; iris dark brown; feet black; weight 17 to 19 grams; stomach contents beetle fragments. One of the unsexed birds is subadult, molting into adult plumage, but still has the 1970 Birds of the Kalinzu Forest 17 upper parts and the breast dotted with white, the abdominal feathers with buffy grayish white ends, and also with the gape whitish. The birds perch on bare branches of tall forest trees from which they make frequent sorties to capture insects. They resemble martins when on the wing. Fraser ia ocreata ocreata (Strickland) Common in the Kalinzu Forest; four specimens taken October 30 to November 5; two of each sex, all in non-breeding state; bill black; iris dark brown; feet blackish; weight 37.5 to 40 grams; stomach contents beetle frag- ments. Three of these forest flycatchers were noted as being collected while they were catching termites on the wing, but the fact remains that only beetles were found in their stomachs! Megahyas fiammulata aequatorialis Jackson Another common Kalinzu species; six specimens taken, three females, two males, October 28 to November 6; all with small gonads; bill black; iris red; feet dull brownish pink to maroon; weight 27.5 to 31 grams; stomach contents beetle fragments, and, in one case, lepidopteran eggs. These birds were seen feeding in trees and were also netted in the undergrowth in exploited parts of the forest. On November 8 an immature male, molting into adult plumage, was collected in the nearby, but lower (3600 feet), Maramagambo Forest. Bias musicus musicus ( Vieillot) Fairly common flycatcher. One of each sex collected, October 31 and November 9 in the Kalinzu, and two others, one of each sex, November 8 in the Maramagambo Forest; bill black; iris yellow; feet pale yellow; weight 21 to 22 grams; stomach contents beetles and other insect fragments, including flying insects. Batis molitor (Hahn and Kuster) This flycatcher was met with only once, on November 6, when a pair were seen running along branches of low trees in search of food. Both birds were in non-breeding state; bill black; iris pale green, lower ring yellow; feet black; weight 11 and 12 grams; stomach contents small beetles. Platysteira castanea castanea Fraser Very common in the Kalinzu Forest; seven examples were taken, October 28 to November 25; gonads large in one, small in the others; bill black; iris chestnut to dark maroon; feet dull maroon; weight 12 to 16 grams; stomach contents insect, largely beetle, fragments. The eye wattles in the females seem to be slightly darker in life than those of the males. These wattle-eyes were caught largely in nets in the undergrowth, close to the ground, in relatively unmolested areas of natural forest. 18 Contributions in Science No. 195 Platysteira blissetti jamesoni (Sharpe) One male, with enlarged testes, taken October 28 and one female, with a small ovary, November 13; bill black; iris dark maroon; feet maroon to dull violet-maroon; weight 10.5 to 11 grams; stomach contents insect fragments, beetles in one case. Netted near a small stream in natural forest. Platysteira concreta graueri (Hartert) This wattle-eye must be very numerous in the Kalinzu Forest, as six specimens were collected November 4 to 21, three of each sex; one of the females with much gonadal enlargement, the other individuals with small gonads. The males illustrate the plumage polymorphism of this species, de- scribed by Prigogine (1969), as one of them has the entire underparts bright yellow while the other two have the whole area bright chestnut; the three females have the chin, throat and breast bright chestnut, the abdomen and under tail coverts yellow, the two being sharply demarcated; bill black; iris very dark maroon-brown with a white inner ring; wattles pale yellow-green; feet dark blue-gray; weight 12.5 to 15 grams; stomach contents small insect, largely beetle, fragments. Although this wattle- eye has been reported from the Semliki Valley on the west side of the Congo-Uganda border, and from Kakamega in western Kenya, we are not aware of any Uganda locality other than Bwamba and Kalinzu Forests. Trochocercus nigromitratus (Reichenow) Two specimens, October 28 and November 13, represent this flycatcher; bill black; iris dark brown; feet gray to dark blue-gray; weight 10 to 11.5 grams; stomach contents beetle and other insect fragments. Terpsiphone viridis ferreti (Guerin) This paradise flycatcher is very com.mon in the Kalinzu Forest; thirteen specimens were obtained October 28 to November 25. The series illustrates the polymorphic nature of the species, two of the females have the entire upper parts, wings and tail dark bluish gray; all the others are reddish brown backed birds but with much variation in the darkness of the blue-gray underparts, which range from fairly pale blue-gray to almost blackish blue. The only males with elongated central rectrices also had enlarged testes; bill dark blue-gray to pale cobalt-blue, the tip and tomium blackish; iris dark brown; feet blue-gray to dark blue-gray; eye skin dark blue-gray; weight 12 to 16.5 grams (one female marked 22 grams) ; stomach contents small beetle and other insect fragments. One of the dark bluish gray backed females was seen together with, and apparently following, a red backed, long tailed male (also collected), a good indication that what we are dealing with here is polymorphism and not a confusion of two different taxa. 1970 Birds of the Kalinzu Forest 19 Turdidae Alethe poUocephala carruthersi Ogilvie-Grant This thrush was common in the Kalinzu Forest; five specimens were taken November 1 to 1 3; gonads small in four, large in one; bill black; iris deep chestnut to dark “soft” brown; feet opaque gray, with or without a pinkish tone; weight 25 to 34 grams; stomach contents small beetle and other insect fragments, one snail. The brown-chested alethe was found in both unspoiled and in heavily exploited parts of the forest. Erithacus aequatorialis (Jackson) Fairly common; four specimens taken, October 31 to November 5; all in non-breeding state; bill black; iris dark brown; feet silvery gray to pale brown- ish gray; toes yellowish; weight 14 to 19.5 grams; stomach contents insect fragments, mainly beetles. These akalats were caught in nets in the under- growth in heavily exploited areas of the forest. Sphenoeacus mentalis mentalis (Fraser) One specimen, a female with an enlarged ovary, was taken November 19; maxilla black above, the tomium and mandible pale gray; iris pale brown; feet pale gray; weight 36 grams; stomach contents insect fragments. The bird was shot while it was skulking in scrub grassland at the edge of the forest. Neocossyphus poensis praepectoralis Jackson Abundant in the Kalinzu Forest; nine examples were collected November 4 to 21; one female with an enlarged ovary, the other specimens in non-breed- ing state; bill black; iris dark brown; feet whitish flesh to flesh pink; weight 45 to 60 grams; stomach contents ants and small beetles. These birds were taken in mist nets in the undergrowth and also shot high up in the forest trees. Stizorhina fraseri vulpina Reichenow Fairly common in the Kalinzu; three specimens were taken October 29 to November 8; all with non-breeding gonads; bill black, the base brownish in a female; iris dark brown; feet pale brown to pinkish brown; weight 37 to 39 grams; stomach contents entirely beetles. Sylviidae Schoenicola platyura alexinae (Heuglin) The great-tailed warbler was met with once, on November 12, when two males were collected, one with small and one with large testes; bill black above, pale gray below; iris brown; feet gray with a brownish wash; weight 16.5 and 17 grams; stomach contents small insect fragments. 20 Contributions in Science No. 195 Prinia leucopogon reichenowi (Hartlaub) The white-chinned prinia was found to be numerous in the Kalinzu where six specimens were collected November 1 to 10; gonads ranging from small to large; bill black; iris chestnut; feet pale pinkish brown; weight 11.5 to 13.5 grams; stomach contents insect fragments. Most of these examples were netted in the undergrowth in heavily exploited areas of the forest. Apalis jacksoni jacksoni Sharpe One specimen, October 28, a male in non-breeding state; bill black; iris brown; feet flesh-brown; weight 10 grams; stomach contents small flies and small insect larvae. Apalis rufogularis nigrescens (Jackson) This warbler is represented in the collection by two specimens, collected November 11 and 16; one a male, the other with sex unrecorded; bill black; iris chestnut to pale brown; feet dull pinkish brown; toes paler; weight 9 and 10 grams; stomach contents small insects and insect larvae. In their dark brownish dorsal coloration these specimens agree with nigrescens from the Kibale, Bugoma and Bwamba Forests, and not with the more grayish backed kigezi of the Impenetrable Forest. Bathmocercus cerviniventris vulpinus Reichenow Very common; seven specimens, October 30 to November 21; gonads small to much enlarged; bill black; iris dark brown; feet dark blue-gray; weight 15.5 to 19.5 grams; stomach contents small insect fragments. The black-faced rufous warbler was netted largely in the undergrowth in heavily exploited parts of the forest. Camaroptera chloronota toroensis (Jackson) Very common; eight specimens, October 28 to November 23; all in non- breeding state; bill black above, pale horn to whitish gray below; iris pale brown to gray-brown; feet pale brown; weight 8 to 13 grams; stomach contents insect fragments. Camaroptera brachyura tincta (Cassin) This little warbler was as numerous as the preceding species; six examples were taken October 29 to November 21; most were in non-breeding state, but one male, November 6, had enlarged gonads; bill black; iris pale brown; feet pale pinkish brown; weight 11 to 12.5 grams; stomach contents insect fragments. Sylvietta leucophrys chloronota Hartert One specimen, a non-breeding female, was taken November 10; bill pale 1970 Birds of the Kalinzu Forest 21 horn-brown, pinkish at the base; iris chestnut; feet deep pink; weight 10 grams; stomach contents insect fragments. Hylia prasina prasina (Cassin) The green hylia was taken once in the Kalinzu Forest, November 13, and a second example was collected in the Maramagambo Forest, November 8; both in non-breeding state; bill black, the tomium and gape yellow; iris dark brown; feet olive to olive-gray; weight 15 and 16 grams; stomach contents insect matter. Laniidae Tchagra minuta minuta (Hartlaub) This shrike was not collected in the Kalinzu Forest, but on November 8 two birds, one of each sex, were taken in a grass covered glade in the nearby Maramagambo Forest; the male had large testes, the female had a small ovary; bill black; iris pale maroon; feet gray; weight 31 and 36 grams; stomach con- tents grasshoppers and beetles. Laniarius liihderi luhderi Reichenow This bush-shrike was common in the Kalinzu Forest; seven specimens were taken October 28 to November 10; gonads ranging from small to much enlarged; bill black; in immature birds it is blackish horn above, paling to pale blue-gray basally on the mandible; feet gray to bluish gray; weight 41 to 47.5 grams; stomach contents grasshoppers, beetles and other insect remains. Telophorus bocagei jacksoni (Sharpe) This shrike is not uncommon in the Kalinzu Forest; two females were collected on October 30 and November 19; bill black; iris dark brown; feet dark gray; weight 27 grams each; stomach contents insect larvae. Lanins coUurio phoenicuroides (Schalow) This winter visitor from Eurasia is represented by one immature in- dividual, taken on November 7; bill horn-gray, the tip blackish; iris very dark brown; feet gray black; weight 26.5 grams; stomach contents small grass- hoppers. It was shot in secondary growth in one area of forest that had been exploited. Lanius mackinnoni Sharpe Not uncommon; three specimens, October 30, November 7 and 8; all in non-breeding state; bill black; iris dark brown; feet black; weight 34 to 37 grams; stomach contents caterpillars, grasshoppers and beetle fragments. 22 Contributions in Science No. 195 Oriolidae Oriolus brachyrhynchus laetior Sharpe One adult male of this oriole was taken in the Kalinzu Forest, November 16, and three immature birds were collected in the Maramagambo Forest, on November 8; bill of adult dull pinkish brown-horn; of immature birds blackish, pinkish brown basally; iris wine-red in adult male, dull red to dark grayish brown in the others; feet blue-gray; weight 48 to 56 grams; stomach contents fruits, small seeds, caterpillars and insect fragments. Oriolus larvatus percivali Ogilvie-Grant One adult male with enlarged gonads was taken on October 31, in a mist net in the undergrowth in an area of heavily exploited forest; bill metallic deep pink; iris dark crimson; feet dark gray; weight 74 grams; stomach contents caterpillars. Paridae Pams funereus funereus ( Verreaux) This tit was common in the Kalinzu, where five examples were collected October 31 to November 2. All were in non-breeding state; bill black; iris red in four cases, yellow-brown in one; feet slate gray; weight 23 to 27 grams; stomach contents caterpillars, beetle larvae and small insect remains. Sturnidae Poeoptera stuhlmanni (Reichenow) Two specimens of this starling, a male with large gonads and a female in non-breeding condition, were taken on October 30 and 31 ; bill black; iris dark brown; with outer ring yellow; feet black; weight 39 (female) and 44 (male) grams; stomach contents remains of fruits. Poeoptera lugubris Bonaparte Five adult males, November 2 to 14, with gonads ranging from small to much enlarged; bill black; iris bright yellow; feet black; weight 42 to 43 grams; stomach contents fruit fragments, seeds, beetles and other small insect remains. Keith (1968) has recently described a race of this starling, P. 1. webbi, from the Impenetrable and Kibale Forests, on the basis of greater wing length and a bluish sheen to the purple gloss of the feathers in the male. He gives the wing length of males of nominate lugubris as 89 to 96 (average 93.13 mm) , of webbi as 94 to 99 (average 97.23 mm) . Our five Kalinzu males bridge this difference as they have wing lengths of 93 to 97 (average 95 mm) . In color they agree with our series of webbi but there is some slight individual variation in the amount of blue or purple. No examples of nominate lugubris from Gabon have been available for comparison and we therefore use a binomial 1970 Birds of the Kalinzu Forest 23 for our specimens. We cannot avoid the conclusion that webhi is a poorly marked form, if a valid one. Onychognathus walleri elgonensis (Sharpe) Four examples of this race of Waller’s chestnut-winged starling, two males and one female with large gonads and one female with a small ovary, were taken November 4 to 14; bill black; iris crimson-red with the inner ring dark brown; feet black; weight 70 to 82 grams; stomach contents small green fruits and fruit seeds. These birds were seen feeding in a fruiting Symphonia tree. Lamprotoniis splendidus splendidus ( Vieillot) This starling was fairly common in the Kalinzu Forest; one male and two females, all with enlarged gonads, were collected on October 31, Novem- ber 12 and 14; bill black; iris yellow-white to cream; feet black; weight 120 to 135 grams; stomach contents fruits and seeds. ZOSTEROPIDAE Zosterops senegalensis toroensis Reichenow Two specimens of this common white-eye, one of each sex, taken on November 19, agree with a series of toroensis from the Kibale and Bwamba Forests; bill black; iris dark brown; feet gray; weight 1 1 grams each; stomach contents insect fragments. Nectariniidae Anthreptes fraseri axillaris (Reichenow) Common in the Kalinzu; six examples of this sunbird were collected October 31 to November 18; gonads ranging from small to large; bill black to dark brown above, pinkish white below; iris brown to reddish brown; feet olive to olive-gray to gray-brown; weight 10 to 13 grams; stomach contents insect remains. This species was found chiefly in natural, undisturbed areas of the forest. Anthreptes rectirostris tephrolaema (Jardine and Fraser) This sunbird was found to be very common in the Kalinzu Forest, where eight specimens were obtained, October 30 to November 19, and one in the Maramagambo Forest, November 8; of five males four had large gonads, while of four females none showed gonadal swelling; bill black; iris red-brown to dark brown; feet black; weight 10 to 12 grams; stomach contents insect frag- ments, small spiders and bits of fruit. The gray-chinned sunbird was found both in natural and in heavily exploited areas of the forest. 24 Contributions in Science No. 195 Anthreptes collaris garguensis Mearns Our collectors met with this sunbird only on October 31 and November 1; on each day they obtained a single example, so the bird may not be very numerous in the Kalinzu; bill black; iris dark brown; feet black; weight 8.5 grams; stomach contents insect remains. Nectarinia olivacea ragazzn (Salvadori) The olive sunbird was found to be very numerous in the Kalinzu, and eleven examples were collected November 2 to 18; gonads ranging from small to large; bill black above, the base of the mandible dull yellow to pinkish brown; feet dark brown to black; weight 9 to 12.5 grams; stomach contents insect fragments. Nectarinia verticalis viridisplendens (Reichenow) This sunbird was also very common in the Kalnizu Forest; six males and one unsexed, but probably female by plumage, were collected November 1 to 10; gonads varying from small to large; bill black; iris dark brown; feet black; weight 12 to 15 grams (one marked 26.5 grams, probably an error for 16.5 grams) ; stomach contents insect fragments, small insect larvae. Nectarinia rubescens rubescens (Vieillot) Common in the Kalinzu; seven specimens October 30 to November 11; gonads small to large; bill black; iris dark brown; feet black; weight 10.5 to 12.5 grams; stomach contents insect fragments. The green-throated sunbird was seen chiefly feeding in fairly tall trees. Nectarinia bouvieri (Shelley) This sunbird was met with but once, on November 24 when an adult male with large testes was collected; bill black; iris dark brown; feet black; weight 10 grams; stomach contents very small insect fragments. The bird was found in the small scrub-bush of a grass covered hillside at the edge of the forest. Nectarinia purpureiventris (Reichenow) The purple-breasted sunbird, a species primarily of highland forest, was seen several times in the Kalinzu by Williams, but he was unable to collect one. Nectarinia kilimensis kilimensis Shelley One female, ovary enlarged, was taken on November 25; bill black; iris dark cream; feet black; weight 16 grams; stomach contents insect fragments. 1970 Birds of the Kalinzu Forest 25 Ploceidae Amblyospiza alhifrons melanota (Heuglin) One adult male of this heavy-billed weaver in breeding condition was taken November 13 from a small flock seen in the undergrowth of an ex- ploited area of the forest; bill horn-black; iris dark brown; feet grayish black; weight 47 grams; stomach contents seeds. Ploceus nigricollis nigricollis (Vieillot) Common in the Kalinzu; three males and four females were collected, October 31 to November 25, all in non-breeding state; bill black; iris dark brown; feet dark gray; weight 21.5 to 29 grams; stomach contents small beetles, green caterpillars and other insect fragments. Several of these specimens were caught in mist nets in the undergrowth in exploited areas of the forest. Ploceus melanogaster stephanophorus (Sharpe) This weaver was very common in the Kalinzu; nine specimens were taken October 28 to November 24; one male with large gonads, all the others with small ones; bill black; iris dark brown; feet dark gray; weight 20.5 to 25 grams; stomach contents insect remains, including beetles and small caterpillars. Four of the specimens are molting into full adult plumage. A number of these black- billed weavers were caught in nets in the undergrowth in a heavily exploited area of the forest, where they came, together with other species, to feed on flying termites as they emerged from the ground. Ploceus superciliosus (Shelley) Not taken in the Kalinzu, but on November 8 one (unsexed) example was collected in the Maramagambo Forest; bill black above, pale gray below; iris dark brown; feet pale brown; weight 21 grams; stomach contents grass seeds. Ploceus insignis insignis (Sharpe) This weaver was met with but once, on November 2, when one adult of each sex, both in non-breeding state, were taken; bill black; iris red-brown in male, bright brick-red in female; feet pale brown; weight 28 grams each; stomach contents insect fragments. The birds were searching for food among the moss and lichen on a small tree. Malimhus ruhricolUs rubricollis (Swainson) This malimbe was common in the Kalinzu Forest; four examples were collected October 31 to November 2; all in non-breeding condition; bill black; iris dark red-brown; feet black; weight 39 to 45 grams; stomach contents insect fragments. 26 Contributions in Science No. 195 Estrildidae Parmoptila woodhousei jamesoni (Shelley) The Kalinzu Forest is an additional Uganda locality for this Congolese flower pecker. One adult male, with enlarged testes, was taken on November 13; bill black; iris very dark red-brown; feet pale huffy brown; weight 10 grams; stomach contents small black ants. Nigrita fusconota fusconota Fraser This negro finch was met with but once in the Kalinzu, on November 25, when a male with large testes and a female in non-breeding state were taken; bill black; iris dark brown; feet dark gray; weight 8 and 9.5 grams; stomach contents very small insects. Nigrita canicapilla schistacea Sharpe This species was not uncommon in the Kalinzu; four specimens were taken on October 30 and 31, but then no more were encountered although the collectors were there for nearly a month; one of our birds was in breeding condition; the others were not; bill black; iris orange-red; feet dark brown; weight 19 to 20 grams; stomach contents small seeds and insect fragments. Nesocharis ansorgei ansorgei (Hartert) It was something of a surprise to find this little waxbill fairly common in the Kalinzu Forest; five specimens were taken October 30 to November 7; all in non-breeding state; bill black, pale gray at base; iris dark brown; feet blackish; weight 8.5 to 10 grams; stomach contents small crushed grass seeds. The birds were found in the tall undergrowth in areas of forest that had been heavily exploited. The Kalinzu Forest is an additional locality to those men- tioned by the senior author (Friedmann, 1968) in his discussion of this little weaver finch. Spermophaga ruficapilla ruficapilla (Shelley) As in so many west Ugandan forests, this bird was found to be common in the Kalinzu; six examples were collected, October 29 to November 10; all in non-breeding condition; bill black with metallic blue wash, the tomium pinkish orange; iris dark brown; feet dark brown to dark gray-brown; weight 19.5 to 26.5 grams; stomach contents seeds and grit. Lonchura bicolor poensis (Fraser) Although this mannikin is not a typical forest bird, the one example obtained was collected in a mist net in an exploited part of the forest; bill pale gray; iris dark brown; feet dark gray; weight 10 grams; stomach contents grass seeds. 1970 Birds of the Kalinzu Forest 27 Fringillidae Serinus frontalis frontalis Reichenow This little finch is represented by a single example collected November 12; bill horn brown above, pale horn below; iris dark brown; feet brown; weight 13 grams; stomach contents insects and small seeds. It was taken in the scrub undergrowth at the edge of the forest. Serinus burtoni tanganjicae Granvik Fairly common in the heavily exploited parts of the Kalinzu Forest. Three specimens were taken November 4 to 20; gonads small to large; bill horn- brown, paler below; iris medium to dark brown; feet pale to dark brown; weight 30 to 32 grams; stomach contents seeds and flower buds. Literature Cited Chapin, J. P. 1953-1954. The birds of the Belgian Congo. Pt. III. Amer. Mus. Nat. Hist. 75A: 1-821; Pt. IV, 75B: 1-846. Friedmann, H. 1968. The olive weaver-finch, Nesocharis ansorgei ansorgei in Uganda. Bull. British Ornith. Club 88: 135-128. Friedmann, H., and J. G. Williams. 1968. Notable records of rare or little known birds from western Uganda. Rev. Zool. Bot. Afr. 77 : 11-36. Keith, S. 1968. A new subspecies of Pooeptera lugubris Bonaparte from Uganda. Bull. British Ornith. Club 88: 119-120. Keith, S., and A. Twomey. 1968. New distributional records of some East African birds. Ibis 110: 537-548. Keith, S., A. Twomey, H. Friedmann, and J. G. Williams. 1969. The avifauna of the Impenetrable Forest, Uganda. Amer. Mus. Novitates 2389: 1-41. Prigogine, a. 1969. Polymorphism of the chestnut-bellied wattle-eye, Dyaphoro- phyia concreta. Ibis 111: 95-96. Rand, A. L. 1960. African Pynonotidae, in Mayr and Greenway, eds., Check-List of Birds of the World, 9: 221-300. White, C. M. N. 1960. A check list of the Ethiopian Muscicapidae (Sylviinae). Pt. I. Natl. Mus. S. Rhodesia, Bulawayo, Occas. Papers Nat. Sci. 3 : 399-450. . 1961. A revised check list of African broadbills, pittas, larks, swallows, wagtails, and pipits. Lusaka, Govt. Printer. 84 p. . 1962a. A check list of the Ethiopian Muscicapidae (Sylviinae). Pt. II. Natl. Mus. S. Rhodesia, Bulawayo, Occas. Papers Nat. Sci. 3: 653-738. . 1962b. Revised check list of African shrikes, orioles, drongos, starlings, crows, waxwings, cuckoo-shrikes, bulbuls, accentors, thrushes and babblers. Lusaka, Govt. Printer. 176 p. . 1963. A revised check list of African flycatchers, tits, tree creepers, sun- birds, white-eyes, honey eaters, buntings, finches, weavers and waxbills. Lusaka, Govt. Printer. 218 p. . 1965. A revised check list of African non-passerine birds. Lusaka, Govt. Printer. 299 p. Accepted for publication April 2, 1970 LOS ANGELES ' COUNTY MUSEUM CONTRIBUTIONS SL IN SCIENCE iJMBER 196 July 1, 1970 TWO NEW SPECIES OF CATFISHES OF THE GENERA NANNORHAMDIA AND IMPARALES (FAMILY PIMELODIDAE) FROM CENTRAL AMERICA I By William A. Bussing Los Angeles County Museum of Natural History • Los Angeles, California 90007 Exposition Park 'i CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sep- arately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions and scien- tists on an exchange basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Virginia D. Miller, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. Robert J. Lavenberg Managing Editor INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM, CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information. Al- though priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections, other technical papers will be considered. All manuscripts must be recommended for consideration by the curator in charge of the proper section or by the editorial board. Manuscripts must conform to those specifications listed below and will be examined for suitability by an Edi- torial Committee. They may also be subject to review by competent specialists out- side the Museum. Authors proposing new taxa in a CONTRIBUTIONS IN SCIENCE must indicate that the primary type has become the property of a scientific institution of their choice and cited by name. MANUSCRIPT FORM.— (1) The 1964 AIBS Style Manual for Biological Journals is to be followed in preparation of copy. (2) Double space entire manu- script. (3) Footnotes should be avoided if possible. Acknowledgments as footnotes will not be accepted. (4) Place all tables on separate pages. (5) Figure legends and unavoidable footnotes must be typed on separate sheets. Several of one kind may be placed on a sheet. (6) An abstract must be included for all papers. This will be published at the head of each paper. (7) A Spanish summary is required for all manuscripts dealing with Latin American subjects. Summaries in other languages are not required but are strongly recommended. Summaries will be published at the end of the paper. (8) A diagnosis must accompany any newly proposed taxon. (9) Submit two copies of manuscript. ILLUSTRATIONS. — All illustrations, including maps and photographs, will be referred to as figures. All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Consult the 1964 AIBS Style Manual for Biological Journals in preparing illustration and legend copy for style. Submit only illustrations made with permanent ink and glossy photo- graphic prints of good contrast. Original illustrations and art work will be returned after the manuscript has been published. PROOF.— Authors will be sent galley proof which should be corrected and returned promptly. Changes in the manuscript after galley proof will be billed to the author. Unless otherwise requested, page proof also will be sent to the author. One hundred copies of each paper will be given free to each author or divided equally among multiple authors. Orders for additional copies must be sent to the Editor at the time corrected galley proof is returned. Appropriate order forms will be included with the galley proof. Virginia D. Miller Editor TWO NEW SPECIES OF CATFISHES OF THE GENERA NANNORHAMDIA AND IMPARALES (FAMILY PIMELODIDAE) FROM CENTRAL AMERICA By William A. Bussing^ Abstract: Two new species of pimelodid catfishes, Nanno- rhamdia lineata and Imparales panamensis, are described from Costa Rica and Panama respectively. Nannorhamdia is recorded for the first time in Central America and each species represents a relict population. The relationships are discussed and habitats described for each species. Recent collections by Horace Loftin (1965) in Panama and by Albert Greenberg in Costa Rica have revealed two undescribed species of small pimelodid catfishes of the genera Nannorhamdia and Imparales. The former was previously known only from South America, whereas Miller (1966) included Imparales sp. in his checklist of Central American fishes. In 1967 Oscar Blanco B. and I obtained additional specimens of the Costa Rican Nannorhamdia. More specimens of Imparales were collected in 1967 in the Chucunaque and Balsas drainages of Panama by members of the Battelle Memorial Institute, Pacific Northwest Laboratory (Battelle NW) as part of their sub-contractual work for the Interoceanic Sea Level Canal Feasibility Study. Material from all these sources was used in this study. Nannorhamdia lineata, new species Figure 1 Holotype: LACM 30688-1 ( 9, 74.1 mm SL) ) , Costa Rica: Puntarenas Prov., Quebrada 36 (elev. 80 m) 12 km W of Pueblo Rio Claro at Inter- american Highway; 5 January 1967; collectors W. Bussing, Oscar Blanco B. Paratypes: UCR 1 1 1-10 (18, 48.5-71.7 mm SL), same data as holotype; LACM 30688-2 (3, 51.8-69.9 mm SL), same data as holotype; BMNH 1967.9.29.1-2 (2, 64.0-68.2 mm SL) , same data as holotype; USNM 204694 (2, 58.4 and 69.5 mm SL) , same data as holotype; USNM 194230 ( 1, 57.0 mm SL) Costa Rica: Puntarenas Prov., Rio Salama Nuevo (?) at Interameri- can Highway; 21 January 1963; collector Albert Greenberg. Type specimens are deposited in the British Museum (Natural History) (BMNH) , Los Angeles County Museum of Natural History (LACM) , Museo de Zoologia, Universidad de Costa Rica (UCR) and the United States National Museum (USNM). Comparative Material : Nannorhamdia nemacheir (Eigenmann and ^Departamento de Biologia, Universidad de Costa Rica, San Jose, Costa Rica; and Research Associate in Ichthyology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 196 Figure 1. Nannorhamdia lineata, new species, lateral view of holotype LACM 30688-1, 74.1 mm SL; dorsal view of paratype UCR 1 1 1-10, 71.1 mm SL. Fisher), USNM 121167 (30, 33.8-68.5 mm SL), Venezuela: Rio San Juan at bridge, tributary to Rio Motatan; 17 and 20 March 1942; collector L. P. Schultz. UCR 318-1 (6, 30.3-55.1 mm SL) , Colombia : Rio Salado, ca. 14 mile upstream from junction with Rio Truando; 27 August 1967; collector H. Loftin. Cetopsorhamdia nasus (Eigenmann and Fisher), USNM 86816 (1, 62.5 mm SL), Bolivia: Tumupasa; December 1921; collector N. E. Pearson. Diagnosis: A Nannorhamdia with pronounced lateral stripe extending from the tip of snout to caudal base; short maxillary barbels not passing pec- toral fin base; subequal caudal lobes; 8-1 1 total gill rakers; 7-9 principal (branched rays plus one simple ray) anal fin rays; no prominent dorsal blotches on body contrasting with ground color; all pectoral and dorsal fin elements unserrated. The new species is immediately separated from N. guttatus Pearson, TV. benedettii Fernandez Yepez, TV. stictnotus Fowler and TV. nemacheir Eigen- mann and Fisher by its lateral stripe and short maxillary barbels; from TV. guttatus and TV. benedettii it also differs in having fewer anal rays (7-9 vs. 10-12) . From TV. macrocephala Miles it differs in having fewer anal rays (7-9 vs. 13) and lacking dorsal and pectoral spines with denticulations on the posterior borders. It differs from TV. schubarti Gomes principally in having fewer gill rakers (8-1 1 vs. 9-15), shorter maxillary barbels (22.4-26.8 vs. 32.1- 55.5), shorter first dorsal ray (14.3-16.9 vs. 19.4-24.6) and shorter first pec- toral ray (13.3-15.4 vs. 16.7-20.0) . The new species is closest to TV. spurrellii of the Rio San Juan drainage. 1970 Two New Species of Catfishes 3 Colombia. The latter is distinguished from N. lineata by its narrower inter- narial distance, interorbital distance and caudal peduncle depth, much longer maxillary and mental barbels, higher gill raker count and much more promi- nent lateral stripe. Mr. Alwyne Wheeler {personal communication, 1968) reported that the type specimen of N. s purr el Hi is quite round-bodied and does not seem to have shrunk, thus discounting unnatural proportional differences due to preservation. Description: Head slightly depressed, snout bluntly rounded. Narrow fontanel in middorsal line extending from before eyes to base of occipital process. Mouth subterminal, snout projecting. Villiform teeth in bands on both jaws; premaxillary patches slightly wider laterally, but without posteriorly projecting angles; dentary patches wider medially, very narrow laterally; no teeth on vomer or palatines. Nostrils wide apart, anterior ones tubular and in line with bases of maxillary barbels; posterior nostrils with an anterior flap and lying equidistant between anterior nostril and margin of eye. Maxillary barbels the longest, one pair lying in a groove along the snout and extending along the first third or half of the length of pectoral fin; two pairs of mental barbels, their bases in a straight line, outer pair reaching to or slightly beyond origin of pectoral fin; inner pair extending three-fourths the distance to pectoral origin. No ontogenetic variation in barbel length on specimens in the present size range. Caudal fin well forked, shortest median rays slightly longer than half the length of the longest rays; upper lobe slightly longer and more pointed than lower. First dorsal ray soft, not extended into a filament, margin of fin trun- cate; origin slightly in advance of pelvic fin origin. Anal fin margin rounded; origin directly below adipose origin. Adipose fin short, slightly longer than anal; height about 4.5 times in length. Pelvic fins short, rounded; arising below third or fourth dorsal ray. First pectoral ray soft, net extended into a filament; appressed fin not quite reaching a vertical throuigli dorsal origin. Anus between middle of length of pelvic fins. Counts: The counts and body proportions in percent of standard length of the holotype and nine paratypes of N. lineata from Quebrada 36 are re- corded below, followed by data for the type of TV. spurrellii in parentheses. Data for the type specimen of TV. spurrellii Regan (BMNH 1913.10.1.41; Rio Condoto, San Juan drainage, Colombia) were kindly furnished by Alwyne C. Wheeler of the Zoology Department, British Museum of Natural History. Proportional measurements of the single specimen from Rio Salama Nuevo, Costa Rica, which in some cases fall slightly out of the range given below due to its shrunken condition, are not included. Dorsal rays i,6; i,5-6 (i,6) . Anal v,8; iv-v,6-8 (ii,6) . Pectoral i,9; i,8-9 (i,9). Pelvic i,5; i,5 (i,5) . Branched caudal rays 15, 15-16 ( 15) . Gill rakers of holotype 2 -f 7 and 1 + 8, total 9; paratypes 1-2 -1- 7-10, total 8-11 (TV. spur- rellii 2 -f 10, total 12) . Comparison of holotype and 16 paratypes of Imparales panamensis and holotype of I. mariai. Proportions expressed in thousandths of standard length. 4 Contributions in Science No. 196 •§ Q o sis < ffi Z c/5 D ON Cj m o\ 2 S f2 z ^ c/5 P S >P ^ - os so :S Is X z c/5 P ^ m VO o(j 00 (N ^ n m ' ^ n ^ (N 00 (N vs o ^ r' CNfTiVSxroOVS^TfOOOOOOn oor^lvooo^'^oofnvo^’rHOs (N ’-H r-H 1— I vs OS m vs r' ON f!— H c« (U Dh q vs On VO vi o r' .-H 00 r— ( n O O vs rp- 1 00 T vs cd vs 6 P 4- m 00 00 p p On Vh cd Ph vd m 00 r s ^ 73 73 p ^ t C c3 it3 Q (U ,.g £ ^ I s "8 i S w p Oh m cj Caudal peduncle, length 255 224-269 212-262 230 205 Predorsal distance 305 317-342 248-353 343 343 Preanal distance 639 648-679 614-693 652 652 Tip of snout to anus 413 426-498 411-491 455 440 Prepectoral distance 178 160-186 178-235 207 171 Preadipose distance 637 625-689 610-674 644 623 Prepelvic distance 358 369-402 356-418 387 377 1970 Two New Species of Catfishes 5 VO r- ON 00 I 1 NO ON O (N (N o CO ON m n 1 »o On ro 1 m (N t-H m ON r- »r> VN r- 00 r-H n n ^ m n ON m o »0 O fO On t-H y—l NO 00 ON NO r-' On fSj r- r- m (N in fO in (n ON r- (N ’—1 cn n »— < m m n T— 1 (N 00 ei) u U T3 Cu o (U kr U-I CO 1 . 1 . £X0 oo O CO O CC 04 CO O 2H G a s Ui G