Vertebrate ecology and systematics: a tribute to Henry S. Fitch

Luis Malaret

We studied the feeding ecology of California Kingsnakes (Lampropeltis californiae) based on stomach contents of 2,662 museum specimens, 90 published records, and 92 unpublished observations. These snakes typically are diurnal, wide-foraging generalists and ingest prey headfirst. Twenty-nine percent of 447 diet items were mammals, 29% were snakes, 25% were lizards, 11% were birds, 4% were squamate eggs, 1% were unidentified squamates, and 1% were amphibians. We detected no differences in diet based on kingsnake sex or color pattern, nor evidence of individual specialization. Rodents, lizards, and birds were eaten more frequently by larger individuals; snakes were eaten with similar frequency independent of predator size. Predation on mammals, birds, and lizards, but not snakes, was seasonally restricted. Kingsnakes from arid bioregions consumed more snakes, fewer rodents, and fewer lizards than did those from non-arid bioregions. Overall frequencies were similar for rodents and snakes, yet snakes accounted for 45% of prey biomass; among snakes, rattlesnakes comprised 24% by frequency and 37% of snake prey biomass and energy. Prey-predator mass ratios averaged 0.24 ± 0.19 (range 0.01-0.73; n = 43); a positive relationship exists between prey mass and snake mass, but larger snakes also consumed small prey items. Rattlesnakes, amounting to only 7% of overall diet and 16% of total biomass and energy value, are available throughout the active season and provide higher payoff per item than other diet types. Our findings thus provide a resolution to the paradox that this generalist predator is specialized (i.e., venom immunity) to feed on rattlesnakes, a rare prey type.

Hibernation sites at higher latitudes must protect snakes from colder conditions for longer periods of time. Because fewer locations are likely to be suitable, hibernation site availability may restrict the northern distribution of snakes. We considered overwinter mortality, hibernation site fidelity, and the abundance of suitable hibernation sites based on surface features to assess whether Massasauga Rattlesnakes are likely to be limited by hibernation site availability on the Bruce Peninsula, Ontario, Canada.We used three years of radio telemetry to locate 46 hibernation sites of 32 individual snakes. Snakes hibernated individually in old root systems, rodent burrows, and rock crevices in forested areas. Hibernation sites could be differentiated from forested areas generally available to snakes but not from locations with holes and crevices in the immediate vicinity of hibernation sites. Few snakes hibernated in the same location in consecutive years, although most (>70%) hibernated within 100 m of their previous location. Overwinter mortality over three years (23%) was similar to mortality during the active season (21%). Our results suggest that Massasauga Rattlesnakes may be limited by the availability of suitable hibernation sites, but sites of similar quality to those used by overwintering snakes are locally abundant. The location of hibernation sites within forests could not be predicted reliably based on surface features. Therefore, efforts to conserve habitat for this threatened species should consider all forested areas on the Bruce Peninsula as potential hibernation habitat.

QL 640.$ .vug HARVARD UNIVERSITY Library of the Museum of Comparative Zoology Vertebrate Ecology and Systematics A Tribute to Henry S. Fitch Edited By Richard A. Seigel Lawrence James L. E. Hunt Knight Luis Malaret Nancy L. Zuschlag The University of Kansas Museum of Natural History o \JkS UNIVERSITY OF KANSAS PUBLICATIONS MUSEUM OF NATURAL HISTORY Copies of publications may be purchased from the Publications Museum of Natural History, University of Kansas, Law- Secretary, rence, Kansas 66045. HARVARD UNIVERSITY H Library of the Museum of Comparative Zoology Front cover: The head of an adult Osage Copperhead (Agkistrodon contort rixphaeofrom Douglas County, Kansas. Drawing © 1984 by Linda Dryden. gaster) University of Kansas Museum of Natural History Special Publication No. 10 21 June 1984 Vertebrate Ecology and Systematics A Tribute to Henry S, Fitch Edited By Richard A. Seigel Lawrence James L. E. Hunt Knight Luis Malaret Nancy L. Zuschlag Museum of Natural History Department of Systematics and Ecology The University of Kansas Laurence. Kansas 66045 University of Kansas Laurence 1984 University of Kansas Publications Museum of Natural History Editor: Joseph T. MU5, Collins COMR ZOOL LIBRARY JUL U HARVARD Special Publication No. 10 pp. i-viii; 1-278; 79 figures 86 tables; 2 appendices Published 21 June 1984 UNIVERSITV Copyrighted 1984 By Museum of Natural History University of Kansas Lawrence, Kansas 66045 U.S.A. Printed By Alltn Press, Inc. Lawrence. Kansas 66044 ISBN: 89338-019-0 CONTENTS PART I. Henry INTRODUCTION S. Fitch in Perspective William Duellman E. 3 The Published Contributions of Henry S. Fitch Virginia R. Fitch PART II. 5 REPRODUCTIVE BIOLOGY AND POPULATION DYNAMICS Growth, Reproduction and Demography of the Racer, Coluber constrictor mormon, Northern Utah William S. Brown and William S. Parker Growth of Bullsnakes {Pituophis melanoleucus sayi) on a Sand Prairie in in 13 South Central Kansas Dwight R. Piatt Communal Denning 41 in Snakes Patrick T. Gregory 57 Parameters of Two Populations of Diamondback Terrapins (Malaclemys terrapin) on the Atlantic Coast of Florida Richard A. Seigel 77 An Ecological Study of the Cricket Frog, Acris crepitans Ray D. Burkett Female Reproduction in <• 89 an Arkansas Population of Rough Green Snakes (Opheodrys aes- tivus) Michael V. Plummer 105 Clutch Size in Iguana iguana in Central A. Stanley Rand Panama Are Anuran Amphibians Heavy Metal Accumulators? Russell J. Hall and Bernard M. Mulhern PART III. 115 123 FEEDING AND BEHAVIOR Energetics of Sit-and-Wait and Widely-Searching Lizard Predators Robin M. Andrews 137 Feeding Behavior and Diet of the Eastern Coral Snake, Micrurus fu/vius Harry W. Greene 147 The Role of Chemoreception in the Prey Selection of Neonate Reptiles Pennie H. von Achen and James L. Rakestraw 163 Ecology of Small Fossorial Australian Snakes of the Genera Neelaps and Simoselaps (Serpentes, Elapidae) Richard Shine 173 Scaphwdontophis (Serpentes, Colubridae): Natural History and Test of a Mimicry-Related Hypothesis Robert W. Henderson Dominance in 185 Snakes Charles C. Carpenter An Experimental Study of Variation Peromyscus maniculatus gracilis John H. Fitch 195 in Habitat Selection and Occurrence of the Deermouse, 203 PART IV. SYSTEMATICS AND BIOGEOGRAPHY Herpetogcography in the Mazatlan-Durango Region of the Sierra Madre Occidental, Mexico Robert G. Webb 217 Systematic Review of the Percid Fish, Etheostoma lepidum Alice F. Echelle, Anthony A. Echelle, and Clark Hubbs 243 Ernest E. New Species of the Anolis aequatorialis Group from Ecuador and Colombia Williams and William E. Duellman Anolis fttchi, a INDEX TO SCIENTIFIC NAMES 257 267 IV Preface This volume the result of a is symposium en- Ecology," held "Perspectives on 9 August 980 in conjunction with the annual meetings of the Society for the Study of Amin Fitchian titled. 1 phibians and Reptiles and the Herpetologists' League at Milwaukee, Wisconsin. The sympo- sium was organized to honor Dr. Henry S. Fitch on the occasion of his retirement in June 1980 tology; Philip S. Humphrey, Director, Museum of Natural History; and Richard F. Johnston. Chairman. Department of Systematics and Ecology, is greatly appreciated. Joseph T. Collins, Editor. Museum Publica- tions, deserves special recognition for his helpful after advice and continued patience in answering our many questions concerning the development and execution of the symposium and this volume. atics The 32 years with the Department of Systemand Ecology at the University of Kansas. Sixteen papers were presented in two sessions during the symposium and. aside from a few cheerful and patient assistance of Rose Etta Kurtz was invaluable. additions, the organizational format of this volume closely follows that of the symposium. Finally, we are most grateful to the following persons for reviewing the manuscripts appearing in this volume: Robert D. Aldridge. Stevan J. Manuscripts were submitted and accepted in late 1980 and 1981. but authors were given an op- Thomas portunity to update their contributions in early 1 983. In organizing the symposium we were surprised by the breadth of research conducted by the participants. Because of Fitch's influence on his past and present students and colleagues, this Arnold, Reeve M. Bailey. Royce E. Ballinger. J. Berger. William S. Brown. Gordon M. Burghardt, Janalee P. Caldwell, Jonathan A. Campbell, David C. Cannatella. David K. Chiszar, Martha L. Crump, Arthur E. Dunham. Donald G. Dunlap. Henry S. Fitch. Darrell Frost. J. Whitfield Gibbons. Peter Gray, Harry not restricted to herpetological contributions. Thus, the topical emphasis of this vol- Wendy Gorman. Harold ume ter Klopfer, Carl Lieb, volume is own research interests. The breakdown by subject of the papers following contained in this volume versus Fitch's published papers: ecology (this volume: 78%, Fitch: (73%); systematics and biogeography (17% vs. 19%): conservation (5% vs. 5%); and by taxonomic emphasis: squamates (this volume: 73%; Fitch: 62%); other amphibians and reptiles (14% vs. 7%); other vertebrates (13% vs. 23%). reflects Fitch's is We a Max A. Nickerson of the Museum and Al Williams of wish to thank Milwaukee Public the University of Wisconsin-Milwaukee and their respective staffs for logistical support in arranging and conducting the symposium. A special note of thanks is extended to Virginia Fitch and other in the members of the development Fitch family for assistance of the symposium. The or- ganizational advice and encouragement of William E. Duellman, Curator. Division of Herpe- W. Greene. Heatwole. James E. Huheey, John B. Iverson, Keith V. Kardong. Pe- Harvey B. Lillywhite. John D. Lynch. Richard Mayden. Roy W. McDiarmid, Lawrence M. Page. William S. Parker, F. Harvey Pough. Rebecca A. Pyles, Steven M. Roble. Albert Schwartz. Richard Shine. Norman A. Slade. Linda Trueb, John Wiens. and Bernard Willard. Without the help of all these individuals this tribute to an outstanding biologist would not have been possible. Richard A. Seigel Lawrence F. Hunt James L. Knight Luis Malaret Nancy L. Zuschlag Lawrence. Kansas October 10. 1981 Contributors Robin M. Andrews. Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061 William S. Brown, Department of Biology, Skid- more College, Saratoga Springs, New York 12866 Ray D. Burkett, Department of Biology, Shelby State Community College, P.O. Box 40568, Memphis, Tennessee 38104 Charles C. Carpenter, Department of Zoology, University of Oklahoma. Norman, Oklahoma 73019 William E. Duellman, tory, University Museum of Natural His- of Kansas. Lawrence, Kansas 66045 Alice F. Echelle. School of Biological Sciences, Oklahoma State University, Stillwater, Okla- homa 74078 Oklahoma Oklahoma 74078 John H. State University, Stillwater, Fitch, Massachusetts Lincoln, Massachusetts Audubon Society, 01773 Virginia R. Fitch, University of Kansas Natural History Reservation, Lawrence, Kansas 66044 Harry W. Greene, Museum of Vertebrate Zoology, University of California, Berkeley, California 94720 Patrick T. Gregory, Department of Biology, University of Victoria, Victoria, British Co2Y2 lumbia, Russell J. Hall, U.S. Fish and Wildlife Service, V8W Patuxent Wildlife Research Center, Laurel, Maryland 20811 V\ . Henderson. Section of Vertebrate Zo- Milwaukee Public Museum, Milwaukee, Wisconsin 53233 Clark Hubbs, Department of Zoology, University of Texas, Austin. Texas 78712 Bernard M. Mulhern. U.S. Fish and Wildlife ology, Service, Patuxent Wildlife Research Center. Laurel, Maryland 208 William S. Parker, Department of Biological 1 1 Sciences, Mississippi University for Women, Columbus, Mississippi 39701 Dwight R. Piatt, Department of Biology, Bethel College, North Newton, Kansas 671 17 Michael V. Plummer, Department of Biology, Harding University, Searcy, Arkansas 72143 James Museum of Natural Hisof Kansas, Lawrence. Kansas L. Rakestraw, tory, University 66045 Anthony A. Echelle, School of Biological Sciences, Robert A. Stanley Rand. Smithsonian Tropical Research Institute, APO Miami. Florida 34002 Richard A. Seigel, Savannah River Ecology Laboratory, Drawer E, Aiken, South Carolina 29801 Richard Shine, School of Biological Sciences, University of Sydney, N.S.W. 2006 Australia Pennie H. von Achen, RD 2, Eudora, Kansas 66025 Robert G. Webb, Department of Biological Sciences, University of Texas, El Paso. Texas 79968 Museum of Comparative Zoology, Harvard University, Cambridge. Massachusetts 02138 Ernest E. Williams, c XS a t_ 00 O Xoa Ih o o 6 o 2 H •3 c >> X) s Xj 00 c I 3 O c E E 00 00 in 00 .J; "5 O O ft a o X i, . --.. i. u o u a </> '-£ c o i. V'v Parti Introduction Fig. 1. Henry S. Fitch in the field. Photograph by David M. Hillis. Vertebrate Ecology and Systematics— A Tribute to Hours S it< h Edited b> R. A. Seigcl. L. E. Hunt. J I Knight. I Malaret and N. L. Zuschlag i "t>84 Museum of Natural History, The University ot Kansas. Lawrence I I Henry S. Fitch in Perspective William E. Who is Henry Sheldon Fitch? This quiet, modest, unassuming man made his first entrance into the world of biologists by publishing on birds in the Condor in 1933. Yes, birds! Oregon Most of us think of Henry S. Fitch as a herpetologist. Yet, 50 published papers in the past 50 years, of his 1 only about two-thirds of them deal with amphibians and reptiles. Twenty others have been on mammals, 12 on birds, and others on spiders, molluscs, and plants. Most of us think of Henry S. Fitch as an ecologist, but 25 of his papers are on systematics and include his classic work on alligator lizards published in 1934 and his highly perceptive study of western garter snakes published in 1940 (doctoral dissertation at the University of California, Berkeley). His more recent systematic work has dealt with Middle American anoles— a field where most systematists have feared to tread. Fitch's best known works are on the natural From his earliest papers on he has provided extensive field observations. In 1 948, he entered a "naturalist's heavhistory of reptiles. reptiles, en"— the University of Kansas Natural History Reservation. There he began intensive studies on the biota of one square mile of deciduous hardwood forest — studies involving population movements, food, growth rates, hiberand reproduction — all substantiated with densities, nation, massive quantities of data. Through his efforts square mile is better known herpetologically than any other in the world. His studies on the natural history of reptiles this are classics. Outstanding examples are the thorough study of the five-lined skink ( 1954) and the Duellman erature research are reflected in his syntheses. These traits combined with dogged determination to learn all is to know about his subcontinued productivity, and there jects of study, his his willingness to share his ideas, knowledge, and enthusiasm with students have assured him of a permanent place in the herpetological hall of fame. At the present time, many biologists commonly are narrow specialists. Henry Fitch doesn't fit modern pigeon hole. He is a naturalist broadest sense of the word. His breadth into a in the is matched by very few of his contemporaries and scarcely imagined by most of his younger colleagues. An analogy can be drawn of knowledge with the story of the hare and the tortoise, with Henry Fitch as the tortoise steadily plodding along his path of scientific endeavor, frequently being passed by various biological bandwagons, only them sometimes morassed or abandoned to find further down the road. He has avoided biopolitics. He has not been a vigorous proponent of controversial theories. Instead, he has continued to be a fine naturalist. But. his published works are among those commonly cited in support of some theories or in the falsification of others. Thus, for half a century Henry S. Fitch has been a major contributor to our knowledge of the natural history of diverse kinds of animals. During this time he has introduced innumerable students to intensive field studies, has thought- guided the research of many graduate students, and has collaborated with a diversity of fully colleagues. A major factor in his remarkable and exhaustive study of the copperhead ( 1 960). More recently he has worked on the interactions of successful career has been a collaborator, assis- behavior and ecology, communities of anoles. and populations and conservation of iguanas. Fitch. In addition to these systematic and ecological works. Fitch has provided us with important syntheses— reproductive cycles in lizards and snakes (1970) and sexual size differences in reptiles (1981). All of his works are characterized by careful and detailed studies on the existence of populations in nature. Vast quantities of such data combined with extensive laboratory and lit- tant, caretaker Few and charming lady — Virginia R. scientists can reflect on such a long and productive career, and yet upon officially retiring maintain such enthusiasm for an active research program. Henry Fitch's careful work on natural is well worth emulating. Our knowledge of animals in nature would be far greater if there history were many more biologists in the world followed in the footsteps of Henry S. Fitch. who Vertebrate Ecolog> and Systematics— Edited by R. A. Seigcl. L. E. Hunt. J. 1984 Museum of Natural Hislorj < A I. I Tnbulc to Henr\ S I itch Knight. L. Malarel and N. L, Zuschlag he Iniversity of Kansas. Lawrence 1 The Published Contributions of Henry S. Fitch Virginia R. Fitch Beginning with his 1933. the writings of first published paper Henry S. of the fence lizard. Univ. California Publ. in Fitch have en- Zool., 44:151-172. Some compassed a wide range of subjects and disciplines, from reptilian ecology to bird behavior, from the economic relationships of rodents to an intensive study of spiders, and include such areas as taxonomy, life history', behavior, and repro- 1940. ductive biology. To date, he has produced 150 papers, all of which appear in the following list. 1941. Geographic variation in garter snakes of the species Thamnophis sirtalis in the Pa- works include as their subjects (19 papers), birds (12), vertebrates in general (5), spiders (3). vegetation and habitats (4), and mollusks ( 1 ), as well as five book reviews, observations on horned owl nests. Condor. 42:73-75. 1941. The feeding habits of California garter snakes. California Fish and Game, 27:232. Coast region of North America. Amer. Fitch's published cific mammals Mid!. Nat.. 26:570-592. but papers on amphibians and reptiles ( 1 00) pre- dominate. His works are widely cited throughout scientific periodicals, and this list is presented both as a service to biologists and to document impressive extent of the knowledge and breadth of interest of Henry S. Fitch. 1942. Interrelations of rodents and other wildlife and predation. California Fish and Game. 32:144-154 (with B. Glading and V. the 1933. Bird notes from southwestern Oregon. Condor. 35:167-168 (with J. O. Steven- House). 1946. Feeding habits of the Pacific rattlesnake. Copeia. 1946:64-71 (with H. Twining). 1 946. Behavior and food habits of the red-tailed son). 1934. New alligator lizards from the hawk. Condor. 48:205-237 (with Swenson and D. F. Tillotson). Pacific Coast. Copeia. 1934:6-7. 1934. A shift of specific names in the genus 1935. An abnormal Ger- Horn). pattern in a gopher snake. Copeia, 1935:144-146. 1935. Natural history of the alligator lizards. Trans. Acad. Sci. St. Louis. 29:1-38. Amphibians and reptiles of the An older name for Triturus similans Twitty. Copeia, 1938:148-149. 1938. A systematic account of the alligator ards in the The 1947. A 191-192. ifornia Publ. Zool.. petologica. 1:152-153. A biogeographical study of the Ordinoides artenkreis of garter snakes (genus Thamnophis). Univ. California Publ. T. of California. Condor. 49:137-151. 1947. Ecology of a cottontail rabbit (Sylvilagus auduboni) population nia. California Fish study of the growth and behavior in central Califor- and Game. 33:159- 184. 1947. Rattlesnakes on the range. Pacific Stockman. 13(6):8-9 (with EC A. Wagnon). 1947. Rattlesnakes on western farm lands. Western Dairy Jour.. Sept.:23. 78-79 (with K. A. Wagnon). 1947. Ground Zool.. 44:1-150. field 48:169-220 (with Rodgers). 1947. Predation by owls in the Sierran foothills fornia. Herpetologica. 1:151-152. 1939. Leptodeira in northern California. Her- A Game. 33:103-123 liz- western United Nat., 20:381-424. 1940. study of a rattlesnake population. (with B. Glading). 1947. Variation in the skinks (Reptilia: Lacertilia)oftheSkiltonianusgroup. Univ. Cal- 1939. Desert reptiles in Lassen County. Cali- 1940. field California Fish and L. Gerrhonotus) States and lower California. Amer. Midi. ( California Ground Squirrel by J. M. Linsdale (Book review). Jour. Mamm.. 28: 1947. Rogue River Basin, Oregon. Amer. Midi. Nat.. 17:634-652. 1938. Ranaboylii in Oregon. Copeia, 1938:148. 1938. F. 1946. Trapping the California ground squirrel. Jour. Mammal.. 27:220-224 (with E. E. rhonotus. Copeia, 1934:172-173. 1936. of the range. Univ. California Agr. Exp. Sta. Bull.. 663:96-129 (with E. E. Horn). 1946. Observations on Cooper's hawk nesting squirrels mean destroyed forage. SPECIAL PUBLICATION-MUSEUM OF Western Livestock Jour.. Oct.:37, 109, 1 10. 1 12. 1948. Further remarks concerning Thamnophis ordinoidcs and its relatives. Copeia, 1948: 1955. Habits and adaptations of the Great Plains skink (Eumeces obsoletus). Ecol. Monogr., 948. Habits and economic relationships of the Tulare kangaroo rat. Jour. Mamm., 29:5- summer tanager in northeastern Kansas. Wilson Bull., 67:4554 (with V. R. Fitch). 1955. The coyote on a natural area in north- eastern Kansas. Trans. Kansas Acad. Sci., 35. 1948. Ecology of the California ground squirrel on grazing lands. Amer. Midi. Nat., 39: 1948. 25:59-83. 1955. Observations on the 121-126. 1 NATURAL HISTORY 513-596. study of coyote relationships on A 58:211-221 (with R. 1956. L. Packard). A field study of the Kansas ant-eating frog, Gastrophryne olivacea. Univ. Kansas Mus. Nat. Hist., 8:275-306. cattle Publ., Management, 12:73- 1956. 1949. Sparrow adopts kingbirds. Auk, 66:368369. An ecological study of the collared lizard (Crotaphytus collaris). Univ. Kansas Publ., Mus. Nat. Hist., 8:213-274. 1956. A range. Jour. Wildlife 78. 1949. Outline for ecological of 1949. reptiles. life history studies Ecology, 30:520-532. Use of California annual-plant ten-year old skink? Herpetologica, 12: 328. 1956. Early sexual maturity and longevity under natural conditions in the Great Plains nar- forage by row-mouthed range rodents. Ecology, 30:306-321 (with J. R. Bentley). 1949. Study of snake populations in central California. Amer. Midi. Nat., 41:513-579. 1949. Road counts of snakes in Mamm., 1951. Sci., 54:548-559 (with W. Tanner). 1951. A simplified type of funnel trap for rep- tiles. 1952. Herpetologica, 7:77-80. in the southeastern United The armadillo States. Jour. Mamm., 33:21-37 Goodrum and C. Newman). 1952. (with P. The University of Kansas Natural History Reservation. Univ. Kansas Mus. Nat. Hist. Misc. Publ., no. 4:1-38. 1952. (Book review) Ecological Animal Geography by Hesse, Allee and Schmidt. Wil- son Publ., Mus. Nat. Hist., 7:309-338 (with L. L. Sandidge). 1953. (Book review) Natural Communities by L. R. Dice. Wilson Bull.. 65:121-123. 1954. Seasonal acceptance mammals. Jour. of bait Mamm., Rainey). 1957. Aspects of reproduction and development in the prairie vole (Microtus ochrogaster). Univ. Kansas Publ., Mus. Nat. Hist., 10: 129-161. 1957. Observations on hibernation and nests of the collared lizard. Crotaphytus collaris. Copeia, no. 4:305-307 (with J. M. Legler). skink, Eumeces Publ., Mus. Nat. Kansas Publ., Mus. Nat. 1958. 8:1-156. Home ranges, Hist., territories, 1 1:1 1-62. and seasonal movements of vertebrates of the Natural History Reservation. Univ. Kansas Publ., Hist., 11:63-326. by 35:39-47. fasciatus. Univ. Kansas Hist., 1958. Natural history of the six-lined racerunner (Cnemidophorus sexlineatus). Univ. small 1954. Life history and ecology of the five-lined Hist., 10: 77-127 (with R. L. McGregor). 1956. The molluscan record of succession on the University of Kansas Natural History Reservation. Trans. Kansas Acad. Sci., 59: 442-454 (with D. H. Lokke). 19 56. Ecological observations on the woodrat, Neotoma jloridana. Univ. Kansas Publ., Mus. Nat. Hist., 8:499-533 (with D. G. Bull., 64. 1953. Ecology of the opossum on a natural area in northeastern Kansas. Univ. Kansas The forest habitat of the University of Kansas Natural History Reservation. Univ. Kansas Publ., Mus. Nat. the systematics of the collared lizard (Crotaphytus collaris), with a description of a new subspecies. Trans. Kansas Acad. 1956. 31:364-365. Remarks concerning am- phibians and reptiles of northeastern Kansas. Univ. Kansas Publ., Mus. Nat. Hist., 8:417-476. western Loui- A new style live-trap for small mammals. Jour. Herpetologica, 12: 1956. Temperature responses in free living siana. Herpetologica, 5:87-90. 1950. frog. 281-282. Mus. Nat. 1959. A patternless phase of the copperhead. Herpetologica, 15:21-24. 1959. Aspects of needed research on North VERTEBRATE ECOLOGY AND SYSTEMATICA 1 American grasslands. Trans. Kansas Acad. Sci.. 62:175-183 (with 5 other authors). 960. Criteria for determining sex and breeding maturity in snakes. Herpetologica, 16:49— 51. 1960. Autecology Kansas copperhead. Univ. Hist., 3:85-288. of the Publ., Mus. Nat. 1960. (Book review). 1 967. Ecological studies of lizards on the University of Kansas Natural History Reser- 1 The Rusty Lizard, a PopW. Frank Blair. Copeia, behavior. Pp. 44-45 in Engineering Re- 1960:386-387. neering Sci. Div., Univ. Kansas. Vol. Ill E. D. Bevan, Ed. (with H. (1967-1968), W. Mountains. Univ. Kansas Publ.. Mus. Nat. 196 1 . An older name for P. Maslin). Thamnophis 1961. W. Milstead). The snake as Shirer. W. Research, (with H. tozoa in the laboratory. Turtox News, 39: 247. A University of Texas Texas Press. Univ. symposium. 1963. Natural history of the racer Coluber constrictor. Univ. Kansas Publ.. Mus. Nat. Hist.. 15:351-468. 1963. Observations on the Mississippi kite in southwestern Kansas. Univ. Kansas Publ.. K. Legler and D. D. Pip- Data acquisition systems for the study of vertebrate ecology. Pp. 19-20 in Research, Vol. IV, The Univ. Kansas, Center for a source of living sperma- 1961. Longevity and age-size groups in some common snakes. Pp. 396-414 in Verte- W. pin.) 1970. cyrtopsis (Kennicott). Copeia, 1961:112 (with Lizard Ecology, a search, Center for Research, Inc.. Engi- 1961. Occurrence of the garter snake Thamnophis sirtalis in the Great Plains and Rocky 13:289-308 (with T. in torial lizards. Herpetologica 24:35-38. 969. Biotelemetric studies of small vertebrate 1 ulation Study by Hist., 30-44 vation. Pp. symposium. Univ. Missouri Press. 1968. Temperature and behavior of some equa- W. Inc.. P. Legler, D. D. Pippitt. F. McMillan. Ed. Shirer, K. Armitage. J. W. D. Pauley and K. J. Downhower.) 1970. Reproductive cycles in lizards and snakes. Univ. Kansas Mus. Nat. Hist., Misc. Publ., brate Speciation: 52:1-247. 1970. Comparison from radiotracking of movements and denning habits of the raccoon, striped skunk, and opossum in northeastern Kansas. Jour. 503 (with H. W. Mamm., 5 1(3):49 1- Shirer). A 1963. Natural history of the black rat snake radiotelemetric study of spatial relationships in the opossum. Amer. Midi. Nat. 84(1): 170-1 86 (with H. W. Shirer). Kansas. Copeia. 1970. Natural history of the milk snake (Lani- Mus. Nat. Hist.. (Elaphe o. obsoleta) 1963:649-658. in 1963. Spiders of the University of Kansas Natural History' Reservation and Rockefeller Experimental Tract. Univ. Kansas Mus. Nat. Hist., Misc. Publ., no. 38:1-202. 1965. 1 970. 12:503-519. propeltis triangulum) in northeastern Kansas. Herpetologica 26(4):387-396 (with R. R. Fleet). 1971. Ecological notes on The University of Kansas Natural History Reservation in 1965. Univ. Kansas Mus. Nat. Hist., Misc. Publ.. no. 42:1- 1971. An ecological study of the garter snake, Thamnophis sirtalis. Univ. Kansas Publ.. Mus. Nat. Hist., 15:493-564. 1965. Breeding cycle in the ground skink, Ly- gosoma laterale. Univ. Kansas Publ., Mus. Nat. Hist., 15:565-575 (with H. W. Greene). 1966. Spiders from Meade County, Kansas. Trans. Kansas Acad. Sci.. 69:1 1-22 (with some common snakes. Co- 1971:118-128 (with H. W. Shirer). 1971. (Book review). A Complete Field Guide to Nests in the United States by R. Headstrom. Jour. Wildlife Mgt., 35:188-189. peia. 1971. A 1971. Echelle and A. F. Echelle). A new anole from Costa Rica. Herpetologica. 27:354-362 (with A. A. Echelle and 1971. A. F. Echelle). Further observations on the demography of the Great Plains skink {Eumeces ob- V. R. Fitch). 1967. Preliminary experiments on physical tolerances of the eggs of lizards and snakes. Ecology, 48:160-165 (with A. V. Fitch). lizards (with A. V. Fitch and C. W. Fitch). A radiotelemetric study of spatial rela- tionships in 60. 1965. some common of southern Mexico and Central America. The Southwestern Naturalist, 15:398-399 comparative analysis of aggressive display in nine species of Costa Rican Anolis. Herpetologica, 27:271-288 (with A. A. SPECIAL PUBLICATION-MUSEUM OF Trans. Kansas Acad. soletus). 98 (with R. 1972. J. Sci.. Iguanidae) and its syntopic congeners at four localities in southern Mexico. 74:93- tilia: Hall). Radio tracking of wild animals in their natural habitat. Pp. 35-36 in Research. Univ. Kansas Center for Research, Inc.. Herpetologica, 31:459-471 (with R. 1975. (Diadophis punctatus) Kansas Mus. Nat. Shirer. L. A. Gold. R. L. scription of a pers. new Mus. Nat. 1976. subspecies. Occas. PaUniv. Kansas, no. F. 1976. 1972. Observations offish-eating and mainte- Echelle and A. F. Echelle). 1973. Observations on the population ecology of the Central American iguanid lizard, Anolis cupreus. Caribbean Jour. Sci., Hist., W. pholcid spider from northeastern Kansas. Bull. Kansas Ent. Soc. (with E. Kansas Mus. little known mainland anoles. Univ. Kansas Sci. Bull.. 51:91-128 (with A. F. Echelle and A. A. Echelle). 1976. A field study of Costa Rican lizards. Sci. Bull., Univ. 50(2):39-126. 1973. Yellow-billed cuckoo nesting at Univer- of Kansas Natural History ReservaKansas Ornith. Soc. Bull., 24(2): 2- tion. 1 von Achen). 1974. Observations on the food and nesting of the broad-winged hawk (Buteo platypterus) in northeastern Kansas. Condor. 15 (with P. 76(3):331-333. Food habits of Basiliscus basiliscus anole (Reptilia: Iguanidae) from 1976. Dragons for dinner. Wildlife Omnnibus, International Wildlife, 6(6): 1 7 (with R. W. A new anole (Reptiha: Iguanidae) from southern Veracruz, Mexico. Jour. Herp. A new Great Corn Island. Caribbean Nicaragua. Contr. Biol. & Geol., Milwaukee Pub. Mus., no. 9:1-8 (with R. W. Henderson). Univ. Kansas, no. 18:1- Ornith. Soc. Bull., 24(4):33-35 (with H. A. Stephens and R. O. Bare). 975. R. A new 1976. Field observations on rare or Henderson). 1977. Age and sex differences in the ctenosaur. {Ctenosaura similis). Contr. Biol. & Geol., Milwaukee Pub. Mus., no. 1 1:1-1 (with 125-1 28 (with R. W. Henderson). 1973. Road counts of hawks in Kansas. Kansas 1 (with Nat. Hist., no. 50:1-21. 7(2): 1974. 10:303-311 Herp., anoles. Occas. Papers. Univ. 41. Kansas 40:1-60. 976. Sexual size differences in the mainland 13(3- 1973. Population structure and survivorship in some Costa Rican lizards. Occas. Papers, Mus. Nat. Hist., study of the rock anoles (Reptilia, Lacertilia, Iguanidae) of southern Mexico. O. Maughan). 1 4):215-229. sity Mus. Nat. field Henderson). Echelle). nance behavior in two species of Basilisens. Copeia, 1972:387-389 (with A. A. A Jour. Hist., 8:1-20 (with A. A. Echelle and A. Kansas. Univ. 1-53. sas 1972. Variation in the Central American iguanid lizard, Anolis cupreus, with the de- in Hist., Misc. Publ.. 62: 1975. Sympatry and interrelationships in Costa Rican anoles. Occas. Papers, Univ. Kan- 21. 1973. A demographic study of the ringneck snake mann, H. W. 1 W. Henderson). Vol. V, P. Nicholas, Ed. (with R. HoffLattis, C. B. Rideout, and R. C. Waltner). 1972. Ecology of Anolis tropidolepis in Costa Rican cloud forest. Herpetologica, 28( ): 1 0- 1973. NATURAL HISTORY 1 W. Henderson). R. 1977. Age and sex differences, reproduction and conservation of Iguana iguana. Contr. Biol. & Geol., Milwaukee Pub. Mus., 13: 1-21 (with R. W. Henderson). 1977. Spatial relations and seasonality in the skinks, Eumeces fasciatus and Scincella laterale in northeastern Kansas. tologica, 33:303-313 (with P. Herpe- von Ach- en). in movements and reproduction Costa Rican bat communites. OcPapers Mus. Nat. Hist., Univ. Kan69:1-28 (with R. K. LaVal). 1977. Structure, Costa Rica. Jour. Herp.. 8(3):260-262 in three (with R. R. Fleet). cas. A preliminary ecological study of the soft- sas, Trionyx muticus, in the Kansas River. Israel Jour. Zoology, 24: 28-42 (with M. V. Plummer). 1975. A comparative study of the structural and 1978. Inter- and intraspecific allometry in a display organ: the dewlap of Anolis (Igua- climatic habitats of Anolis sericeus (Rep- 1978. Behavioral evidence for species status of shelled turtle, nidae) species. Copeia, 1978(2):245-250 (with A. F. Echelle and A. A. Echelle). VERTEBRATE ECOLOGY AND SYSTEMATICS Anolis uniformis (Cope). Hcrpetologica 34(2):205-207 (with A. F. Echcllc and A. 1982. Reproductive cycles in tropical reptiles. Occas. Papers. Mus. Nat. Hist.. Univ. of Kansas 96:1-53. A. Echelle). 1978. Dragons: 25c/lb. Animal Kingdom, Feb./ March: 12-17 (with R. W. Henderson). 1978. A study of the red-tailed field hawk 1982. Resources of a snake community in prai- rie-woodland habitat of northeastern Kansas. Pp. 83-97 in Herpetological Communities (N. J. Scott Jr., ed.). U.S. Fish and Wildlife Serv.. Wildl. Res. Rep. in eastern Kansas. Trans. Kansas Acad. Sci.. 8 1(1): 1-1 3 (with R. O. Bare). 1978. Sexual size differences in the genus Sce- 13. Sci. Bull., 51(13): 1983. Exploitation of iguanas in Central America. Pp. 397^ 1 7 in Iguanas of the World: 1978. Ecology and exploitation of Ctenosaura Their Behavior, Ecology, and Evolution (G. M. Burghardt and A. S. Rand, eds.) loporus. Univ. Kansas 441-461. similis. Univ. Kansas Two new anoles (Reptilia: Noyes Press (with R. W. Henderson and D. M. Hilhs). 1983. Thamnophis elegans. Cat. American waukee Pub. Mus.. 20:1-15. 1983. 483-500 (with 1 978. Sci. Bull.. 51(15): R. W. Henderson). Iguanidae) from Oaxaca, with comments on other Mexican species. Contr. Biol. & Geol.. Mil- A 1978. field study of the prairie kingsnake (Lampropeltis calligaster). Trans. Kansas Acad. Sci.. 81:353-363. 1978. The plight of the iguana. LORE, Milwaukee Pub. Mus., 28(3):2-9 (with R. W. Jones). 1983. A 20-year record of succession on ed of tallgrass prairie on the RockeExperimental Tract. Univ. Kansas Mus. Nat. Hist.. Spec. Publ., 4: 1-1 5 (with area of northeastern Kansas. Proc. Sev- enth North American Prairie Conf., Au1 980. Pp. 117-121 (with W. D. Ket- Ctenosaura similis (Reptilia: Iguanidae) at Belize City, Belize. Brenesia, 16:69-80 1 W. Henderson). gust tle). In press. Remarks concerning certain western garter snakes of the Thamnophis elegans complex. Trans. Kansas Acad. Sci.. 83: 106-113. 980. Reproductive strategies of reptiles. Pp. 2531 in Reproductive Biology of Captive Reptiles T. Collins, eds.), (J. B. SSAR In J. ural area in northeastern Kansas. Occas. Geographic variation in clutch size and size in North American reptiles. Univ. Kansas Mus. Nat. Hist. Misc. Publ. In press. Cont. Herpetol.. 1981. Sexual size differences in reptiles. Univ. Kansas Mus. Nat. Hist. Misc. Publ.. 70: litter In press. Ecological patterns of relative clutch mass 1-72. in snakes. Oecologia (with R. A. Sei- gel). a species distinct from C. constrictor. Trans. Kansas Acad. Sci. In 84:196-203 (with W. In press. mormon, S. Brown and W. S. Parker). 1981. Interspecific vari- and morphological associations with habitat. Copeia (with D. M. Hillis). press. Succession in small mammals on a natPapers, Mus. Nat. Hist., Univ. Kansas (with V. R. Fitch and W. D. Kettle). and Diseases Murphy and The Anolis dewlap: ability 1:1-277. 1981. Coluber Costa Rican Chicago Press. 1979. Notes on the behavior and ecology of 1980. in 1983. Ecological succession in vegetation and small mammal population on a natural E. R. Hall). (with R. 422-425 Natural History. (D. H. Janzen, ed.) Univ. reseed- fields feller cherriei (Escincela Par- Sphenomorphus da, Skink). Pp. Henderson). 1978. Rept., 320.1-320.4. Ctenosaura similis (Garrobo. Iguana Negra. Ctenosaur). Pp. 394-396 in Costa Rican Natural History. (D. H. Janzen, ed.) Univ. Chicago Press (with J. Hackforth- Amph. Thamnophis Amph. Rept.. sirtalis. Cat. 270.1-270.4. American press. Thamnophis Amph. couchi. Cat. American Rept. Intergradation of Osage and broadbanded copperheads in Kansas. Trans. Kansas Acad. Sci. (with J. T. Collins). Part II Reproductive Biology and Population Dynamics Vertebrate t'cology and Systemalics— A Tribute to Henry S. Fiteh Edited by R. A. Seigcl. L. £ Hunt. J I Knight. L. Malaret and N. L. Zuschlag c 1984 Museum of Natural History. The University of Kansas. Lawrence Growth, Reproduction and Demography of the Racer, Coluber constrictor mormon^ in Northern Utah William S. Brown and William The Introduction S. Parker present study focuses on the biology of the Western yellow-bellied racer. Coluber con- Considerable interest has developed recently strictor and demographic studies because the data point up a number of evolutionary strategies taken by separate interand intraspecific populations. To date, the data have been effective mostly in illustrating the selection and adaptive basis for the life histories of lizards, birds, and mammals among the vertebrates (Stearns 1976; Hutchinson 1978). Rarely have data on snakes perfused the general literature even though a number of sound field studies of snake populations have been completed (Blanchard et al. 1979; Branson and Baker in comparative 1974; life history Brown 1973; Carpenter 1952; Clark 1970, mormon Baird and Girard, hereinafter called simply "racer." Our approach has been empirical and autecological and has concentrated on one large population of this snake at a Utah over a four-year period. This paper treats growth, maturity, reproduction, population structure, and demogsingle locality in northern raphy of the racer. North America, Widespread and abundant in C. constrictor lends itself well to a study of its adaptive biology in several parts of its geographic range. Our attempt is to provide ecological comparisons of populations in Utah and Kansas. This study reveals different life his- tory strategies at the intraspecific level. 1974; Clark and Fleet 1976; Feaver 1977; Fitch 1949, 1960, 1963, 1965, 1975; Gregory 1977; Hall 1969; Parker and Brown 1974, 1980; Piatt Methods 1969; Prestt 1971; Spellerberg and Phelps 1977; Stewart 1968; Tinkle 1957, 1960; Viitanen 1967). Snakes were captured autumn 969 through 1 W Coluber constrictor (Serpentes, Colubridae) is to occur from Guatemala to southern known Tooele County, Utah (40°36'N, 1 12°32'W, elevation 1580 m), ca. 58 km of Salt Lake City. This area is our primary study locality (area M) where all long-term mark and recapture field ville, WSW Canada (Conant 1975; Etheridge 1952; Stebbins 1966). The species is polytypic, with 10 described subspecies (Wilson 1970). In the United States, eight of the nine recognized geographic races occur east of the in spring 1973 at a communal denning area in a desert shrub habitat located 4 km of Grants- work was conducted. Rocky Mountains (Auf- We recorded a total of 1 694 captures of 1046 racers at this site. Originally studied by Woodbury and his co- fenberg 1955; Fitch 1963; Wilson 1970). C. c. mormon occurs west of the Continental Divide. workers in the 1940's (Woodbury et al., 1951), "main den" (den M) was later sampled in mid-1960's by Hirth and King (1968) and This subspecies has been recorded in most of the states in the western third of the U.S. (Auffenberg the 1955; Stebbins 1966; Wilson 1978:218.1). An extensive range hiatus in the Rocky Mountains, again in the early 1970's by us. We discovered other actively used dens near den M; these were lack of intergradation, and differences in the mor- considered part of a discrete group which we called "M complex." A separate series of newly- phology and ecology between the midwestern subspecies, C. may c. flaviventris, and warrant elevation of the species rank (Fitch, C. c. latter discovered dens located 0.8 designated "S complex" (Parker and Brown and Parker 1976a). Brown and Parker 1981). was Brown 973; to the south 1 The technique we used to capture snakes was them with a screen wire fence erected aound their hibernaculum. As the dens were sin- Aside from several brief reports on various aspects of the biology of C. c. mormon (see review of the literature in Fitch 1963), no comprehensive ecological study of this wide-ranging western form has been conducted. Fitch's (1963) study in Kansas of C. c. flaviventris is, to date, the most to intercept gle small rock piles located in fairly level terrain with sandy soils, it was possible to encircle each den completely. We sank steel reinforcing bars around a den, attached screening (ca. 95 cm high) to the stakes, and buried the base of the fence by extensive ecological investigation of any population of km mormon taxon to Coluber constrictor. 13 SPECIAL PUBLICATION-MUSEUM OF 14 it with soil from a perimeter trench. Captures occurred almost daily in favorable weather as snakes attempted to enter a den in covering autumn and leave it in spring. The chronology M of sampling Coluber and other snakes at area is summarized in Parker and Brown ( 980). Our 1 969- 972 from and autumn dens each spring by sampling autumn 1969 through spring 1973. Data preresults pertain to the four- year period 1 1 sented for a given calendar year were derived from sampling in the autumn of that year (den and only) and the spring of the next (den M M other dens). Individuals were processed in the laboratory and most were released within 24 h after capture. Each snake was permanently marked by clipping (Brown and Parker 976b). Snoutvent length (SVL = distance from tip of snout to posterior edge of anal scute) and tail length to ventral scutes the nearest 0.5 nearest body mm 1 cm (snakes > 1 year old) or to the (hatchlings and juveniles) and weight to the nearest 0. 1 g (all live snakes) were Reproductive condition of males was determined by obtaining cloacal smears and examining them microscopically for the presence of spermatozoa. Snakes in spring were released outside of their den fence, those in autumn were released inside. Snakes caught by hand on their summer range were released at the capture site. Other racers were collected from two nearby localities in northern Utah. Most snakes from these areas were sacrificed for food and reproductive data. These localities are designated as area SLC, vicinity of Salt Lake City, Salt Lake recorded for each individual at all captures. County, Utah; and area RB, Red Butte Canyon, 5 km E of Salt Lake City, Salt Lake County, Utah. Both areas SLC and RB provided data on clutch size. Female racers were marked and released in and NATURAL HISTORY spring, snakes placed in a designated year from their actual age by about 7 class differed we No growth occurred during hibernation assigned an equivalent age to autumn and months. so spring-captured snakes as follows: hatchling (age 0),juvenile(1.5and8.5 months), -year-old (13.5 and 20.5 months), 2-year-old (25.5 and 32.5 1 months), and so on. The simpler age designation in years corresponds to the number of full 5- month growing seasons which through and aspects of the > a facilitates analysis life snake had been of age-specific history. year old were sexed visually by the thick (males) or thin (females) tail base. relatively Juveniles lacked external sexual differences and those in 1972-1973 were sexed using a blunt Racers 1 probe to detect presence (males) or absence (females) of hemipenial sacs. In earlier sample periods juveniles were not sexed and numbers of male and female juveniles were apportioned assuming a 1:1 sex ratio. Some of these juveniles later by recapturing them as marked -year-olds after they had attained sufficient discernible sexual dimorphism as yearlings. were sexed 1 Assignment of males and females to specific age classes was based on size and growth of marked individuals. Sample means and 95% confidence intervals of length and weight were calculated for recaptured 1 -year-olds marked initially as juveniles. Snakes in all sampling periods that closely to these values were as-year-olds. Records for these initial compared signed as 1 juveniles and 1 -year-olds that were later recaptured were then used to determine preliminary length and weight characteristics for 2- and 3-yearolds. Some individuals were thus followed from age 3 in 1969-1970 to age 6 in 1972-1973. By working in this step-wise procedure, many individuals were aged through 6 years and a few RB and provided data on body weight changes. Some other females killed for examination of reproductive tracts were from area M. through 7 years. Lacking prior captures made some error possible in assigning ages of 4 and 5 These snakes included several casualties from our marked population and a few others taken > 2 km from the study dens and beyond the maximum dispersal limits of racers from area M. In most years at area hatching occurred around mid-August. Juveniles normally arrived in early October at an average age of at den ca. 1.5 months. Winter dormancy lasted ca. 7 months (Oct. -Apr.) and the activity season ca. 5 months (May-Sept.) (Brown 1973; Parker and Brown 1980). As our sampling was in autumn of comparing sizes to known-age statistical values was consistent and uniformly applied over all ages. We tended to be conservative in cases area M M years to snakes early in the study, but our method involving a size intermediate between two ages, snake was between the two- and three- e.g., if the year-old size, we designated it as a 2-year-old. Snakes too large for age determination, whether recaptured or not, were pooled as older adults (>6 years old). Yearly individual length and weight changes are based on successive spring or successive au- VERTEBRATE ECOLOGY AND SYSTEMATIC S 15 tumn captures. Annual growth increments thus include one intervening period of hibernation. Weight losses during winter dormancy did not 120 from year to year so both the spring-to-spring and autumn-to-autumn intervals used for determining annual growth rates differ significantly are considered equivalent. Proportional annual increases or decreases in SVL or weight were calculated as the 100 Y = -100 80 2 (females amount increased or decreased 93 X ) during the year divided by the initial size at the beginning of the year. For example, if a 1 -yearold male increased from 31.9 to 48.9 g (an absolute increase of 17.0 g/yr), the proportional increase would be 17.0 -5- 31.9 = 80 0.533/yr, or 53%. I Survival rates were measured over two major periods in the annual cycle of Coluber at area M: 60 winter period of hibernation and (2) the year. Like growth rate calculations, annual (1) the full survival rates include one intervening winter period and were calculated from spring-to-spring Y= -82 65 / o or autumn-to-autumn capture records. Population size estimates based on capturerecapture were calculated using the Jolly-Seber ( 2 males 57 X ) 40 method following Caughley (1977) and Krebs (1978). Eight censuses at den provided data for the Jolly-Seber analysis over three years (1970-1972). Snakes recaptured following their movement to a different den of complex were included in the tabulations as were den individuals that were experimentally displaced from that den in autumn 1971 (cf. Brown and Parker 1976a). Thus, bias due to these factors was eliminated. Population sizes were calculated separately for juveniles (both sexes combined) and for yearling and older (> 1 year inclusive) males and females. by least squares. At the time of year these data were obtained, females were nonreproductive. methods in this paper follow Sokal 969) and Woolf ( 1 968). Mean values are followed by ± one standard error of the mean P< (SE) with the extremes in parentheses. female regression coefficients, indicating stochastic - M % ,^° °8V u ° 20 M 40 60 50 M SNOUT-VENT LENGTH 70 (cm) Fig. 1. Relationship between snout-vent length (X) and body weight (Y) in an autumn sample of 73 c5<3 (solid circles) and 72 99 (open circles) Coluber constrictor mormon in northern Utah. Size records were cho- random from September and October 969— 1972 samples at den M. Regression lines were fitted sen at 1 Statistical and Rohlf ( 1 was obtained between the male and 0.01) statis- divergence between the This resulted from large females tically significant slope Results Sexual Dimorphism. — Weights of 73 male and 72 female snakes > year old collected during 1 autumns of 1969-1972 were regressed on snout-vent lengths (Fig. 1). There was a highly = 0.96, 22 r= 0.97; P < 0.01) significant (<?<5 r correlation between body length and weight in both sexes; 93% (66) and 95% (22) of the variation the in body weight is explained by linear regression on body length. A significant difference (/ = 3.6, two lines. weighing more than large males. Mean length) relative tail length (percentage of total was greater in significantly (r = 12.1, /><0.01) males (26.76 ± 0.11%. range 23.9= 73) than in females (25.07 ± 0.09%. 28.9%, N range 23.1-27.0%. N= 72). Although cally significant, this distinction in visual sex statisti- could not be used determination. Size of Snakes of Known Age.— Snout-vent lengths and weights of 1236 Coluber of known SPECIAL PUBLICATION-MUSEUM OF 16 Table J 1. Sizes of Coluber constrictor = juvenile, recorded 1-7 = in spring at mormon of known method of designating emergence from hibernation. Mean ± years (see text for 1969-1972. Ages: H = hatchling (not sexed), Except for hatchlings, all measurements were SE, sample size and extremes in parentheses. age, age). 1 NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS AGE ( yr 17 ) Growth in snout-vent length of Coluber constrictor mormon, 1969-1972. Data for hatchlings (H) include both 66 and 99 (sexes combined). Except for hatchlings. all records pertain to spring = sample mean; solid rectangles = 95% confidence limits for population mean: open only. Horizontal lines = ± standard deviation (SD); vertical line = range. Means of 66 connected by dashed line, 99 by rectangles solid line. Sample sizes indicated above each bar diagram. Fig. 2. and juveniles (J) 1 involved a decrease in weight during the interval (Table 2). Proportions of weight loss records for 78 males (95.5%) and 1 55 females (93.5%) were 1 similar, as were proportions for both sexes over four winters (1969-1970, 86.8%; 1970-1971, 92.5%; 1971-1972, 96.3%; 1972-1973, 97.2%). Females, averaging larger in size than males, lost = 3.5, P < 0.0 1 ) more weight than significantly (/ did males (Table 2). Eleven of 13 juveniles lost an average of 0.67 ±0.13 (0.3-1.6) g/snake. On a relative basis, juveniles lost 7.7% of their au- tumn body weights, not significantly more (F = 1.44, P > 0.05) than males (7.4%) and females (7.3%). Analysis of variance also showed that there were no significant between-years differences in weight loss in males (absolute F = 1.15, P> 0.05; relative females (absolute F= 2.49, P> F= 0.30, P > 0.05) and in F= 0.66, P > 0.05; relative 0.05). Annual Age- specific Growth. — Absolute and relative yearly rates of increase in snout-vent and weight are summarized in Tables 3As no recapture records were available to mea- length 6. sure growth in the season of hatching directly, growth calculations were based on differences between means of hatchling (mid-August) and juvenile (October) sizes. Young 16.0% in SVL and 36.3% 1.5-month interval. in racers increased weight during this Weight increase during the first year was rapid. Males increased an average of 225% and females 223% of initial juvenile weights (3.2-fold inOne-year-old females nearly doubled their weight again in their second year (mean creases). proportional increase 82%), achieving a growth 1 -year-old males. By rate 1.2 times greater than the time females reached an age of 3 years and most became sexually mature, they were 1.3 times heavier than an average 3-year-old male and 1 1 times heavier than the average hatchling. The SPECIAL PUBLICATION-MUSEUM OF 18 NATURAL HISTORY 100 80 a< 60 H I O UJ $ 40 20 -i H AGE Fig. 3. Growth in r- 1 4 3 2 ( yr weight of Coluber constrictor mormon, T" -r- 5 6 ) 1969-1972. Explanation and symbols as in Fig. 2. first full growing season was clearly the period of greatest rate of increase; thereafter growth rates declined steadily with age. Both absolute and growth rates in males were less than corresponding rates in females at all ages. Unequal growth rates between years are in- relative dicated (Tables 3-6). We compared all age-specific absolute rates of growth recorded in 1970 and 1971 against those in 1972. Five age intervals from 1-2 yr to 5-6 yr for each sex were (P < 0.05) between-years differences appeared in seven of 10 Mests of SVL increases and in eight of 10 Mests of weight increases. In particular. Coluber that were <5 years old grew significantly faster in 1970 and 1971 tested. Significant proportion of snakes gained weight during 1971 (85% of 142 records) than in either 1970 (70% of 46 records) or 1972 (44% of 162 records). Proportions of total annual rainfall in the 5-month were 40% and 19% (1972). Amounts of weight gained and lost are shown in Table 7. Individual weight gains were significantly greater in 1970-1971 than in 1972 in males (t = 4.75, P < 0.00 1 and females (/ = 3.40, P < 0.0 ). Individual weight losses were not significantly greater in 1972 than in 1970-1971 in males (t = 0.30, 0.80 > P > 0.70) and females activity period (May-Sept.) each year (1970). 39% (1971), ) 1 (t = 1.84, 0.10 > P > 0.05). Female Reproductive Cycle. — Females con- tained enlarged preovulatory oocytes in late April. than in 1972. Weight Changes. — To include records of individuals of unknown age not analyzed above, proportions of all large snakes May, and early June (Brown, unpubl. data). At other times of the year ovaries were small and contained no enlarging oocytes. Available evi- (males mostly >6 years old, females >4 years old) that increased in weight each year were compared to each year's rainfall (Fig. 4). A greater dence indicates production of a single clutch of eggs/2 per year in northern Utah. Clutch Size. — Clutch size was determined from Annual Variation in VERTEBRATE ECOLOGY AND SYSTEMATICA Winter weight losses in Coluber constrictor mormon > 1 year old. Absolute last autumn capture and first spring capture at den M; relative loss between weight SE. sample size and extremes in parentheses. weight. Mean ± Table 2. 19 loss is is difference in percentage of body autumn 1 Absolute weight loss (g Relali\e weight loss snake) l".,i " Year 1969-1970 4.04 ± 0.56(15) 1970-1971 4.72 ± 0.34(56) 1972-1973 4.10 4.23 ± 0.36(60) 0.33 (49) a sample of 43 reproductive females (Fig. 5). At the mean number of eggs/2 was 5.78 ± area M mode 5 (N = 18). ± 0.55(60) 7.53 7.40 For these females, a significant (r = 0.53. 0.05 > P > 0.01) linear correlation existed between body size and clutch ± 0.38(49) (1.7-14.2) (0.2-21.9) (0.1-20.7) ± 0.63(35) (0.2-16.9) 6.61 7.56 ± 0.28(145) 5.39 7.16 7.54 ± 0.42(56) (1.0-21.3) (0.3-15.6) ± 0.20(170) ± 0.68(39) 9.17 ± 0.74(18) (3.2-14.3) (0.3-20.1) ± 0.57(43) 5.11 (0.1-15.1) 0.24 (4-8), 7.25 (0.9-10.4) (0.1-15.1) All snakes ± 5.25 (0.3-12.2) ± 0.80(15) (0.9-11.4) 6.09 ± 0.72(35) (0.1-20.7) (0.2-9.5) 1971-1972 6.59 (1.6-12.5) ± 0.34(39) 3.81 ± 0.75(18) 5.11 (0.7-8.6) ± 0.61 (43) (0.5-18.6) ± 0.29(170) (0.2-21.9) 7.33 ± 0.29(145) (0.2-18.6) tween 8-25 July after 9-28 days in an environmental chamber maintained at 29°C. Three gravid females from area collected between 27 June3 July oviposited in the laboratory between 12- M 54.3) mm in length. 18.00 ±0.14 (15.9-20.0) mm in width, and 7.80 ±0.17 (5.9-10.8) Data 15 July. Hatching in the 1971 clutches occurred between 19-27 August, after a mean incubation period of 42.6 (41-44) days at 29°C. Nine area females had enlarged ovarian oocytes between 3-7 June 1972. Four collected between 18-26 June oviposited in the laboratory between 26 June and 9 July. Eggs in three 1972 clutches hatched between 8-23 August after 44-45 days on egg of incubation at 29°C. size; SVL explained 28% of the variation in clutch size. — Measurements of 54 eggs in Size of .Eggs. nine clutches were recorded after oviposition in the laboratory. Eggs averaged 37.78 ± 0.75 (29.2g. size were not recorded in ten additional lab-deposited clutches. Eggs in three of these were weighed indirectly by dividing the female's ovipositional weight loss by her clutch size. Mean egg weights calculated in this manner were 7.6. 7.8, and 9.8 g (overall, 8.4 g/egg, N = 18). demonstrated a significant difference between clutch means of all three measurements of egg size (length F = 20.6, width F = 16.8, weight F = 38.2; P < 0.0 ). A model II analysis of the components of variance showed that relatively more of the total Analysis of variance (model I) 1 variation occurred among clutches (73-86%) than The small- among eggs within clutches ( 1 4-27%). (mean weight 6.2 g/egg) differed from the largest (mean weight 10.6 g/egg) by a mean difference of 4.4 g/egg. There was no significant est clutch correlation between female size (SVL) mean weight of eggs 0.05; N = 9). in her clutch (r = and the 0.19. P> Incubation and Hatching. — Between 27 June July 1971 seven gravid females were col- and 1 lected in area RB. These females oviposited be- M We followed three gravid females with imin 1972 planted radio transmitters at area (Brown and Parker 1976a). Two of these females M oviposited on 21 and 23 June. Eggs of one clutch were excavated 36 days later and were lab-inat 29°C an additional 12-13 days; hatching occurred on 11-12 August after 48 and 49 days incubation. At the second field site a hatch- cubated ling was captured by fencing on 10 August. 50 days after oviposition. The third site was excavated on 6 August, 4 1 days after oviposition. and one freshly-hatched egg was recovered. In 1971. timing of reproduction between areas M RB and (located 65 km apart) was similar. If had oviposited between most females at area 5-15 July 1971 and between 20-30 June 1972, with a probable natural incubation period of 4550 days, most hatching in the field around the communal dens occurred between 20-30 August 1971 and between 10-20 August 1972. Hatching Success. -In 1971 and 1972. 20 fe- M males oviposited in the laboratory. A total of SPECIAL PUBLICATION-MUSEUM OF 20 Table 3. Mean ± Aee 1 NATURAL HISTORY Age-specific growth in snout-vent length of 199 66 Coluber constrictor SE, sample size and extremes in parentheses. mormon during 3 years. VERTEBRATE ECOLOGY AND SYSTEMATICA Table 3. 21 Continued. Proportional increase scar 1970 + 1971 .589 ± 029(2) .502 (.560 -.618) .178 ± 009 ± 006 .124 ± (37) .041 (43) ± 009(16) .030 ± 009 .027 ± 009 .024 (8) .019 (6) ± 004(16) weight prior to ovi- 1 1 ) eraged 30.1 (14.6-48.8) g/snake during the prereproductive interval; the mean rate of weight July 1971 (3 22) and 1972 (2 22) (Table .022 One female oviposited in the laboratory both four females were later recaptured in The M For some area females, additional weight records were obtained in a following year as they were again recaptured emerging from hibernation. on One female 5 Sept. (No. 4. Table 9), weighed 1 14.0 1971 after reproduction. 100.5 g in known reproductive year as in the spring preceding that year are available for three females in Table 8. lowing a Proportional increase year .041 (3) ± .005(16) (.000-.064) Continued. ± (29) spring 1972. and 109.2 g in spring 1973. Data showing very similar weights in the spring fol- After ovipositing in the laboratory between 114 July, parturient females were released in the .644 ± .004 (.000-.085) weights (Table 9). Mean postreproductive weight recovery was 0.92 (0.45-1.38) g/day. g gain was 0.97 ± 0.08 g/day. 1970 + 1971 ± .004(29) August and early September, 31-53 days after release. These spent females recovered an average of 53% (24.7-71.1%) of their parturient by subsequent oviposition in the field or laboratory. Recaptures occurred 7-47 (mean 3 days after release during which time these females had increased by an average of 32.6% of their initial body weights. Absolute increases av- 4. .029 (10) 9). .005 (59) (.000-.085) years. cytes or Table .040 (.000 -.053) in ± (.010-. 160) ± .003(21) position were recorded. Ten females captured in spring at emergence and released between 26 April-20 May were later recaptured between 426 June (Table 8). These animals were gravid as determined by palpation of enlarged ovarian oo- field in .077 ± .003(19) ± 005 ± .008(53) (.032-.293) (.000-.057) (.000 -.064) was determined, changes .162 010(16) (.009 -.057) (.000-.085) ± .045(7) (.404-. 72 7) (.010 -.064) (.000 -.127) .042 .527 (.054 -.187) (.029 -.160) .053 ± 059(5) (.404 -.727) (.032 -.293) .090 1970-lv": 1972 19" SPECIAL PUBLICATION-MUSEUM OF 22 Table sample 5. NATURAL HISTORY Age-specific growth in weight of 179 6$ Coluber constrictor si/e and extremes in parentheses. mormon during 3 years. Mean ± 1 SE. VERTEBRATE ECOLOGY AND SYSTEMATICS TABLE 5. 23 Continued. Proportional increase- year 1970 + 1971 2.153 ± .195(2) (1.958- -2.348) .647 ± .035(37) (.137- -1.081) .283 ± .016(43) (.124- -.579) .171 ± .021 (16) (.003- -.409) .136 ± .022(8) (.030- -.223) .099 ± .018(9) (.019- -.188) 1970-1972 19" 1.744 ± .177(5) .384 ± .047(16) .568 .243 (.004- -.314) ± 017(56) (.004- .579) ± .014(12) .123 (.002- -.157) ± 017(28) (.002- .409) ± .016(15) .088 (.010- -.232) .046 ± 033(53) (.076- 1.081) .110 ± .025(13) .062 147(7) (1.316- 2.363) (.076- -.756) .058 - 1.882 (1.316- -2.363) ± 015(23) (.010- .232) ± .004(3) .086 (.038- -.053) ± 015(12) (.019- .188) dens between 1969 and 1972 (Table 10) constitute a direct census which was influenced by following egg laying (weight lost at oviposition/body weight prior to oviposition). Mean gravid weight of the 12 females was 111.6 ± 4.47 (82.3-137.2) g; mean parturient diately (1) the snakes' fidelity to the several communal hibernacula and (2) the effectiveness of our encircling fences in capturing them. We believe both RCM weight was 62.7 ± 2.73 (43. 1-77.0) g. Mean was 43.8 ± 1.03% (37.9-49.2%). Females were possible sources of error were minimal, assuring of our direct counts of individ- weighed an average of 6 (1-13) days prior to oviposition during which time some weight loss a high reliability would be expected through dehydration (although water was supplied ad libitum), so the measured relative weight loss due to oviposition was probably slightly higher than actual losses had weighing immediately preceded oviposition. However, there was no significant correlation between weighing interval and percentage weight marked snakes of all ages were caught loss (r = 0.30. P> uals captured. Nonetheless, each year Structure"). 6. all snakes one year old or 26-29% at dens M, and 5 were captures of unmarked individuals (Table 1). In 1972 at dens M and S3, new captures comprised older in 1971, 1 , 1 22% and 33% of the samples, respectively. Mark-recapture population estimates using the 0.05). Jolly-Seber Estimates of Population Size. — Total numbers of all individual racers captured at the various Table Among some un"Age (see method resulted in population esti- mates for males and females > slightly higher than the actual 1 year old only number of snakes Continued. Proportional increase >car 1970 + 1971 2.653 ± .384(3) (.022 -3.348) 1.000 ± .038(39) (.399 -1.510) .442 ± .028(38) (.159 -.873) .264 ± .047 (20) (.010 -.789) .160 ± .019(13) (.049 -.297) ± .027(8) (.010 -.247) .106 19^0-1972 1972 2.055 ± .252(7) (1.156- -3.313) .675 ± .076(10) (.281- -1.061) .219 ± .037(18) (.052- -.754) .098 ± .021 (13) (.001- -.285) .078 ± .022 (8) (.003- -.219) .081 ± .044 (2) (.037- -.124) 2.235 ± .219(10) (1.156- -3.348) .934 ± .038 (49) (.281- -1.510) .370 ± .026 (56) (.052- -.873) .199 ± .032 (33) (.001- -.789) .129 ± .017(21) (.003- -.297) .101 ± .022(10) (.010- -.247) SPECIAL PUBLICATION-MUSEUM OF 24 NATURAL HISTORY M complex, 1.8 km from (1976a) (1.6 km from S complex) as radii of circular areas, and assuming uniform dispersal in all directions from each _ 40 E wo den complex, areas occupied by the Coluber populations were 804 ha at M complex and 1017 ha at S complex. In autumn 1971 and spring 1972, when sampling was most complete, 528 Coluber 30 2 20 2 M complex, weighing 29.728 kg were recorded at and in spring 1972, 271 Coluber weighing 15.795 kg were recorded at den S-3 in S complex (Table 10 M and 10). Population and biomass densities at S complexes in 1971 were 0.66 and 0.27 snakes/ ha and 37 and 16 g/ha, respectively. Population ^ o w o ? 1.0 census data were adjusted by calculated difference factors (Table 12) to estimate total populations. Adjusted population densities were 0.79 .8 snakes/ha ct (Table at 13). are located ca. 875 m 600-ha region of overlap, encomof the S dispersal area and 75% of The two den complexes 84 78 apart. Thus, a .4 passing additive densities. The overlap densities were 0.78 snakes/ha and 39.6 g/ha. These are the most representative estimates - .2 60% M, could contain the o CL o complex and 0.32 snakes/ha M .6 O M at S complex. Adjusted biomass densities were 39.8 complex and 16.7 g/ha at S complex g/ha at of these parameters under the conditions of sam- q: pling 970 972 97 Annual proportions of Coluber Fics. 4. Population Changes. constrictor mormon that increased in weight in three successive years (1970-1972) compared to yearly rainfall. Upper histogram shows total annual rainfall (unshaded), May- and June-Aug. total (hatched) Grantsville, Utah. Weight change records Sept, total (stippled), recorded at (lower histogram) are for 1 79 <53 > 50.0 g (hatched bars) and 1 7 1 99 > 60.0 g (stippled bars) initial body weight; sample sizes above each bar. M in 1970 and 1971 (Table 12). was greater for juveniles, reflecting the greater difficulty of capturing them and their higher mortality rate. These factors lowered recapture proportions and tended to raise the es- caught The at den disparity timated population of juveniles relatively more than the estimates for adults. The relatively low "difference factors" for older snakes indicated that the sampling technique effectively captured a high proportion of the adult population. Population Density. — Using maximum persal distances recorded by and assumptions employed in the calcu- lations. dis- Brown and Parker — Population changes during our sampling are shown in Table 14. The 16.5% in 1970(den M), and by 16.7% (den M) and 18.9% (dens 1 and 5) in 1971. The population increases noted in 1 970 and 97 1 were not sustained during 1 972 racer population increased by 1 22.2% (den M) and 20.3% (den S3). Sex Ratio. — For each den and sampling period, proportions of total numbers of males (822) and females (725) >1 year old were 0.531 and 0.469, respectively. In all but two sampling periods, males outnumbered females (Table 10). Sex ratios were never significantly different from when the populations declined by expectation as tested by chi-square for any den or sampling period. Sex ratio at birth was determined by eversion of hemipenes after injection for 18 lab-reared hatchlings randomly preserved in 1971. There were 9 males and 9 females in this sample. A sample of 1 7 juveniles in autumn 1972 and spring 1973 was sexed. There were 10 males and 7 fea 1 : 1 males in this sample 2 (x = 0.24, 0.70 > P > VERTEBRATE ECOLOGY AND SYSTEMATICS Annual absolute weight changes of 350 Coluber > 50.0 g and 171 29 > 60.0 g that gained or sample size and extremes in parentheses. Table 7. are for 179 36 constrictor lost mormon 25 in three successive years. weight during a yearly interval. Mean ± Data 1 SE. SPECIAL PUBLICATION-MUSEUM OF 26 NATURAL HISTORY 90- 6 o80 I Io z D A A A UJ 70 z > o o i IZ> O 60z 50 co UJ -10 o o 5 -IT 4 Li r 6 - ^t— —i— —r 7 8 9 oo 10 CLUTCH SIZE Fig. 5. Clutch sizes-9 body size (SVL) relationship for Coluber constrictor mormon in 1971 and 1972. Circles represent laboratory oviposition records; triangles represent enlarged preovulatory oocytes. Solid symbols in scatter diagram (upper portion of figure) and corresponding shaded bars in histogram (lower portion of figure) from area M. Open circles and triangles (upper) and unshaded bars (lower) represent records from area RB, and squares area SLC. The regression line shows clutch size (Y) vs. snout-vent length (X) for area females and is described by the equation Y = -0.56 + 0.10X. are records M VERTEBRATE ECOLOGY AND SYSTEMATICS Table 8. Prereproductive (late spring-early summer) weight increases in 10 gravid female Coluber constrictor mormon at area M, 1971 and 1972. O = ovarian = oviducal eggs. Last two snakes were reoocytes, E captured prior to being tracked by telemetry. 27 SPECIAL PUBLICATION-MUSEUM OF 28 NATURAL HISTORY 10. Numbers, live-weight biomass, and sex ratios of Coluber constrictor mormon captured at five hibernacula in complex (dens M, 1,2, 3, 5) 1969-1972 and at one hibernaculum in S complex (den S3) in is total different individuals for both autumn and spring sampling; for 1972. Number of snakes at den all other hibernacula totals are different individuals in spring only. Weights are for animals in spring unless > 1 year old, juveniles (J) include both only autumn capture was recorded. Males and females include all ages sexes < 1 year old (see text). Table M M VERTEBRATE ECOLOGY AND SYSTEMATICA 12. Comparison of sampled and estimated population sizes of Coluber constrictor mormon at den in 970 and 1971. Difference factor is a proportion Table M 1 calculated by dividing the Jolly-Seber population estimate by the actual number of snakes caught. 29 Table 13. Population and biomass densities of Coluber constrictor mormon in 1971. Difference factors used to adjust population sizes and total weights were calculated from data in Table 12. Population density Year Age 30 SPECIAL PUBLICATION- MUSEUM OF NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICA FEMALE 95(2) 31 SPECIAL PUBLICATION-MUSEUM OF 32 NATURAL HISTORY MALE FEMALE 6+ c 5 (161) 4 (209) i 7•2^^^^^^^^x^ *•'•'•'^'^'•'•'''^'^'J''-*-'-'-** 3 2 1 -i ——————————— i i i i i i i i i < i " J " ' 6+ 5 (380) 4 (419) pyyKTgggg?W5 WW^T?TC??WWTCTCOTWB ' 3 2 J 1 — —————— ——— — -i i i i i i i i i i i J r <~ i i ——— —— i i i < 971 > ' i ' 6* 5 (187) (199) 4 1 3 2 1 J "I 1 1 1 1 1 1 1 1 1 1 1 1 1 ' ' i ——— —— ————— — — i i S i i i i i i i ' '970 i i i" VERTEBRATE ECOLOGY AND SYSTEMATICS Overwintering survivorship of Coluber 1 year old estimated by recapture proportions at den M. Table 16. constrictor mormon > Table tor 17. 33 Annual survivorship of Coluber mormon estimated by dens(M, 1,5) in constric- recapture proportions at three 1970 and 1971 (years combined) and SPECIAL PUBLICATION-MUSEUM OF 34 ^ .8 • '/ a* *" .7 O A^ I- 2 .6 / /A' / / a: id W .4 .3 / NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS Table 19. Schedule of age-specific survivorship and fecundity of female Coluber constrictor mormon, x = = age-specific survival rate; l x = surviage (years): P x = number of female eggs produced vorship to age x: m each year by a female of age x; R„ = net reproductive rate. See text for assumptions and for adjustment factors of m, schedule. v Unad- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Px lx .170 .736 .815 .767 .875 .769 .787 .787 .787 .787 .787 .787 .787 .787 .787 .787 .000 .170 .125 .102 .078 .068 .053 .041 .033 .026 .020 .016 .013 .010 .008 .006 justed Adjusted ITU m. 2.50 2.60 2.75 2.85 2.90 3.00 3.10 3.20 3.30 3.40 3.50 3.50 3.50 3.50 0.18 .023 1.84 .188 .178 2.28 2.36 2.41 2.49 2.57 2.66 2.74 2.82 2.91 2.91 2.91 2.91 R = I v m x .160 .128 .102 .085 .069 .055 .045 .038 .029 .023 .017 1.187 accrue to females during their increased movements in search of oviposition sites (Brown and Parker 1976a; Parker and Brown 1972. 1980) so it seems reasonable to suppose that a higher mortality may occur during the shorter prereproductive phase than during the longer postreproductive phase. In both a "good" and a "bad" year (1971 and 1972. respectively), three-yearold females had lower survival rates than either of the adjacent age classes. As most females matured and presumably began reproduction for the first time at age 3, our data on higher mortality in 3-year-old females support this argument. To our knowledge, the only other attempt to measure the components of mortality in a snake population is that of Feaver 1977). Adult female ( N. sipedon suffered their heaviest losses (50% of the total annual mortality) in summer, adult males in spring (47% of the total). In each sex, mortality was higher in the season of most active reproductive behavior, i.e., spring mating activity in males, summer gestation and parturition in females (Feaver 1977). Of the total annual mor- N. sipedon, 32% occurred over the winThis value is almost identical to our data (33% of the annual mortality was overwintering tality in ter. 35 Table 20. Age-specific body size and fecundity in two populations of Coluber constrictor. Data are for C. flaviventris in Kansas (Fitch 1963) and C. c. mormon < Utah (present study). numeral = years. in H = hatchling, J = juvenile, SPECIAL PUBLICATION-MUSEUM OF 36 NATURAL HISTORY Table 21. Comparison of the major life history traits in two populations of Coluber constrictor. Where possible a measured value is given for each trait. Data from Fitch (1963) for Kansas (C. C. flaviventris) and present study for Utah (C. c. mormon). Life history Population density* Body Utah Kansas parameter size** higher (5.0/ha) lower (0.8/ha) larger (66 123 g) smaller (66 56 g) 69 g) (25 155 g) Growth (92 slower faster rate Reproduction Sexual maturity 1 year (83) 1 year (<3<5) 3 years (22) 3 years (22) larger (1 1.7 eggs) smaller (5.8 eggs) smaller (5.7 g) larger (7.8 g) Hatchling size smaller (4.2 g) larger (6.0 g) Relative clutch massf lower (.40) higher (.62) Clutch size Egg size Demography Age distribution younger (72% 1-3 (28% 4+ older yr) yr) (52% 1-3 yr) (48% 4+ yr) Relative contribution to R by female of age x higher over ages 2-6; peak at age 3 (19.3%) lower over ages 2-6; peak at age 3 (15.8%) Juvenile survivorship higher (31 %/yr) lower (23%/yr) Adult female survivorship lower (62%/yr) higher (79%/yr) Generation time shorter (5.1 yr) longer (6.9 yr) * Value for Kansas from Turner (1977). Mean body weight, random samples > 1 year old C. c. flaviventris (N = 50 each mormon (136 <5<5, 114 22, den S3, spring 1972). t Mean clutch weight/mean body weight of non-gravid 22; mean RCM value for ** somewhat higher relative contribution to tween ages 2-6 years (Fig. 1 2). The R be- distributions indicate that 3-year-old females contribute the most to R in each population. Utah racers have late trends. apparent that there are several prominent reproductive and demographic differences bein Kansas and Utah superimposed on a basic plan of biological similarities. Both subspecies of C. constrictor exhibit an identical growth pattern in which females mature later and tween racers grow larger than males. Feaver (1977) placed C. Rhabdophis tigrinus, Thamnophis and Nerodia sipedon in this group as con- constrictor, butleri, trasted with Crotalus viridis, Agkistrodon contortrix, grow and Elaphe quadrivirgata in which males There are several impor- to the larger size. (2-6+ years). maturity, high adult survivorship, small and large young as contrasted to the Utah racers have a somewhat longer estimated generation time, suggesting a less frequent It is ages c. and behavioral differences between the two groups of snakes (cf. Shine 1978); generally the last group of species tends to show clutches, turnover of the population. An overall summary of the major life history comparisons is presented in Table 21. 5 1963) and C. tant reproductive higher adult survival rates than do Kansas racers. Life tables developed for each population show that sex; Fitch which shows the opposite one could apply a "K-selected" label to the second group and an "r-selected" label to the first. Whereas such a comparison may help to visualize the broad first species group Viewed at this level, it is less capable of showing differences an intraspecific comparison. C. c. flaviventris and C. c. mormon each seems to possess some "K" and some "r" attributes (cf. Pianka 1970; strategies, in Stearns 1976). Our data on survivorship of C. c. mormon show that there were considerable between-years on survival in adult females and lesser males when a dry year (1972) followed wetter, more favorable years (1970 and 1971). Juvenile survivorship, on the other hand, was not as strongly reduced in 1972 from 1970— 1971 levels. Under this regime (with adult mor- effects effects in adult VERTEBRATE ECOLOGY AND SYSTEMATICS 4 3 5 AGE (YR) Fig. 10. Mean age-specific relative clutch mass in two populations of Coluber constrictor calculated as the clutch weight as a proportion of the mean nongravid 9 body weight. Mean clutch weight was calculated from clutch size and mean weight of eggs for C. c.flaviventris in Kansas (K; Fitch 1963) and C. c. mormon in Utah (U; present study). tality variable), a stable vor such environment should fa- fewer young, longer life span, smaller reproductive effort, and slower develtraits as opment (Stearns 1976). It is not clear from the data available whether C. c. flaviventris has a more variable adult or juvenile survivorship and which environment, Kansas or Utah, is the more "stable." The Kansas habitat appears to be trophically more diverse. Insects (grasshoppers, by C. c. mormon almost ex- crickets) are eaten clusively, KANSAS n ' 242 whereas C. c. flaviventris takes a mod- 37 SPECIAL PUBLICATION-MUSEUM OF 38 these data with those of a population of C. c. flaviventris in Kansas studied by Fitch (1963). Racers were captured at their dens with encircling screen fences each autumn and spring NATURAL HISTORY In all samples of snakes >1 year old, comprised 53% and females 47% of the lation. Age structure favored younger (<5 animals which comprised 62-76% of the males popuyears) popu- 1969-1973. Snakes were measured, lation in different years. Large proportions of weighed, sexed, and permanently marked by scale clipping. The age of each individual was deter- 1-year-olds in 1969 (27.4%) and juveniles in 1971 (11.1%) indicated that 1 968 and 1971 were years of high productivity. In contrast. 972 was a poor year for recruitment of juveniles (4.6% of the between mined by comparing its size to confidence inand weight of recaptured knownage snakes. A total of 1046 racers was captured tervals for length 1694 times. Males became sexually mature at an age of <13.5 months. In females, 8% of 2-year-olds, 77% of 3-year-olds, and 90% of 4-year-olds were considered mature. Mean weight of hatchlings was 6.0 g, and juveniles 8.5 g. At an age of 1 year, females = 28.4 weighed significantly more than males (x = 27.0 g) and females continued to be significantly larger in both snout(.v g) vent length and weight at all ages. Body weight declined in 95% of the snakes over the winter; losses averaged 7.4% of initial autumn weight in both sexes. In 1971. a year of relatively high rainfall. 85% of racers gained weight over the summer, whereas in 1972. a dry year, only 44% gained weight. Age-specific growth rates were significantly higher than in in 1970 and 1971 1972. Females produced a single clutch per year avand eraging 5.8 eggs. Eggs averaged 38 x 18 mm 7.8 g. ratio Mean clutch weight/female body weight was 44%. Oviposition occurred in late June through early July: hatching occurred in mid to late August after an incubation period of 45-50 days. Hatching success was 92%. Sex ratio at 1 population). Overwintering survival rates averaged 93% in both sexes. Annual survivorship in juveniles was 23%. First year survival (egg to age 1) was estimated to be 7%. Adult survivorship in favorable years was 78% in males and 79% in females. In an unfavorable year adult survivorship was 62% in males and 56% in females. Two other species of sympatric colubrid snakes in Utah had annual survival rates of around 80% per year. In 1 contrast, literature reports for ing (7%), prereproductive, and postreproductive suggest that prereproductive mormortality. in females from exposure to adis higher tality We ditional risks associated with egg laying. A life table for C. c. mormon calculated using the combined female survival randomly-collected females in early parturient. dens yielded 528 racers. The largest number recorded at a single den in one season (spring 1972) was 271 snakes. UsIn 1971, sampling at six maximum dispersal distances and assuming uniform radial movement pattern from the dens, population density was 0.8 snakes/ha and biomass density was 40 g/ha. The population at den increased from 139 to 189 individuals 8%/yr) over two successive favorable years (1970, 1971) and declined to 147 individuals (21% decrease) in an unfavorable year (1972). ing a M 1970 and . 1 Three-year-old females contributed the highest proportion (15.8%) to R Compared to the life history of C. c. flaviven. tris in Kansas. C. c. mormon in Utah is distinct lower growth rates and ways: smaller adult size; (2) lower age-specific fecundity; (3) larger eggs summer, 88% were gravid or rate in 1971 showed a net reproductive rate (R ) of 1 87, a value indicating an increasing population. summer averaged ca. g/day as did postreproductive weight recovery in late summer. Among species of col- 50% per year. The 21% annual mortality in C. mormon may be partitioned into overwinter- in the following 1 1 c. hatching did not differ significantly from 1:1. Weight increases in prereproductive females in early 1 ubrids indicate an average survivorship of ca. ( 1 ) and hatchlings; (4) higher body weight ratio; (5) lower and higher adult survivorsurvivorship juvenile ship; and (6) older age distribution and longer generation time. These life history traits appear to fit some "r" and some *'K" strategies in each population. Without more detailed work on resource levels, environmental stability, and preclutch weight/female dation. we caution against simplistic interpretations in contrasting the two populations. Acknowledgments ( 1 At the University of Utah our studies were supported by American Museum of Natural His- VERTEBRATE ECOLOGY AND SYSTEMATICS Memorial Fund) tory (Theodore Roosevelt 39 Niche dimensions and resource partitioning in a Great Basin desert snake communitv. 1982. Support Grant FR070902, a Graduate Research Fellowship to Brown, and an NDEA Fellowship to Parker. At tological Communities. U.S. Fish life Service. Wildl. Res. Rep. 13. Skidmore College, Denton W. Crocker, Chairman, Biology Department, provided an opportunity for portions of this work to be completed. Eric J. Weller, Dean of the Faculty at Skidmore Carpenter, C. C. 1952. Comparative ecology of the common garter snake (Thamnophis s. sirtalis), the ribbon snake Thamnophis s. sauritus), and Butler's garter snake (Thamnophis butleri) in mixed grants, a Biomedical Sciences College, allocated numerous ways in the lab and in the field in Utah, we thank George C. Douglass, Richard J. Douglass. Thomas C. Juelson, Arthur C. King. John M. Legler, Grady W. Towns, and Robert M. Winokur. We thank Paul E. Feaver, Henry S. Fitch, Harold Heatwole, Richard Shine, Fredand an anon- script for critically reading the manu- and suggesting improvements. Brown populations. Ecol. Monogr., 22:235-258. Analysis of vertebrate populations. John Wi& Sons. N.Y. 234 p. Clark, D. R.. Jr. 1970. Ecological study of the worm snake. Carphophis vermis (Kennicott). Univ. Kansas Publ. Mus. Nat. Hist.. 19:85-194. 1974. The western ribbon snake {Thamnophis proximus): ecology of a Texas population. 1977. ley Herpetologica, 30:372-379. R.. Jr. and Fleet, R. R. The rough earth snake (Virginia striatula): ecology of a Texas population. Southwest. Clark, D. 1976. Con ant, Nat., 20:467-478. R. A 1975. appreciates the assistance and support of his wife Betsy and children Amy, Lee, and Bonnie, and Parker similarly thanks his wife Beth. Elaine C. Rubenstein photographed some of the figures and Edie Brown competently typed the manuscript. Herpeand Wild239 p. (Ed.). Caughley, G. erick B. Turner, Stephen C. Stearns, ymous reviewer J.. Jr. ( support to allow financial Brown's participation in the 1980 annual herpetology meetings and provided funds for manuscript preparation through a Mellon Foundation Grant for faculty development. For assisting us in Pp. 59-81. In Scott. N. field guide to reptiles and amphibians of and central North America. 2nd Houghton Mifflin Co.. Boston. eastern Deevev, ed. E. S. 1947. Life tables for natural populations of animals. Q. Rev. Biol., 22:283-314. Etheridge, R. E. The southern range of the 1952. racer Coluber con- with remarks on the Guatemalan species Coluber ortenburgeri Stuart. Copeia, 1952:189-190. strictor stejnegerianus (Cope), Literature Cited Feaver, AUFFENBERG, W. 1955. A reconsideration of the racer. Coluber con- strictor, in P. E. 1977. Zool., 2:89-155. Blanchard, F. N., Gilreath, M. R. and Blanchard, F. C. 1979. The eastern ring-neck snake (Diadophis punctatus edwardsii) in northern Michigan (Reptilia, Serpentes, Colubridae). J. Herpetol., 13:377-402. Branson, B. A. and Baker, E. An ecological study of the queen snake. Re1974. C gina septemvittata (Say) in Kentucky. Tulane Stud. Zool. Bot., 18:153-171. Brown, W. 1973. Fitch, H. S. 1949. Study of snake populations in central California. Am. Midi. Nat., 41:513-579. 960. Autecology of the copperhead. Univ. Kansas Publ. Mus. Nat. Hist.. 13:85-288. 1963. Natural history of the racer. Coluber constrictor. Univ. Kansas Publ. Mus. Nat. Hist., 15:351-468. 1965. An ecological study of the garter snake. 1 S. Ecology of the racer. mormon Coluber constrictor (Serpentes, Colubridae). in a cold 1975. temperate desert in northern Utah. Ph.D. Thesis. Univ. Utah, Salt Lake City. 208 p. Brown, W. 1976a. S. and Parker, W. S. Movement ecology of Coluber constrictor near communal hibernacula. Copeia, 1976: 1976b. A 225-242. ventral scale clipping system for permanently marking snakes (Reptilia, Serpentes). J. Herpetol.. 10:247-249. The demography of a Michigan population of Natrix sipedon with discussions of ophidian growth and reproduction. Ph.D. Thesis. Univ. Michigan. Ann Arbor. 131 p. eastern United States. Tulane Stud. Thamnophis sirtalis. Univ. Kansas Publ. Mus. Nat. Hist.. 15:493-564. A demographic study of the ringneck snake (Diadophis punctatus) in Kansas. Univ. Kansas Mus. Nat. Hist. Misc. Publ.. 62:1— 53. Fitch, H. S., Brown, W. S. and Parker, W. S. 198 1. Coluber mormon, a species distinct from C. constrictor. Trans. Kansas Acad. Sci., 84: 96203. 1 Gregory, 1977. P. T. Life-history parameters of the red-sided NATURAL HISTORY SPECIAL PUBLICATION-MUSEUM OF 40 garter snake ( Thamnophis sirtalis parietalis) in an extreme environment, the Interlake region of Manitoba. Natl. Mus. Canada Publ. Spellerberg, Zool.. 13:1-44. Hall, R. Ecological observations on Graham's watersnake {Regina grahami Baird and Girard). Am. Midi. Nat.. 81:156-163. 1969. Hirth, H. 1 Stearns, J. F. Weight changes and mortality of three species of snakes during hibernation. Herpetologica, 966. 22:8-12. Hirth, H. 1 968. F. and King, A. C. Biomass densitites of snakes 1 Univ. Press, Krebs, C. New Haven, Conn. 260 1 northern Utah. Herpetologica, 29:319-326. Mortality and weight changes of Great Basin rattlesnakes (Crotalus viridis) at a hibernaculum in northern Utah. Herpetologica, 30: Comparative ecology of two colubrid snakes, Masticophis t. taeniatus and Pituophis mel- 980. anoleucus deserticola, in northern LHah. Biol. Geol. No. Milwaukee Public Mus. Publ. 7:1-104. PlANKA, E. R. 1970. On r and K selection. Am. Nat., 104:592597. Platt, D. R. 1969. Natural history of the hognose snakes Heterodon platyrhinos and Heterodon nasicus. Univ. Kansas Publ. Mus. Nat. Hist.. 1 8:253420. Prestt, 1971. I. An ecological study of the viper, Viperabesouthern Britain. J. Zool. (London), rus, in 164:373-418. Shine, R. 978. Sexual size dimorphism and male combat in snakes. Oecologia (Bed.), 33:269-277. Sokal, R. R. and Rohlf, F. J. 1 1969. 1 A 960. ia: 1977. Species composition and population changes in two complexes of snake hibernacula in 974. 2:71-86. Ecology, maturation, and reproduction of Thamnophis sauritus proximus. Ecology, 38: 1957. Turner, p. S. 234-239. 1 phis). J. Herpetol., Tinkle, D. W. Telemetric study of movements and oviposition of two female Masticophis t. taeniatus. Copeia, 1972:892-895. 973. Some observations on the natural history of two Oregon garter snakes (genus Thamno- 69-77. Row, New York. 678 S. and Brown, W. 1 51:3-47. Biol., 279 p. Stewart, G. R. J. Parker, W. 1972. C. S. Rev. p. Ecology: the experimental analysis of distribution and abundance. 2nd ed. Harper and 1978. E. Stebbins, R. C. 1966. A field guide to western reptiles and amphibians. Houghton Mifflin Co.. Boston. cold des- ert of Utah. Herpetologica, 24:333-335. Hl'TC hinson, G. E. 978. An introduction to population ecology. Yale and Phelps, T. Life history tactics: a review of the ideas. Q. 1976. 1968. in the F. I. Biology, general ecology and behaviour of the snake, Coronella austriaca Laurenti. Biol. J. Linn. Soc. (London), 9:133-164. 1977. Biometry: the principles and practice of statistics in biological research. W. H. Freeman Co., San Francisco. 776 p. population of Opheodrys aestivus (ReptilSquamata). Copeia, 1960:29-34. F. B. The dynamics of populations of squamates, crocodilians and rhynchocephalians. Pp. 157-264. In Gans, C, and Tinkle, D. W. Biology of the Reptilia, Vol. 7. Academic Press, New York. (Eds.), Vial, J. 1977. L., Berger, T. J. and McWilliams, W. T., Jr. Quantitative demography of copperheads, Agkistrodon contortrix (Serpentes, Viperidae). Res. Popul. Ecol., 18:223-234. Viitanen, P. Hibernation and seasonal movements of the 1967. viper, Vipera berus berus (L.), in southern Finland. Ann. Zool. Fenn., 4:472-546. Vitt, L. J. and Congdon, J. D. 1978. Body shape, reproductive effort, and relative clutch mass in lizards: resolution of a paradox. Am. Nat., 112:595-608. Wilbur, H. M., Tinkle, D. W. and Collins, J. P. 1974. Environmental certainty, trophic level, and life history evoluNat., 108:805-817. resource availability in tion. Wilson, Am. L. D. The racer Coluber constrictor (Serpentes: Colubridae) in Louisiana and eastern Texas. Texas J. Sci., 22:67-85. 1978. Coluber constrictor. Cat. Amer. Amphib. Rept., 218.1-218.4. Woodbury, A. M. 1951. Introduction— a ten year study. Pp. 4-14. //; Woodbury. A. M., et al., Symposium: A Snake Den in Tooele County, Utah. Her- 1970. Woolf, 1 968. petologica, 7:1-52. C. M. Principles of biometry. D. Van Nostrand Co., Inc. Princeton, N.J. 359 p. Vertebrate Ecology and Systematica— A Tribute to Henry S Fitch N. L. Zuschlag Edited by R. A. Seigel. L. E. Hunt. J. L. Knight. L. Malaret and 984 Museum of Natural History. The University of Kansas. Lawrence i 1 Growth of Bullsnakes (Pituophis melanoleucus sayi) on a Sand Prairie in South Central Kansas Dwight R. Platt growth records from marked, released and recaptured individuals; and 2) determining size at different ages, usually up to one year old, by an analysis of size frequencies in population sam- Introduction Growth rates of snakes in natural populations have been studied for fifty years. Blanchard and Finster (1933) presented limited data on growth rates of recaptured garter snakes and water snakes. KJauber (1937) derived a growth curve for the ples. My study of growth rates in snakes was part of a larger study of the ecology and population dynamics of sympatric species of snakes on the Sand Prairie Natural History Reservation in western Harvey County in south central Kansas. southern pacific rattlesnake (Crotalus viridis helleri = C. v. oreganus) by analyzing a collection of preserved specimens and pointed out that the The growth of captive snakes may be distorted. Seibert and Hagen ( 1 947) presented growth data for the plains garter snake (Thamnophis radix) and smooth green snake (Opheodrys vernalis) from a mark-recapture study of populations in Illinois. objectives of the present study were: 1) to investigate the range and patterns of variability in growth rates and the effects of prey availability and of age and sex on growth; 2) to compare the growth rates and strategies of five species of snakes living in the same general environment: 3) to compare the two methods (above) of determin- Henry S. Fitch (1949) was a pioneer in the study of free-living snake populations with his field work in central California. His analysis of growth talus viridis and his students growth ing growth rates. This paper reports results based on 709 captures of 471 bullsnakes (Pituophis melanoleucus northern pacific rattlesnake (Crooreganus) has been widely cited. He in the have provided many reports on sayi) (Clark 1970, 1974; Clark and Fleet 1976; Fitch 1960, 1963a, 1963b. 1965, 1975; Fitch and Fleet 1970; Platt growth rates 1969). and on nine young which were hatched in the laboratory. Subsequent papers will describe other species studied and comparative aspects of rates of snakes in natural populations the study. Other notable studies on in free-living populations of colu- Methods Brown and Parker (1984), Carpenter (1 952), Feaver (1977), Fukada (1959, 1960, 1972, 1978), Heyrend and Call (1951), Imler (1945) and Parker and Brown (1980). Growth has previously been studied in brid snakes include those by This study used standard mark-recapture techniques with snakes trapped alive in drift fence traps (Fitch 1951, 1960; Platt 1969). Thirty to area fifty trapping stations were used on a study of 80 acres (32.4 hectares) in 1966. 1967 and early 1968. From late 1968 through 1974. traps at 100 to 120 stations were in operation on the study area and at up to 20 stations on adjacent pastures. A trapping station consisted of a low metal drift fence with a funnel trap fitted under the bullsnake (Pituophis melanoleucus sayi) in Nebraska by Imler (1945), in the pacific gopher snake (P. m. catenifer) in California by Fitch (1949) and in the great basin gopher snake (P. m. deserticola) by Parker and Brown (1980). Growth of several species of elapid and viperid each end. These traps set without bait intercepted movement of the animals. Most measurements were made in the labo- snakes has been studied, including studies by Gibbons (1972), Heyrend and Call (1951). KJauber ( 1 956), Prestt( 1971), Shine (1978, 1980). Volsoe (1944) and Wharton (1966). Most of these investigations have indicated a high degree of ratory. Snout-vent length (SVL) and tail length were measured to the nearest one millimeter with the snake stretched along a metal tape until it relaxed. Weights to the nearest 0.1 gram were individual variability in growth rates. Two methods have been used to study growth summarizing of snakes in natural populations: measured on a 1 ) 41 triple beam balance. The snakes SPECIAL PUBLICATION-MUSEUM OF 42 were released within three to four days at the site of capture. NATURAL HISTORY May to the end of October or early November over a nine year period, 966-1 974. Growth was not continuous throughout the year. On my study 1 Marking was accomplished by clipping or branding subcaudals or ventrals so that each snake was individually recognizable. Individual variations in color pattern and scutellation were also recaptured snakes were individually identified with certainty. Sex- recorded so that almost all ing was accomplished by probing through the vent for the hemipenial sacs and was checked later by body proportions. Food records were it was usually most rapid in early summer but occurred throughout the period of activity and trapping; probably little or no growth oc- area curred during dormancy. Therefore the mean growth rates in this study were calculated using the 184 days from the first of May to the end of vent lengths, total lengths and/or weights as a measure of size. I used snout-vent length (SVL) October as the growth season. Absolute growth rates were calculated as growth increment in SVL per month (30 days) excluding November to 30 April. Relathe period from tive growth rates were calculated as the growth median SVL. increment per month per 100 The median SVL was defined as the midpoint between the lengths at two successive captures or between the mean lengths of two successive measures by the stage of population samples. Some studies (Carpenter 1952; Fukada 1959, 1960, 1978) have used the fecal matter from the intesand forcing stomach contents back up the gullet into the mouth for identification and then obtained by palping tine repalping into the stomach. Previous studies of growth have used snout- because of size. it is one of the Weight is more least variable affected the feeding cycle or the reproductive cycle while total length is affected by partial loss or differ- 1 mm initial length at growth rates. first capture to calculate relative The median length is more similar growth of the tail. Only measurements of live snakes were used in the analysis. Growth rates from recapture records were cal- to the length culated by averaging growth increments during the period between captures for samples of recaptured snakes. Recapture records were used to of the period between captures. Mean values in this paper are usually accompanied by one standard error. Homogeneity of ential calculate growth rates only if three weeks or more had elapsed since the previous capture. Although bullsnake eggs probably hatch on the study area in August, young snakes were not caught in traps of the snake during the growing period. Relative growth rates calculated from median lengths are less affected by the duration variances was tested by an F test. Differences in the central tendency of different samples were tested by Student's / test for samples having similar variances and by the Mann-Whitney U test the variances were heterogeneous (Cox until September. First-year snakes were defined in this study as those caught between September when of their hatching year and the end of the next August. Records of recaptured first-year and old- were calculated by were readily distinguished by plotting the SVLs of snakes with respect to capture Rodent populations, principal prey of bullsnakes, were sampled by the same drift fence traps used to capture snakes and by 100 baited er bullsnakes date. Growth changes in rates were also calculated from the mean SVL of population samples of identified age. Frequency distributions were culated for the lengths of cal- bullsnakes caught in each two-week interval throughout the trapall ping season in each year. First-year snakes were readily identified in these frequency distributions and they did not overlap samples of older snakes in size until they had completed their first full year of growth. Mean snout-vent lengths were 1980). Regression equations of weight et al. on length method (Simpson 1960). mammal live traps (constructed like traps described by Fitch 1950) set in a grid 1 50 meters on a side. Drift fence traps were operated continuously from May through October while baitsmall ed live traps were operated for a few nights per month through the summer (May-August). Rodents caught in drift fence traps were recorded as number caught per 100 trap station days (TSD) while those caught in baited traps were recorded as calculated for these first-year snakes captured in each month and growth rates were calculated from the means of these monthly samples. Snakes were trapped from late April or early Bartlett's number caught per 100 trap nights (TN). Study Area is The Sand Prairie Natural History Reservation 80 acres (32.4 hectares) of prairie on sand dunes VERTEBRATE ECOLOGY AND SYSTEMATIC Tabii Rodents trapped on the Sand 1. Prairie Nat- ural History Reservation in Kansas. Hailed traps Dnl'i fence traps No rodents 100 TN No. rodents No cil Year TSD 100 No. nap Medium station Medium Small of trap days sized sized niehls si/ed sized speeies species (TN) species species i rsD) S T\nn 2. Proportions of bullsnakes (Pituophis melanoleucus) containing recoverable food items in the stomach or residues in the intestines. Chi square tests were run on the differences in proportions of snakes containing food in successive years. N = number of snakes examined. Small Summer (Ma>-Aug.) Per- 1967 1968 1969 1970 1971 1972 1973 1974 7317 9989 0.2 6.4 3.2 9.8 19.775 17.076 19,962 15.272 17.014 17,908 0.3 1.1 managed 0.2 1.9 0.1 0.4 0.3 4.5 2369 1.3 161 3.4 1430 1088 2.0 1061 1518 1.5 1 0.1 0.8 2134 0.2 1.1 1747 as a natural area. Prior to 1966 1967 1968 1969 1970 1.8 0.6 acquisition by Bethel College in 965. it was used as a pasture but was never cultivated. All snakes used in my analyses were captured on this study area or on 1 immediately adjacent pastures. The Sand Prairie Reservation is in a band of wind-blown sand deposits, the Hutchinson Dune Tract of the Great Bend Lowland physiographic division (Frye and Leonard 1952;Schoewe 1949). The upland grass communities on the reservation are dominated by little bluestem (Andropogon scoparius). Forbs and other genera of grasses Triplasis, Aristida and Panicum) also occur. The unliooded lowlands have dense tall grass communities dominated by switchgrass (Panicum ( virgatum), sand bluestem (Andropogon hallii), indiangrass (Sorghastrum avenaceum). eastern gammagrass Tripsacum ( dactyloides) and prairie cordgrass (Spartina pectinata). Thickets of chickasaw plum (Primus august ifolia) are common on the uplands and buttonbush (Cephalanthus occidentalis) and black willow (Sa/ix nigra) in the lowlands. The area is poorly drained and its low depressions between sand dunes are relatively wet, having ponds, shallow marshes or dry ground depending upon the amount of recent rainfall. A more complete description of the study area can be found in Piatt (1973. 1975). Results — Prey of bullsnakes on the Prey Populations. area were predominantly rodents. Trapstudy success (Table provides a rough measure ping of the size and activity of rodent populations. 1 ) Medium-sized rodents, centage contain ing loud Year 1.3 its 43 prairie voles (Microtus 46 43 31 1971 55 65 1972 1973 1974 42 44 15 61% 72% 52% 58% 26% 33% 52% 68% x" 2.43 7 54** 0.66 27.94** 1.72 6.85** 4.46* \utumn (Sept. -i K i i SPECIAL PUBLICATION-MUSEUM OF 44 Table 3. Absolute and relative growth rates of bull- snakes {Pituophis melanoleucus) in Kansas ( 966-1 974) determined from recapture records. N = number of useable recapture records. Mean followed by ± 1 stan1 dard error. NATURAL HISTORY Absolute and relative growth rates of bull4. snakes (Pituophis melanoleucus) older than one year in Kansas determined from recapture records. The probable ages are based on size (see text). N = number of useable recapture records. Means followed by ± stan- Table 1 VERTEBRATE ECOLOGY AND SYSTEMATICS Table 45 Snout-vent lengths of bullsnakes (Pituophis melanoleucus) in successive months during their first 5. year in Kansas. Young snakes were first caught in September and were dormant between October and May. Means were calculated only for sample sizes >4. Mean is followed by ± 1 standard error. Sample size is listed under the mean. The range of lengths is listed in parentheses. The year 1971-1972 was omitted because few first-year snakes were caught. SPECIAL PUBLICATION-MUSEUM OF 46 NATURAL HISTORY Tabi 6. Absolute and relative growth rates of bullsnakes (Pituophis melanoleucus) during their first year in Kansas, determined from the data in Table 5. Values in parentheses are based on mean snout-vent lengths of i small samples (<5). VERTEBRATE ECOLOGY AND SYSTEMATICS Table 47 Snout-vent lengths and success of feeding for juvenile bullsnakes caught in half-month intervals post-hatching autumn (September and October) in Kansas. From samples arranged in order of decreasing SVLs, mean and extreme SVLs are listed for the approximate upper and lower one-thirds of samples >4. Sample sizes for A and B are listed in C. The year 1971 was omitted because few juveniles were 7. in their first caught. Mean SVL Year S 1-15 A. Lower h of l Extreme Half-moruh inters als S 16-30 O 1-15 SVL Half-month intervals O 16-30 S 1-15 S 16-30 O 1-15 O 16-30 SPECIAL PUBLICATION-MUSEUM OF 48 TABLE 8. Absolute growth rates ofbullsnak.es (Pituophis melanoleucus) trapped more than once during a = number of useable recapture recyear in Kansas. N ords. 1969 and 1972 were omitted because of the small number of recaptures. Means followed by ± 1 standard error. Age: first year summer only Year Age: >one year NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS Table 10. 49 Absolute growth rates (GR) of recaptured bullsnakes (Pituophis melanoleucus) in Kansas after between captures. N = useable recapture records. Means followed by ± standard different lengths of intervals error. 1 SPECIAL PUBLICATION-MUSEUM OF 50 NATURAL HISTORY II00-- I000-- ]> 900-- — 800+ to 700 UJ _J > 600 CO 500-- 400-I f I A S FIRST ————— -I— — — —h MJJASO MJJA S i I I I t- I SECOND I I I -H M 1 1 J J 1 1 H- A S FOURTH THIRD GROWING SEASON Fig. 1. Growth curve for bullsnakes (Pituophis melanoleucus sayi) in south central Kansas. Tick marks on the abcissa indicate the mid-point of each month. Vertical dashed lines represent the dormant period of six months. The solid growth line estimates the average growth of bullsnakes at different ages in nine years (1966-1974). Hollow circles represent designations). Dashed mean lengths of snakes in lines represent mean growth monthly samples of first-year snakes (see text poor year (1970-1971) and good years rates in a for age ( 1 968— 1969). ably higher than in 1969 and 1974 when rodent trapping was much less successful. Growth would not be expected to increase in proportion to the increase in prey populations if the snakes were already satiated. Also, at high feeding levels, a large proportion of the increased food intake probably goes into fat reserves rather than into increased growth in length. In 1968, weights of bullsnakes for any given length were higher than in any other year (Table 9). In 1969, although growth rates were relatively high (Table 6), feeding activity was much lower (Table 2) and weights were relatively low (Table 9). Young bullsnakes that accumulated extra fat reserves in the autumn of 1968 were able to maintain high growth rates in 1969 when rodent populations were probably only moderately high. The year 1974 had higher feeding activity (Table 2) and relatively high growth rates and weights (Tables 6, 8 and 9). Growth rates of juveniles in the autumn were variable. Low production, low survival and/or low growth of juveniles occurred in the autumn in 1970, 1971, and 1972. Feeding rates were also generally low (Table 7). A growth curve for bullsnakes for the years studied is presented in Fig. 1. The first year's growth was determined from the mean lengths of monthly samples while the growth curve be- VERTEBRATE ECOLOGY AND SYSTEMATICA yond the first year was estimated from the growth The lower dashed rates of recaptured snakes. line 1970-1971. a represents the average growth in line reppoor growth year, and the upper dashed resents growth determined from the combined samples for 1967-1968 and 1968-1969. good years. The data were inadequate quency distribution was not included in a growth study, normal adult si/e was taken from the growth curve or from the author's statements about adult size. Other biological parameters modify the relation between first-year growth rate and size: to estimate the for snakes older than variation in a) growth monthly one year so the growth rate is applied uniformly inthrough the active season. This growth curve dicates that bullsnakes were approximately 790 SVL at one year of age in August. 950 in Michigan with a short growing season reported lower first-year growth than studies of related forms in Kansas. The growth two years and 1030 mm SVL at three estimates may be slightly low for These years. the second and third years, since growth was probably more rapid before August than after of 35August. If older snakes grew at the rate 40 at mm per year (Table 4). a bullsnake mm 1200 mm 1 rates of six years. mm; ca. mm SVL) in 1100 b) regopher snakes (P. m. desert kola) in Utah SVL. Fitch quired 8-20 years to reach 200 949) estimated that P. m. catenifer in central California reached a SVL of more than 800 growth c) mm are presented in The first year species are ar- order of increasing hatchling size. rate of snakes in their first year is posand to itively related to both size of hatchlings was size normal adult size (Table 12). Hatchling either listed in the reports cited or was deter- ranged in sublonger to mature than the faster growing of species from Kansas. The smallest species Growth mined from the growth curves reported and was rounded to the nearest five mm. Normal adult size was usually determined from length distributions as the largest size mode group SVL was rounded snakes, closest 50 and closest 100 of the snakes forming the population sample. The in small to the closest 10 in moderate-sized snakes in a mm mm in mm large snakes. If a fre- first-year in Information for some 12. between low growth rates, Lampropeltis triangulum Kansas and Nerodia sipedon in Michigan, take two to three years to mature. Coluber constrictor mormon in Utah with lower growth rates is both smaller as an adult and takes on the basis of the general magnitude of growth Table ative growth There is a relationship Elaphe, Pituophis, Coluber and Masticophis, usually take more than one year to mature and growth rates are not as high relative to size. Females of moderate-sized species with Different studies of growth rates in snakes cannot be precisely compared because of differences in methods used. But comparisons can be made life. rel- growth rate and age at sexual maturity. Females of most of the species of moderatesized snakes studied become sexually mature at one year of age. First-year growth amounts to at least half of the growth from hatchling to adult size. Females of large snakes, such as two years of age. Growth of bullsnakes in central Kansas was comparable to that of bullsnakes in Nebraska but more rapid than populations of the same species in Utah and California. of and rates. at in the first year Thamnophis have high rates in proportion to hatchling Lampropeltis studied appear to have low 1 colubnd species on growth during the Pi- Utah, adult size while the species of Elaphe and mm 1 1 ( in species in Kansas. Taxonomic differences in growth are evident. Species of the genus five to Parker and Brown (1980) found that mormon and where ecosystem primary production is lower, were much lower than those of related sub- 1 inches (1245 Coluber constrictor tuophis melanoleucus deserticola 100 be five years old and one be seven to eight years old. This is similar to the estimate by Imler ( 945) that bullsnakes in Nebraska reach a total length of 49 SVL would SVL would Geographic variation in growth rate probably was mediated through environmental limitation. The studies of growth of Thamnophis and Nerodia mm mm SVL 51 Elaphe studied. E. quadrivirgata, matures in one year but the larger species, E. climacophora and E. obsoleta, mature in three years and have relatively slower growth rates. from First-year growth in viperid snakes ranged 1949. mm/month) (Fitch 70-370 (10-45 mm 960; Gibbons 972: Klauber 1956: Prestt 1971; Volsoe 1944; Wharton 1966). Growth rates in months of elapid snakes in Australia up to 12 in a growth season 4 1 70 to from age ranged 1 1 mm 52 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 12. Growth increments (mm SVL) for colubrid snakes during the first year. Sex symbols are used to designate growth increments when authors reported different growth rates for males and females. An X designates growth rates of a pooled sample of the two sexes. The two numbers following each symbol are SVL of hatchling VERTEBRATE ECOLOGY AND SYSTEMATICS Table 12. 53 Continued. and normal adult SVL (see text). Listed in parentheses are the number of active or growing months in the year and the age in years of females at sexual maturity. The geographic location of the population studied is listed. Millimeters growth in first year SPECIAL PUBLICATION- MUSEUM OF 54 of eight months (Shine 1978. 1980). It appears that bullsnakes on the Sand Prairie Reservation in Richard NATURAL HISTORY Piatt, and William Schmidt assisted with the data analysis. Kansas have one of the highest first-year growth been reported for snakes in free- rates that has Literature Cited living populations. Blanc hard. Summary 1933. F. N. and Finster, E. B. A method of marking living snakes for future some probEcology. 14:334-349. recognition with a discussion of The growth of bullsnakes (Pituophis melanoleucus sayi) was studied by mark-recapture techniques on the Sand Prairie Natural History Res- lems and Brown. W. 1984. bullsnakes were made. Growth rates were calculated from the records of recaptured snakes and from the mean snout-vent lengths of 1 first-year bullsnakes in monthly samples. No significant sexual differences in size or growth rates Bullsnakes grew rapidly for the first year of life (absolute growth rate of 70.4 mm/month) and mm SVL at one year of age. Growth 40% of the first-year rate in the 20% in the third year and to 0% rates declined to second year, to and less in Carpenter, C. C. 1952. 1 older snakes. Growth rates were significantly lower in 1971 when prey populations and feeding rates of bullsnakes were low. Prey populations and feeding rates were very high in 1968 and growth rates 237-243. 1 970. The western ribbon snake (Thamnophis proximus): ecology of a Texas population. Herpetologica. 30(4):372-379. Clark, D. R., Jr. and Fleet. R. R. 1976. The rough earth snake (Virginia striatula): ecology of a Texas population. Southwest. Nat., 20(4):467-478. Cox. G. W. 1 980. Laboratory Manual of General Ecology. Wm. Brown Co., Dubuque. Iowa, viii + 237 C. pp. from recapture records were consistently lower than growth rates calculated from the change in length of samples of first-year snakes. This discrepancy was due to a temporary decrease in growth after capture and to changes in the monthly samples caused by rates calculated differential mortalitv R.. Jr. Ecological study of the worm snake Carphophis v£7vw.s(Kennicott). Univ. Kansas Publ. Mus. Nat. Hist.. 19(2):85-194. 1974. were highest then. Growth Growth and maturity of three species of Thamnophis in Michigan. Copeia. 1952(4): Clark. D. of bullsnakes were found. reached 790 S. and Parker, W. S. Growth, reproduction and demography of the racer. Coluber constrictor mormon, in northern Utah. In R. A. Seigel et al. (ed.). Vertebrate Ecology and Systematics: A Tribute to Henry S. Fitch. Univ. Kansas Mus. Nat. Hist. Spec. Pub., 10. ervation in south central Kansas. During the nine years of the study, 1966 through 1974, 709 captures of 47 results. of smaller snakes. Feaver, 1977. P. E. The demography of a Michigan population of Natrix sipedon with discussions of ophidian growth and reproduction. Ph.D. Thesis, Univ. of Michigan. Fitch, H. S. 1949. Study of snake populations in central California. Am. Mid. Nat.. 4 1(3):5 13-579. A new style live-trap for small mammals. 1950. Jour. 1951. Acknowledgments A Mammal.. 31(3):364-365. simplified type of funnel trap for reptiles. Herpetologica. 7:77-80. Financial support for this study came from National Science Foundation Grants GB-7830 and GB-35441. The Nature Conservancy fur- nished support to initiate these studies. Bethel College provided financial support for initiation and completion of this study. The following persons assisted with the field study and were indispensable to the completion of this investigation: Victor Claassen, Steven G. Hetzke, Marilyn Johnson. Mark Matthies, Scott Matthies. Ka- 1 960. Autecology of the copperhead. Univ. Kansas Publ. Mus. Nat. Hist.. 3(4):85-288. 963a. Natural history of the black rat snake (Elaphe o. obsoleta) in Kansas. Copeia. 1963(4):649658. 1963b. Natural history of the racer Coluber constrictor. Univ. Kansas Publ. Mus. Nat. Hist.. 15(8):35 1-468. An ecological study of the garter snake, 1965. Thamnophis sirtalis. Univ. Kansas Publ. Mus. Nat. Hist., 5(10):493-564. A demographic study of the ringneck snake 1975. 1 1 1 Diadophis punctatus) in Kansas. Univ. Kansas Mus. Nat. Hist. Misc. Publ.. 62:1( mala Piatt. ert C. Waltner and James Wedel. Douglas Harms, Stanley Senner, Patricia Senner. Rob- 53. VERTEBRATE ECOLOGY AND SYSTEMATICS Fitc h, H. S. and Fleet, R. R. 1970. Natural history of the milk snake Masticophis t. taeniatus and Pituophis mclanoleucus descrticola, in northern Utah. (Lampro- triangulum) in northeastern Kansas. Herpetologica. 26(4):387-396. Frye, J. C. and Leonard, A. B. Pleistocene geology of Kansas. Univ. Kan1952. sas, State Geol. Surv. Kansas Bull.. 99:1- Milwaukee Public Mus. Publ. + 104 pp. peltis Fl 230. IKADA, H. 1959. Biological studies on the snakes. VI. Growth and maturity of Natrix tigrina tigrina (Boie). Bull. Kyoto Gakugei Univ., Ser. B, 15:25Biological studies on the snakes. VII. Growth and maturity of Elaphe quadrivirgata (Boie). Bull. Kyoto Gakugei Univ.. Ser. B, 16:6-21. Growth and maturity of some Japanese 1960. 1972. snakes (Review). The Snake, 4:75-83. Growth and maturity of the Japanese rat snake, Elaphe climacophora (Reptilia, Serpentes, Colubridae). J. Herpetol.. 12(3):269274. 1978. Gibbons, J. 1972. 1951. 7:28-40. Imler, R. H. 1945. Bullsnakes and their control on a Nebraska wildlife refuge. Jour. Wildl. Mgt., 9(4):265- Rattlesnakes, their habits, on mankind, life histories, and Univ. California Press, Berkeley, xxx + 708 pp. vol. 1. Parker, W. S. and Brown, W. S. 980. Comparative ecology of two colubrid snakes. 1 Birds, 29(6): 1146-1151. 1. An ecological study of the viper Vipera berus southern Britain. Jour. Zool. (London). 164:363-418. SCHOEWE, W. H. 1949. The geography of Kansas. Part II. Physical geography. Trans. Kansas Acad. Sci.. 52(3): 1971. in Seibert, H. C. and Hagen, C. W. 1947. Studies on a population of snakes in Illinois. Copeia, 1947(l):6-22. Shine, R. 1978. Growth rates and sexual maturation in six species of Australian elapid snakes. Herpetologica, 34:73-79. Reproduction, feeding and growth in the Australian burrowing snake Vermice/la annulata. J. Herpetol.. 14(l):71-77. Simpson, G. G., Roe, A., and Lewontin, R. C. 1960. Quantitative Zoology. Harcourt. Brace. & 1980. New York, viii + 440 pp. Volsoe, H. Statistical study of the rattlesnakes. IV. The growth of the rattlesnake. Occ. Papers San Diego Soc. Nat. Hist.. 3:1-56. influence 1 American Prestt, Co., 273. Klal'ber, L. M. 1956. History Reservation, Harvey County. KanTrans. Kansas Acad. Sci.. 76( ):5 1—73. Breeding birds of Sand Prairie Natural History Reservation, Harvey County, Kansas. sas. 261-333. F. L. and Call. A. Symposium: a snake den in Tooele County, Utah. Growth and age in western striped racer and Great Basin rattlesnake. Herpetologica, 1937. Platt, D. R. 1969. Natural history of the hognose snakes Heterodon platyrhinos and Heterodon nasicus. Univ. Kansas Publ. Mus. Nat. Hist.. 15(8): 351-468. 1973. Vascular plants of the Sand Prairie Natural W. Reproduction, growth and sexual dimorphism in the canebrake rattlesnake (Crotalus horridus atricaudatus). Copeia, 1972(2):222226. Heyrend. Biol. Geol., 7: vii 1975. 41. 55 1944. Structure and seasonal variation of the male reproductive organs of Vipera berus (L.). Spolia Zool. Mus. Hauniensis, 5:1-157. Wharton, C. H. 1966. Reproduction and growth in the cottonmouth, Agkistrodon piscivorus Lacepede, of Cedar Keys, Florida. Copeia, 1966(2): 149161. Vertebrate Ecology and Systematics — A Tribute to Henry S. Fitch Edited by R. A Seigel, L. E. Hunt. J. I, Knight, L. Malaret and N. L. Zuschlag © 1984 Museum of Natural History. The University of Kansas. Lawrence Communal Denning Snakes in Patrick T. Gregory come extremely abundant Introduction in both fall and spring, the beginning and end of hibernation, respec- The temperate tudes, is zone, especially at higher laticharacterized by a pronounced season- Relatively few species of reptiles occur in such environments and those that do must be adapted to contend with this climatic variability. ality. The most season for reptiles in the temwinter, when conditions may be- critical perate zone come much is too cold for continued activity. In areas where winters are long and cold, reptiles must find shelter from the prevailing conditions and hibernate. Recent research has revealed that many species have important physiological adaptations which enhance their chances for survival over winter, mainly by depressing metabolic rate in hibernation below that predicted on a simple Q ln basis (Moberly 1963; Mayhew 1965; Aleksiuk 1976; Gatten 1978; Patterson and Davies 1978; Johansen and Lykkeboe 1979). This depression is interpreted as an important mechanism for conserving energy at a time when losses cannot be replaced (Gregory 1982). In addition to these physiological adaptations, however, the need to hibernate has had other important In ecological and evolutionary effects on reptiles. this paper, I discuss ecological aspects of one phenomenon in reptilian communal denning of snakes. particular ing: overwinter- Communal denning occurs mainly among reptiles, although it has been in This conspicuousness sometimes works of snakes by advertising their tively. to the disadvantage presence to predators, including man. Rattlesnakes in particular have suffered major declines due 1 to raids on denning populations (Klauber Dunson 1974) and it is likely 972; Galligan and that preservation of some species depends in part on protection of communal dens (e.g., Crotalus Brown et al. 1982). On the other hand, the conspicuousness and abundance of snakes at some dens have provided us with an opportunity horridus. much about the structure and dynamics of snake populations that we might not otherwise have learned (Brown 1973; Gregory 1977a; Par- to learn Brown 1980). The den plays a central role in the annual cycle of some species of snakes. More than half the year is spent at the den in some cases, and mating ker and often occurs in the vicinity of the den in spring or fall. The den may even function as a base of operations for part of the population during the summer months. In this paper, I want to emphasize this central role played by the den in communally hibernating species by describing the major ecological and behavioral features of communal denning, using studies of garter snakes {Thamnophis) in Canada as main examples. I snakes hope described to provide at least a partial answer to the "Why do snakes den communally?" question: some lizards and turtles (Woodbury and Hardy 1949; Weintraub 1968; Ataev 1974; Vitt for Major Features of Communal Denning 974). Communal hibernation has been reported for a few snakes in the southern hemisphere (e.g., 1 Although considerable variation exists from one case to another, several important features emerge from an examination of studies of communal dens. These are: type of site used for hibernation, spatial relationship of den to summer Aparallactus capensis in southern Africa, FitzSimons 1 962, p. 30; Demansia reticulata and Pseudechis porphyriacus in Australia, Kinghorn 1964; see also Shine 1979). but this behavior is most pronounced in the northern hemisphere, especially at higher latitudes. However, not all habitat, size tions, denning northern snakes hibernate communally. In communal denning, large and numbers of oth- 1. erwise solitary animals aggregate at localized sites to pass the winter. Both single and mixed species and structure of denning populaand spring activity of snakes at fall sites. Type of site. — Communal dens relatively permanent are usually structures with cavities or passageways which allow the snakes access below the frostline to pass the winter. For example, in southern British Columbia, the western rattle- aggregations occur (Gregory 1982). This type of behavior is often conspicuous. Snakes which ordinarily are not locally abundant suddenly be- snake {Crotalus 57 viridis) usually hibernates com- SPECIAL PUBLICATION-MUSEUM OF 58 Table 1 . Dispersal distances of snakes between hibernacula and NATURAL HISTORY summer ranges (modified from Gregory 1 982). VERTEBRATE ECOLOGY AND SYSTEMATICA 2. Relationship of den to summer habitat.— Snakes which hibernate singly or in very small (Fitch use sites within the may groups and Glading 1 summer 947; Naulleau 1 966). range Com- munal dens of snakes, however, are frequently distances from the sumseparated by fairly long mer range, necessitating an annual migration back and between the two. Distance travelled forth m to several km (Taranges from a few hundred ble 1): perhaps length of migration is inversely related to the availability of sites suitable for 59 have been reported for many snakes, especially the young (Bailey 1948; Carpenter 1953; Hirth 966a; Viitanen 967: Lang 969, 1971; Gregory 1977a, 1982; Parker and Brown 1980). Parker and Brown (1974, 1980) suggest that high mortality figures in such studies may be an artifact of handling and marking snakes, but the evidence in support of this contention is minimal. In cases where individuals generally return to the same den in successive years and where mating usually occurs at the den site (see below), 1 1 1 housing snakes in winter, but this idea has not been tested. In Coluber constrictor, mean dispersal distance may be correlated with popula- communal denning produces presumably, in years with high numbers, individuals which disperse farther escape intraspecific competition for resources (Brown and Parker 1976). Some snakes which move between discrete denning and summer areas show a highly directional form of dispersal (Gregory and Stewart portant contributing factor to differentiation within species (Gannon 1978). At the local level, 1975). while others do not (Parker 1976). Individual snakes often return to the same den or ing occurs denning area year after year; measures of den fidelity of snakes in successive years are often in mating has been observed in some cases (Brown 1973; Brown and Parker 1976). Finally, there is no particular reason to believe that young snakes hibernating at a communal den for the first time necessarily use the same den as their parents, except when the young are born at the den. 3. Size and structure of denning populations.— Sizes of overwintering aggregations of snakes have been reviewed by Klauber (1972). Parker and Brown (1973). and Gregory (1982). Most aggregations probably consist of much fewer than 100 individuals of all species combined. Some denning populations, however, may include a few to several hundred individuals of a tion density; the 90-100% range (Fitch 1960; Viitanen 1967; Lang 1971; Brown and Parker 1976; Gregory 1977a. 1982; Parker and Brown 1980). Other authors have concluded that den fidelity is low (Noble and Clausen 1936); however, the definition of what constitutes a den or denning area varies from study to study so that results are not necessarily comparable. In addition, distance between neighboring dens may affect fidelity but is not always reported; Lang (1971) found lower fidelity and greater annual interchange between dens that were closer together. Nevertheless, a remarkable ability to home is shown by some a large departure from panmixia. Over large areas, isolation of different denning sites might ultimately be an im- however, populations at particular dens are probably never completely isolated demes. Den fidelity is rarely 100% so that some interchange occurs between dens. In addition, even in species which normally mate at the den. occasional mat- from away from the den different hibernacula when individuals may come into con- tact (Gregory 1977a). Inter-den den complexes less in Coluber constrictor \n Utah (Brown 1973; Brown and Parker 1976) and homing Thamnophis sirtalis in Manitoba must apparently pass close by other dens en route to their own dens each fall (Gregory and Stewart 1975). Other similar examples are given by Viitanen ( 967) and Lang (1971). given species (Criddle 937; Viitanen 967; Lang 1969; Klauber 1972; Parker 1976). The largest denning populations known are those of Tham- The exact mechanisms used in homing are not known, but there is presumably selective value in returning to a den in which overwintering has been successful previously, even where other hi- abundance is smaller and often more secretive than adults. Nevertheless, it seems clear that young-of-year species: Homing than 1000 m to specific is apart almost 100% 1 bernating sites are abundant. This is important since high rates of mortality during hibernation 1 1 nophis sirtalis in Manitoba, where numbers at one den fluctuated between about 4000 and 8000 in a four-year period (Gregory 1977a). Sampling the snake population different size age groups in a proportion to their relative difficult because young snakes are in and/or juveniles are frequently absent (Viitanen 1967; Gregory 1977a: Sexton and Hunt 1980) or greatly underrepresented (Hirth et al. 1969; NATURAL HISTORY SPECIAL PUBLICATION-MUSEUM OF 60 Klauber 1972; Parker and Brown 1973, 1980; Parker 1976; Brown and Parker 1976) at communal hibernacula. Prestt (1971), however, in contrast to Viitanen's ( 967) observations of the 1 However, young Thamnophis sirtalis hibernate at a communal hibernaculum in British Columbia despite occasional predation on them by adult T. elegans at the same den (Gregwith adults same species ( Vipera berus), found young hibernating with the adults. In some small species, the ory, unpubl. obs.). Quantitative assessment of young also apparently hibernate with the adults (Noble and Clausen 1936; Lang 1971). Why young snakes often do not use the same dens as the adults is puzzling. Perhaps whether or not they do depends to some extent on the distance between the den and the summer hab- that there young are born itat. If the a long way from the den, summer range may simply be too make the journey if in the it expensive energetically to they can find suitable hibernacula closer by. This seems quite likely as smaller snakes can use sites which are inaccessible to the adults because of probably the case for Thamnophis sirtalis in Manitoba, where the adults, but no young-of-year, hibernate in limestone sinks several kilometres from the summer habitat their size. This is (Gregory and Stewart 1975; Gregory 1977a). Young of this species are known to hibernate communally with two species of small snakes in mounds in nearby Minnesota (Lang 1971); presumably what occurs in the Manitoba summer habitat, where ant mounds are abundant. Young born or hatched closer to the den. on the other hand, may be more likely to hiber- ant these ideas awaits further study, but it is clear an important ontogenetic change in hibernation behavior in many species of snakes. is and spring activity at dens. — Previous studies have revealed a great deal of variation in patterns and timing of entry into and emergence from hibernation of snakes at communal dens, including differences between species, sexes, and age/size groups at the same den (Viitanen 1967; 4. Fall Lang 1971; Prestt 1971; Brown 1973; Landreth 1973; Gregory 1974, 1977a, 1982; Brown and Parker 1976; Parker and Brown 1980). In most cases, fall and spring activity periods at dens span several days or weeks, but individual animals may be active above ground only briefly (e.g., In other cases, however, individual remain active in the vicinity of the may den for a large part of the fall and/or spring period Lang 1971). snakes Viitanen 1967; Prestt 1971; Gregory 1974; Parker and Brown 1980), usually without feeding. The advantages of remaining active above (e.g., at the den, rather than seeking cooler conditions below ground, are not clear, since such this is ground nate with the adults (but see Viitanen and Brown 1980). The significance of activity dens is obvious in cases where snakes mate 1967). is activity energetically very expensive (Parker at at Gravid females of many snake species show a the den site or nearby (Viitanen tendency to aggregate in areas of localized shelter (Gregory 1975a); in some cases, this may occur at or near the den site (Viitanen 1967; Prestt Gregory 1974, 1977a; Bennion and Parker 1976; Parker and Brown 1980). Some species, however, apparently mate away from the den (Brown 1973; Brown and Parker 1976; Parker and Brown 1971; Gregory, unpubl. obs.; see section "Communal Denning and Mating Behavior on in Thamnophis"). Other examples of gravid females occasionally being found at dens are given by Preston (1964) and Galligan and Dunson (1979). intra- and this case, interspecific predation by adults (in Masticophis taeniatus and Coluber constrictor). The hypothesis that there is some young snakes in hibernating with supported by the observation that disadvantage to the adults is most young Masticophis using communal dens do not survive to age one and most one-yearolds at dens are not known to have used the dens the previous year (Parker and Brown 1980). 967; Prestt 1971; 1980). Fall mating occurs in some snakes (Trapido 1940; Rahn 1942; Saint Girons 1957; Greg- ory 1977a), but in most cases sional phenomenon and which less is only an occa- intense than in the major breeding season for most temperate zone snakes. Prolonged activity at dens in fall is therefore generally not explained spring, Parker and Brown (1980) argue that young snakes hibernate elsewhere as a defense against 1 is by mating behavior. Perhaps spring and fall activity is related in other ways to the reproductive cycle. For example, male Vipera berus bask at dens in fall to promote spermatogenesis, which will be completed during basking the following spring (Volsoe 1944). Vipera species, however, are different from all other temperate zone snakes, in which is completed well before hiber- spermatogenesis VERTEBRATE ECOLOGY AND SYSTEMATICS nation (Aldridge 1979a). Thus, it is not obvious why individual male Masticophis taeniatus, which breed in spring, remain active at the den for up to 37 days before entering hibernation (Parker and Brown 1980). Females of some species of snakes undergo part of secondary vitellogenesis in fall (Aldridge 979b), but presence or absence of this pattern has not been correlated with fall activity or lack of it. In females of all 1 temperate zone species, all or part of secondary vitellogenesis occurs in spring (Aldridge 1979b). If basking is important to this process, snakes in some cases may trade off the lack of food at the the advantage of readily available shelter at times when cold weather could arise den site for suddenly. We do not yet know enough about of reproductive cycles (and factors details af- 61 currence, but large aggregations probably have a different basis. There are at least three possible reasons, not mutually exclusive, for communal low availability of suitable hibernating sites; 2. aggregation of snakes in hibernation to minimize losses of endogenously produced heat; 3. enhancement of mating success in the breeding season. A fourth possible advantage of denning: 1. communal denning efficient utilization is that it may lead to more of resources around the den; the area occupied by a dispersed population may change according to annual changes in snake population density and/or resource abundance (the "refuging" hypothesis, Parker and Brown 1980). However, it is not clear to me that communal denning is necessary for this system to operate and even if so, it is more likely to be a them) or fall and spring activity periods of most snakes to be able to correlate these fea- consequence of communal hibernation rather tures. Shortage of suitable hibernacula is undoubtedly the main cause of communal overwintering fecting Why Do Snakes Den Communally? than a reason for in many cases. its occurrence in the first place. This argument has been used to winter aggregations of some lizards (Weintraub 1 968; Vitt 1974) and the rattlesnakes Crotalus viridis (Gannon 1978) and C. horridus explain Certain disadvantages appear to be inherent of communal hibernation. First, an- in the habit imals at dens in spring and fall may be very conspicuous because of their abundance, and may therefore attract predators. For example, crows take a fairly heavy dens Manitoba toll of Thamnophis when sirtalis at (Brazaitis 1 980) and the occurrence of more than in large communal dens (e.g.. Car- one species penter 1953; Hirth el al 1969; Lang 1969). Smaller species of snakes may be less influenced vegetation cover is sparse (Gregory 1977a). Individuals hibernating singly at isolated locations by this factor than larger snakes since they are presumably capable of using cavities unavailable to the latter because of size. The problem of lim- would be much ited availability of hibernacula is in in early spring less the ground conspicuous. [Professional collectors for biological supply companies make even greater inroads in populations at these dens (Gregory 1977b). The problem of collection and/ or slaughter by humans at dens is also great for rattlesnakes, since these animals are often ac(Klauber 1972; Galligan and However, human collection is a relatively recent phenomenon and cannot be considered a long-term selective force.] Another tively persecuted Dunson 1979). possible disadvantage of communal denning is related to the fact that the den and summer habitat may have be quite far apart. In such cases, snakes through unfavorable hab- to migrate, often expending energy and possibly exposing themselves to a higher risk of predation. itat, The question therefore arises as to why snakes den communally. Very small aggregations of snakes (see examples in Parker and Brown 973) may simply be fortuitous and irregular in oc1 expected to be particularly serious in cold climates, where hi- bernation at considerable depth is critical for survival. This correlates well with the observation that communal denning is an especially well de- veloped phenomenon at higher latitudes. Gannon (1978) feels that availability of hibernacula is an important factor limiting the distribution of Crotalus viridis in southern Saskatchewan and On the other hand, several authors have noted that there may be apparently usable hibernacula which go unused in any winter, even Alberta. at high latitudes (Viitanen 1967; Lang 1971; 977a). Lang (1971) concluded that availability of ant mound hibernacula was therefore not a limiting factor on num- Klauber 1972; Gregory 1 bers of three species of small snakes in Minnesota. This could be true, however, even if all hiber- nacula were used since there might still be space for more animals within individual hibernacula SPECIAL PUBLICATION- MUSEUM OF 62 (Parker and Brown 1973). In addition, what appears to be a suitable den to the observer may we need to know more about what qualities make a good hibernating site and to assess these qualities at potential sites before we can reach a conclusion renot be seen as such by snakes; garding availability of hibernacula. The argument that snakes hibernate commun- and ally so that they can aggregate below ground reduce heat loss is difficult to support. It was put forward by White and Lasiewski (1971). with particular reference to rattlesnakes. In favor of NATURAL HISTORY numbers of snakes involved, especially for species which mate at or near the den in fall or spring. Even in species which do not mate right at the den site, communal hibernation may still enhance mating chances since individuals dispersing from a small area should come into contact more often than when widely scattered (Parker and Brown 980). As in the case of the "refuging" hypothesis above, it may be argued that high 1 probability of reproductive success is not a primary reason for the occurrence of communal large denning, but simply a secondary advantage of it. This problem is somewhat circular, however, since it is also possible that the prior evolution masses of animals (KJauber 1972); however, such behavior could be due to disturbance. Aleksiuk of early spring mating has resulted in selection for individuals that seek hibernacula used by (1977) has also shown that Thamnophis sirtalis tend to huddle under cold conditions, but there conspecifics, or that the is no evidence that this actually happens in the den during hibernation. Snakes hibernating in tion. this idea is the observation that rattlesnake blasted open in winter communal dens are sometimes reveal dens frequently not in contact with one another, although small groups may be formed (Noble and Clausen 1936; Carpenter 1953; Lang 1971; Brown et al. 1974). and isolated individuals do not differ in body temperature from grouped individuals (Brown et al. Use of energy reserves during hibernation probably very low in most cases (Parker and Brown 1980), consistent with the observa1 974). per se is low 1974; Sexton and tion that hibernation usually takes place at a (e.g.. Brown et al. Hunt 1980; Brown 1982; Gregory temperature 1982), not a may be very high in and dens fall spring, but this during is not taken into account in most studies, yielding considerable overestimates of the energetic cost of hibernation (Bartlett 1976; Parker and Brown 1980). Finally, an important physiological adaptation of many hibernating reptiles seems to high one; use of energy reserves activity at be that metabolism is it is not surprising that snakes hibernate at fairly low temperatures, contrary to White and Lasiewski (1971). the predictions of Snakes hibernating with conspecifics at communal dens presumably have greater chances of finding mates in the mating season than they would have if they hibernated singly. This idea is difficult to test in the field, but it is obvious mating opportunities at communal sites should be frequent simply because of the large that studies two have evolved jointly. which aim to unravel this ques- In some cases, the advantages of communal denning in terms of mating extend beyond mere numbers. Once the mass overwintering habit is established, an opportunity is presented for mating behaviors to evolve which take advantage of An example is provided by the Thamnophis sirtalis. In this species, this situation. garter snake, the different seem mating strategies of the two sexes to be reflected in significant differences in the dynamics of their behavior at the den during the breeding season. This example is examined in detail in the next section. Although the data analysis is largely a posteriori, its main function is to suggest testable hypotheses for further study and points for comparison with other communally denning species which show different behaviors. Communal Denning and Mating Behavior in Thamnophis significantly depressed at low temperatures (e.g., Aleksiuk 1976; Johansen and Lykkeboe 1979; Gregory 1982). If this is interpreted as an energy-saving device during hibernation, then We need The common garter snake, Thamnophis sirthe most widespread species of snake in talis, is North America. While this species does not den range, such behavior communally throughout is well developed in the northern parts of its its range. The study area in question here is in the Interlake region of Manitoba, near the northern limit of distribution of T. sirtalis. This region has a continental climate, with long cold winters and variable summers (Gregory 977a). Only four species of snakes occur in the study area, and T. sirtalis is by far the most abundant of these. Communal dens of T. sirtalis in the Interlake 1 VERTEBRATE ECOLOGY AND SYSTEMATICS 63 are mainly limestone sinks, formed by the collapse of the ground surface into subterranean The major den examined caves. (Den 1 is ) in this studs a large, bowl-like depression m m about 20 m wide x 3 deep; the bottom of long x 12 the bowl is riddled with cavities leading underground. These dens occur on ridges between large marsh belts. Dens are abundant in such areas km apart. Popuand are frequently less than lations using dens may be very large. Den is estimated to have housed as many as 8000 snakes during one winter, but population size fluctuates drastically from one year to another, apparently 1 1 response to variations in weather (Gregory Den populations are exclusively adult (Fig. 1); it is not known where the young hiberin 300 540 620 700 780 860 SVL(mm) 380 460 1977a). nate in this area. snakes is in the The summer habitat of these marshes between the ridges. In- Fig. Den 1 of 20 1 . Size frequency distribution of T. sirtalis at Animals are grouped into intervals in fall 1972. mm snout-vent length (SVL). n = sample dividuals may move as much as 18 km between den and summer range; migrations are unidirectional, with all animals moving south in summer Data from Gregory (1977a). despite the fact that suitable marshes are also found in other directions (Gregory and Stewart Gartska 1975). Despite these long migrations and the rel- size. et al. ( 1 982) indicate that females, which probably have an active role in mating, are more likely to mate when still cold from emergence; not be sex- warmed up may ative closeness of dens to one another, females which have success of individual snakes to the ually receptive. In both sexes, sexual behavior is independent of gonadal activity (Crews and Gartska 1982; Gartska et al. 1982). The environmental problem faced by these snakes is that of a very short, and sometimes homing same den in successive years is about 96% (Gregory 1977a). The den is a central feature in the annual cycle T. sirtalis. The hibernation period be as long as six months; in addition, the and spring activity periods in the vicinity of of Interlake may fall the den may occupy up to h months each (Greg1 y ory 1977a). In extreme cases, therefore, individual snakes may spend only three months away from the den during the year, and this is the only in which feeding takes place (Gregory and Time cold, active season. to premium. Under such for bernation. spring after emergence (Gregory 1974, 1977a). often as possible since this Mating activity of males is apparently stimulated by the change from the cool conditions ex- of increasing above ground in spring (Aleksiuk and Gregory 1974; Hawley and Aleksiuk 1975; Crews and Gartska 1982; Gartska et al. 1982). Male tions courtship activity is directly related to temperature (Hawley and Aleksiuk 1975), but declines and Gregory 1974; Camazine et al. 1980). The temperature change associated with emergence from as the mating season progresses (Aleksiuk hibernation also stimulates sexual receptivity of females, but does not affect their attractivity to males (Licht and Bona-Gallo 1982). However, therefore at a we should ex- pect the evolution of a mating system which maximizes the efficiency and success of mating Stewart 1975). Although occasional fall mating occurs, virtually all mating occurs at the den in warmer condi- is conditions, time perienced in hibernation to the reproduce and per- form other essential functions both sexes early The two in the season following hisexes have, in effect, different reproductive strategies: Males should mate as fitness; is their only means females, on the other hand, need mate only once per season (but see Gibson and Falls 1975 for evidence of multiple insemination of females, and discussion below) and should spend a minimum of time involved in mating activities per se, devoting instead more time to other activities critical to successful reproduction. Differences in behavioral dynamics of the two sexes of Interlake T. sirtalis during the mating season appear to reflect these differ- ences in mating strategies. The relationships described above are difficult to test directly for lack of an appropriate control reasonable substitute for a true consituation. A SPECIAL PUBLICATION-MUSEUM OF 64 Numbers of individuals of each sex of T. in fall and spring of four overcaught at Den wintering seasons (data from Gregory 1977a). Table sirtalis 2. 1 NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS SEPT. 7 13 65 OCT. 25 19 13 25 19 1 .8 6 o o 4 V 4 Q- .2 ki J 25 .. II 1 Witt iln I 25 19 13 i i * JUNE MAY APR. 6 31 Fig. 2. Proportion of females (P-99) in daily samples of T. sirtalis at Den 1 for fall 1971 (open circles) and spring 1972 (closed circles). Vertical lines are 95% confidence limits calculated on basis of binomial distribution. Daily sample sizes range from 3-55 for fall 1971 and 2-125 for spring 1972. However, the numbers involved daily samples does not vary greatly over the spring period (Fig. 2). In contrast to males, however, activity in the Interlake are females apparently emerge throughout the spring period and spend little time at the den, dispersing those reported elsewhere, obscuring actual copulation. Mating almost always occurs on the very soon after emergence; road counts of dispersing snakes indicate mostly females leaving early in the spring and increasing proportions of ground, but males may follow females into low bushes and mate there (Gregory 1975b). Following copulation, the mating group breaks up rap- males leaving as the season progresses (Fig. 6). Females also emerge later in the day than do males, but they emerge progressively earlier as the season continues (Gartska et al. 1982). Except for the early part of spring when weather is sometimes cool, females are courted as soon as they emerge, or even while emerging ( Aleksiuk and Gregory 1974). Typically, many males si- males seem to have no further interest which becomes unattractive for a day or more and even intolerant of further courtship (Crews and Gartska 1982; Gartska et al. multaneously court a single female, creating a writhing mating "ball'"' (Aleksiuk and Gregory 1974). Not surprisingly, the head-to-head ori- between mature, non-mated and recently mated females and only court the former. The cues used entation of male and female shown by many col- ubrid snakes is not required for successful courtship in this species (Gillingham and Dickson 1980). Courtship and mating take several min- not usually possible to see which male manages to copulate with the female. This utes, but it is contrasts with the observations of others (e.g., Devine 1977), in which unsuccessful males leave before the successful male has finished copulat- ing. much in mating larger than idly; the in the female, 1982), but turn to other emerging females instead. Devine shown ( 1 that 977) and Ross and Crews (1977) have male garter snakes can distinguish are apparently pheromonal. The female attractiveness pheromone is a non-volatile lipid, related to vitellogenin, the precursor of yolk which is manufactured in the liver and circulates in the blood (Gartska and Crews 1981; Crews and Gartska 1982; Gartska et al. 1982). This pher- omone is presumably brought to the skin via a dermal vascular bed and is forced to the body surface through the thin skin between the dorsal and lateral scales. It is a contact pheromone. de- SPECIAL PUBLICATION-MUSEUM OF 66 NATURAL HISTORY OCT. SEPT. 25 19 13 7 13 25 19 i 1.0 .8 .6 Q UL a u I 11 flu £ 1.0 .8 .6 .4 ii ii .2 6 25 APR. 6 1 7 19 13 25 6 31 JUNE MAY Fig. 3. for fall Proportion of recaptures from same season (P-recaps) in daily samples of T. sirtalis at Den 1971 (upper level) and spring 1972 (lower level). Open circles represent females and closed circles males. Vertical lines are 95% confidence limits calculated on basis of binomial distribution. Daily sample sizes range from 330 (males) and 1-25 (females) for fall 1971, 2-1 10 (males) and 1-15 (females, plus some days with no captures) for spring 1972. 1 tected by the male via the vomeronasal system, and may or may not be the same as the trailing pheromone, which allows species-specific trailing of females by males and has its most pronounced effect during the spring mating period (Ford 1978, 1981, 1982; Ford and Low 1982). In any case, males are not sensitive to the female pheromone early in the season when mating opportunities are very low; however, as the season progresses, females become slightly more abundant relative to males, and males become sensitive to the pheromone and attractiveness more discriminating about (Gartska et a/. potential mates 1982). Mated females are unattractive to males be- cause of a male-inhibiting pheromone. Following copulation, a plug forms in the cloaca of the female (Devine 1975); this copulatory plug is apparently manufactured in the renal sex segment of the male (Crews and Gartska 982). The 1 male-inhibiting pheromone is probably made by the male at the same time as the plug (Ross and Crews 1977; Crews and Gartska 1982), although Devine (1977) suggests that the female produces VERTEBRATE ECOLOGY AND SYSTEMATICA 67 2000 1.0r 1500 .6 Q a u . .4 <u N 1000 .2 - 666666466666 666 500 23 17 11 29 MAY nit Proportion of recaptures from same season in daily samples of T. sirtalis at Den 1 for spring 1971. Symbols as in Fig. 3. Daily sample sizes range from 11-123 for males and 1-6 (plus some days with no Fig. 4. captures) for females. 25 1 APR. 19 13 pheromone. In any case, females with plugs The plug is generally mechanism for preventing insemination by rival males and there- are unattractive to males. interpreted as a temporary fore 31 5. Estimated above ground daily population for spring 1972. (N) of male T. sirtalis at Den Estimates from Jolly-Seber mark-recapture analysis (Jolly 1965). Vertical lines are 95% confidence limits (for method of calculation see Gregory 1974. 1977a). Data from Gregory (1974); note incorrect date of two data points in Fig. 3 of Gregory (1974). Fig. sizes the 25 MAY sperm competition, but not 1 for ensuring sole paternity (Devine 1975; Gibson and Falls 1975). Since the plug is expelled by the female a few attractive sexually. Multiple matings of females days after mating (Devine 1975), mated females may be only briefly unavailable or unattractive preclude the possibility of multiple insemination from retention of sperm from a copulation the sexually. previous fall or earlier (Gibson and Falls 1975) or from a subsequent mating during dispersal, An interesting question therefore not females mate is more than once spring, especially since Gibson and whether or in a given Falls (1975) provide indirect evidence of multiple insemination in T. sirtalis from Ontario. These authors argue that because ovulation and fertilization in in spring are unlikely in this case. This does not but these are probably rare events. Another possibility is simultaneous mating of a female by two males (polyandry. Gartska this is et al. 1982). but likely also rare. What are the advantages to individuals of both occur a few weeks after copulation Gregory 1977a), all reproductive females sexes of the system of emergence and mating shown by the Interlake snakes? The emergence represent mating opportunities for males and multiple insemination in a given mating season of exposure to predators) than continually avoid- of males before females is a widespread phenomenon in snakes (Duguy 1963; Viitanen 1967; Lang 1971; Prestt 1971; Parker and Brown 1980: Gregory 1982). but other species do not show the same details of spring behavior as described ing suitors (Gibson and Falls 1975). Multiple insemination is therefore most likely to occur in sirtalis where females do not disperse rapidly from the denning area (e.g.. Devine 1977). In the Interlake, however, females disperse almost immediately after emerging and mating, when they possess copulatory plugs and are presumably un- Fagerstrom (1971), which is an attempt to account for the existence of protandry in butterflies. Wiklund and Fagerstrom conclude that protandry is a reproductive strategy of males which are capable of multiple mating, and is a result of T. sirtalis (e.g., should be expected. For females, submitting to remating may simply be less risky (e.g., in terms situations here. The pattern of emergence of Interlake T. appears to fit the model of Wiklund and NATURAL HISTORY SPECIAL PUBLICATION-MUSEUM OF 68 1CL 10. o t 1_ o»- I 0.5. 4-9 10-15 16-21 22-27 28-29 MAY Fig. 6. Proportion of female (P-22) T. sirtalis on roads in vicinity of Interlake dens in spring (1971 and 1972 combined). Data are grouped into 6-day intervals; vertical lines are 95% confidence limits calculated on basis of binomial distribution. Sample sizes range from 7-93. Data modified from Gregory (1974). competition for mates (see also Gibson and Falls 1975). As long as the competitive ability of all males is equal, so that the number of receptive females encountered and mated by a male is a direct function of the proportion of such females males that emerge before females should mate with more females, on average, than should late emerging males. Competitive ability and relative mating success of individual male T. sirtalis in the Interlake have not been measured, but empirical observation of mating behavior in the field does not suggest any in the population, obvious variation in competitive ability. Perhaps mating success is reflected in the length of time individual males stay at the den in spring (i.e., males which mate several females early in the season might disperse earlier). Other species of snakes have different mating systems (e.g., Masticophis taeniatus, Bennion and Parker 1976; Parker and Brown 1980; Vipera berus, Viitanen 1 967; Prestt 1971), but earlier emergence of males is still probably best explained in than females terms of competition (male-male aggression in these cases) among males for emerging females (Crews 1975; Parker and Brown 1980). Male Vipera berus also complete spermatogenesis by VERTEBRATE ECOLOGY AND SYSTEMATICA males (Crews and Gartska 1982; Gartska 1982). Since the pheromone is el al. related to vitel- logenin and larger females produce more yolk because they have bigger broods (Gregory 977a), larger females may be more attractive because they produce more pheromone (Crews and Gart1 ska 1 982; Gartska el al. Table male 69 Distribution of mated and non-mated 3. T. sirialis in fe- reproductive and non-reproductive Samples from summer habitat and from categories. roads in vicinity of dens in April and May 1972; reproductive and mating status determined by dissection (data from Gregory 1977a). Mated Non-maled 20 4 1982). Males might even choose mates on the basis of a previous year's reproductive output (Crews and Gartska 1982) since lipid may be stored in the skin of females (Gartska and Crews 1981; Crews and Gartska 1982; Gartska el al. 1982). In any case, males can select the potentially most fecund mate.] A second advantage to females of this pattern of activity is that they are mated almost immediately upon emergence, reducing their time of exposure to predators (Crews and Gartska 1981) and allowing them to disperse quickly to the summer habitat and begin feeding. They thus spend a minimum of time active without feeding. is important because summers are very short This Potentially reproductive 7 Non-reproductive H„ (no difference in proportion of mated females in reproductive and non-reproductive categories) rejected with P < .001 (x 2 contingency table). haps the fall activity period case. In other respects, is very brief in this however, activity den seems basically similar at this to that described for Spring collections from den are again heavily biased in favor of males in both species (Figs. 8 and 9). Apparently, 77. sirialis leaves the den earlier in spring than 77. T. sirialis in the Interlake. this in this area and gravid females do not feed in advanced stages of gestation (Gregory and Stewart 1975). Late spring and early summer may elegans; therefore be an important time of year for re- again at the den in the same season, whereas males are recaptured frequently over the spring producing females to balance their energy budgets. Males are not under such energetic constraints as reproductive females and may obtain additional benefits from being near shelter at the den if cold weather strikes in spring. Females apparently trade off this advantage for the others mentioned above. It is presumably also advantageous for non-reproductive (usually smaller) females to leave the den soon after emergence since they would then extend their feeding season and might reach a by the end of sumproduce bigger broods larger size mer; larger females tend to (Gregory 1977a). The observations and conclusions reported here period. 1979 I probably also emerges earlier. In 77. Mating of both species occurs at the den although mating balls are seen much in spring, than in the Interlake. More often, evidence of spring mating is obtained from occasional females found with copulatory plugs in less frequently their cloacas. The talis is spring activity pattern of Interlake 77. sirtherefore probably typical of communally denning garter snakes. If so. it may be an important part of the suite of adaptations allowing garter snakes to be so successful in the rigorous environments which limit the northern distri- bution of most other North American are probably not unique to communal dens of garter snakes in the Interlake. Partly to answer this question, in it elegans, as in Interlake 77. sirialis, the females that are caught in spring are usually not seen reptiles. Questions began monitoring activ- and T. elegans at a Chilcotin-Cariboo region of British Columbia, also an area with long, cold winters. Data for only the first year and a half of Although various aspects of communal denning in snakes have been studied in some detail, there remain many gaps in our knowledge of this the study are presented here, but some trends are apparent. Unlike the Interlake dens, this den is to be ity patterns of communal den occupied in the T. sirialis in the summer by gravid females (Figs. they apparently give birth there and the young remain at the den for their first winter. Few adult snakes are seen at the den in fall; per8 and 9); phenomenon. Many of the questions which need answered are interrelated and include the following: What are the important physical feado distures of suitable hibernating sites? How persion and abundance of suitable hibernating sites in a given area affect the distance snakes migrate between hibernacula and summer range? SPECIAL PUBLICATION-MUSEUM OF 70 10 o NATURAL HISTORY In a _ _ _ _ _ o o in ff7yi-n nj c CO 22-23/4/30 n=60 29/4-1/ty80 n=17 2-3/5/79 n=16 8/ty80 n=6 15-16yty79 n*15 22/5/80 n=2 __ 5-6/5/79 o 6 _ _ ^ I i — i I n=1 18-W^BO n=1 V7/79 n-1 n=12 __ 25/V79 L \ i =10 Snakes 200 300 400 500 600 700 800 SVL(mm) Fig. 8. Size frequency distribution of T. sirtalis at den in Chilcotin-Cariboo region for various times of year snout-vent length (SVL). Dates given (1979 and 1980 combined). Animals are grouped into intervals of 25 are day/month/year; n = sample size. Data above each line represent males (open areas) and unsexed juveniles (hatched areas); data below line represent females (dark areas represent obviously gravid females). mm Why do some snakes disperse in a particular di- rection, especially if suitable habitat able in other directions? Why often poorly represented at are is also avail- young snakes communal dens and where do they hibernate? Where, in relation to the den, are young snakes born or hatched, and does this influence the likelihood of them using the same den as the adults? How do individual snakes find their way back to the same den over long distances year after year, especially where several dens are present in the same general area? Are new dens occasionally colonized (or old ones recolonized following a disturbance) and if so, how and at what rate? What is the extent of genetic isolation among populations at dens in a given area and how is new genetic material introduced to a den? Why are snakes sometimes active at dens for long periods of time in fall and/ or spring without feeding or mating? Several of these questions are discussed by Parker and Brown (1980), who also suggest possible ap- proaches to some of them. Underlying all of this is the question of why snakes den communally. Complex, apparently co-ordinated patterns of emergence and mating, such as that shown by Interlake Thamnophis sirtalis, can probably function only in a denning situation. communal An important hypothesis therefore is that snakes which hibernate communally have a reproductive advantage over those which hibernate as isolated individuals. This hypothesis should be testable. The ideal way to make such a test would be to make direct comparisons of communal and non-communal hibernators within the same population, but know of no examples which both types occur. in I Comparison of the same species in widely separated parts of its range is somewhat risky because the environmental pressures may differ markedly in the two locations. A more useful VERTEBRATE ECOLOGY AND SYSTEMATICS 71 20- 22-23/V80 n=23 Ll P7W. 29/4 -1/5/80 n= 50 20 2-3/5/79 n=88 6/5/80 n=18 15 -1^5/79 n-56 22/5/80 n=19 28/5/79 n=9 5-6/E/79 n=17 18-19/^80 n-10 11/7/79 n-U 25/S/79 n.13 g*q CO o _ - -=^- - »-»• 300 400 500 600 t 100 200 ]»10 i . Snakes 700 SVL(mm) Fig. 9. Size frequency distribution of T. elegans (1979 and 1980 combined). Symbols as in Fig. 8. den at in Chilcotin-Cariboo region for various times of year separate by obapproach to this problem therefore is probably through carefully planned manipulative experiments on specific cases. I am now designing such studies. approach might be to compare the ecology of exclusive and communally and non-communally hibernating species within the same region; if the species in- servation. volved are similar ature tolerance, in etc., it abundance, might temperbe possible size, at least may to eliminate the alternative hypothesis that communal hibernation results from a shortage of overwintering may sites. However, be The most difficult to fruitful Summary different species use different strategies to solve the environmental problem (Wilbur ct al. same 1974; Stearns 1976) and different species may hibernate communally for different reasons. In fact, the hypotheses put forward in this paper to account for communal denning are not mutually In northern regions, where winter may be sev- months long, many snake species hibernate communally in large aggregations of up to a few eral thousand individuals munal hibernacula tures, in extreme cases. Com- permanent strucoften used annuallv bv the same individare usually SPECIAL PUBLICATION-MUSEUM OF 72 and are sometimes a considerable distance from the summer habitat. Denning populations frequently consist mainly of adults and the snakes may be active at dens for some time each spring uals, and Communal hibernation probably relow availability of overwintering sites in fall. flects many cases. Another advantage, however, may be that individuals hibernating with conspecifics have enhanced chances of successfully mating the NATURAL HISTORY computer program limits for proportions of garter snakes (Thamnophis sirtalis) in spring at dens in Manitoba indicates that the two sexes have different behavior patterns consistent with their different reproductive strategies (i.e., males mate more than once per season, females probably once only). Males emerge in fairly large numbers early in spring, whereas females emerge numbers throughout the spring. Most reproductive females mate immediately upon emergence and then disperse to the summer habitat, thereby presumably maximizing the length in smaller of their trast, summer remain activity period. Males, in conat the den for longer periods in spring and continue mating. Thus, emergence patterns are co-ordinated in such a way that mating opportunities are maximized for all individuals in the population. Such a system can operate of communal denning. In fall, when mating is rare, the two sexes do not behave differently from one another. Preliminary data only in the context from a garter snake (T. sirtalis and T. elegans) den in British Columbia suggest similar behaviors. Other communally denning snakes do not show these kinds of behavior patterns, but this does not negate the presumed adnecessarily and Gary Caine assisted preparing figures. Funds for computing were in provided by a UVic Faculty Research Grant and other funds by an operating grant from the Natural Sciences and Engineering Research Council of Canada. The manuscript was typed by Barbara Waito. early in the active season, an important adaptation where summers are short. Analysis of activity for calculating confidence Literature Cited Aldridge, R. D. 1979a. Seasonal spermatogenesis in sympatric Crotalus viridis and Arizona elegans in New Mexico. J. Herpetol., 13:187-192. 979b. Female reproductive cycles of the snakes Arizona elegans and Crotalus viridis. Herpetologica, 35:256-261. 1 Aleksiuk, M. Metabolic and behavioural adjustments to 1976. temperature change in the red-sided garter snake ( Thamnophis sirtalis parietalis): an integrated approach. J. Thermal Biol., 1:153156. 1977. Cold-induced aggregative behaviour in the red-sided garter snake (Thamnophis sirtalis parietalis). Herpetologica. 33:98-101. Aleksiuk, M. and Gregory, P. T. 1974. Regulation of seasonal mating behavior Thamnophis sirtalis parietalis. in Copeia, 1974: 681-689. Ataev, Ch. 1974. Characteristics of hibernation of the Caucasian agama in the Kopet-Dag Range. Soviet J. Ecology, 5:159-161. Bailey, R. M. Winter mortality in the snake Storeria de1948. kayi. Copeia, 1948:215. Bartlett, P. Winter energy requirements of Sceloporus 1976. occidentalis in the Mojave Comp. desert. vantages of such behaviors. It should be possible to design experiments which test hypotheses arising from the garter snake model. Biochem. Physiol., 55A: 179-181. Bennion, R. S. and Parker, W. S. Field observations on courtship and aggres1976. sive behavior in desert striped whipsnakes, Acknowledgments Masticophis t. taeniatus. Herpetologica, 32: 30-35. Brazaitis, P. Denizens of a secret underworld. Animal 1980. Kingdom, 82:14-18. I thank Henry Fitch for stimulating me to study snake ecology through his many papers on the natural history of various species of snakes. I also Brown, W. 1973. mormon thank Linda Gregory for her enthusiastic criticism of my work and Malcolm Macartney, Anna Roberts, and Neil Dawe for their interesting discussions about garter snake dens and mating behavior. Malcolm Macartney also allowed me access to his preliminary data Pat Konkin (UVic Statistics on Crotalus viridis. Laboratory) wrote S. Ecology of the racer. Coluber constrictor (Serpentes, Colubridae), in a cold temperate desert in northern Utah. Ph.D. Dissertation, Univ. of Utah, Salt Lake City. Brown, W. 1982. S. Overwintering body temperatures of timber rattlesnakes (Crotalus eastern Brown, W. 1976. S. New York. J. and Parker, W. Movement homdus) Herpetol., 1 in 6: 1 north- 45-1 50. S. ecology of Coluber constrictor VERTEBRATE ECOLOGY AND SYSTEMATICS near communal hibernacula. Copeia, 1976: 225-242. Brown, W. S., Parker, W. Thermal and 1974. S. and Elder, spatial J. A. relationships of two species of colubrid snakes during hibernation. Herpetologica, Brown, W. S.. 982. 1 30:32-38. J. 1979. Gartska, W., Tokarz, R. and of castration and androgen replacement on male courtship behavior in the redsided garter snake Thamnophis sirtalis parietalis). Horm. Behav., 14:358-372. Carpenter, C. C. A study of hibernacula and hibernating as1953. sociations of snakes and amphibians in 1978. Michigan. Ecology, 34:74-80. Crews, D. 1975. Psychobiology of reptilian reproduction. Science, 189:1059- 1065. Crews, D. and Gartska, W. R. 1982. The ecological physiology of a garter snake. Duguy, 1963. V. Factors limiting the distribution of the praiBlue Jay, 36:142-144. 982. 1 Copulatory plugs, restricted mating opportunities and reproductive competition among male garter snakes. Nature, 267:345-346. Biologie de la latence hivernale chez Vipera Camazine, B. and Crews, D. and physiology dur- ing the annual reproductive cycle of the redsided garter snake (Thamnophis sirtalis parietalis). Herpetologica, 38:104-123. Gartska, W. R. and Crews, D. Female sex pheromone 1981. in the skin and cir- culation of a garter snake. Science, 2 4:681683. 1 Gatten, R. 1978. E., Jr. Aerobic metabolism in snapping turtles, Chelydra serpentina, after thermal acclimation. Comp. Biochem. Physiol., 61A:325- 337. Gibson, A. R. and Falls, J. B. Evidence for multiple insemination 1975. in the common Can. 1980. J. garter snake, Thamnophis sirtalis. Zool., 53:1362-1368. C. and Dickinson, J. A. Postural orientation during courtship in the eastern garter snake, Thamnophis s. sirtalis. Gillingham, J. Behav. Neur. 1 974. Biol., 14:358-372. P. T. Gregory, R. R., Interactions of behavior Am., 247(5):159-168. S. 1937. Snakes from an ant hill. Copeia, 1937:142. Devine, M. C. 1975. Copulatory plugs in snakes: enforced chastity. Science, 187:844-845. 1977. Am. Zool., 22:852. (abstr.). H. and Di nson, W. A. Biology and status of timber rattlesnake J. Gartska, W. ( Sci. biological mechanism for determining direction of pheromone trails having low vol- rie rattlesnake. Effects Criddle, A (Crotalus horridus) populations in Pennsylvania. Biol. Conserv., 15:13-58. Crews, D. 1980. (Thamnophis). Copeia, 1982:10-13. B. and Low, J. R. atility. Gannon, 161. B., Species specificity of sex pheromone trails of sympatric and allopatric garter snakes Galligan, B. Movements and temperature relationships of timber rattlesnakes (Crotalus horridus) in northeastern New York. J. Herpetol., 16:151- Camazine, Ford, N. Pyle, D. W., Greene, K. R. and Fried- LAENDER, 1982. 982. 1 73 Patterns of spring emergence of the red-sided garter snake ( Thamnophis sirtalis parietalis) in the Interlake region of Manitoba. Can. J. aspis L. Vie et Milieu, 14:31 1-443. Fitch, H. S. 1 960. Autecology of the copperhead. Univ. Kansas Publ. Mus. Nat. Hist., 13:85-288. 1963a. Natural history of the black rat snake (Elaphe o. obsoleta) in Kansas. Copeia, 1963:649658. 1963b. Natural history of the racer Coluber constrictor. Univ. Kansas Publ. Mus. Nat. Hist., 15:351-468. An ecological study of the garter snake, 1965. 1975a. Aggregations of gravid snakes in Manitoba, Canada. Copeia, 1975:185-186. 1975b. Arboreal mating behavior in the red-sided garter snake. Can. Field-Nat., 89:461-462. 1977a. Life-history of the red-sided garter snake (Thamnophis sirtalis parietalis) in an extreme environment, the Interlake region of Manitoba. Publ. Zool. No. 13, Nat. Mus. Thamnophis sirtalis. Univ. Kansas Publ. Mus. Nat. Hist., 15:493-564. Fitch, H. S. and Glading, B. 1977b. Rare and threatened snake species of Canada. Pp. 122-126. In Mosquin.T. and Such- 1947. A field Zool., 52:1063-1069. Canada. study of a rattlesnake population. and Game, 33:103-123. FitzSimons, V. F. M. 962. Snakes of southern Africa. Purnell and Sons, Capetown, S.A. Ford, N. B. al, 1 1978. Evidence mone 1981. 1 982. for species specificity of phero- two sympatric garter snakes, Thamnophis. Herp. Review, 9:10-11. Seasonality of pheromone trailing behavior in two species of garter snake, Thamnophis (Colubndae). Southwest. Nat., 26:385-388. trails in C. (eds.), Canada's Threatened Species 6, Canadian Nature Federation. Ottawa. Reptilian hibernation. Pp. 53-1 54. In Gans. C. and Pough. F. H. (eds.). Biology of the Reptilia. Volume 13, Physiology D. Academic Press, London. and Habitats. Special Publ. No. Calif. Fish Gregory, 1975. P. T. and Stewart, K. W. Long-distance dispersal and feeding strategy of the red-sided garter snake (Thamnophis sirtalis parietalis) in the Interlake of Manitoba. Can. J. Zool., 53:238-245. SPECIAL PUBLICATION-MUSEUM OF 74 Hawley, A. W. L. and Aleksiuk, M. 1975. Thermal regulation of spring mating behaviour in the red-sided garter snake {Thamsirtalis parietalis). nophis Can. J. NATURAL HISTORY Noble, G. K. and Clausen, H. J. 1936. The aggregation behavior of Storeria dekayi and other snakes, with especial reference to the sense organs involved. Ecol. Monogr., Zool., 53: Sexual receptivity in the female red-sided garter snake ( Thamnophis sirtalis parietalis). 1976. Parker, W. F. 966a. Weight changes and mortality of three species of snakes during hibernation. Herpetologica, 22:8-12. 966b. The ability of two species of snakes to return to a hibernaculum after displacement. Southwest Nat., 1:49-53. Hirth. H. F., Pendleton, R. C, King, A. C. and 1 1 57. Parker, W. 1 974. 1 980. Downard, d-Hb Physiol., Kinghorn, J. R. 964. The snakes of Australia. Rev. 1 ed. Angus and Robertson, Sydney, Australia. Klauber, 972. 1 Lang, J. British Orientation and behavior of the rattlesnake, Crotalus atrox. Copeia, 1973:26-31. Prestt, of the north- An ecological study of the viper, I 'ipera besouthern Britain. J. Zool., Lond., 164: rus, in and mating in the garter snake, Tham- nophis sirtalis parietalis. Am. Zool., 22:859. (abstr.). Mayhew, W. W. Hibernation in the horned soma Columbia. M.Sc. Thesis, Univ. of Columbia, Vancouver. I. 1971. W. atures m'calli. lizard, Comp. Biochem. Phryno- Physiol., 16: 103-119. MOBERLY, W. R. Hibernation in the desert iguana, Dipsosaurus dorsalis. Physiol. Zool., 36:152-160. Naulleau, G. Etude complementaire de Factivite de pera aspis dans la nature. Vie et Milieu, 171 7: Male reproductive der, Vipera berus, 373-418. Rahn, H. 1942. The reproductive Crotalus viridis. cycle of the prairie rattler, Copeia, 1942:233-240. Reinert, H. K. and Kodrich, 1982. W. R. Movements and habitat utilization by the massasauga, Sistrurus catenatus catenatus. J. Herpetol., 16:162-171. Ross, P., Jr. and Crews, D. 1977. Influence of the seminal plug on mating behaviour in the garter snake. Nature, 267:344345. Saint Girons, H. 1957. Le cycle sexuel chez Vipera aspis (L.) dans l'ouest de la France. Bull. Biol, de France Belgique, 91:284-350. Sexton, O. J. and Hunt, S. R. 1980. Temperature relationships and movements of snakes {Elaphe obsoleta, Coluber constrictor) in a cave hibernaculum. Herpetologica, 36:20-26. 461-509. NlLSON, G. 980. facial pit British F. northwestern Minnesota. M.S. Thesis, Univ. of North Dakota, Grand Forks. Licht, P. and Bona-Gallo, A. 1982. Dependence of vitellogenesis on low temper- 1 B. The importance of the 964. Biol., ifornia Press, Berkeley. 1971. 1966. Preston, W. lizard Rattlesnakes. 2nd ed. (2 vols.). Univ. of Cal- Hibernation and movements of Storeria oc- 1963. dormancy in the J. Thermal Lacerta vivipara. Jacquin. 3:183-186. M. cipito-maculata in northern Minnesota. J. Herpetol., 3:196-197. (abstr.). Overwintering of three species of snakes in 1965. tion during winter 1 969. 1 Patterson, J. W. and Davies, P. M. C. 1978. Energy expenditure and metabolic adapta- ern Pacific rattlesnake {Crotalus viridis oreganus) under natural conditions in southern L. Landreth, H. 1973. anoleucus deserticola, in northern Utah. Publ. Biol. Geol. No. 7, Milwaukee Pub. Mus. 130B:293-300. Jolly, G. M. 1965. Explicit estimates from capture-recapture data with both death and immigration — stochastic model. Biometrika, 52:225-245. S. 234-239. Comparative ecology of two colubrid snakes, Masticophis t. taeniatus and Pituophis mel- binding in the snake, Yipera be- Comp. rus. J. and Brown, W. northern Utah. Herpetologica, 29:319-326. Mortality and weight changes of Great Basin rattlesnakes {Crotalus viridis) at a hibernaculum in northern Utah. Herpetologica, 30: 1969. and S. Species composition and population changes in two complexes of snake hibernacula in 1973. 1 T. R. Dispersal of snakes from a hibernaculum in northwestern Utah. Ecology, 50:332-339. Johansen, K. and Lykkeboe, G. 1979. Thermal acclimation of aerobic metabolism S. Population estimates, age structure and denning habits of whipsnakes, Masticophis t. taeniatus, in a northern Utah Atriplex-Sarcobatus community. Herpetologica, 32:53- 976. 1 Copeia, 1976:401-404. Hirth. H. 6: 269-316. 768-776. cycle of the European adits relation to annual and activity periods. Copeia, 1980:729-737. Shine, R. 979. Activity patterns in Australian elapid snakes (Squamata: Serpentes: Elapidae). Herpetologica, 35:1-1 1. 1 VERTEBRATE ECOLOGY AND SYSTEMATICA C. Life-history tactics: a review of the ideas. Stearns, S. 1976. Quart. Rev. Biol., 51:3-47. Trapido, H. 940. Mating time and sperm viability Copeia. 1940:107-109. 1 VllTANEN, 1967. 1 F. 1971. N. and Lasiewski, R. C. Rattlesnake denning: theoretical considerations on winter temperatures. J. Theor. Biol., 30:553-557. in Storeria. Wiklund, P. Wilbi k. W. and Con ins. J. P. Environmental certainty, trophic level, and H. M., Tinkle, D. 1974. resource availability in tion. J. Winter aggregations, breaks size classes in the tree lizard, and relative Urosaurus or- 1 A. M. and Hardy, R. The dens and behavior of the desert 940. Spolia. Zool. 1968. J. W. and Sanders, tortoise. D. Zar, orcutti, Stejneger. lizard. Copeia, 1968: J. 1974. F. L.. G.. Gi issmeyer, Call. A.. Smart. E. R. T. den in Tooele County. Utah. Herpetologica. 7:1-52. 1951. Symposium: Mus. Hauniensis, 5:1-157. Winter behavior of the granite spiny Sceloponts 708-712. evolu- Science. 92:529. H. R.. Heyrend, Structure and seasonal variation of the male reproductive organs of Vipera bents (L.). Weintraub, life history Nat.. 108:805-817. Woodbury. A. M., Vet as, B., Julian, 182-183. VOLS0E, H. 944. Am. Woodbury, nat its (Sauria: Iguanidae). Herpetologica. 30: 1 Fagerstrom, T. 31:153-158. logia, Hibernation and seasonal movements of the viper, 'ipera bents (L.). in southern Finland. Ann. Zool. Fenn.. 4:472-546. tail C. and Whvdo males emerge before females? Oeco- 1977. I Vitt. L. 974. White, 75 a snake H. Biostatistical analysis. Prentice-Hall. Engle- wood Cliffs. N.J. Vertebrate Ecology and Systematics— A Tribute to Henry S. Fitch Edited by R. A. Sergei, L. E. Hunt, J. L. Knight. L. Malaret and N. L. Zuschlag 1984 Museum of Natural History. The University of Kansas. Lawrence ! Parameters of Two Populations of Diamondback Terrapins (Malaclemys terrapin) on the Atlantic Coast of Florida RlC HARD Introduction A. SEIGEL of canals and ditches which are permafilled with water. A more detailed description of the area is presented elsewhere (Seigel series The diamondback terrapin, Malaclemys seem ter- rapin has several attributes which would to make it an interesting subject for ecological study. These include a unique habitat for che- lomans (brackish water), an extremely wide linear range (Massachusetts-Texas), and the distinction of being the "most celebrated of American turtles," a reflection of the popularity of this turtle as a gourmet food item in the early 20th century (Conant 1975). Despite these feaour knowledge of the life history of this species remains surprisingly limited. Studies of terrapins in the wild have dealt mainly with retures, nently 1979). For the purposes of this study, terrapins were collected primarily from the northern ends of two lagoons, known locally as the Indian and Banana rivers (Fig. 1). Indian River turtles were collected by deploying small mesh (maximum diameter = 6 cm) gill nets along a narrow canal bordering a dike road. Two nets were set perpendicular to the shoreline to block off a 100 section of the canal. Turtles moving up and down m the canal became entangled in the nets and were removed within two hours of capture. Turtles from the Banana River were collected by walking production (Finneran 1948; Reid 1955; Burger and Montevecchi 1975; Montevecchi and Burger 1975; Burger 1976a, 1976b, 1977; Auger and surveys around a small man-made spoil island. Turtles were captured by hand while they basked Giovannone 1979; Seigel 1980b. 1980c). epizoic fouling (Jackson and Ross 1971; Ross and Jack- along the shoreline, or while they swam and fed in the clear waters surrounding the island. son 1972; Jackson et al. 1973), mortality (Seigel 1978, 1980a) and hibernation (Lawler and Musick 1972; Yearicks et al. 1981). Data on the population biology of Malaclemys are few, es- under natural conditions. Cagle (1952) reported growth rates and age at maturity for Louisiana intergrades (M. t. pileata x littoralis), pecially and Hurd et al. (1979) described the size structure and population size of M. t. terrapin from Delaware. Most data on population biology are based on captives (Hildebrand 1929, 1932; Allen and Littleford 1955), and must be viewed with caution due to the unnatural conditions under which the turtles were maintained (Carr 1952; The following straight-line measurements were 0. 1 cm using vernier calcarapace (CL) and plastron (PL) length, length of the right abdominal scute, and medial length of visible abdominal annuli. Wet body recorded to the nearest ipers; weight was recorded to the nearest 1 g with a spring balance. All turtles were given an individual mark (Ernst et al. 1974) and released at point of capture. Plastral annuli have been used to estimate the rate and age of several species of turtles, using a variety of techniques (see Graham 1979 for review). In my study, age was estimated using growth method of Sexton (1959). Growth was estimated using Sergeev's 937) formula of L,/L 2 = C,/C 2 where C, represents the annuli length, C 2 the abdominal scute length, L, the plastron length when the annuli was formed, and L 2 the current the Burnley 1969). From 1977 to 1979 I studied the life history and ecology of the Florida east coast terrapin. ( 1 , M. t. tequesta, at the Merritt Island National Wildlife Refuge, Brevard County, Florida. This plastron length. Since large, female Malaclemys >16 cm PL often lacked one or more annuli. paper presents data on the growth rates, population structure, and age at maturity for two populations of Malaclemys under natural condi- they were excluded from this analysis. Statistical tests follow Ott (1977). Means are followed by ± one standard deviation. tions. Materials and Methods Results and Discussion The Merritt Island refuge consists of three large, brackish water lagoons, each surrounded by a Growth and Sexual Maturity. — One hundred Malaclemys were examined from the thirteen 77 SPECIAL PUBLICATION-MUSEUM OF 78 NATURAL HISTORY 80° 40' Fig. site is 1. Merritt Island National Wildlife Refuge. Shaded areas represent lagoonal waters. Indian River study cross-hatching. Banana River study site by cross. shown by VERTEBRATE ECOLOGY AND SYSTEMATICA 79 12- 8 - E o X —r- H 3 4 T ~r- 3 4 2 O cr h- 20 i < _J CL 16 12- 8 H nr 6 5 AGE Relationship between age and plastron length River. Vertical bars represent sample range. Fig. 2. Indian River and 44 from the Banana River. Fifty-three of the Indian River turtles bore distinct growth annuli, but heavy shell damage from barnacles (Seigel 1983) obliterated most annuli on terrapins from the Banana River. Ontogenetic change in the relative size of the abdominal scute, such as that noted by Moll and Legler 97 1 ) for ( 1 Pseudemys scripta, was minor in this study < 2%), so no correction factor was needed. Fig. 2 shows the relationship between age and plastron size. The wide variability in size within tropical ( in 53 female and 13 male Malaclemys from the Indian a particular age class observed in Malaclemys frequently occurs in other turtles (Gibbons 1 968: Ernst 1971. 1975. 1977: Growth of Plummer 1977b). and two years of life, but begins to diverge after age three, when male growth rates decline, but females continue to grow at a steady rate. The curve for both sexes shows a marked the sexes similar for the is relatively constant first decline in growth as sexual maturity is reached (see below). Fig. 3 shows the relationship be- tween percent growth/year and plastron size. Most SPECIAL PUBLICATION-MUSEUM OF 80 100 NATURAL HISTORY -| 80 £ > LL! 60 - h- < DC \- 40 - <> o DC CD 4) II 20 - T I) II -L .L _L <> I 1 3.5 5.5 1 1 1 1 7.5 1 1 9.5 15.5 13.5 11.5 PLASTRON LENGTH (CM) Relationship between growth rate (%/year) and plastron length bars represent sample range. Fig. 3. rapid growth occurs at PL 3-3.9 cm, followed by and then a more gradual decline a sharp decrease, growth to <5%/year in mature individuals. This pattern is similar to that of most other freshwater turtles, especially the genera Chrysemys and Pseudemys (see Bury 1979 for review). Limited data from turtles recaptured after six months or more support the above growth estimates. Two mature females of 13.8 and 14.6 cm PL grew at annual rates of 5.4% and 2.9% respectively. Six mature females of > 1 5.0 cm PL grew at a mean annual rate of 2.2% (range = 0-7.1%). Based on these values, the largest female in the Indian Rivin er population (PL = 17.7 cm) would be approx- imately 15 years old. Longevity in this population is estimated to be about 20 years. Fig. 4 compares the PL/age relationships of Malaclemys from different parts of the range. Florida Malaclemys grow at a slightly faster rate for Indian River Malaclemys. Vertical than terrapins from North Carolina or Louisiana (Cagle 1952). Although Florida Malaclemys are larger at hatching than turtles from the other populations (Seigel 1980c), this difference tial size is insufficient to account for the in ini- differ- 4. Gibbons ( 1 967) showed that even populations of Chrysemys picta varied widely in growth rates because of differences in ences in Fig. local local feeding habits and food quality. Most data suggest that the feeding habits of Malaclemys are relatively similar throughout its range (Cagle 1952; Wood 1976; Hurd et al. 1979; R. Seigel, Cochran 1976), with no comparable dramatic differences such as Gibbons 967) noted. It therefore seems unlikely that the differences in growth rates seen in Fig. 4 are due to differences in local feeding habits. However, the North Carolina turtles were captives, and were fed fish as supplements to their normal food pers. obs.; but see ( 1 VERTEBRATE ECOLOGY AND SYSTEMATICS 81 14 12-| ?o E10 o z LLt _l Z 8- o cc l(/) < c 4 3 14 5 J J AGE Fig. 4. Comparison of growth rate of Malaclemys from different parts of the range (sexes combined). See text for data sources. (mollusks), so their growth may have been somewhat affected. The differences in Fig. 4 may reflect M. I. the longer activity and growing season of tequesta, which at Merritt Island is active from mid-February to late November (Seigel, unpub. data), whereas North Carolina captives were only active from May to October (Hildebrand 1932). No data on the activity season of Louisiana terrapins are available, but from a climatic viewpoint, it is probably more similar to Florida than North Carolina. The smallest female showing evidence of sexual maturity (oviducal eggs or corpora lutea) was 13.5 cm PL, and all females > 14.0 cm PL were mature. Fig. 2 shows that most females reach 13.5-14.0 cm by age four, but that some may not attain maturity until age five. The smallest male considered mature (based on secondary sexual characteristics and enlarged testes) was 9. 1 cm PL, and all males >9.5 cm PL were consid- ered mature. According to Fig. 2, males may reach this size as early as the second year of life, but most probably do not mature until age three. Hildebrand (1932) suggested that sexual matu- Malaclemys was related to size rather than and my results support this idea. Table shows the size and age at maturity for Malaclemys from different parts of the range. Size at rity in age, 1 maturity is rather uniform for both sexes, while age at maturity is more variable. Bury (1979) summarized the data on growth and sexual maturity for freshwater, mainly northtemperate turtles, and made the following conclusions: 1) males often mature earlier and at a smaller size than females; 2) growth is most rapid before maturity is reached; 3) in temperate regions, individuals in southern populations mature earlier than northern conspecifics: 4) sexual SPECIAL 82 Table 1. Size range. Subspecies (locality ) and age PUBLICATION-MUSEUM OF NATURAL HISTORY at sexual maturity for male and female Malaclemys terrapin from different parts of the VERTEBRATE ECOLOGY AND SYSTEMATICS 20-i BANANA RIVER N 10 =44 - \° o\ > 2 O r^ LU D O LU DC io - 83 SPECIAL PUBLICATION-MUSEUM OF 84 >o s 40 - NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS movement). Plummer ( 977a) found that temporary movements of Trionyx muticus out of his Kansas study site greatly increased the variability of his population size esbarriers restraining timates. Although showed 1 at Merritt Island Malaclemys relatively long-term (ca. 18 months) delity to a particular area (Seigel 1979), able that short-term both study somewhat sites, it is movements took so the above estimates fi- prob- place at may be biased. These population estimates and the size limits of the two sampling areas were used to construct density estimates. The Indian River sampling area covered 2.27 acres, yielding a density of 178.3 individuals/acre; the Banana River sam- benefit from large body size via increased reproductive potential, whereas males attain only a small body size, but reach maturity earlier than The two study populations females. water turtle populations (Bury 1 differed sig- Banana Rivmore individuals nificantly in size structure, with the er population having relatively This may reflect higher mortality among Indian River females. The sex ratios of both populations were significantly different from 1:1, with females outnumbering males by at least 5:1. Schnabel population size estimates for the Indian and Banana rivers were in the larger size classes. 404.7 and 212.5, respectively, and it appears that Malaclemys may attain a considerable density and biomass pling area was 1.62 acres, yielding a density of 131.1 individuals/acre. These figures are some- what higher than most reports of density 85 in local areas. Acknowledgments in fresh- 979), but are not field was provided by E. Scott Timothy R. Claybaugh, John D. Galluzzo, Mary T. Mendonca, Boyd Thompson, and Sherry Williams. Thanks go to the U.S. Fish and Assistance in the as high as the 239 individuals/acre reported by Ernst (1971) for C. picta. Biomass estimates, based on the above figures and wet body weight Clark, were 390.0 kg/ha for the Indian River, and 355 kg/ha for the Banana River. Both the density and biomass estimates may be somewhat inflated as a result of a) an arbitrary and possibly unrealistically low estimate of the population boundaries, and/or b) the tendency of Malaclemys to Wildlife Service for proved this manuscript. Particular appreciation form goes to my wife, Nadia, for constant help both large aggregations during the breeding season (Seigel 1980b). However, it seems clear that Merritt Island Malaclemys may attain a consid- erable density and biomass in local areas, at least during certain times of the year. and making for logistical support. field sites accessible, I also thank Schlager for statistical advice. views of William Henry The Gunther critical re- Duellman, Carl H. Ernst, Fitch, and an anonymous reviewer im- S. E. and during preparation of the manuto the late James D. Anderson for and script, advice and encouragement. This research was supported by NASA contract NASI 0-8986, to in the field M. L. Ehrhart. Summary The growth Literature Cited age at maturity, population size and population structure of the Florida east coast terrapin, studied from rates, terrapin tequesta were 1977 to 1979 at the Merritt Island Malaclemys National Wildlife Refuge, Brevard County, Flor- Data from two areas (Indian and Banana rivers) are presented. Growth was most rapid ida. immediately after hatching, declining to <5%/ year in mature turtles. Females matured at a plastron length of 13.5-14.0 cm, at an age of 4-5 years. Male terrapins reached maturity at a plastron length of 9.0-9.5 cm, at an age of 2-3 years. Female terrapins attain a much larger body size than do males, with a mean FMR (female to male size ratio) of 148 for length and 313 for weight. Such dimorphism probably reflects divergent reproductive strategies between the sexes; females Allen, and Littleford, R. A. Observations on the feeding habits and growth of immature diamondback terrapins. J. F. 1955. Herpetologica, 11:77-80. Auger, 1 P. J. and Giovannone. P. On the fringe of existence. Diamondback ter- 979. rapins at Sandv Neck. Cape Natur.. 8:44- 58. Berry. J. 1980. F. and Shine, R. Sexual size dimorphism and sexual selection in turtles (Order Testudines). Oecologia, 44: 185-191. Bull, C and Vogt, R. Temperature-dependent sex determination J. J. 1979. in turtles. Science, Burger. 206:1 186-1 188. J. 1976a. Behavior of hatchling diamondback terrapins (Malaclemys terrapin) in the field. Copeia. 1976:742-748. SPECIAL PUBLICATION-MUSEUM OF 86 1976b. Temperature relationships in nests of the northern diamondbaek terrapin. Malaclemys terrapin terrapin. Herpetologica, 32:4 2- Graham, 1979. mondbaek Malaclemys terrapin, in Amer. Midi. Natur.. 97:444-464. and Montevecchi, W. A. toise Soc. Carr, A. 1952. L. A Louisiana terrapin population (Malaclemys). Copeia, 1952:74-76. F. Handbook of Comstock turtles. Publ. As- in Malaclemys Maryland. Bull. t. Md. terrapin (SchoepfF) Herpetol. Soc, 14: 100. 1973. field guide to reptiles and amphibians of ton Mifflin Co., Boston. 429 in southeastern tologica, Hough- Lawler, A. R. and Musick, J. A. 1972. Sand beach hibernation by a northern diamondbaek terrapin, Malaclemys terrapin Moll, E. 1971. p. turtle, Pennsylvania. Herpe- 27:135-141. Montevecchi, W. A. and Burger. 1975. L. C. Diamond-back terrapin in Connecticut. Co- mys 1977. 1 968. J. W. Variation in growth rates in three populations of the painted turtle, Chrysemys picta. Herpetologica. 23:296-303. Population structure and survivorship in the turtle, Chrvsemvs picta. Copeia, 1968:260-268. Sex ratios in turtles. Res. Popul. Ecol., 12: 252-254. painted 1970. terrapin terrapin. An Amer. Mildl. Natur., analysis. 730 p. Overton, W. S. 1969. ratios. introduction to statistical methods and data peia, 1948:138. Fitch, H. S. 1981. Sexual size differences in reptiles. Misc. Publ., Mus. Natur. Hist., Univ. of Kansas 70: 1-72. J. Aspects of the reproductive biology of the northern diamondbaek terrapin. Malacle- 94:166-175. Nichols, J. D. and Chabreck, R. H. 1980. On the variability of alligator sex Amer. Natur.. 116:125-137. Ott, L. 1 1967. J. M. The life history of a neotropical slider turtle, Pseudemys scripta (SchoepfF). in Panama. Bull. Los Angeles Co. Mus. Natur. Hist. 1: O. and Legler, 1-102. Chrysemys pic- 1971b. Population dynamics and activity cycles of Chrysemys picta in southeastern Pennsylvania. J. Herpetol., 5:151-160. 1975. Growth of the spotted turtle, Clemmvs guttata. J. Herpetol., 9:313-318. 1977. Biological notes on the bog turtle, Clemmys muhlenbergii. Herpetologica, 33:241-246. Ernst, C. H., Barbour, R. W. and Hershev, M. F. 974. A new coding system for hardshelled turtles. Trans. Kentucky Acad. Sci., 35:27-28. Gibbons, Malaclemys terrapin Amer. Midi. Natur., 89:495- terrapin, macrospilota. 497. 1 Growth of the painted ta. 1948. Epifaunal invertebrates of the ornate dia- mondbaek Ernst, C. H. Finneran, 9: 551-563. E., Smedes, G. W. and Dean, T. A. A ecological study of a natural population of diamondbaek terrapins (Malaclemys t. terrapin) in a Delaware salt marsh. Estuaries, terrapin (SchoepfF). Copeia, 1972:389-390. A eastern and central North America. 1971a. terrapins size at- and longevity. Zoologica, 2:28-33. CONANT, R. 1975. culture Bull. U.S. Bur. Jackson, C. G., Jr. and Ross, A. 1971. Molluscan fouling of the ornate diamondback terrapin, Malaclemys terrapin macrospilota. Herpetologica, 27:341-344. Jackson, C. G., Jr., Ross, A. and Kennedy, G. soc, Ithaca. 542 p. Cochran, J. D. 1978. A note on the behavior of the diamondbaek terrapin, artificial 45:25-70. Growth of diamondbaek tained, sex ratio 1979. F. R. 1952. York. Fish., 1932. Hurd, 3:32-34. Population ecology of freshwater turtles. Pp. 571-602. In Harless, M. and Morlock, H. (eds.), Turtles, Perspectives and Research. Wiley-Interscience, New York. 1979. Cagle. J.. B. (eds.). Turtles, Per- and Research. Wiley-Interscience, HlLDEBRAND, S. F. 1929. Review of experiments on of diamond-back terrapin. the terrapin Malacle- site selection in mys terrapin. Copeia. 1975:1 13-1 19. Burnley, J. M. Diamondbaek terrapin. Int. Turtle and Tor1969. Bury. R. New dia- terrapin. J. Nest 1975. M. and Morlock. H. spectives Determinants of hatehing success Burger, T. E. Life history techniques. Pp. 73-95. In Harless, 1 418. 1977. NATURAL HISTORY Duxbury Press. Belmont. Estimating the numbers of animals in wildlife populations. Pp. 403-455. //; Giles, R. Wildlife Management Techniques. (ed.). Wildlife Plummer, M. 1 Soc Washington. V. 977a. Activity, habitat and population structure in the turtle, Trionyx muticus. Copeia. 1977: 431-440. 1977b. Reproduction and growth in the turtle, Trionyx muticus. Copeia, 1977:441-447. Reid, G. K. 1955. Reproduction and development in the northern diamondbaek terrapin, Malaclemys terrapin terrapin. Copeia, 1955:310-31 1. VERTEBRATE ECOLOGY AND SYSTEMATICA Ross, A. and Jackson, C. G.. Jr. Barnacle fouling of the ornate diamondback 1972. Malaclemys terrapin macrospilota. Crustaceana, 22: 203-205. Seigel, R. A. 978. Simultaneous mortality in the diamondback Occurrence and effects of barnacle infestaon diamondback terrapins (Malaclemys terrapin). Amer. Midi. Natur.. 109:34- 1983. tions terrapin, 1 terrapin. Malaclemys terrapin tequesta Schwartz. Bull. N.Y. Herpetol. Soc, 14:3132. Reproductive biology of the diamondback terrapin. Malaclemys terrapin tequesta. Master's thesis, Univ. of Central Florida, Orlando. 40 p. 1980a. Predation by raccoons on diamondback ter1979. rapins, Malaclemys terrapin tequesta. J. Herpetol., 14:87-89. 1980b. Courtship and mating behavior of the dia- mondback terrapin, Malaclemys terrapin of Florida. Trans. Kansas Acad. 246. Sci., 39. Sergeev, A. Some 1937. 83:239- materials to the problem of reptilian post-embryonic growth. Zool. 723-735. Sexton, O. J. A 1959. method J. Moscow, 1 6: for estimating the age of painted demographic studies. Ecol- turtles for use in Wood, ogy. 40:716-718. R. C. Outdoors. 3:14-15. 26. R. C. and Johnson, W. S. Hibernation of the northern diamondback 1976. $25 per Yearicks, E. F., 1981. egg. N.J. Wood, terrapin, te- questa. J. Herpetol., 14:420-421. 1980c. Nesting habits of diamondback terrapins (Malaclemys terrapin) on the Atlantic Coast 87 tuaries, Yntema, 1979. Malaclemys terrapin terrapin. Es- 4:78-80. C. L. levels and periods of sex determination during incubation of eggs of Chelydra serpentina. J. Morphol., 159:17- Temperature 28. Vertebrate Ecology and Systematics— A Tribute to Henry S. Filch Edited by R A. Seigel. L. E. Hunt. J. L. Knight. L. Malaret and N. L. Zuschlag 1484 Museum of Natural History. The University of Kansas. Lawrence < An Ecological Study of the Cricket Frog, Acris crepitans Ray D. Burkett The pond Introduction at the Reservation was created by the construction of an earthen The cricket frog, Acris crepitans, subject for population studies since it is a useful is generally abundant throughout most of the year and tends to form separate and distinct populations. It occurs in a variety of habitats, such as along lakes, pond, streams and occasionally tempo- rivers, rary ponds or rain pools and even relatively dry stretches of intermittent streams. Most Acris aggregate on relatively level, bare areas at the water's edge, avoiding steep, vegetation-covered slopes in most instances. Cricket frogs venture into water away from the shore line only when mats of algae are present on the surface. Earlier knowledge of the ecology of Acris was based mainly on short notes summarized by Wright and Wnght (1942). More recent studies include those by Turner (1960b) and Ferguson el al. (1965) on Acris gryllus\ and those by Pyburn (1958, 1961a, 1961b), Blair (1961), Fer- guson ( 1 et al. (1967), Bayless (1969b), Labanick and Johnson and Christiansen 976), ( 1 976), on Nevo (1973a, 1973b) has studied both species and Bayless (1969a) studied sympatric populations of both species. A. crepitans. Some comparisons were made with popula- from other locations, but the main this of study was to determine if there objective were any differences in the ecology of popula- tions of Acris water on its dam impounding northeastern side. Water overflowing pond drains down a stream to the southwest and into a small creek that empties into the Kanthe km east of Lawrence. The maximum circumference of about sas River about four pond has a 435 m. The northeastern end of the pond is shallow and swampy with numerous willows (Salix) along its edge. locust (Gleditsia triacan- Honey thos) borders much of the northern edge and northwestern edge of the dam. The southeastern end of the pond is almost always shaded by large oaks (Quercus velutina), elms (Ulmus ameri- cana) and ash (Fraxinus americanus). Much of the northwestern edge of the pond and dam are bordered by small trees, shrubs, herbs and grass- Algae are common in the pond in a zone from about 0.3 to 0.9 m from shore. For a detailed description of the Reservation see Fitch (1952, es. 1965) and Fitch and McGregor (1956). At the Fish Lab the reservoir is on a southponds are located about facing slope and the 1 90 1 m south of the reservoir. Each pond is drained through pipes that empty into a small stream south of the ponds. The stream continues south it reaches the Wakarusa River, which enters km east of Lawrence. Kansas River about until the 1 1 The reservoir fluctuates considerably in depth since water is used to fill the ponds. The maxi- mum circumference of the reservoir during my study was about 365 m, and the minimum 230 m. The only trees around the reservoir are small tions of cricket frogs in different habitats separated by only a few kilometers. saplings of Populus and Salix, which occur about equal numbers. Description of Study Areas Populations of Acris were studied in and near Lawrence, Douglas County, Kansas, in the fall of 1 96 1 and from fall 1 963 through spring 1 966. Two populations were studied intensively by capture/recapture and toeclipping: one in a wooded pond and stream at the University of in Methods , A total of 2492 frogs were captured at the Natand 1077 were captured at the Fish Laboratory. Owing to the large numbers of individuals that were sometimes ural History Reservation, Kansas Natural History Reservation (KUNHR). about 1 km northeast of Lawrence, and the other in an open reservoir and 1 1 rectangular ponds at the University Fish Laboratory (FL) on the southwestern part of the campus. The Kansas present, frogs were 1 River lies serially rather than m m All individuals captured in each area were given a unique mark for that area and date. Frogs cap- between the two populations as a pos- sible barrier to marked long individually. Areas (not exceeding 100 and 4.5 wide) were marked off at each locality. tured during the gene flow. 89 initial sampling period and sub- 90 SPECIAL PUBLICATION-MUSEUM OF Table 1. Estimated reproductive output for. 4cris crepitans in two populations in northeastern Kansas. See text for explanation. Location NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS 35 91 54 288 361 148 JO I69 30h 890 142 33ri n U) c -J 220 65 274 ' 91 25 47 (D II J J C ^ J 1 i u 32 ir 20 +-» ||. =J o " J- 127 151 II 10 ii JASON J L MAM L L J J J KUNHR 35 141 T 97 56 JZ 74 176 819 30 74 U) c 1c i 35 86 ft <D 25h ill 223 =r |! v 51 -4-> -J \ 20 |i Z3 1 66 46 o c is 10 -II- JASONDMA J I I I I FISH I LAB L I M J J 92 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. Stomach contents of four samples of Acris crepitans. Abbreviations are the same as in Table 3. The average number of items per frog was calculated as the total of each item divided by the total number of frogs in each sample and not by the frequency for each item. Thus it represents the average for the sample, not the average of those frogs feeding on a particular category. VERTEBRATE ECOLOGY AND SYSTEMATICS Table 3. KUNHR 93 Relationship between number of metacercariae and size of frogs in four samples of Acris crepitans. = University of Kansas Natural History Reservation: RET = Rockefeller Experimental Tract: Fish Lab. = University of Kansas Fish Laboratory. Average below the date of collection. size (in mm) of each sample is given in parentheses SPECIAL PUBLICATION-MUSEUM OF 94 NATURAL HISTORY - OH UJ I— UJ 2000 -1500 Z q: w Cl -1000 l/) O cr U_ -500 J A S O N D J F M A M J MONTH Fig. 2. Estimates of density of frogs per meter of shoreline (left), and population estimates (right) at the Estimates of University of Kansas Natural History Reservation and University of Kansas Fish Laboratory. sites. both areas for inhabited assume size equal population vation at this time of year (late September and early October) but were much smaller than frogs viduals captured in later samples. The MLP, then, serves as a basis for comparing other estimates found at the Fish Lab. An inverse relationship was found between the number of metacercariae and frog body size (Table 3). Owing to higher densities and (perhaps) increased competition for food, frogs at the Reservation may have been of population numbers; if those estimates are lower than the MLP. they are known to be too low; if higher, accuracy of the estimates is more difficult to interpret. Since the MLP is based upon more at the susceptible to fluke infestations than frogs Fish Lab. number of individuals captured, it reflects of activity of the frogs as well as effort degree of the collector(s). the Several methods were used to estimate the size of the populations throughout each season. In most instances population estimates were lower than the MLP and were therefore invalid, or they were extremely high and believed to be unreli- be able according to calculations based on other Estimation of Post-Metamorphic Density and Survival The MLP (minimum living population) minimum number of individuals known is to reduce alive at methods. The Modified Lincoln Index in positive bias) produced more usable results than did the Schnabel or Haynes methods (see Smith and most reliable estimates are 1974). The any one time, and can be calculated because the approximate age of all frogs is known. It is based on the number of individuals captured any sample plus all previously marked indi- MLP (to VERTEBRATE ECOLOGY AND SYSTEMATICS 95 MLP— 4. minimum living population at Population estimates of two populations of Acris crepitans. estimate based on density along shore line (no. frogs meter); H — Hayne method beginning of specified period; of estimating population; MLI — modified Lincoln index; S— Schnabel method of estimating population. See text Table D— for further explanation. Date Location KUNHR Pond 1963 Oct. 1260 956 633 Nov. 230 Apr. 170 92 86 27 Sept. Early Late 1964 Oct. May June July Stream 1964 MLP Oct. 256 240 Oct. 120 Pond Mar. 423 1965 Apr. May May 411 318 163 June 91 Early Late Early Late Stream 1965 Sept. July 20 Early Oct. Late Oct. 553 335 216 93 Nov. 1966 Apr. Fish Lab Mar. Ponds Apr. May- 115 100 68 June 14 1965 July Aug. 5 1 Oct. Dec. 1966 Apr. Reservoir 1965 Apr. May June 5 July 4 July-Aug. Early Late Sept. Oct. Oct. Nov. 887 695 649 492 Apr. 163 86 66 30 Mav 24 Dec. 1966 75 47 Mar. Population cstimate(s) 700 H-32 SPECIAL PUBLICATION-MUSEUM OF 96 J 215 NATURAL HISTORY ASONDMAMJ 76 75 827 171 37 56 92 140 11 J 11 = top, brown = middle, green = bottom) for combined Fig. 3. Frequency of the three color morphs (gray samples of Acris crepitans at the University of Kansas Natural History Reservation and University of Kansas Fish Laboratory. Letters at top indicate month; numbers indicate sample size. Broken lines and numbers on right side indicate average values for each color morph. sons: 1 ) small, young frogs are more susceptible to desiccation than larger frogs because of a higher surface/volume ratio; 2) increased density at the time of metamorphosis may have attracted pred- ators that normally feed on a wider variety of species; 3) the young of most predators appear at approximately the same time, resulting in heavier predation on small frogs; 4) greater density at the time of metamorphosis may enhance the spread of disease; and 5) cricket frogs tend to jerk violently when held, in an attempt to escape. Injuries were noticed occasionally, and in many instances these frogs were not recap- tured. It is likely that small frogs received more due to handling than did large frogs. the above possibilities, the first three are fatal injuries Of probably more important in accounting for the rapid disappearance of young frogs in late summer and early autumn. Although most meta- morphosis occurred in July and August, causing a population peak at that time, few juveniles were marked during those months because most field work was concerned with the study of adults. Vegetation became dense at that time, especially at the Reservation pond, causing parts of become inaccessible and allowing more it to frogs to VERTEBRATE ECOLOGY AND SYSTEMATICS Table 97 Comparison of percentage of frogs having different colored vertebral stripes in populations from The average is followed by the extremes in parentheses, then the number in the sample. 5. northeastern Kansas. Green Combined samples from Fish Lab Reservation (open) (wooded) 86 5.0(0-15.0) 26.2 (9.5-35.2) Brown Gray 450 68.8(56.5-84.2)1154 219 10.6(6.2-15.6) 23.9(18.5-49.2) 495 65.5(44.6-70.2) 1358 escape detection. Furthermore, increased rainfall, high temperature and the usual swarms of arthropod pests resulted in decreased efficiency in collecting. Due to rapid recruitment of young into the of adults during and rapid mortality population the latter part of the breeding season, composition of the population shifts constantly at that time. The change from adult populations to those consisting almost entirely of juveniles takes less than a month. Those frogs that metamorphose and since metamorphosis continues for over two months, the range in size of young frogs in any sample is large. Thus, samples taken in autumn do not show distinct size early grow classes, rapidly, even though a few adults may be present. By September, more favorable conditions for what appeared to be the end of a trend in high mortality rate. The number of frogs marked in autumn and recaptured in spring was extremely low, indicating that winter weath- collecting revealed 5.8(0-12.1) only one age class is represented in a breeding population, and those members of a population that survive to breed have all been exposed to relatively similar conditions. 2000 to 6000/year respectively). at the Fish Lab ponds and swamp probably equal that at the reservoir. In both populations, approximately 50% of the frogs alive September die before mid-October. Almost 95% do not survive winter, and the survivors are reduced even more as the breeding season progresses. Less than 0.1% may live into in early life anurans, most common species live through more than one breeding season as adults, and a breeding population consists of animals among representing several age classes, such as in the ( 1 in Blair isiana (Bayless 1 969a). It is such wide variation in the not surprising to find life pattern of Acris crepitans, since this species occupies an extensive Under most circumstances geographical range. natural selection can be expected to favor rapid develoment, and the sooner an organism matures, the better are its chances of reproducing before it dies. However, the climatic conditions and are such in Kansas that sufficient growth maturation cannot occur (at least in females) be- The production of sperm by young early autumn is of questionable value fore winter. in seems unlikely that any adult females would be gravid at that time. If mating did occur, it development of tadpoles could not occur before winter, and chances of survival would be nil. With the relatively short breeding season and the sudden appearance of large numbers of young in northern Kansas, death of the adults removes one of the main sources of possible intraspecific competition for food and may allow for more rapid growth of the young through an increased food supply for each individual. Mortality pattern, consisting of annual turnover, differs from the pattern in most vertebrate animals that have been studied. Even study by Turner ideal 1961), and some adults survive two through breeding seasons in southern Lou- burn the following September or October. This general is the-year may mature and reproduce before the end of the breeding season in central Texas (Py- males to This species for studies of both life patterns, since young-of- since Recruitment 4 11.4(6.7-15.0) 14.3(13.3-15.0) 5 74.3 (70.0-80.0) 26 ever, in populations of Acris crepitans in Kansas, Comparison of the population estimates at both localities indicates that the Reservation population is considerably larger than the Fish Lab compared (wooded) 7 24.8(11.9-45.5)29 69.4(42.5-83.3)81 er contributes appreciably to mortality. (Reservoir) population (5000 to 26,000 egg/year other populations (open) 960a) on Rana pretiosa. How- The causes of mortality between populations be the same (i.e., desiccation, predation, may parasitism, winter kills and natural death), but the specific interactions of each of these are prob- ably unique to each population and also from year to year within a population. Since tadpoles are difficult to find, the causes of mortality of SPECIAL PUBLICATION-MUSEUM OF 98 NATURAL HISTORY Table 6. Number of captures of 2244 cricket frogs in the 1964 and 1965 year classes at the University of Kansas Fish Lab and Natural History Reservation. Elapsed time in months is given in parentheses below the heading for each year Number of class. VERTEBRATE ECOLOGY AND SYSTEMATICA Table 7. Distances moved between captures by cricket frogs in the populations of 1963. 1964. and 1965 (A, B, and C, respectively) at the University of Kansas Natural Historv Reservation. A. 1963 KL'NHR population Number Distance moved in meters 0-7.7 7.8-23.0 23.1-38.2 38.3-53.4 53.5-68.7 68.8-83.9 84.0-99.1 Sample avg. = 20.1 m of 99 SPECIAL 100 Table last 8. PUBLICATION-MUSEUM OF NATURAL HISTORY Movements of cricket place of capture is frogs in the 1965 University of Kansas recorded for frogs captured several times. fish laboratory population. Only the VERTEBRATE ECOLOGY AND SYSTEMATICS morphs that green are protected on certain back- that most distant movements were toward more favorable microhabitats. grounds. As time moved by most C). Howmovements occurred elapsed, the distance frogs gradually increased (Table Movements Of 2244 of the frogs marked, a total of 547 (68.9%) were not recaptured, and the remaining 697 were recaptured from one to five times (Ta1 ble 6). Because of few multiple recaptures it was impossible to establish "preferred" activity ranges of individuals, as shown by Pyburn ( 1958). However, preferences for certain areas ponds are indicated around the some extent by patterns of distribution along their shores. to Dispersal occurred in all directions during and following rains. On one occasion several Acris and one bullfrog were found in a small roadside puddle more than a quarter mile west-northwest of the Reservation pond, which was the closest permanent body of water. Movements discussed hereafter refer only to movements within the study areas, and were measured as distance moved around the ponds between two successive captures. Movements were compared in relationship to and sexual difremain in fairly of the recaptures were distance, time, habitat, rainfall Most frogs tended to between captures is given in Table 9. Although greater distances usually were moved after rain, the majority of frogs were still captured within 7.6 m of the site of previous capture. The average distance moved by frogs from both populations was 20.7 m. Rainfall alone cannot account for long movements; relative humidity, temperature conditions and breeding activity are rainfall also important. Summary Populations of cricket frogs, Acris crepitans, were studied near Lawrence, Kansas. The two major study areas consisted of a pond and stream woodland habitat (Reservation) and a reservoir and eleven ponds in grassland habitat (Fish Lab). Acris is usually the most abundant anuran in near Lawrence, being active from March until or December. November Most spawning occurs from late May to early and newly metamorphosed frogs are found between mid-July and late September. The sex small areas; nearly 50% within 7.6 of the previous place of capture July, (Table 7). The number of frogs moving greater distances steadily decreased. Since group mark- ratio varied m 7B and many of the longest over extremely short periods of time. Pyburn (1958) made similar observations in Texas. Comparisons of distance moved with amount of ever, ferences. 101 from about four females per male in 53% or more males in adults, indi- ings were used on most samples in autumn, the time interval between captures could not be determined for the 1963 year class at the Reser- juveniles to 964 year class at the Reservation, recaptured remained in the area of the pond where they were originally captured (<30 m), 21% moved to adjacent areas, and 16% moved vation population. Estimates of reproductive vation. In the 1 63% to 90 more m distant areas. Movements of more than in this population usually moving from the stream movements occurred during and may have been involved frogs to the pond. These the breeding season in response to calling. Av- movements recorded in the three year classes were: 1963-20.1 m; 1964-25 m; 1965erage 19.5 m. In the 1965 year of the Fish Lab % 61 of the recaptured frogs of the pond where originally moved to an adjacent area, and population (Table remained class 8), in the area 24% 5% moved to more captured, distant areas. Large aggregations at the southwest corner of the reservoir, 1 where soil was almost always damp, indicated cating higher mortality rates in females. The percentage of males was usually higher in the Reserpotential indicate an average annual production of from 5000 to 26,000 eggs at the Reservation pond compared to about 2000 to 6000 at the Fish Lab reservoir and about 7000 at the Fish Lab ponds. Average life expectancy is about four months, about 5% of the population survives the winter, and complete population turnover occurs in about sixteen months. Density was greater at the Reservation, suggesting that the wooded habitat there is more favorable than the grassy habitat at the Fish Lab. Two periods of rapid growth were observed: from July until late September and from March through July. Size and growth rate of females exceeded those of males at all ages, and the Fish Lab population contained frogs that were con- SPECIAL 102 PUBLICATION-MUSEUM OF NATURAL HISTORY siderably larger than individuals in the Reservation population. The small size of individuals at the and to greater competition for food. Three classes of vertebral stripe coloration were distinguished: green, brown and gray. In all populations sampled, gray morphs were predominant (usually accounting for more than 60% of the sample), while brown morphs were less com(about 25%); in open areas green comprised nearly 5%, and 10%. in wooded morphs areas, about Most frogs occupied shore lines having muddy, beach-like areas, and in dry periods they tended to remain in these areas. However, following rains and in mild, humid weather, they dispersed in all directions; movements of more than 100 meters were not uncommon. Nearly Reservation. Univ. Kansas Mus. Nat. Hist. Misc. PubL 4:1-38. movements were less than 8 m, and recorded movements of entire year classes averaged between 19 and 25 m. Move- 1956. ranges, territories, and seasonal vertebrates of the Natural History Reservation. Univ. Kansas Publ. Mus. Nat. Hist., l(3):63-326. 1965. The University of Kansas Natural History Reservation in 965. Univ. Kansas Mus. Nat. Hist. Misc. Publ., 42:1-60. Fitch, H. S. and McGregor, R. L. 1956. The forest habitat of the University of Kansas Natural History Reservation. Univ. Kansas Publ. Mus. Nat. Hist., 10(3):77-127. Johnson, B. K. and Christiansen, J. L. 1976. The food and food habits of Blanchard's cricket frog, Acris crepitans blanchardi (Amphibia, Anura, Hylidae), in Iowa. J. Herpe1 1 tol., eral separate into sev- 1976. Prey availability, consumption and selection the cricket frog, Acris crepitans in phibia, Anura, Hylidae). Blair, W. F. 1961. Calling and spawning seasons in a mixed population of anurans. Ecology, 42(1 ):99— E. 167. Role of temperature and fat deposition in hibernation and reproduction in two species of frogs. Herpetologica, 25:105-113. Burkett, R. D. 1969. An ecological study of the cricket frog, .4cm crepitans, in northeastern Kansas. Ph.D. Thesis, Univ. of Kansas, Univ. Microfilms, No. 69-21497. Ferguson, D. E., Landreth, H. F. and Turnipseed, R. 1972. T. man-made Natural and in the Missouri River basin. Ph.D. Thesis. Univ. Kansas, Univ. Microfilms, No. 72-32920. Savage, R. M. 1962. The ecology and life history of the common frog (Rana temporaria temporana). Hafner Publ. Co., New York, vii -I- 221 pp. Smith, A. K. 1977. Attraction of bullfrogs (Amphibia, Anura, Ranidae) to distress cricket frog, Acris gryllus. Copeia, 1965(1): J. Herpetol., 1 calls of immature frogs. 1(2):234-235. Smith, H. M. 58-66. E., conditions deter- mining the range of Acris crepitans Astronomical orientation of the southern Ferguson, D. Landreth, H. F., and McKeown, Handbook of amphibians and reptiles of Kansas. Univ. Kansas Mus. Nat. Hist. Misc. Publ. No. 9, 2nd. Ed. 1-356. 1956. J. P. 1967. Size Regan, G. J. 1969. M. F. and local movements of a population of cricket frogs {Acris crepitans). Texas Jour. Sci., 10(3):325-342. 1961a. The inheritance and distribution of vertebral stripe color in the cricket frog. Pp. 235-261. In Blair, W. F. (Ed.), Vertebrate Speciation. Univ. Texas Press, Austin. 1961b. Inheritance of the green vertebral stripe in Acris crepitans. Southwest. Nat., 6(3-4): 164— 1958. 110. 1965. (Am- Herpetol., 10(4): 293-298. Nevo, Pyburn, W. Bayless, L. E. 1969a. Ecological divergence and distribution of sympatric Acris populations (Anura: Hylidae). Herpetologica, 25(3): 18 1-1 87. 1969b. Post-metamorphic growth of Acris crepitans. Amer. Midi. Nat., 81(2):590-592. F. J. 1973a. Adaptive color polymorphism in cricket frogs. Evolution, 27(3):353-367. 1973b. Adaptive variation in size of cricket frogs. Ecology, 54(6): 127 1-1 280. ponds. Literature Cited Brenner, 10(2):63-74. Labanicr, G. M. Fish Lab tended to be greater than those at the Reservation, probably as a result of Lab Home movements of at the division of the habitat at the Fish Temperature responses in free-living amphibians and reptiles of northeastern Kansas. Univ. Kansas Publ. Mus. Nat. Hist., 8(7): 417-476. 1958. half of the recorded ments S. The University of Kansas Natural History Reservation was attributed to inhibition of growth by heavy infestations of metacercariae mon Fitch, H. 1952. Sun-compass orientation of the northern Anim. Behav., cricket frog, Acris crepitans. 15(649):45-53. Smith, 1 96 1 P. . W. The amphibians and reptiles of Illinois. Bull, VERTEBRATE ECOLOGY AND SYSTEMATICS Illinois Nat. Hist. Surv.. 28(1): 1-298. 252 figs. Smith, R. 1974. Stewart, 1960. ogy, 41(2):258-268. Row. xii field + 850 biology. 2nd. Ed. Harper pp. P. L. Lung-flukes of snakes, genera Thamnophis and Coluber, in Kansas. Univ. Kansas Sci. Bull.. Vol. Turner, 1960b. Size and dispersion of a Louisiana population of the cricket frog, Acris gryllus. Ecol- L. Ecology and & 103 XLI(8):877-890. F. B. 1960a. Population structure and dynamics of the western spotted frog. Rana p. pretiosa Baird and Girard. in Yellowstone Park, Wyoming. Ecol. Monogr., 30(3):25 1-278. Wendelken, 1978. On P. W. prey-specific hunting behavior in the western ribbon snake, Thamnophis proximus (Reptilia. Serpentes. Colubridae). J. Herpetol. 12(4):577-578. Wright, A. H. and Wright, A. A. 1942. Handbook of frogs and toads. Comstock Publ. Assoc. Ithaca, N.Y. — A Tribute Vertebrate Ecology and Systematica Edited by R. A. Seigel. L. E. Hunt. J. c 1984 Museum to Henry S Fitch Knight. L. Malarel and N. L. Zuschlag of Natural History. The University of Kansas. Lawrence L. Female Reproduction in an Arkansas Population of Rough Green Snakes {Opheodrys aestivus) Michael V. Plummer Introduction micrometer, ovarian follicles, corpora lutea. and oviducal eggs. The diameter of the oviducts was Because of the kinship between lizards and snakes the tendency when reviewing certain aspects of snake ecology is to compare them to lizards (e.g.. Fitch 1970; while lizards". for ecology as a . . Turner 1977). However, may well become paradigmatic whole" (Schoener 1977), such is hardly true for snakes. Indeed the nocturnality, and frequent periods of inactivity of most snakes make them less of an ideal subject than the conspicuous, mostly measured lomic weighed clearly reveal the need for more information on the latter. Although highly cryptic. Opheodrys aestivus a non-secretive, diurnal arboreal abundant, has low vagility, and is snake that is is easily collected 1981). Consequently, some of the attributes of lizards that contribute to successful (Plummer study also subject. make this snake species a favorable Opheodrys aestivus ranges from south- New Jersey to southern Florida, west to eastern Kansas and Texas, and south to southern ern Tamaulipas, Mexico (Conant 1975). In this paper I report on female reproduction from a population in the central part of the species' range. to the nearest .01 g. Percent body fat (g)/body wgt. (g) x 100) was used to control for size-induced variation in fat body weight. In the laboratory snakes were housed ina 1.1 4.9 L x 2.8 H cage. A thin layer of moist wood Wx m shavings was maintained on the floor. Ten 30 x 30 cm plywood boards were placed on the floor under which clutches were deposited. The cage was maintained at 28 ± 2°C and on a 14L:10D photoperiod. Crickets and water were provided ad libitum. On the day of oviposition females and eggs were measured, weighed, the eggs were individually marked with a felt-tipped pen. and each egg was incubated individually on top of a thin layer of moist vermiculite in a small glass jar at 28°C. Some eggs were sacrificed immediately following oviposition in order to stage the embryo according to Zehr 962). On the day of emergence from the egg hatchlings were weighed, measured (SVL) and sexed. They were returned to the field at a later date. Data are reported as ( 1 mean ± 1 SE. Results Sexual Maturity and Mating. — The presence Methods of oviducal eggs, corpora lutea. enlarged ova, or convoluted oviducts indicated sexual maturity. Female O. aestivus (N = 167) were collected May 1977-October 1979 from a population at Bald Knob Lake in White County, Arkansas. Snout-vent length (SVL) and body weight were measured. Cloacal smears were made on 43 snakes during April-October 1979 and were examined for sperm under 100x magnification. One hundred and twenty-seven snakes were preserved and autopsied for reproductive condition. Forty gravid snakes, collected 14 June- 14 July 1979, were returned to the approximate midpoint. Coebodies were removed, blotted, and at their (fat secretiveness. great vagility. diurnal lizards. Reviews of reproductive ecology in lizards and snakes (Fitch 1970; Turner 1977) fat Most snakes mature between 36-40 cm SVL (Ta). The largest immature was 45.0 cm whereas the smallest mature measured 33.5 cm. Immature snakes had straight, narrow (0.5-1 .5 mm), ribbon-like oviducts and follicles <3.0 mm in ble 1 diameter with greater interfollicular distance than mature snakes. Sperm were present in most mature females in spring but were not detected in other parts of the year (Table 2) indicating that mating is limited to spring. No sperm was found in any females <35.0 cm SVL. Ovarian Cycle. — In adults follicles measuring in greatest diameter were present in 1-5 field after their clutch- were deposited in the laboratory. Observations were made on snakes in the field which were not collected. Autopsy included counting and measuring with vernier calipers or an ocular es mm 105 SPECIAL PUBLICATION-MUSEUM OF 106 Table 1 . Opheodrys of females. SVL(cm) aestivus: size at sexual maturity NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICA I 15 CO Ld _l CJ APRMAY 16-31 107 MAY JUN 12 10 16 8 O 2 C7> I- b 2 w=u* tr UJ DD JUL >- AUG io oct HI2 O O CO iioh HP < !±! SEP- I 6h r 2 — -I i i i— I I II _ < +' OE II -8 I I OE 3 II I I OE DIAMETER OF FOLLICLES (mm) Fig. aestivus: Number of various sized follicles and amount of body fat in mature females at of the year. The horizontal lines of the dicegrams are means; vertical lines are ranges: rectangles confidence limits. OE = oviducal eggs. Opheodrys 1. different times delimit 95% were either partially or totally unshelled. For modal embryonic stage oviposition was 25 (N = 37, range totally shelled eggs the (Zehr 1 962) at The range of stages 21-27). was 25-26. For 7 eggs in a single clutch of partially shelled eggs or those that were inviable at or soon after oviposition (determined by the rapid loss of tonicity and growth of mold) the mode was 18 (range = 14-19; N = 10). Statistics relating to egg size are Table 5. given Incubation ranged 36-43 days and averaged in Table 3. Opheodrys aestivus: Number of snakes (>35.0 cm SVL) determined to be gravid by palpation in the field Period during various times in 1978. SPECIAL PUBLICATION-MUSEUM OF 108 NATURAL HISTORY X Ll) i.o- > Q O CD X UJ X o I_J ° 35 40 50 45 SNOUT-VENT LENGTH (cm) Fig. 2. Opheodrys aestivus: The relationship of clutch weight/post-reproductive body weight and snout-vent = .00 IX + .588 (r = .024, P > .75). length for snakes collected in 1979. The regression equation is Y 1.36 the .02, ± 0.23 g; same SVL P> / = .01, (13.9 ± P> .90) and have about ± .10 cm; / = .10, 14.0 .90). Reproductive Effort. — Reproductive effort of a female is that organism's total investment in a current act of reproduction (Pianka 1976). In snakes reproductive effort has been crudely es- timated using the ratio of clutch weight to nonreproductive female weight (C/B) (Clark 1970; Fitch 1975; Pianka and Parker 1975; Shine 1977). no parental care brooding or oviducal retention) most of the Because (egg in O. aestivus there is reproductive investment should be contained in the egg itself and therefore the ratio C/B should be representative of reproductive effort (but see and Congdon 1978). The risks involved in transporting the enlarged ova and eggs in the maternal body are assumed to be negligible. In O. aestivus C/B averages .64 and does not change with body size (Fig. 2). Less than .06% of the Vitt variation in Table 5. C/B is explained by body size. Larger Opheodrys aestivus: Wgt. Shelled eggs Hatchlings 1.62 1.37 Egg and hatchling (g) ± .015(190); 1.17-2.26 ± .016(144); .82-1.76 snakes produce both larger eggs and larger clutch3). There is a possible trend toward small- es (Fig. er eggs with increasing clutch size (Fig. 4) al- though there the regression is 2 great variation (r = 1.7%) and not significant. Larger eggs pro- duce significantly larger hatchlings (Fig. 5). Discussion Opheodrys aestivus appears to have a typical female reproductive cycle for a temperate oviparous snake. From the limited data available for a comparison of geographic variation in reproductive attributes, other reports appear to consouthern Louisiana Morris (1982) found similar results in O. aestivus with regard to size at sexual maturity, the ovarian cycle, and repro- form with Tinkle ( 1 this population. In 960) and in Illinois is limited fall mating ductive potential. Apparently, mating to spring in this population although may occur in other populations (Richmond statistics. All Max. width (cm) 9.9 is ± .04(190); 8.4-11.9 data are expressed as x ± 1 SE (N); range. SVL Max. length (cm) 1956). (cm) 24.8 ± .23(190); 16.2-34.2 ± .07(144); 10.7-16.1 13.9 VERTEBRATE ECOLOGY AND SYSTEMATICS 10- _ LiJ M 8 CO 5 6 h_l o 4 2 r- X - - 1.8 C5 UJ 5J w .6- 1.4 < LxJ 1.2- 109 110 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS 16 1.4 Fig. Y= Opheodrvs .79X + .115 (r= 5. aestivus: .85. P < The 2.0 1.8 EGG WEIGHT 11 (g) relationship of hatchling weight and egg weight. The regression equation is .001). frequency of female reproduction in several snakes (summarized in Wharton 1966; Gibbons 1972). Production of equal numbers of male and female hatchlings of similar size and weight is in accordance with Fisher's sex ratio theory and is the usual situation in snakes (Shine and Bull and O. aestivus (present study) C/B remains constant with body size. In Notechis scutatus and Pseudechis porphyriacus (Shine 1977) C/B debody size. Pianka and Parker 975) creases with ( 1 and Pianka ( 1976) suggested that correlations between reproductive effort and reproductive value might be greater in multiple-brooded species than Because metabolism decreases with body weight in snakes (Galvao et al. 1965) propor- where proximal factors such as resource availability might have a greater effect. In all of the above studies the snakes were more energy may be available for reproThe risks involved when time and en- single-brooded. However, in a study of annual reproductive variation in O. aestivus (Plummer ergy are allocated to reproduction may decrease survivorship and therefore the expectation of fu- 1983) it was shown that C/B and other reproductive attributes did not vary between years in which snakes stored greatly different quantities in single-brooded species 1977). tionally duction. ture progeny (reproductive value). Therefore, a younger snake with a higher expectation of future progeny might be expected to devote less time and energy to reproduction than an older snake which has less expectation (Pianka and Parker 1975; Pianka 1976). Tests of this hypothesis in snakes have shown diverse results. In Carphophis vermis (Clark 1 970) C/B increases with body size(=age). In Diadophis punctatus (Fitch 1975), Masticophis taeniatus (Pianka and Parker 1975), of lipids. Even if reproductive effort remains con- stant with age (as in O. aestivus), the absolute energy allocated to reproduction actually increases. The increased energy available in O. aestivus is reflected in the and production of larger eggs larger clutch sizes (Fig. 3). Fecundity in snakes body size (Fitch 1970; Shine 979; present study). Shine ( 1 978) found that in about 66% of species (including O. is often related to 1977; Aldridge 1 SPECIAL PUBLICATION-MUSEUM OF 112 aestivus) females attain a larger size than body males. Shine suggested that one reason for this disparity was that selection has favored large body sizes in the females because of greater fecundity. might be that larger snakes produce larger eggs which produce larger hatchlings (Fig. 5). In general, larger hatchlings should enjoy higher survivorship and Another reason for increased body size be better competitors (Pianka 1976). In the lizard Sceloporus undulatus (Ferguson and Bohlen 1978) larger hatchlings from late broods enjoy greater survivorship than do smaller hatchlings, but larger hatchlings from early broods have survi- vorship similar to smaller hatchlings. Although to my knowledge there are no comparable data was increased by proshould favor then selection ducing larger eggs either larger parental body size or decreased clutch for snakes, if female fitness Smith and Fretwell 1974), Pianka (1976), and Stewart ( 1979) discuss models which predict that with a constant reproductive effort, an increased ( female size may (range 21-27). Incubation averaged 39 days. The sex ratio of hatchlings was not significantly different from 1:1. Male and female hatchlings are and weight. Reproductive effort change with body size. Larger females produce both larger clutches and larger similar in length = (jc result in either larger clutches or larger sized eggs. These models assume a negative correlation between clutch size and egg .64) did not eggs. Larger eggs produce larger hatchlings. Acknowledgments I thank several students who were involved in various aspects of this project. They are T. M. Baker, F. W. Brown, D. B. Farrar, M. W. Patterson, D. E. Sanders, and M. White. W. B. Rob- erson assisted in the laboratory. R. A. Aldridge, J. S. Jacob, R. Shine and an anonymous reviewer manuM. Groves and J. Huckeba willingly typed numerous versions of the manuscript. I owe made helpful suggestions regarding the script. the sizes (Pianka 1976). NATURAL HISTORY much my to H. S. Fitch who interests in living To him conditions. stimulated and refined organisms under natural this paper, and this entire This study was from in Sigma Xi and part by grants supported volume, is rightfully dedicated. Harding University. weight. In O. aestivus there is no statistical relationship between clutch size and egg weight (Fig. 4). Although the correlation between and egg size is not strong = SVL appears that in this population correlates of female body (/' .46), Literature Cited it size are selection for increased clutch size as well as for increased egg size. Aldridge, R. D. 979. Female reproductive cycles of the snakes Arizona elegans and Crotalus viridus. Herpe1 tologica, Summary Press. Various aspects of female reproduction in Opheodrys aestivus were examined by specimen autopsy and from the study of living snakes in the field and in the laboratory. These snakes ma- 36-40 cm SVL and breed annually thereafter. Ovarian follicles measuring 1-5 mm in diameter are present in mature snakes throughout ture at Rapid yolking of follicles occurs in the spring and ovulation begins in late May. Extrauterine transfer of ova is common. One clutch = 6. eggs). Coelomic fat is produced per year (.v bodies cycle annually and presumably provide energy for vitellogenesis and ovulation. Oviposition occurs in late June and July. Ninety percent of the eggs laid were fertile and 90% of fertile eggs hatched in the laboratory. At oviposition the modal embryonic stage (Zehr 1962) was 25 35:256-261. Anderson, P. The Reptiles of Missouri. Univ. Missouri 1965. Blanchard, 1933. F. 330 p. N. Eggs and young of the smooth green snake, Liopeltis vernalis (Harlan). Papers Michigan Acad. Sci., Arts, Letters, 17:493-508. Carpenter, C. C. 1958. Reproduction, young, eggs and food of Okla- homa snakes. Herpetologica, 14:1 13-1 15. Clark, D. R. 1970. "reproductive effort"' in the snake Carphophis vermis (Kennicott). Trans. Kansas Acad. Sci., 73:20-24. Age-specific worm the year. Conant, R. 1938. The 1 Conant, 1975. Reptiles of Ohio. Amer. Mid. Nat., 20: 1-200. R. A field guide to reptiles and amphibians of eastern and central North America. Houghton Mifflin Co., 429 p. Conant, 1940. R. and Downs, 2nd ed. A., Jr. Miscellaneous notes on the eggs and young of reptiles. Zoologica, 25:33-48. VERTEBRATE ECOLOGY AND SYSTEMATICS Richmond, N. D. Curtis, L. 1950. A case of twin hatching in the rough green snake. Copeia, 1950:232. Ferguson, G. W. and Bohlen, C. H. 1978. Demographic analysis: a tool for the study of natural selection of behavioral traits. In Greenberg, N. and MacLean, P. D. (eds.). Behavior and Neurology of Lizards. NIMH. Publ. No. 77-491. DHEW Fitch, H. S. 1970. Reproductive cycles in lizards and snakes. Univ. Kansas, Mus. Nat. Hist., Misc. Publ., 52:1-247. 1975. A demographic study of the ringneck snake {Diadophis punctatus) in Kansas. Univ. Kansas, Mus. Nat. Hist., Misc. Publ., 62:153. Galvao, P. E., Tarasantchi, J. and Guertzen- STEIN, P. 1965. Gibbons, 1972. Heat production of tropical snakes in relationship to body weight and body surface. Am. J. Physiol., 209:501-506. J. W. Reproduction, growth and sexual dimorphism in the canebrake rattlesnake (Cwtalus horridus atricaudatus). Copeia, 1972:222- 1953. E. McCauley, R. H., Jr. The Reptiles of Maryland and 1945. 25:655-666. Sexual size dimorphism and male combat in snakes. Oecologia, 33:269-277. Shine, R. and Bull, J. J. 1977. Skewed sex ratios in snakes. Copeia. 1977: 1978. 228-234. Smith, C. C. and Fretwell, S. D. 974. The optimal balance between size and number of offspring. Am. Nat., 108:499-506. Smith. P. W. 1961. Amphibians and Reptiles of Illinois. Bull. Illinois Nat. Hist. Survey, 28:1-298. 1 Stewart, coeruleus. Herpetologica, 35:342-350. the District Colubridae). Nat. Hist. Misc., 214:1-1 R. H. Turner, 1977. and embryos, and the evolution of viviparity within the class Reptilia. Biol. Rev.. 52:71- and rhynchocephalians. Pp. 157-264. In Gans, C. and Tinkle, D. W. (eds.). Biology of the Reptilia. Academic Vitt, L. 1978. Age-specific reproductive tactics. Webb, R. G. 1970. Amer. Nat., 109:453-464. 1966. Habitat utilization, diet and movements of Herpetol., 15:425-432. variation in stored lipids 370 Univ. Oklahoma p. C. H. Reproduction and growth in the cottonmouth Agkistrodon piscivorus Lacepede. of Cedar Keys, Florida. Copeia, 1 966: 149-1 6 1 564 a temperate arboreal snake (Opheodrys aes- of Oklahoma. Reptiles Wright, A. H. and Wright, A. A. 1957. Handbook of Snakes. Vol. I. Comstock Plummer, M. V. Annual and Congdon, J. D. Body shape, reproductive effort, and relative clutch mass in lizards: Resolution of a paradox. Amer. Nat. 112:595-608. J. Wharton, E. R. Natural selection of optimal reproductive tactics. Am. Zool.. 16:775-784. Pianka, E. R. and Parker, W. S. 1983. The dynamics of populations of squamates, Press, 1976. tivus). J. F. B. crocodilians. 105. 1981. population of Opheodrys aestivus (ReptilSquamata). Copeia, 1960:29-34. Press. The Reptiles and Amphibians of Alabama. Auburn Univ. Agric. Exp. Station, 347 p. Packard, G. C, Tracy. C. R. and Roth, J. J. 1977. The physiological ecology of reptilian eggs 1975. W. A ia: 1. 1975. PlANKA, R. J. The balance between number and size of young in the live bearing lizard Gerrhonotus Tinkle, D. 1960. of Columbia. Hagerstown. Maryland, 194 p. Morris, Michael A. 1982. Activity, reproduction, and growth of Opheodrys aestivus in Illinois (Serpentes: Mount. of the rough green snake. Herpetologica, 12:325. Sabath, M. and Worthington, R. 1959. Eggs and young of certain Texas reptiles. Herpetologica, 15:31-32. Schoener, T. W. 1977. Competition and the niche. In Gans. C. and Tinkle, D. W. (eds.). Biology of the Reptilia. Academic Press, pp. 370. Shine, R. 1977. Reproduction in Australian elapid snakes II. Female reproductive cycles. Aust. J. Zool., V. Herpetological notes from Southeastern Texas. Herpetologica, 9:49-56. Autumn mating 1956. 1979. 227. Guidrv, 113 and repro- Publ.. p. Zehr. D. R. 1962. Stages in the normal development of the garter snake, Thamnophis sirtalis common duction in green snakes (Opheodrys aestivus). Copeia. 1983:741-745. . sirtalis. Copeia, 1962:322-329. Vertebrate Ecologs and Systematics — A Tribute to Henry S. Fiteh Edited by R. A. Sergei. L. E Hunt. J. L. Knight. L. Malaret and N. L. Zuschlag c 1984 Museum of Natural History- The l niversil\ of Kansas. Lawrence ! Clutch Size in Iguana iguana A. Stanley Introduction Central Panama Rand animals were not distributed normally nor do they suggest that the population can be divided into age classes. Even though animals were col- Iguana iguana, laying up to 6 dozen eggs in a one of the most prolific lizards in the new world. Prized as food and heavily hunted in many parts of its range, its conservation and the possibilities for sustained yield harvesting have been discussed (Fitch et al. 1983). single clutch, is emphasize the extremes, only one feSVL. Because sampling male was below 300 was not random, ranges are probably more aclected to mm curate representations of the population than are means. Female weight is closely correlated with Though reproductive potential is important in any understanding of population dynamics, only one detailed study of the number of eggs which SVL (N = 30, ticularly if logs of both are plotted (N = /• 0.86. P < .001) (Fig. = 1) par- 30, r P < .001). Number of eggs per clutch ranged from 71 (N = 30, mean = 40.6) and was closely = 0.91, female iguanas produce has been published. Fitch and Henderson (1977) for Nicaragua. The present paper describes the size and weight of clutches produced by female iguanas and their relation to female size in Panama. The clutch size and reproductive investment in Iguana iguana is compared with in itively correlated tion of egg P < with female size. number with SVL (N = .001) (Fig. 2) is 9 to pos- The correla- 30, r = 0.78, about equal to that with female weight (N = 30, r = 0.79. P < .001). A better predictor of the number of eggs that a that described for other lizards. female will lay, and one that can be used in the her weight before she has laid her eggs her own body weight plus the weight of her field, is Materials and Methods (i.e., (N = 28. r = 0.88, P < .001) (Fig. 3). The weight of 28 clutches ranged from 84 to 1086 g (mean = 538 g) and is closely correlated with female size (N = 28, r = 0.83. P < .001). Clutch size data were collected from females clutch) caught during the nesting season, late January to early March, between 1968 and 1980 in the vi- Panama City and Gamboa, Republic of Panama. Some females were caught and allowed cinity of Mean egg weight per clutch (clutch weight/number of eggs) ranged from 9.3 to 16.0 g (N = 28, mean = 13.1). Eggs within a clutch appear quite outdoor enclosure, others were killed or found as fresh road kills and the eggs removed from the oviducts. The sample was not randomly selected from the population, rather, because of my interest in the relationship between female size and clutch size, the few females deliberately shot were selected because they were to nest in a large uniform in size. Larger females tended to lay larger eggs but the correlation of mean egg weight to SVL, though significant (N = 28, r= 0.55, P < not as high as the correlations already cited. The single very small female with her verysmall eggs contributes greatly to this correlation; .01), is very large or very small. The following measurements were taken on 30 females: snout-vent if she is excluded the correlation is lower (N = = (number of eggs). In some cases clutch 0.35. .01 < P < .05). The water content of the eggs varied little, 5767% (N = 11, mean = 62.3%). There was no sig- volume, clutch dry weight (oven dried at 105°C) were also measured. Regressions were compared using covariance egg weight or female size. The percentage that the clutch contributed to (Snedecor 1956). the length (SVL), female wet weight without eggs, clutch wet weight before significant hydration and clutch size 27. /• nificant correlation of water content either with combined weight of female and clutch (relmass of Vitt and Congdon 1978) = 28. mean = ranged from 19.7 to 39.9% (N with correlation a weak It shows positive 30.3%). ative clutch Results The results female SVL (N = 28. r = 0.30. .01 < P < .05) which depends heavily on the single small fe- from examination of females with oviductal eggs are given in Table 1 . Sizes of these 115 SPECIAL PUBLICATION- MUSEUM OF 116 Table 1. Clutch NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS 117 2000 x 1000 — x x X x J_ 250 ± ±. 300 1. The so that females of the relationship between weight same weight in two pop- ulations lay about the same number of eggs but have a lower total clutch weight in Nicaragua. egg mass and the investment per not significantly different between the two relative clutch is populations. No other equally extensive data on Iguana iguana clutch sizes has been published. Hirth 7 females ( 1 963) reported SVL and clutch size for from Tortuguero, Costa Rica (Fig. 2). These clutch sizes are intermediate between those of Panamanian and Nicaraguan females of similar sizes. This suggests the possibility of a geographical trend in reproductive strategies that would be worth exploring. However, it does not appear to continue into South America. from 24 Hoogmoed ( 1973) Surinam. Muller (1972) reports clutches from 14 to 70 at Santa Marta, Colombia, and egg weights averreports clutch sizes to 57 in aging 13.0 g (12.4-14.0). Detailed comparison with South American populations awaits more data. Wiewandt (1983) has compared reproductive among iguanine lizards. He distinguished three groups of genera on ecological grounds: 1) those in mainland deserts of tempatterns L WO 350 SNOUT VENT LENGTH, The x - 400 Fig. v x X X x X x IN 450 MM and snout-vent length in Panamanian iguanas. perate North America (Dipsosaurus and Sauromalus); 2) those on dry subtropical islands (Cy- and 3) those in mainland tropical areas (Iguana and Ctenosaura). The mainland tropical group grows the most rapidly, matures earliest, and has the largest clutch sizes and the lowest ratio of egg weight to female weight. clura); In Iguana iguana, though its eggs are small relative to female size, the weight of clutch, relative to that of the female, same as it is is its total about the for the three other iguanine species which Wiewandt gives data (Sauromalus obesus, Cyclura carinata and Cyclura coronuta stejnegeri). The marine iguana of the Galapagos was not classified in his scheme but is extreme for within the iguanines in having very few, verylarge eggs with a high investment per offspring but a low investment per clutch (Carpenter 966). Wiewandt attributes the reproductive pattern 1 in Iguana iguana and Ctenosaura similis to the on young lizards in these species. Tinkle et al. (1970) have reviewed the reproductive strategies of a wide taxonomic and geographical representation of lizards. That survey included few iguanines. or other large tropical herbivorous lizards and it is relatively high predation pressure SPECIAL PUBLICATION-MUSEUM OF 118 80 60 X X CO U3 5/|0 LLl 00 20" + NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS 119 HO.. 30-- oo CD £20-- ++ 10- *t + +4 + ++ — ' lol ' ' —— "" ' 260 300 SNOUT VENT LENGTH AT MATURITY, IN = Fig. 4. The relationship between clutch size and minimum snout-vent length at maturity. + single brooded = J 'vblyrhynchus cristatus, (Carpenter 1966), 2 = Sauromalus obesus, 3 = Cteniguanas (Tinkle et al. 1970), 1 osaura similis, 4 = Cyclura cahnata, 5 = Cyclura coronuta stejnegeri (2-5, Wiewandt 1983), 6 = Iguana iguana from Panama, 7 = Iguana iguana from Nicaragua (Fitch the second group. In contrast to the largely tropical iguanines, most of Tinkle et al.'s single brooded species are temperate in distribution. Fig. 4 plots the relationship between minimum size at first reproduction and mean clutch size for the species that Tinkle et al. included in their brooded group as well as the data for Iguana iguana from Panama and Nicaragua and 4 other iguanines. As Wiewandt noted, Amblyrhynchus cristatus, Cyclura cahnata and C. coronuta stejnegeri and Sauromalus obesus have few eggs for their size. Fig. 4 shows that Ctenosaura similis and Iguana iguana have clutches close to but still slightly below those predicted from the smaller lizards. We do not have enough data to plot other iguanines but those for which single and Henderson 1977). 1978); in part, because a female iguana expends a great deal of energy in traveling to a nest site, digging a nest burrow, defending it, filling the burrow and returning to her home range (Rand and Rand 1976). Even for the clutch itself, clutch mass or calorific content is a better measure of reproductive effort than is number of eggs. We do not have calorific data for iguanas but Ballinger and Clark (1973) and Vitt (1978) have shown that calorific content per unit weight is quite constant for the eggs of a variety of numbers of lizards. Vitt (1978) has shown that the ratio of ca- content of clutch to calorific content of female is similar to the ratio of wet weight of lorific clutch to wet weight of female. That our ratio of dry to wet weight of iguana eggs lies within the we do have some data (Brachylophus faciatus, Sauromalus varius and S. hispidus, Conolophus subcristatus, and Iguana delicatissima) all seem range that Vitt reported for other lizards is support for our assumption that iguana eggs are probably not too different from other lizards in have clutch sizes below those oflguana iguana and Ctenosaura similis and those of the smaller, single brooded lizards. Number of eggs per clutch is an important calorific to parameter in a reproductive strategy. It is not, however, a very good index of reproductive effort (Tinkle and Hadley 1975; Vitt and Congdon content per unit wet weight. For Iguana iguana, in Panama, wet clutch weight averages 30.3% of the total wet weight of female and clutch (relative clutch mass): this is close to the mean of 27.7% that Vitt and Congdon (1978, Table 2) give for 17 much smaller North American iguanid lizards. Not surprising- SPECIAL PUBLICATION-MUSEUM OF 120 Table 2. A comparison between clutch sizes of NATURAL HISTORY Iguana iguana from Panama and Nicaragua. VERTEBRATE ECOLOGY AND SYSTEMATICS Table 2. Continued. 121 SPECIAL PUBLICATION-MUSEUM OF 122 Conservation. Garland York. Hikih. H. 1963. STPM Press. New F. Some aspects of the natural history of Igua- na iguana on a tropical strand. Ecology. 44(3): 613-615. HOOGMOED, M. 1973. Tinkle. D. W., Wilbur, H. M. and Tillev, S. G. 1970. Evolutionary strategies in lizard reproduction. Evolution, 24(l):55-74. Tinkle, D. W. and Hadlev, N. F. 1975. Lizard reproductive effort: caloric estimates and comments on 427-434. S. Notes on the herpetofauna of Surinam. IV. The lizards and amphisbaenians of Surinam. W. Junk. The Hague. I-IX: 14-19 pp. Vim, 1 131. Pianka, 1 975. E. R. S. Age-specific reproductive tactics. Rand. W. M. and Rand, A. 976. Amer. Nat.. J. J. and Congdon, Herpetol., 12:65-72. J. D. Body shape, reproductive effort, and relative clutch mass in lizards: resolution of a paradox. Amer. Natur.. 12(985):595-608. WlEWANDT, T. A. S. Agonistic behavior in nesting iguanas: a stochastic analysis of dispute settlement dominated by the minimization of energy cost. Z. Tierpsychol. 40:279-299. Statistical methods. Iowa State Univ. Press, Ames. Iowa. Evolution of nesting patterns in iguanine lizG. M. and Rand, A. S. (eds.). Iguanas of the World: Behavior, Ecology and Conservation. Garland STPM Press, New York. Williams, G. C. 1966. Adaptation and natural selection. Princeton 1 983. ards. In Burghardt, Snedecor, G. W. 1956. Vitt, L. 1978. 1 and Parker. W. 109:453-464. 1 Caloric content of lizard and snake (Reptilia) eggs and bodies and the conversion of weight to caloric data. Ukologische und ethologische studien an Iguana iguana L. (reptilia: Iguanidae) in Kolumbien. Zoologische Bertrage N. F. 8: 109— evolution. Ecology, 56: its L. J. 1978. MlLLER, H. 1972. NATURAL HISTORY pp. 1-534. University Press, Princeton, 1-307. New Jersey, pp. Vertebrate Ecology ami Systematics— A Tribute to Henry S Fitch Malaret and N. L. Zuschlag Edited by R. A. Seigel, L. E. Hum. Knight. he 'inversus of Kansas. Lawrence 1984 Museum of" Natural Histoiv I I I I .. I Are Anuran Amphibians Heavy Metal Accumulators? Russell J. Hall and Bernard M. Mulhern Methods Introduction Concern about heavy metals in the environment has increased recently, partly as a result of increased awareness of their potential effects, and also because of the prospect of fossil fuels in expanded use of processes that release metals. From time to time amphibians have been examined as possible indicators of contamination by heavy metals. Their habitats, abundance, and ease of sampling have made them convenient subjects for such purposes. There have been indications that amphibians may be unusual in their ability to accumulate metals. A survey of the copper content of the livers of a wide range of vertebrate species (Beck Adult amphibians collected at the Patuxent Research Center were generally obtained from the Island Marshes, artificial habiWildlife waterfowl management, or they were picked up on service roads on wet nights. Larval amphibians and fish were captured by seine or tats built for dip net from Harding Spring or Mabbott ponds: both are shallow, moderate-sized artificial ponds surrounded by wooded areas. The research cenis not known to be contaminated by heavy metal residues. Iron is naturally abundant in the soil and groundwater. Possible alteration of Harding Spring Pond by runoff from a nearby landfill has led us to undertake a program monter 1956) indicated extremely high levels (up to 1 640 in Bufo marinus; average copper concen- itoring organochlorine and heavy metal levels in ppm) certain animals found in different areas of the trations in livers were generally much lower in other species, although one species of marine fish had higher average concentrations. Surprisingly center. high concentrations of lead in the livers of some frogs from a remote and apparently uncontam- also analyzed for inated area were reported by Schroeder and Tipton (1968). Gale el al. (1973) found up to 1590 ppm (dry weight) of lead in tadpoles from a con- 5 g portion was weighed into a crucible for heavy metals analysis. A separate 5 g portion was weighed into a round-bottom flask to determine mercury levels. Digestion for mercury analysis used the method described by Monk (1961). Mercury was determined by cold vapor atomic iens) Health Laboratory, Madison. Wisconsin was heavy metal levels. Tissue samples were homogenized in a blender life and a taminated area and these results suggest that they have a much greater ability to concentrate environmental lead than do the other species sampled. The iron content of one sample of Rana absorption spectrophotometry using the method of Hatch and Ott (1968) with a Coleman model catesbeiana tadpoles analyzed in our own laboratory reached the startling level of 19.000 ppm MAS-50 mercury analyzer. The lower limit of reportable residues was 0.02 ppm. The sample used to determine other metals was dried in an (dry weight). It is the purpose of this paper to present data from our own work at the Patuxent Wildlife Re- search Center (PWRC) and oven and then charred in a muffle furnace where the temperature was raised to 550°C at the rate of 100°/hr and left overnight. The cooled ash was dissolved over a hot plate in approximately 2 ml ml of concenof concentrated nitric acid and trated hydrochloric acid, transferred to a 50 ml polypropylene centrifuge tube, and diluted with dionized water. Analysis was by flame atomic information from the literature in order to ascertain A sample of 10 leopard frogs (Rana pipobtained from the National Fish and Wild- whether the con- centrations of metals in amphibians fall outside the normal range of variation of other animals. 1 This paper will assemble data which may bear on the questions of whether amphibians are par- heavy metal pollution, whether they can accumulate levels which may be hazardous to their predators, and whether they can be of value as monitors of heavy metal con- absorption spectrophotometry with a Perkin-Elmer model 703 equipped with a deuterium arc tamination. a ticularly susceptible to background corrector, an AS-50 autosampler. and PRS-10 printer. The lower limit of reportable 123 SPECIAL PUBLICATION -MUSEUM OF 24 Table 1. Cadmium in amphibians. NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICA Table Copper 2. in 125 amphibians. PPM Sample Tadpoles (sp.) Area Tissue Wet weight Dp. ueight Reference SPECIAL PUBLICATION-MUSEUM OF 126 Tabi f 3. Lead in amphibians. NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICA Table 4. Mercury in amphibians. 127 SPECIAL PUBLICATION-MUSEUM OF 128 Table 5. Zinc in amphibians. NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS Table 6. Other elements in amphibia. 129 SPECIAL PUBLICATION-MUSEUM OF 130 and Pickering 1958). Pendleton and Hanson NATURAL HISTORY (1958) looked at cesium- 137 uptake in a variety of organisms following addition of 6 pCi/ml of seen in Brungs' data on adult bullfrogs; they accumulated much lower levels of all the metals than did tadpoles, and less than detritivores such the radionuclide to the water of a concrete-lined as crayfish. pond. They analyzed concentration factors after approximately 90 days when levels of Cs-137 in Relatively high concentrations of Cs- 37 were found in tadpoles in ponds experimentally dosed by Pendleton and Hanson (1958), but the levels were lower than those reported in sunfish, shrimp, and adult frogs. These comparisons were based on data collected some months after the addition of Cs-137 to the system. The authors stated that tadpoles are among organisms which take up the the water had apparently stabilized. They found high concentrations in bullfrog tadpoles, with the bulk of the material stored in the gut fraction. higher levels were found in spadefoot toad (Scaphiopus hammondi) tadpoles and adult bullStill frogs. Comparison of Anuran Amphibians with Other Animals Brungs (1963) published a number of useful comparisons of the abilities of aquatic animals to take up radionuclides. The highest recorded tissue levels of Co-60, Zn-65, Sr-85, and Cs-137 were all recorded in tadpoles. Somewhat lower levels were found in pelecypods (Co-60, Zn-65, Sr-85) and gastropods (Cs-137). Concentrations in bluegill sunfish and carp tended to be much lower except for Zn-65 and Sr-85 which tended to accumulate in bone. One possible explanation for the high body burdens in tadpoles is their relatively large gut capacity and the chance that a large part of the metals recorded was in the gut cavity and had not actually been assimilated. Separate analyses of gut and the remainder of the carcass confirmed the presence of high levels in the gut fraction, but, with the exception of Sr85, body remainders still had greater accumulations that most other animals. Also of interest is the fact that the highest levels of radioactivity in tadpoles occurred relatively soon after exposure; other species usually took longer to reach maximum levels and they maintained high levels longer than did the tadpoles. Brungs suggested that the high levels of radio- nuclides recorded were the result of the vertical distribution of the contaminants in the experimental ponds and the tendency of tadpoles to feed on fine sediments. Shortly after addition to the aquatic system, the radionuclides become at- tached to fine particles and 1 metal rapidly, accumulating it faster than do adult fish, frogs, or seed plants, but Pendleton and Hanson (1958) did not present specific data on the speed of uptake by tadpoles. The apparent differences between these results and those of Brungs are due to the different time spans between dosing and observation; both the relative amounts of Cs- 1 37 in tadpoles compared to other animals and its absolute concentrations declined as the time after dosing approached 80 days (Brungs 1963). Most of the lead, zinc and copper in tadpoles from a lead-contaminated area were in the gut (Jennett et al. 1977). However, concentrations in the rest those in of the body tended to be higher than fish from the same waters. These contents with those of bass and bluegills indicates an approximate 10-fold greater concentration of the three metals in the amphibian samples, also contention that feed(1963) supporting Brungs' ing habits produce the higher levels in Getz et al. (1977) compared lead in different freshwater animals in urban and rural areas. They pointed out that lead levels were higher when the animals (fish and invertebrates) were more closely associated with silt substrata; analyses showed that the uppermost layers of sediment were highest in lead. Getz et al. uptake. They believed high levels before various processes had distributed the contaminants more generally through- fish collected produced the high levels observed in tadpoles is (1977) concluded that physical contact with silt and the direct ingestion of lead in silt and detritus were important in tration did not occur. out the the system. Support for the assertion that feeding habits rather than physiological factors amphib- ians. bottom. Tadpoles consume them there and accumulate settle to the results support the idea that the uptake of the metals is through the diet. Comparison of tadpole gut and that food chain concen- Pooled samples of tadpoles of two species, and from two nearby ponds, are comThese results do not closely corpared in Fig. respond to those metal levels reported by Gale et al. (1973) and Jennett et al. (1977) nor the 1 . radionuclides documented by Brungs ( 1963) be- VERTEBRATE ECOLOGY AND SYSTEMATICA 3i H B G Mg 100- 1- M 300i 200 2- 2 o- Cu Pb 131 M P B G M P B G SPECIAL PUBLICATION-MUSEUM OF 132 groups of animals are much less than those re- ported earlier. The concentrations of metals re- ported in Fig. 1 seem to show some real differ- them NATURAL HISTORY excellent indicators of contaminated en- vironments. Metals transported into an aquatic ecosystem would first collect in sediments where tadpoles could accumulate them, as has been sug- ences, but they indicate that conditions favoring the uptake of specific metals do not always result gested in the case of lead (Getz in the greatest nett et uptake occurring in tadpoles. Presumably the availability of metals to the different animals and their potential for uptake are influenced by the habits of the animals (see al. et al. 1977; Jen- 1977). Residual metals in uncontam- perhaps because our areas were essentially uncontaminated and had stable levels of most of inated areas, or those which have been in the ecosystem for some time, should tend to become more widely dispersed (Brungs 963) and to produce patterns similar to those seen in samples analyzed in our laboratory. Thus because of their apparent tendency to selectively accumulate those metals adsorbed to surface sediments, it might be possible to use tadpoles .to identify ongoing contamination. the metals rather than a single treatment (Brungs 963) or a continuous (Jennett et al. 1977) influx Acknowledgments Steinwascher 1978, 1979) and the distribution of the metals within the environment. Distribution of metals in the ponds seems to differ from that in the systems examined by other authors, 1 1 of contaminants. The result would be a greater dispersion of the metals and less tendency for tadpoles to accumulate them. This apparent ten- dency for tadpoles to selectively take up contam- inants which have only recently entered an aquatic system, or which enter on a more or less contin- uous basis, would seem to make them good H. M. Ohlendorf and C. Brand collected some of the samples. D. Brown and P. McDonald helped with preparation of the manuscript. Drafts of the manuscript were reviewed by E. H. Dustman, and J. C. Lewis. Conclusions Adult amphibians of certain species can accumulate extremely high levels of copper in the liver. It seems likely that dietary imbalances or metabolic factors, rather than high environmental levels, result in this accumulation. It has been shown that some anurans are protected from these high copper levels, but individuals with such accumulations may be toxic to their predators. There is little evidence that adult amphibians can 1 ) concentrate other metals to a greater extent than other vertebrates. Tadpoles accumulate high levels of certain metals, including lead, zinc, copper, cobalt, ce- sium, strontium, iron, and manganese, because of their contact with them in sediments and sus- pended particles. There is Literature Cited Bec k, A. B. 1956. The copper content of the liver and blood of some vertebrates. Aust. J. Zool., 4:1-18. Brungs. W. A. which indicates that these organisms are susceppoisoning by metals. Doubtless their unusual powers of accumulation can sometimes re3) The apparent tendency for tadpoles to pick up metals from surface sediments might make relative distribution of multiple radionuclides in a freshwater pond. Ph.D. Thesis. Ohio State Univ., Columbus. Ohio: 97 p. R., Kosta, L. and Stegnar, P. The occurrence of mercury in amphibia. En- Byrne. A. 1975. viron. Lett., 8:147-155. Dmowski, K. and Karolewski, M. A. 1979. Cumulation of zinc, cadmium and lead in invertebrates and in some vertebrates according to the degree of an area contamination. Ekologia Polska, 27:333-349. Domby, A. H., Paine, D. and McFarlane, R. W. 977. Radiocesium dynamics in herons inhabiting a contaminated reservoir system. Health Physics, 33:523-532. 1 Dustman, 1 970. E. H., Stickel, L. F. and Elder, J. B. wild animals: Lake St. Clair 1 970. Pp. 46-52. In Hartung. R. and Dinman, B. D. (eds.). Environmental Mercury Contam- Mercury in ination. Ann Arbor Michigan. Fleischer, M., Sarofim, A. Hammond, tible to metals in tissues reaching toxic levels. The 1963. extensive literature, not reviewed here, on the toxic effects of metals on amphibians and other aquatic vertebrates sult in Beyer, in- dicators of environmental contamination. 2) W. N. P., Sci. Publ., F., Ann Fassett, Shacklette, H. Arbor, D. W., T., Nisbet, C. T. and Epstein, S. Environmental impact of cadmium: A review by the panel on hazardous trace subI. 1974. stances. Environ. Hlth. Perspectives. Exp. Issue No. 7:253-323. VERTEBRATE ECOLOGY AND SYSTEMATICS Gale, N. 1973. L., Wixson, B. G.. Hardie, M. G. and Jennett, J. C. Aquatic organisms and heavy metals in Missouri's New Lead Belt. Water Resour. Bull.. National Resear( h Coun< 1977. 1979. 9:673-688. Getz, Haney, A. Q., Larimore, R. W.. McNurney, J. W., Leland. H. V.. Price, P. W.. Rolfe, G. L., Wortman, R. L., Hudson, J. L., Solomon, R. L. and Reinbold, K. A. Transport and distribution in a watershed L. L., 1977. ecosystem. Ch. 6 In Boggess, W. R. (ed.). Lead in the environment. National Science RANN Foundation Program NSF/RA 770214. Schiller, B. and Sternwieb, I. Copper in hepatocyte lysosomes of the toad, Bufo marinus L. Nature, 228:172-173. Goldfisc her, 1970. S., Hatch, W. R. and Ott. W. Ireland, M. Pasanen, 1974. Rolfe, G. 1977. PHILL, D. D., Gale, N. and Tranter, L. W. H. Transport and distribution from mining, milling, and smelting operations in a forest ecosystem. Ch. 7. In W. R. Boggess (ed.), Lead in the Environment. National Science 1977. Foundation RANN Program NSF/RA 770214. LOVETT. R. H., D.. Lisk. D. and Harris, E. J. A 1972. GUTENMANN, W. J., Youngs, W. survey of the total 406 fish ters. J. from 49 J.. PAKKALA, I. S.. Burdick. G. E. cadmium New York Fish. Res. Brd. content of State freshwa- Canada, 29:1283- W. and Pickering, D. J. C. porary. Nature, 182:1242-1243. Monk, H. 1961. E. Recommended methods of analysis of pesfood stuffs. Report by the Joint Mercury Residues Panel Anal., 86:608614. ticide residues in Comp. Environmental contamination by lead and Illinois Urbana-Champaign. Studies in the biochemistry of copper. XXX. Seasonal changes in the amount and distribution of copper in tissues of the cultivated bullfrog. Japan. J. Med. Soc, 4:65-69. trace elements in man: Arsenic. Chron. Dis., 19:85-106. Schroeder, H. A., Balassa, J. J. and Tipton, I. H. 1962a. Abnormal trace elements in man: nickel. J. Chron. Dis., 15:51-65. 1962b. Abnormal trace elements in man: chromium. J. Chron. Dis.. 15:941-964. Schroeder, H. A. and Tipton, I. H. The human body burden of lead. Arch. En1968. vir. Hlth., 17:965-978. Singh, K. 1978. Serum iron level of the common Indian frog Rana tigrina Daud. Experientia. 34:433-434. Steinwascher, K. The effect of coprophagy on the growth of 1978. 1966. Abnormal J. Rana caiesbeiana tadpoles. Copeia. 1978: 130-134. Direct absorption of dissolved strontium-90 and yttrium-90 by tadpoles of Rana tem- 1958. L. Schroeder, H. A. and Balassa, J.J. 1290. Lucas, Rana temporaria RANN Pollut., 2:85-92. C, Wixson, B. G.. Lowsley, I. H.. Jennett. PURUSHOTHAMAN, K... BOLTER, E., HEM- frog, Energy, 18:419-422. L., Haney, A. and Reinbold, K. A. Univ. Sarata, U. Xenopus laevis fed increasing levels of lead-contaminated J. 248 p. S. and Koskela, P. Seasonal changes in calcium, magnesium, copper and zinc content of the liver of the other heavy metals, Vol. II. Ecosystem analysis. Final Rept. National Science Foundation Program, Inst, for Envtl. Stud. in toads earthworms. Environ. . Biochem. Physiol., 48A:27-36. Pendleton, R. C. and Hanson, W. C. 1958. Absorption of Cesium-137 by components nd United of an aquatic community. Proc. 2 Nations Conf. on Peaceful Uses of Atomic 1938. P. Lead retention 1977. ii Copper. National Academy of Sciences, Washington, DC. 115 p. Iron. University Park Press. Baltimore. MD. common L. Determination of sub-microgram quantities of mercury by atomic absorption spectrophotometry-. Anal. Chem., 40:2085-2087. 1968. 133 Competitive interactions among tadpoles: Responses to resource level. Ecology. 60: 1172-1183. Wagemann, R., Snow, N. B.. Rosenberg, D. M. and Lutz, A. Arsenic in sediments, water and aquatic bio978. ta in lakes from the vicinity of Yellowknife. Northwest Territories, Canada. Arch. Environ. Contam. Toxicol., 7:169-191. 1979. 1 Part III Feeding and Behavior Vertebrate Ecolog> and Systematics— A Tribute to Henrj S Fitch Edited by R. A. Seigel, L E Hunt. Knight. I. Malarct and N. L. Zuschlag IV84 Museum of Natural Historv The University of Kansas. Lawrence .1 I • Energetics of Sit-and-Wait and Widely-Searching Lizard Predators Robin M. Andrews Introduction The foraging tactics of insectivorous lizards, of most other predators, appear to be like those dichotomous (Pianka 1978). In North America, iguanid lizards exemplify the "sit-and-wait" tactic in which prey are sought passively from a fixed perch site. Sight of a elicits ambush moving prey item or pursuit. Teiid lizards, on the other hand, exemplify the "widely-searching" tactic in which prey are sought actively while the lizard moves through the habitat. These two tac- represent a fundamental means of partitioning the food niche (Pianka et al. 1979). Each tics may apparently gives maximal foraging efficiency (in time or energy units) under conditions of varying prey abundance (Norberg 1977) or structural configuration of the habitat (Stamps tactic 1977). The and Gorman 1979). In contrast, (macro) tends such as Cnemidophorus not only have high searching costs (Bennett and Gleeson 1979). but their foraging tactic is associated with preferred nett and the widely-searching taceach associated with an "adaptive syndrome" of predator characteristics (Eckhardt 1979). In addition to characteristics strictly related to foraging, the adaptive syndromes of igsit-and-wait body temperatures of about 40°C (Asplund 970; and Gorman 1979). Thus, 1 Schall 1977; Bennett for many iguanids (notable exceptions are desert Holbrookia and Callisaums), the lizards such as metabolic cost of foraging is low compared to low levels that of teiids not only because of the of activity associated with the sit-and-wait tactic but because of low activity temperatures. The major objectives of this study were to an- What are the relative enswer two questions: ergy intakes of lizards using sit-and-wait and widely-searching tactics when both forage in the 1 ) same habitat? 2) Does the proportional alloca- tion of assimilated energy to production and metabolism differ for lizards using the sit-and-wait tactic and the widely-searching tactic? tics are uanid and teiid lizards differ markedly Lizard Subjects and Field Sites Field studies were conducted in the Chiricahua Mountains of Arizona. The in several ways. Iguanids have more stereotyped responses to novel items in their environment than do teiids and Cnemidophorus exsanguis (parthenogenetic. Cole and Townsend 1977). These species are an comparative studies of feeding behavior and energetics. First, they are broadly ideal pair for Iguanids escape predators by cryptic behavior and, once discovered, by the use of known routes sympatric in oak-pine-juniper woodland. Second, they are of similar size; females of both species reach a maximum weight of about 20 g. Third, their ecology is comparatively well known (Simon 1975; Congdon 1977; Schall 1977; Ruby on rapid escape their predators (Vitt and Congdon 1978; Schall and Pianka 1980). Clutch size per unit body weight is higher for iguanids than for flight to teiids (Vitt were Sceloporus jarrovi (viviparous, Goldberg 1971) (Regal 1978). Iguanids are strongly territorial, teiids lack home range defense (Stamps 1977). to hiding places. In contrast, teiids rely lizard subjects and Congdon 1978). 1977; Ballinger 1979). Observations were made from July to 8 of fat reserves the At this 1979. time, August Although many aspects of the adaptive syndromes of sit-and-wait and widely-searching predators have been described, the energetic costs and benefits of each tactic are unknown. For example, the low searching costs of the sit-and-wait tactic are often associated with relatively low preferred body temperatures. Sceloporine and anoline iguanids have preferred body temperatures of 35°C or less even in well insolated environments (Blair 1960; McGinnis 1966; Andrews, unpublished data; Huey and Webster 976; Ben- 1 1 both species are increasing rapidly (Goldberg 1972; Schall 1978). Since energy stored by female lizards prior to winter inactivity contributes directly to the development of offspring or eggs that be produced the following spring (Hahn and Tinkle 1965; Gaffney and Fitzpatrick 1973), en- will ergy available for fat storage is directly related to the reproductive effort of both S. jarrovi and C. exsanguis. Moreover, adult S. jarrovi females. 1 137 138 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 1. Prey items used in laboratory feeding experiments, their dry weights, ash contents, and proportional representation in the feeding regimes of the lizard subjects. Drv Prey taxa wt. (%) (%) S jarrovi 59 Coleoptera Tenebrio molitor, adult 34.0 Dermestes caninus, adult Dermestes caninus, larvae Phyllophaga sp., adult Chaulognathus pennsylvanicus, adult Lepidoptera (various moths) Orthoptera Blattella germanica. adult Ash male C. exsanguii 61 VERTEBRATE ECOLOGY AND SYSTEMATICS their lengths were estimated roughly at mm 5 intervals. A major assumption of this method of estimating food intake is that lizards are active every day and that they defecate regularly. From observations made near my site B, Simon and Middendorf 1976) found that the percent of adult S. jarrovi active every day was 75% in July and Table 2. Prey items of Sceloporus jarrovi and of Cnemidophorus exsanguis in July-August 1979. Proportion of total prey August. Thus, the assumption of daily probably valid for S. jarrovi but has in activity is and C. exsanguis maintained in large cages under simulated field conditions suggest that defecation occurs at least every morning following the attainment of preferred body temperatures (see also Cowles and Bogert 1944). Ash contents of faeces, urinary wastes, and the 5. jarrovi given for each species followed by in parentheses. Pre\ laxa S \UTTOVi unguis Coleoptera .306(5-10) .518 (<5) (adults) Formicidae .189(5-10) .500(5-10) Lepidoptera 0.0 (adults) .122(10-15) Hymenoptera not been tested for C. exsanguis. Observations on is modal length category ( 100% 139 .082(5-10) .023 (<5) .023(10-15) .047 (<5) (adults) Araneida Orthoptera Miscellaneous* 0.0 .067 (<5) .078(10-15) .044(5-10) 90 * 2 S. jarrovi: 4 Homoptera-Hemiptera; C. exsanguis: Homoptera-Hemiptera. 1 mantid. 1 Chilopoda. various prey types used in the laboratory experiments were measured by heating samples for 1 h at 550°C materials is in an ashing oven. The mass of all presented as ash-free dry weight. Activity Periods and Body Temperatures by both frewere probably the most lizard species (Table 2). Judging quency and size, beetles important component of the diets of both S. jarrovi and C. exsanguis. Orthoptera were probably the second most important component of the 5. jarrovi of S. jarrovi and Lepidoptera were probably the second most important component of the all diet of C. exsanguis. Any lizard seen was considered active. Because individuals were readily found during daylight hours. I assumed that their activity diet Ants were not used in the period potentially spanned 10-12 h. In contrast. C. exsanguis individuals were encountered most feeding experiments although they comprised about half of the items eaten by both species in frequently in the morning. To define the activity period of C. exsanguis, a series of 30-minute censuses was conducted on 3 and 4 August. All individuals encountered while I slowly walked the through site A were counted. Body (cloacal) temperatures (T b ) were measured immediately after capture with a Schultheis quick-reading thermometer. Temperatures of lizards which avoided capture for more than about 2 km minute were not taken to avoid bias. Shaded air temperatures were taken at m and at cm above the place where the lizard was first seen. field. Because of their small of C. exsanguis in New Mexico. Medica (1967) also found the major items (by volume) to be beetles and Lepidoptera. with Hymenoptera (mostly ants) to be relatively unimportant. Food intake of field-collected S. jarrovi and C. exsanguis females was estimated as a 1 I, Results 0.83 and where S. jarrovi Various species of beetles made up about 60% of the insects eaten by lizards in the feeding experiments and moths and cockroaches made up the other 40% (Table This particular feeding 1 -= F*CFF-'*W- 1 I„ Food Intake by Free- ranging and C. exsanguis size (bulk) their contribution to total energy intake was probably low. Using stomach contents to evaluate the diet ). regime was similar to the natural diets of the two I, and = U*CFU-'*W-°- 83 Iu are the respective estimates of food intake based on faecal and urinary produc- F and U are faecal and urinary production (mg dry wt) during the 48 h of confinement, respectively. CFF and CFU are the factors which convert F and U to food intake for faecal and 3 is urinary production, respectively, and tion. W live body weight in g raised to a power of 0.83 to adjust for weight specific metabolic rates (Ben- SPECIAL PUBLICATION-MUSEUM OF 140 Table 3. Daily food intake (I, and u of S. jarrovi and C. exsanguis based on production of faecal and I ) urinary material (sec text for details). In Spec ies and I, In I„ ± SE (mg-g-°" d ') ± SE (mg'g~°-"-d" site jarrovi— A jarrovi— B C. exsangi<is—A 2.58 2.72 S. S. 3.06 ± 0.122 ± 0.067 ±0.128 (Table NATURAL HISTORY 3, Fig. 1) were statistically significant (P < 0.05, analysis of variance). A posteriori tests showed that C. exsanguis females had a signifi- < 0.05) greater I, and I u than both the populations, and that the S. jarrovi populations did not differ from one another for cantly (P S. jarrovi > 0.05, Duncan's multiple range I, or I u (P Therefore, in subsequent analyses the data for S. jarrovi females have been combined. The either 2.94 ± 0.097 tests). and Dawson 1976). This latter procedure elimated a positive and significant relationship nett between I, or I u and W. CFF and CFU were determined as CFF = F lab /I two estimates of food intake for C. exsanguis differed by only 4% on a In scale (Table 3) and by only 1% on an arithmetic scale. Since the correlation between U and I 3 for S. jarrovi was not statistically significant, I u was not deter1 mined. 3 Activity Periods and CFU = U lab /I 3 where F lab and U lab are the respective production of faeces and urine during the 48 h confinement following the laboratory feeding experiments and I 3 is the food intake on the final (3rd) day of the feeding experiments. I, was used to determine CFF and CFU because correlations between F both species and between U and I, for C. exsanguis were statistically significant (P < 0.05). Correlations between F or U and sum- and I, for mations of food eaten on the last 2 days and the total 3 days of the feeding experiments were generally not significant, and all had lower correlation coefficients (r) than did the correlations be- tween F and I 3 and U and I,. Since neither CFF nor CFU varied as a function of lizard weight for either species (P > 0.05), mean values were used to estimate food intake. Respective mean and C. exsanguis were CFF = 0.56 ± 0.053 and 0.24 ± 0.024 and CFU = 0.2 ± 0.027 and 0. 9 ± 0.027. Natural log transformations were used to normalize the Some I, and the I u data for statistical analyses. female-sized male S. jarrovi were included in the analyses (5 of 9 and 9 of 29 individuals on Sites values ( ± SE) for S. jarrovi Scleoporus jarrovi and C. exsanguis differed considerably in the apparent length of their activity periods. I made observations from about 0800 to 1730 h with comparable times spent the field in the morning and A and B, respectively) since the faecal production of these males did not differ from females on (P > 0.05, two-tailed /-tests). Although individuals were captured at various times during the day (see below), regression analyses indicated that time of capture was not related to in in the afternoon. The number of S. jarrovi individuals observed in the morning and the afternoon was very similar. In contrast, C. exsanguis individuals were active primarily in the morning; only 3 of 37 individuals collected were caught in the after- noon. The census data also indicated that peak was in the morning (Table 4). Body temperatures of S. jarrovi were depen- activity dent on weather conditions (Figs. 2 and 3). On A where temperatures were measured under sunny conditions, S. jarrovi individuals main= tained relatively constant T h s (Mean ± SE oneB about on site In contrast, ± 34.2 0.36°C). half of temperature measurements were taken site under overcast or intermittently cloudy condiAt these times, T b s averaged 31.1 ± 0.6 PC. tions. 1 1 either and Thermoregulation During sunny conditions T bs averaged 35.8 ± 0.34°C. Body temperatures of C. exsanguis were in- dependent of ambient temperatures (Fig. 1), averaging 40.0 ± 0.3 PC. The one individual with a T b of 34°C had probably just emerged from a burrow. site food intake (P > 0.05). Differences in I, and I u among the C. exsanguis females and the two populations of S. jarrovi Discussion During the July-August study period, 5. jarhad a sig- rovi females, using sit-and-wait tactics, nificantly lower intake of food than did C. exsan- guis females which were using widely-searching VERTEBRATE ECOLOGY AND SYSTEMATICA 141 100i CO 00 60 o E A A A a* • A A 20 A A 4 a A* A 10 Fig. 20 eg) Food intake based on faecal production (I ) of field-collected Sceloporus jarrovi (Site A. filled circles, and Cnemidophorus e.xsanguis (Site A, open triangles) as a function of body weight. 1. r Site B. filled triangles) The tactics. result is particularly interesting in was active more than twice as long as C. e.xsanguis. Moreover, since lizards were foraging in the same habitat, individuals of both species potentially had the same kinds and abunthat S. jarrovi dances of prey available to them. Thus, the widely-searching tactic appears to be more efficient both in terms of time spent and energy acquired. In order to and compare the energy C. e.xsanguis females duction, jor * 15 W and A. A I, and Iu I I is for pro- metabolism of a 12 g and my sites were located within a few km of one another, I have used his July-August determinations directly. The field metabolism of C. e.xsanguis was estimated from metabolic data collected on Cnemidophorus murinus, a West Indian species. Metabolic rates of C. murinus were determined under standard conditions for both resting individuals and for individuals moving were partitioned into their ma- = R + P + FU food intake. R is 4. Numbers of Cnemidophorus individuals seen during 30 min censuses conducted on 3 and 4 August 1979. Shaded air temperatures m above ground are shown for the time the census was began. Both days were sunny. Table metabolism. P is 1 production, and urinary wastes. was determined FU is the combined faecal and The energy value of food intake as times 5800. the mean caloric I value for a variety of insects (Griffiths 1977). Digestion and assimilation efficiencies of small insectivorous lizards are quite similar (Harwood 1978; Johnson and Lillywhite 1979). Therefore. FU was estimated as 20% of I for both S. jarrovi and S. jarrovi female are from Table 3 and Appendix A. Table 1. of Congdon (1977). Since his Ash Spring site as components where that S. jarrovi have available to estimate field C. e.xsanguis (Johnson Andrews and Asato 1977). and Lillywhite 1979; The parameters used Census period SPECIAL PUBLICATION-MUSEUM OF 142 NATURAL HISTORY 45 A A ZA AA A A O o A A A A A & LU DC Z) h- 35 • • < T DC LU Q_ ~ v> 4 6. 25 010, <P *-o S) LU 25 -7-1 10 —r- I- 4 12 11 1 5 TIME at 1 m (filled circles) and of Cnemidophorus exsanguis (open triRegression lines represent the relationship between shaded air temperature and time (in hours, e.g.. 1200. 1300. etc.). Body temperatures of Sceloporus jarrovi Fig. 2. angles) on site A as a function of time. for each lizard capture site speeds at which foraging normally takes place (Bennett and Gleeson 1979; Bennett and Gor- at man 1979). In order to extrapolate their results I adjusted metabolism for dif- to C. exsanguis, ferences in exponent body weight using b = 0.705 in the relationship as the between metabolic rate and weight and I adjusted for differences in temperature using a Q 10 of 2.64 (both values from Bennett and Gorman 1979). Judging from field observations of activity periods, the daily progression of soil temperatures in Cnemidophorus burrows, and the body temperatures of two individuals observed in burrows during the day, I calculated metabolism assuming an individual 27°C 22°C (night). In addition, since the metabolic rates determined murinus were measured on fasted indifor to be active 4 h at 40°C, inactive for 6 h at (day), and inactive for 14 h at C of S. jarrovi these females lute caloric intake 71% located 60-90% fewer calories to production than did The prediction that widely-searching lizard predators might have a greater proportion of their energy budgets allocated to metabolism appears to be incorrect, at least for the S. jarrovi-C. exsanguis comparison. The explanation involves two species. Although locomotion of Cnemidophorus lizards is energetically costly (Asplund 1970; Bennett and Gleeson 1979), this activity by C. exsanguis was confined to a 3-4 h period in the morning. The rest of the day was spent in burrows and under activity patterns of the rocks where body temperatures presumably ap- proximated soil temperatures (27°C). In contrast, active S. jarrovi individuals maintained temperatures of about 35°C for count for the additional metabolic increment due to recent feeding (Andrews and Asato 1977). body temperatures of I C exsanguis females. applied a correction factor of 1.8 (day) and 1.6 (night) for inactive individuals to acviduals. al- fewer calories to metabolism and the day. To some at least 8 body h during extent, then, the relatively C low exsanguis when they were inactive during the day compensated for Energy budgets of S. jarrovi and C. exsanguis females were similar in their proportional allocation of food energy to metabolism (59 versus the high costs of activity and thermoregulation that were incurred over a relatively short period. 54-61%) and production (22 versus 19-26%) (Table 5). However, because of the lower abso- exsanguis than by S. jarrovi production by females means that both growth and fat storage The greater absolute allocation of energy to C VERTEBRATE ECOLOGY AND SYSTEMATICA 143 40 P LU 0C Z) < 8 30 DC LU Q_ t.O&l .0^ ^ ?v* V LU h- 20 10 12 11 TIME Fig. 3. Body temperatures of Sceloponis jarrovi on site B. Filled triangles represent temperatures taken under sunny conditions and open diamonds represent temperatures taken under overcast or intermittently sunny conditions. Regression lines as in previous figure (for sunny conditions only). should occur at at a greater rate for C. exsanguis This prediction cannot be tested present because of a lack of information on than S. jarrovi. C. exsanguis. However, the prediction is at least consistent with the observation that C. exsanguis becomes inactive in late August or early Septem- ber (Schall 1978; C. J. Cole, pers. comm.) while 5". jarrovi females continue to fatten during Sep- tember and October simultaneously with the initiation of yolk deposition (Goldberg 1972). During the time frame of this study, the widelysearching tactic was more efficient than the sitand-wait tactic. Even with a restricted foraging period, C. exsanguis individuals had a greater daily energy intake than S. jarrovi individuals. This result suggests several approaches for future Table 5. Energy budgets (cal/d) for 12 g female S. and C. exsanguis. See text for explanation of how each compartment was estimated. jarrovi SPECIAL PUBLICATION-MUSEUM OF 144 48 h confinement. The relationship between food intake and faecal and urinary production was established by measuring the faecal and urinary production of lizards fed known quantities of normal prey items. Active metabolism was estimated from recently published studies on Cne- midophorus minimis and Food S. jarrovi. intake of C. exsanguis females was sig- (P < 0.05) than that of 5. jarRespective food energy values calculated for 12 g females were 972 and 669 cal/ nificantly greater rovi females. d. Thus, the widely-searching foraging tactic was efficient than the sit-and-wait foraging tac- more tic. Although C. exsanguis and S. jarrovi had metabolism (54 and 59%) and to production (26 and 22%), C. exsanguis had a greater absolute allocation of energy to metabolism (528 versus 387 cal/d) and to production (250 versus 149 cal/d) than did S. jarrovi. Thus, the high metaby Biology of the Reptilia, Volume 5 (Physiology A). Academic Press, New York, New York, USA. Bennet, A. F. and Gleeson, T. T. 1979. Metabolic expenditure and the cost of C. exsanguis fe- males was associated with a greater absolute allocation of energy to metabolism than for 5". jarrovi females which foraged passively. The greater absolute allocation of energy to production by C. exsanguis than by S. jarrovi females suggests that both growth and fat storage should occur at a greater rate for C. exsanguis than S. Oecologia, 42:339-358. Blair, W. F. The rusty lizard: A population study. Univ. 1960. Texas Press, Austin. Cole, Ch. J. and Townsend, C. R. Parthenogenetic reptiles: new subjects for laboratory research. Experientia, 33:285-289. 1977. Congdon, jarrovi: a I montane lizard Sceloporus measurement of reproductive ef- Ph.D. fort. thesis, Ann al, Cowles, R. 1944. A Arbor, Mich. and Bogert, C. M. B. preliminary study of the thermal require- ments of desert like to initiate my thank Dr. C. studies J. guilds of insectivorous birds in the Colorado Rocky Mountains. Ecological Monographs. 49:129149. Gaffnev, 1 973. F. G. and Fitzpatrick, tigris. Copeia, 1973:446-452. R. S. Reproductive cycle of the ovoviviparous iguanid lizard Sceloporus jarrovi Cope. Her- Goldberg, 1972. L. C. Energetics and lipid cycles in the lizard, Cne- petologica, 27: Cole for helping David Johnston for assistance in the field. This study was funded with a grant from the American Am. Museum ECKHARDT, R. C. 1979. The adaptive syndromes of two 1971. on Cnemidophorus and reptiles. Bull. of Natural History, 83:261-296. midophorus would Arizona State University, Tempe. University Microfilms Internation- Goldberg, me D. J. Energetics of the 1977. jarrovi. Acknowledgments for- aging in the lizard Cnemidophorus murinus. Copeia, 1979:573-577. Bennett, A. F. and Gorman, G. C. 1979. Population density, thermal relations, and energetics of a tropical lizard community. a similar proportional allocation of food energy to bolic cost of active foraging NATURAL HISTORY 1 23-13 1 . R. S. Seasonal weight and cytological changes in the fat bodies and liver of the iguanid lizard Sceloporus jarrovi Cope. Copeia, 1972:227232. Philosophical Society. Griffiths, D. 977. Caloric variation in Crustacea and other animals. Journal of Animal Ecology, 46:59.3605. 1 Literature Cited Andrews, R. M. and Asato, T. Energy utilization of a tropical lizard. Comparative Biochemistry and Physiology, 58A: 57-62. Asplund, K. K. 1970. Metabolic scope and body temperatures of 1977. whiptail lizards (Cnemidophorus). tologica, Ballinger, R. E. 1979. 26:403-41 Herpe- 1. Hahn, W. 1965. and Tinkle, D. W. adaptive significance to ovarian folin the lizard Via stanshuriana. Jour. Expt. Zool.. 158:79-86. for its licle Harwood, 1978. development R. H., Jr. The of temperature on the digestive of three species of lizards, Cnemidophorus tigris. Gerrhonotus multicarinatus. and Sceloporus occidentalis. Ph.D. Thesis, Univ. Calif, Los Angeles. University Microfilms International, Ann Arbor, Mich. effect efficiency Intraspecific variation in demography and history of the lizard Sceloporus jarrovi, life along an altitudinal gradient in southeastern Arizona. Ecology, 60:901-909. Bennett, A. F. and Dawson, W. R. 1976. Metabolism. Pp. 127-223. In Gans, C. (ed.). E. Fat body cycling and experimental evidence Huev, R. 1976. B. and T. P. Webster Thermal biology of Anolis lizards in a com- VERTEBRATE ECOLOGY AND SYSTEMATICS plex fauna: the cristatellus group on Puerto Rico. Ecology, 57:985-994. Johnson, R. N. and Lillywhite, H. B. 1979. Digestive efficiency of the omnivorous lizard Klauberina nversiana. Copeia. 1979:431- 1 Ri in. D. E. Winter activity in Yarrow's Spiny Lizard. Sceloporus jarrovi. Herpetologica. 33:322- 1977. 333. Sc H \1 I . Ginnis, S. M. 966. Sceloporus occidentalis: preferred body temperature of the western fence lizard. Science. logica, 1978. 152:1090-1091. Medic a, P. habits, habitat preference, reproduc- and diurnal activity in four sympatric species of whiptail lizards (Cnemidophorus) in South Central New Mexico. Bulletin So. tion, Academy Calif. Sciences, 66:251-276. An Row, New York. Huev, R. B. and Lawler. L. R. Niche segregation in desert lizards. Pp. 6715. In Horn, D. J.. Stairs, G. R. and Mitchell, R. D. (eds.). Analysis of Ecological Sys- 1979. 1 tems. Ohio State University Press, Colum- bus. Regal, 1978. . 1980. 1978:108-116. and Pianka, E. R. Evolution of escape behavior diversity. American Naturalist. 15:551-566. J. J. 1 and per and E. R., i 1975. ecological theory on foraging time energetics and choice of optimal foodsearching method. J. Anim. Ecol.. 46:51 1529. PlANKA, E. R. 1978. Evolutionary ecology. Second Edition. Har- Pianka, Schai Simon, C. A. Norberg, R. A. 1977. 33:261-272. Reproductive strategies in sympatric whiptail lizards (Cnemidophorus): two parthenogenetic and three bisexual species. Copeia. A. Food 1967. J. J. Thermal ecology of five sympatric species of Cnemidophorus (Sauna: Teiidae). Herpeto- 1977. 437. M( 145 Ecology. 56:993-998. Simon, C. A. and Middendorf. G. A. Resource partitioning by an iguanid lizard: 1976. temporal and microhabitat aspects. Ecology. 57:1317-1320. Stamps. J. A. 1977. Social behavior and spacing patterns in hzards.Pp. 265-334. In Gans. C. and Tinkle. D. W. (eds.). Biology of the Reptiha. Vol. 7 (Ecology and Behavior A). Academic Press. New York. Tinkle. D. W. and Hadlev, N. F. 1975. P. Behavioral differences between reptiles and mammals: an analysis of activity and mental capabilities. Pp. 183-202. In Greenberg, N. and MacLean, P. D. (eds.). Behavior and Neurology of Lizards. N.I.M.H. Rockville. Marvland. influence of food abundance on territory size in the iguanid lizard Sceloporus jarrovi. The Lizard reproductive effort: caloric estimates and comments on its evolution. Ecology. 56: 427-434. Vitt, L. J. and Congdon. J. D. 978. Body shape, reproductive effort, and relative clutch mass in lizards: resolution of a paradox. Amer. Natur.. 112:595-608. 1 Vertebrate Ecology and Systematics— A Tribute to Henr> S Fitch Edited h\ K V Scigel. L. E Hunt. J L Knight. L. Malaret and N. L. Zuschlag c 1984 Museum of Natural Hislon. The I'nncrsin of Kansas, Lawrence Feeding Behavior and Diet of the Eastern Coral Snake, Micrurus fulvius Harry W. Grefne With few exceptions Introduction pids; Voris et Snakes are prominent predators in many terrestrial, aquatic, and tropical marine communities, and exhibit some unusual morphological and behavioral modifications for this role. They rely heavily on chemical senses for locating food (Burghardt 1970; Chiszar and Scudder 1980) and their usual method of locomotion (lateral unis energetically more efficient than teand Taylor 1973). Perhaps (Chodrow trapody most importantly, these "limbless tetrapods" possess an extremely flexible jaw apparatus that dulation) permits the ingestion of large prey items without the assistance of limbs or mastication (Gans 1961). Although many species swallow prey alive struggling, others immobilize it by constric- and venom combination of these methods (Gans 1978; Greene and Burghardt tion, 1978; It injection, or a Kardong is now 1980). clear that venom delivery systems grades of structural comcomprise have evolved indepenthese that and plexity dently in several lineages of snakes (see Gans and at least three Gans 1978; Savitzky 1978; 1980; Kardong 1980; Cadle. in press, for extensive discussion and reviews). Opisthoglyphs (many species of colubrids) possess enlarged, grooved teeth on the pos- ends of otherwide normal, elongate, toothed maxillae. Proteroglyphs (elapids and hydro- terior phiids) have one or two enlarged, canaliculate, anterior teeth on each short, nonmobile or slight- know that these snakes often strike also Greene. MS). paper I provide a description of an ecological characterization behavior, feeding of the food habits, and a discussion of factors In the present affecting diet in a composition venomous coral snake. Micrurus fulvius. This species occurs in the southeastern United States and northeastern Mexico, in habitats ranging from subtropical forests to semiarid scrub swamps and lowland (Wright and Wright 1957). It is a northern representative of an essentially Neotropical radiation of the cosmopolitan front-fanged family Elapidae(Roze 1967; see Savitzky 1978. and Cadle and Sarich 198 1 for contrasting views on the relationships of coral snakes). Eastern coral snakes have been found crawling on the surface and in or under rocks, logs, stumps, litter, and burrows (Wright and Wright 1957; Gentry and Smith 1968). There is perhaps seasonal and geographic variation in diel activity, but these snakes are predominantly diurnal (cf. Neill 1957; Wright and Wright 1957; Jackson and Franz 1981). An average adult is ca. 50-85 cm long and weighs 20-55 g. Wright and Wright 1957), Shaw (1971). Campbell (1973). Greene 973a. 973b). Quinn (1979). and Jackson and Franz (1981) summa( ( rized some 1 1 aspects of the biology of this species. and it before swallowing Klauber 1956; Duellemeijer 1962; Nalleau 1966; Minton 1969; Kardong 1975; Chiszar and Methods (e.g., Behavioral Observations. — Sixty-five complete feeding sequences on live and dead prey byfour captive coral snakes were observed (one female, three males; total lengths 52.5-85.0 cm; from Dallas. Hidalgo, and Nacogdoches Counties, Texas). The snakes were individually housed in glass terraria that measured 32 x 32 x 62 cm Scudder 980). Although there are isolated notes on the feeding behavior of opisthoglyphs and 1 proteroglyphs in the literature (e.g.. Armitage 1965, Lambins 1967, for African elapids). the only extensive accounts are for certain sea snakes et al. we little on proteroglyphs, particularly terrestrial forms, hampers broader considerations of functional morphology, adaptive radiation, and community structure in snakes (cf. Arnold 1972; Rabb and Marx 1973; Kardong 1980; Savitzky 1980; release prey, then relocate (Voris very ies mobile maxilla. Solenoglyphs (viperids and atractaspids) have a single, very elongate hollow fang on each highly movable maxillary bone. Studies on several solenoglyphs of the family Vi- show Shine 1977, for ela- (e.g.. 1978. for hydrophiids), about the dietary ecology of proThis general lack of descriptive studteroglyphs. ly peridae al. 1978; Radcliffe and Chiszar 1980). 147 SPECIAL PUBLICATION-MUSEUM OF 148 or 27 x 32 x 52 cm. Each cage had a gravel substrate covered with leaf litter, a water bowl, and one large piece of bark for cover. Water was sprinkled over the leaves two or three times each week. The snakes were kept in a dark room that usually had a temperature of 22-24°C, but bulb on top occasionally rose to 30°C. A 100 at least W NATURAL HISTORY on the basis of a tail or a tail and posterior portion of a body, by comparisons with published information and intact reference specimens. Additional records were obtained from conversaand from tions or correspondence with collectors the literature (Matthes I860; Hay 1893; Mitchell 1903; Strecker 1908; Schmidt 1932; Loveridge of the perforated metal cover of each tank raised 1938, 1944; Klauber 1946; Ruick 1948; Minton the temperature at one end to ca. 24-26°C for 10 hr each day. Observations were timed with a Highton stop watch and recorded on audio tape or with camera and electronic flash. a 35 mm Captive coral snakes were offered live or Eumeces tetragrammus, E. fascia- Scincella lateralis, Carphophis amoenus, Coluber constrictor, Diadophis punctatus, Elaphe obsoleta, Heterodon platyrhinos, Nerodia erytits. throgaster, N. rhombifera, Opheodrys aestivus, Sonora semiannulata, Storeria dekayi, TantiUa gracilis, T. nigriceps, Thamnophis proximus, Tropidoclonion lineatum, and Virginia striatula. Live prey was released in a cage as far from the coral snake as possible. Dead prey was held with forceps ca. 20 cm from an active snake and jiggled to simulate prey movements; if there was no response, the prey was moved closer until it was seized. Trail Following.— \ used a modified version of the arena used by Gehlbach et al. (1971), consisting of an 80 x 80 cm piece of white duck cloth (28 strands/cm 2 ) in a plastic Neill 1968; Chance 1970; Malloy 1971; Fisher 1973; Jackson and Franz 1981). Myers 1965; Snout-vent (SV), dead prey, as available, of the following species: Anolis carolinensis, 1949; Clark 1949; Telford 1952; Curtis 1952; 1956; Martin 1958; Kennedy 1964; swimming pool. octagonal trail lane with segments 20 cm on an outer side and 1 cm wide was marked on the An tail, and head lengths of pre- served coral snakes were measured when possi- many museum specimens had damaged heads, SV was used for comparisons with prey TL. estimated the weights of common prey ble. Because I items from published statements and from live measurements of four Scincella lateralis, one Leptotyphlops dulcis, three Storeria dekayi, three Tantilla sp., six Tropidoclonion lineatum, and eight of I all 'irginia striatula. The average total lengths snakes in east Texas were taken as the midpoints of the ranges for adults given in Conant (1975). In a few cases I weighed preserved coral snakes and intact prey after blotting them on paper towels. I evaluated geographic variation in food habits by grouping records for Texas in four subsamples: "east Texas" (mixed deciduous and pine forests), "north central Texas" (tall grass-prairieforest ecotone), "central Texas" (forested hill country of the Edwards Plateau and the extreme eastern edge of the Chihuahuan Desert), and cloth with small, faint broken lines of indelible "south Texas" (semiarid thorn scrub and sub- an experiment a potential prey item was restricted to the trail lane by a portable 8 cm high cardboard alley and allowed to crawl around for one or two circuits. Then the prey animal and the cardboard alley were removed. Next a coral snake was confined in the center of the arena for three minutes in a bottomless 1gal plastic jar. The snake was released by lifting the jar, and its behavior observed under a 60 red light positioned so that the arena was very dimly lit. The cloth arenas were machine washed, rinsed, and dried after each test. tropical forest, see ink. Prior to W — Museum specimens were opened with a ventral incision and the orientation of each prey item in the gut was recorded. The identity and approximate total length (TL) of each item was determined if possible, often Diet Studies. Gould 1969, for vegetation regions). Records from elsewhere in the species range are grouped as "Florida" and "other" (Arkansas, Louisiana, South Carolina, and Mexico). Feeding Behavior The description that follows incorporates pub1898;Ditmars 1907, 1912; lished accounts (Grijs Clark 1949) and my observations. Feeding behavior is discussed in six groups of sequentially and functionally related motor patterns to facilitate future comparisons with other snakes. Encountering Prey. — Methods of encountering prey should be included in discussions of feeding behavior, because snakes use speciestypical postures and strategies for obtaining food. VERTEBRATE ECOLOGY AND SYSTEMATICA 149 Prey might be located by some type of searching, following, "sitting and waiting," (Pianka 1966), or a mixed strategy (Tollestrup 1980; an important component of coral snake antipredator behavior (Gehlbach 1972; Greene 1973b), is often very low for this species Chiszar and Scuddcr 1980); each of these techniques might be enhanced by behavioral or mor- (pers. obs.), phological specializations. For example, search- brational stimuli. trail and following utilize stereotyped poking behavior (in coral snakes, see below) and highly ing trail specialized receptor systems (e.g., facial pits in some viperids). "Sitting and waiting" probably more efficient when accompanied by boids and is camouflage (Fitch 960) or caudal luring (Greene 1 a coral snake had not fed for several crawled slowly over the substrate and head in and out of the leaf litter. This poked involved repeated forward and lateral head days, Neill and perhaps the snake observed by was responding defensively to tactile or vi- Several species of small snakes deposit chemserve as attractant pheromones ical trails that (Burghardt 1970; Gehlbach et al. 1971). and there are indications that these trails release searching and trail following behavior by coral snakes. Once two small earth snakes ( Virginia striatula) were moss for several days before and moss were put in a coral snake's The coral snake was crawling on the leaves kept in ajar of wet and Campbell 1972). When for tail waving, it its movements, and was accompanied by frequent flick clusters. At times a snake crawled the snakes cage. and encountered the moss. It moved its head back and forth over the moss for approximately five minutes and frequently flicked its tongue. tongue Then slowly beneath a large leaf or a small piece of bark and soon emerged from the opposite side, lowing the route taken by one of the earth snakes. The coral snake soon found the prey in a corner still its moving its head from tongue. side to side and flicking When a coral snake was searching, any movement of an it crawled across the cage, generally fol- and ate it. During staged encounters with ground skinks {Scincella lateralis), a coral snake fre- object in the terrarium elicited it was not a large object, ap- quently paused for several seconds in the exact spot where a skink had recently rested and point- When an acceptable prey item caused the approach, it was seized and eaten. Unsuccessful attempts to capture prey were followed by more searching behavior. F. R. Gehlbach (pers. comm.) observed similar ed and tongue-flicked before searching again. Experiments with coral snakes on cloth arenas provide additional evidence that they respond to prey trails. For two trails with each of two coral crawling and poking movements by two freeliving coral snakes on the Santa Ana Wildlife Refuge. Hidalgo County. Texas, one of which I dekayi or pointing and, if proach behavior. later used for behavioral studies. Neill (1951) described what was perhaps foraging behavior by a coral snake in Clay County, Florida. The snake crawled rapidly, to side, and poked its moved its head from side head into the surface Neill also stated that the snake's tail and times "the hind part of the creature was stant rapid, probing that at litter. made "con- motions" in the leaves, snakes, a small colubrid snake (adult Storeria was allowed to crawl one time. In each case the coral snakes crawled away from the central release point, paused briefly and pointed at the trail, and moved off the cloth. A second block of trials used I around the 'irginia striatula) alley by a small snake or a skink (adult female Eumeces fasciatus) making four circuits of the octagon in five minutes. One coral snake trails laid trails with pointing and then escape behavior, but followed a skink trail for one complete circuit and two additional turns responded to two snake thrown nearly as far forward as the head." He observed similar behavior in a captive snake, and suggested that the head and tail movements on the octagon. The other served to flush small reptiles and amphibians from cover. These observations suggest that lane segments, and three lane segments, respectively. It followed two lane segments of a skink crawling and head-poking in ground litter are motor patterns normally used by coral snakes to trail locate potential prey items. However, neither Gehlbach nor I observed use of the tail in foraging, and I doubt that it is a normal behavior, at least for coral snakes in Texas. The threshold trails laid by S. coral snake followed dekayi (two tula (one trial) for trials) one complete and V. stria- circuit, seven before crawling off of the cloth. These observations suggest that known prey species can leave trails which are perceived and followed by coral snakes. Additional experiments using more and more trials coral snakes, more prey species, are required before comparisons with the exten- SPECIAL PUBLICATION-MUSEUM OF 150 sive study by Gchlbach (1971) are war- pulled slowly but were attacked when pulled more rapidly. In 10 incomplete feeding sequences, a was stereotyped and Gehlbach et al. 97 ) prey item was grasped and immediately released, or maneuvered for a short time and then re- et al. ranted. Trail following behavior similar to that described by ( 1 for blind snakes. Leptotyphlops dulcis. A 1 coral snake crawled slowly from the release site, pointed and flicked its tongue at the trail, then turned 90° and began following it. The snake's head remained elevated while it crawled, and there were frequent tongue-flick clusters. At each corner it overshot 2-4 cm, paused, pointed and tongue- moved head from side to and resumed the cloth in If was on a wire jiggled crawling. front of a coral snake it pointed and approached flicked at the cloth, its turned back onto the side, trail, rapidly. The available captive NATURAL HISTORY and field observations leased. This suggests that a coral snake continues from the prey after it is seized, perhaps via either oral sensory papillae (Burns 1969; Greene, unpublished) or the Jacobson's to receive input Organ (cf. Burghardt 1970). Capture and Immobilization. — Approach was usually slow if the prey snake was moving slowly, and rapid if it crawled away quickly. Prey was seized with a quick forward movement of the anterior part or entire body of the coral snake, usually from a distance of several centimeters. In some cases a coral snake crawled parallel to a moving snake, flicked its tongue several times, imply that coral snakes actively search for prey, but the frequency and extent of foraging movements are unknown. There is no evidence that free-living coral snakes use a "sit and wait" strategy to ambush prey, but the behavior of captives suggests that they might. My snakes were frequently seen coiled with head raised and protruding from beneath the edge of a piece of bark or pile of leaves. Such snakes responded to near- and then seized the prey by turning its head sharply to the side and down. Coral snakes have relatively small eyes (Marx and Rabb 1972) and apparently cannot strike by movements by pointing, tongue-flicking, and an approaching predator at a distance of several centimeters and often slipped away unseen. Dur- approaching. Recognition and Approach. — Recognition of prey probably begins as soon as a coral snake points toward a stimulus, and incorporates visual and chemical cues. Captives approached any small movement, such as a wire jiggled in the very accurately. Live Scincella lateralis proved difficult for them to seize, perhaps because of the coral snakes' relatively poor vision and the skinks' small size and erratic escape behavior (Lewis 1951). Also, ground skinks seemed to perceive 1 1 attempts on these lizards by a coral snake, observed eight misses, two tail autotomies (skink ing I escaped unharmed), and one capture. These were during staged confrontations on a 32 x 62 cm substrate of gravel and scattered leaves, and the leaves or a finger moved against the glass from outside of the terrarium. Larger moving objects, only capture occurred when the snake trapped a skink in a corner. Small live prey snakes pre- such as a hand or a piece of bark, usually elicited pointing and then rapid head withdrawal and sented a slower and more elongate target, and were captured without difficulty; each of 23 attempts was successful. Ditmars (1907) and Clark (1949) stated that Micrurus fulvius immobilizes its prey with venom before swallowing, but Ditmars (1912) re- seemed especially likely if the obwas moved suddenly. Approach was accompanied by tongue-flick clusters, which evidently convey the necessary crawling. This ject stimuli for seizing or avoiding a potential prey item. Coral snakes quickly approached to within cm of large coleopteran larvae, cricket frogs (Acris crepitans), and newborn mice, but then withdrew without seizing them. Small live water 2 snakes (Nerodia rejected, and in were also approached and most cases they had discharged sp.) the cloacal sac contents. movements seemed However, rapid prey to result in a quicker attack chemical cues. Dead Nerodia were usually refused when stationary or and to override aversive marked that the venom is of little duing "cold blooded" animals. value in sub- My observations indicate that this species typically holds prey at the point of seizure until paralysis and then begins pre-ingestion maneuvers (see below). Slight movements of the prey were sometimes seen even as the tail was swallowed, suggesting that it is immobilized but not immediately killed by the venom. Coral snakes usually dragged their prey a few centimeters backward or forward before pausing, seemingly in response to its struggles. VERTEBRATE ECOLOGY AND SYSTEMATICA Table 1. 151 Pre-ingcstion latencies (in seconds) for coral snakes. Micrurus fulvius, dealing with live and dead means, standard deviations, and sample sizes are given. prey. Ranges, Snake no. 4 Latency Time between and onset of pre-ingestion maneuvers (live prey) Time between seizure (jc and onset of maneuvers (dead prey) = N= =28.6 ± (a- last prey body movement and onset of pre-ingestion maneuvers N= Time between This tended to untangle a small, writhing snake, might also imbed the fangs more deeply. During envenomation. the temporal region of the coral snakes sometimes appeared shriveled: this was probably caused by contraction of the M. adductor mandibulae externus superficialis, which has been shown to force venom out of the main venom gland in an elapid, Bungarus cae- and it ruleus (Rosenberg 1967; see also Savitzky 1978). In two instances a coral snake bit and quickly Eumeces fascial us that One of the skinks was imstruggled violently. mediately recaptured. The other lizard crawled released an adult female slowly for several centimeters and went under a piece of bark. It was soon followed by the coral snake and regrasped. Both skinks subsequentlymade only feeble movements and were eventually eaten. 6 = 400.9 ± N=8 — Coral snakes norPre-ingestion Maneuvers. do not release mally prey prior to swallowing it. Pre-ingestion maneuvers are probably evoked by and/or chemical cues (cf. Nalleau 1966) tactile and inhibited by prey movements. If prey movements inhibit the coral snake, the time between seizure and the onset of pre-ingestion maneuvers should be longer with live prey than with dead The mean pre-ingestion handling times with live and dead prey (Table 1) differed significantly for each of two coral snakes (P < .01, MannWhitney c'test). If prey movements inhibit the prey. snake, the time between the last prey movement and the onset of preingestion maneuvers should be similar for live and dead prey. These times were significantly different for one snake (P < (a= 63-190 (a the feeding comparison 86.5) 73.3) trials. Captive and free living coral snakes almost always swallowed prey head first, and scale overlap its on the prey item was used as a cue in locating anterior end (Greene 1976). Alternating jaw movements, used to typical of snakes (Gans 1961). were along the prey's body prior to swallowing. In one instance a small stick in the mouth of a coral snake prevented it from shifting over shift a snake's snout to begin swallowing. The coral snake released the prey, removed the stick by jaw movements and rubbing its head on the substance, regrasped the prey by the snout, and swallowed it. In all other feeding sequences, prey snakes were not released before they were swallowed. Swallowing.— After the prey's head had been down the throat, it was swallowed by repeated series of alternating jaw movements. These were separated by brief pauses and accompanied by lateral movements of the entire head. According to McDowell (1970), Micrurus belongs to a group of elapids in which "the palatine is shifted erected along with the maxilla during maximum protraction of the palate." This presumably occurs when a coral snake's head is rotated back and forth across a prey snake's long axis during swallowing movements. I could not observe the action of the palatine bones in live coral snakes, but frequently saw the maxillary fangs depress (and penetrate?) a prey snake's skin during swallowing. inertia to achieve better contact last 10 as resulting from individual differences and from the use of different sizes and species of prey in and the equivocal of the ± N= =71.8 ± N=4 rolled about results 73.7 0-152 52.6) Mann-Whitney c'test) but not for the other snake (P > .90). I interpret the large variances .01, 334.9) 0-290 32.3) 10 = 99.2 ± N= 5 (v 5 70-940 (a- 0-85 seizure pre-ingestion Snake no. 290-595 434.2 ± 132.5) During swallowing a coral snake sometimes its long axis, perhaps using the prey's between its teeth SPECIAL PUBLICATION-MUSEUM OF 52 Table east 2. Texas NATURAL HISTORY Frequency of prey items by taxon in eastern coral snakes, Micrurus fulvius. Abbreviations refer to (E), north central Texas (N), central Texas (C), south Texas (S), unknown localities in Texas (U), Florida (F), other parts of the species range (O), and total for prey for each sample is in parentheses. E Prey species all localities (T). Number of coral snakes containing VERTEBRATE ECOLOGY AND SYSTEMATICS Table 2. Continued. 153 SPECIAL PUBLICATION-MUSEUM OF 154 NATURAL HISTORY Seasonal incidence of three prey types in coral snakes, Micrurus fulvius, from Texas. Number of coral 3. snakes containing prey per season are in parentheses following months. Decimal fractions indicate contribution of each prey type to the total prey sample for each season. Table Juvenile Skinks Seasons Spring. March-May (33) Summer. June-August (16) Fall, September-December Total (82) 12 (.29) (33) large snakes Other prey Total VERTEBRATE ECOLOGY AND SYSTEMATICA 500- 155 SPECIAL PUBLICATION-MUSEUM OF 156 UJ 4 O HI a o 3 LL O 2 cc LU CO ^ PREY WEIGHT Fig. 2. Costs and benefits for coral snakes feeding on skinks (SK) versus snakes (SN). The straight line (F) indicates food value. See text for details. because of their absence of effective anti- size, predator behavior, and preferred microhabitat. Predation on juveniles of larger species of snakes is presumably more restricted because of their seasonal availability in temperate climates (Table 3; Fitch 1970). Size, defensive capabilities, and microhabitat probably also influence coral snake predation on lizards. It appears that skinks are not as vulnerable as snakes because of their smaller total length, agility, and capacity for tail autotomy. Whiptails {Cnemidophorus sp.) are sympatnc with Micrurus fulvius throughout its range, but these highly mobile lizards prefer hot, open areas (Fitch 1958) and are probably rarely encountered by coral snakes. Small iguanids {Anolis sp., Sceloporus sp.) are abundant in some places and anoles are sometimes accepted as food by captives (pers. obs.); however, these lizards are probably not important in the diet of coral snakes because they are largely arboreal and would not often be found by a foraging Micrurus. What follows is a post hoc consideration of "ideal" prey size, "ideaF prey type, and two aspects of variation in the diet of Micrurus fulvius (see Appendix, Note 1). For this purpose, loca- tion costs include the energy expenditure and risk required to bring a snake within attack distance of its prey, and handling costs include the energy expenditure and risk involved in capturing and ingesting an item (these terms include search time and pursuit time, respectively, of MacArthur and Pianka 1966). Food value includes the energy and other nutritional factors present in a prey 1 NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS coral snakes sometimes eat small prey: (i) Be- cause of the negative allometry of metabolic rate that obtains in most snakes (Bennett and Dawson 1976), an item of a particular relative weight might contribute proportionately more to the total actually true for large not known, (ii) and small numbers of coral snakes is haps surprising, because at all Texas they contain several species of very small, moderate, and large snakes, but very few medium-small species relative to adult M. fulvius Appendix, Note 3). In other words, be(ii) and (iii), large coral snakes in the southeastern United States probably rarely en(Fig. 3: cause of counter prey snakes proportionately as large as those eaten by small individuals. Skinks are more heavy bodied than small snakes and as a result their food value-handling cost intercept occurs at a lower weight (Fig. 2). is increased by the lower vulnerof skinks (see above and Vitt et al. 1977) disparity ability and perhaps by their capacity for inflicting a powerful bite on the predator. In other words, skinks are probably more costly to handle than small snakes of equivalent weight and provide less food value than small snakes of equivalent handling cost. Ideally, coral their diets only snakes should add skinks to when location costs are reduced proportionate to the increased handling costs these lizards impose. This suggests an explanation for the increased predation on skinks in east Texas: quantitative data are lacking, but my field experience is that skinks are much more com- monly encountered there than in other parts of the state where coral snakes occur. In any case, the stomach contents and behavioral observations certainly imply that Micrurus fulvius often attacks skinks and that these encounters fre- quently result in little or no net energy gain for adult coral snakes (three of 12 records of skink for very small M. fulvius, for which might have been proportionately large items). Either skinks (or skink tails) are proportionately more valuable than small snakes in per gram food value (cf. Clark 1 97 1 B. E. Dial, pers. tails were they ; comm.) or the overall expectation of finding make them worth per attempt. These considerations suggest that eastern coral snakes attack and sometimes eat substantial relatively small snakes). Occurrence of young individuals to of the very low average payoff is times of the year, (iii) The size configurations of terrestrial snake communities in temperate forests can be discontinuous; in east The low enough this it Whether of large prey snakes is seasonally restricted, and they are thus not a predictable resource for coral snakes is in spite would energy budget of a large snake than to that of a smaller individual. "better" items chasing 157 feed intuitively non-ideal prey (skinks, infrequently (Greene 1983, MS) That they do so is per- snakes apparently on relatively heavy items and such predators might be many especially able to defer feeding until a highly profitable prey could be located. There are at least two plausible, non-exclusive reasons why Mi- crurus fulvius does not meet this prediction: (i) Coral snakes might forage so as to minimize the time required to find and consume a given food, rather than to maximize the amount of intake of energy in a given time period or preyencounter (Schoener 1969; Morse 1980). In doing so they would reduce the time of exposure to predators and gain time for other activities, but the importance of either factor in coral snake biology is unknown. According to MacArthur (1972:62), "an animal should elect to pursue an item if and only if, during the time the pursuit would take, it could not expect both to locate and to catch a better item" (MacArthur 1972:61, included "capture and eating" in "pursuit"). This paradigm underlies much subsequent literature on optimal foraging (e.g., Pyke et al. 1977; Krebs and Davies (ii) 1978; Morse 1980), although 59) noted that it assumed "a expectation of the resources MacArthur (1972: fairly clear statistical [a predator] will come upon." However, I suspect that location costs/ item very greatly exceed handling costs/item for many terrestrial snakes (and perhaps some other predators), and that in most cases the predictability of finding a "better" item nearby is extremely low (see also Godley 1 980). If this is true, MacArthur's formulation is trivial for such predators, at least in. the practical sense of specifying the occurrence of a narrowly defined item in the diet. I conclude that diet breadth in coral snakes is probably constrained primarily by naive feeding preferences and perhaps minor experimental modifications (Appendix, Note 4), by morpho- (Appendix, Note 2), and by relaabundances (through their effects on encounter rates), rather than by more complex strategic "decisions" on an item by item basis logical factors tive prey NATURAL HISTORY SPECIAL PUBLICATION-MUSEUM OF 158 (Krebs and Davies 1978:23). The feeding rule for coral snakes seems to be, if it is an elongate reptile, not too large or dangerous, and can be caught, eat it. Acknowledgments This paper is extensively revised from part of submitted to The University of Texas at a thesis Arlington in partial fulfillment of the requirements for a Master of Arts degree. I am partic- Summary ularly indebted to the Tais paper reports the first extensive survey of feeding biology in a New World proteroglyphous snake, Micrurus fulvius. Foraging behavior was described on the basis of anecdotal field reports and detailed observations on four captive snakes. Literature records and stomach analysis of museum specimens provided information on 221 items from 177 coral snakes. Eastern coral snakes used stereotyped head poking movements and chemical cues to search for prey and to follow prey trails. Visual and chemical stimuli elicited attack, and prey was held until it was immobilized by venom. Preingestion movements were apparently inhibited by the prey's struggles and directed by scale overlap. Prey was almost always swallowed head first, by means of lateral shifts of the entire head of the coral snake and by unilateral jaw movements. Occasionally prey were bitten, released, relocated, and reseized before ingestion. This variable prey handling repertoire combines elements of a simple pattern seen in colubrids and some pro- teroglyphs with a more complex sequence seen in other proteroglyphs and some solenoglyphs. Micrurus fulvius of all sizes feed almost entirely on small, terrestrial snakes, elongate lizards (especially scincids and limbless anguids), and amphisbaenians. Other lizards and the young of large colubrid and viperid snakes make up the remainder of the diet. Taxonomic variation in the diet largely reflects the distribution and seasonal availability of particular prey species, rath- of prey taken. Large coral snakes sometimes eat larger prey than do smaller individuals, but they also continue to er than shifts in the general types feed on relatively small items. Behavioral observations and the diet analysis demonstrate that coral snakes often feed on two kinds of items, skinks and relatively small snakes, that are perhaps non-ideal in terms of average payoff per attack. These results and other considerations suggest that diet breadth in Micrurus fulvius might be constrained by naive prey preferences, morphological constraints, and relative prey abundance rather than by strategic "decisiGns" on an item by item basis. W. chairman of my commit- Pyburn, for his encouragement and counsel over the years, and to F. R. Gehlbach tee, F. for serving as visiting ing me committee member, loanand the gift of the live his trailing arenas, coral snake that began the study. The following curators and institutions allowed me to examine J. Cole, C. W. Myers, and R. G. American Museum of Natural History; Drewes and A. E. Leviton, California Acad- specimens: C. Zweifel, R. L. J. McCoy, Carnegie Museum of Natural History; W. J. Voss, Fort Worth Museum of Science and History; M. A. Nickerson, Milwaukee Public Museum; B. Hinderstein, Sam emy of Sciences; C. Houston State University; F. L. Rainwater, SteAustin State University; B. C. Brown, Strecker Museum, Baylor University; A. H. Cha- phen F. A&M ney, Texas University; J. R. Dixon, Texas Cooperative Wildlife Collection, Texas A&M University; P. Meylan, Florida State Museum; H. Marx and H. K. Voris, Field Museum of Natural History; A. G. Kluge and R. A. Nussbaum, University of Michigan Museum of Zoology; E. E. Williams, Museum of Comparative Zoology, Harvard University; G. Zug, National Museum of Natural History; R. F. Martin, Texas Memorial Museum, University of Texas; A. C. Echternacht, University of Tennessee; and Pyburn, University of Texas at Arlington. thank R. L. Anderson, K. F. Barnes, J. R. W. I F. also Dixon, Hendricks, B. Hinderstein, F. L. Rainwater, and R. A. Thomas for donating live coral snakes; J. B. Murphy for access to live snakes in the F. S. Dallas Zoo; S. Brums, G. H. S. M. Burghardt, R. Franz, Harris, D. R. Jackson, nedy, R. L. Lardie, J. H. D. E. Joy, J. P. KenLehmann, R. F. McMahon, and J. A. Roze for other assistance; and J. E. Cadle, B. E. Dial, F. M. Jaksic, G. A. Middendorf III, W. S. Parker, S. E. Riechert, and R. L. Seib for comments on the manuscript. G. M. Christman and L. Hunt prepared the figures. This work was supported by the Department of Biology, The University of Texas at Arlington; Foundation for Environmental Education; Karl P. Schmidt Fund, Field Museum of Natural History; Theodore Roosevelt Memorial Fund, American Museum of Natural History; the Na- VERTEBRATE ECOLOGY AND SYSTEMATICS and the Committee on Research, University of California, Berkeley. tional Science Foundation: 159 Clark, D. R. 1964. Reproduction and sexual dimorphism population of the rough earth snake. striatula (Linnaeus). Literature Cited Texas J. Sci.. I in a 'irginia 16:265- 295. Ardrey, R. The 1970. 1971. social contract. New Atheneum Publishers. Clark, R. R. Observations on differences and behavior of Dendroaspis D. polvlepis. J. in morphology angiisticeps and Herp. Assoc. Africa. (1):12- Species densities of predators and their prey. Amer. Nat., 106:220-236. The microevolution of feeding behavior. Pp. 409-453. In Kamil, A. C, and Sargent. T. 980. D. Foraging Behavior: (eds.). Ethological. Garland Beavers, R. A. 1 Food 976. Ecological, and Psychological Approaches. STPM Press. New diamondback rat- tlesnake, Crotalus atrox, in Texas (Viperi- Bennett. A. F. and Dawson, W. R. 976. Metabolism. Pp. 27-233. In Gans, C. and Dawson, W. R. (eds.). Biology of the Rep1 1 tilia. Vol. 5 (Physiology A). New York. Academic Press, Bl RGHARDT, G. M. Chemical perception in reptiles. Pp. 2411970. 308. In Johnson, J. W., Moulton, D. G. and (eds.). ical Signals. Communication by Chem- Appleton-Century-Crofts, New York. Burns, Oral sensory papillae in sea snakes. Copeia. 1969:617-619. J. Problems and approaches in the interpretation of the evolutionary history of venomous snakes. Mem. Inst. Butantan. Cadle, 1 98 J. 1 J. Campbell, 1973. Chance, 1970. and Sarich, V. M. E. The phylogenetic position and historical biogeography of the New World coral snakes. Zool., 195:157-167. J. A field guide to reptiles and amphibians of Hough- Garden City. The feeding habits of serpents. Zoologica. 1: Herp.. 7:136-137. Fitch, H. S. 1954. Life history and ecology of the five-lined skink, Eumeces fasciatus. Univ. Kansas Publ. Mus. Nat. Hist'., 8:1-156. 1958. Natural history of the six-lined racerunner, (Cnemidophorus Publ. Mus. Nat. 1 960. sexlineatus). Univ. Hist.. Kansas 11:11-62. Autecology of the copperhead. Univ. Kansas Publ. Mus. Nat. Hist., 13:85-288. Reproductive cycles in lizards and snakes. Misc. Publ. Univ. Kansas Mus. Nat. Hist.. 52:1-247. Frazzetta, T. H. 1970. Studies on the fossil snake Dinilysia patagonica Woodward Part II. Jaw machinery of the earliest snakes. Forma et Functio. 3:2051970. 221. captive hatching of Micrurus Julvius (Serpentes. Elapidae). J. Herp.. 7:312-315. Gans, C. 1961. B. note on the feeding habits of Micrurus Maryland Herpetol. Soc, 1978. The feeding mechanism of snakes and its possible evolution. Amer. Zool.. 1:21 7-227. Reptilian venoms: some evolutionary conGans, C. and Gans. fulvius fuhius. Bull. siderations. Pp. 1-42. In 6:56. Biology of the Reptilia. Vol. 8 (Physiology B). Academic Press, New York. and Gans, K. A. (eds.) Biology of the Reptilia. Vol. 8 (Physiology K. A. Chemosensory searching by rattlesnakes during predatory episodes. Pp. 125-139. In Muller-Schwarze, D., and Silverstein. R. M. (eds.). Chemical Signals. Vertebrates and aquatic invertebrates. Plenum Press. New York. Chodrow, 1973. Texas coral snake. Her- Fisher, C. B. 1973. Status of the flat-headed snake. Tantilla gracilis Baird and Girard, in Louisiana. J. Chiszar, D. and Scudder, K. M. 1980. in the petologica. 8:27. R. 197-238. Dullemeijer, P. 1962. Some remarks on the feeding behaviour of rattlesnakes. Proc. Kon. Ned. Akad. Wetensch.. 64:383-396. A. A A J. ton Mifflin Co., Boston. Ditmars, R. L. The reptile book. Doubleday. Page and Co.. 1907. E. In Press. Cannibalism eastern and central North America. B. 1969. Cadle, 1975. York. dae). Southwest. Nat.. 20:503-515. Turk, A. CONANT. 1912. habits of the western Snakes of the hill parishes of Louisiana. Tennessee Acad. Sci.. 24:244-261. Cirtis, L. S. J. 1972. 1 1949. 1952. 16. Arnold, strategy of tail-autotomy in the ground skink, Lvgosoma laterale. J. Exp. Zool.. 1 76: 295-302. York. Armitage, W. W. 1965. The R. E. and Taylor, C. R. Energetic cost of limbless locomotion snakes. Fed. Proc. 32:422. Gans, C, 1978. B). Gehlbach. 1 972. (eds.). Academic Press, New York. F. R. Coral snake mimicry reconsidered: the strategy of self-mimicry. Forma et Functio. 5: 311-320. in Gehlbach, F. R., Watkins, J. F. and Kroll, J. C. 1971. Pheromone trail-following studies of typhlo- SPECIAL PUBLICATION -MUSEUM OF 160 pid, leptotyphlopid, and colubrid snakes. Behaviour, 40:282-294. Gentry, J. B. and Smith, M. H. Food 1968. habits 1 J. Mamm., 49:562-565. F. Texas Kennedy, 1 Foraging ecology of the striped swamp snake, Regina alleni, in southern Florida. Ecol. 969. A life history study of the Great Plains ground snake, Sonora episcopa episcopa Kennicott. Monogr., 50:411-436. Gould, E. F. 1961. J. S. 1980. Kassing, and burrow associates of Pero- myscus polionotus. Godley, NATURAL HISTORY 946. L. M. The glossy snake, Arizona, with descriptions of new subspecies. Trans. San Diego Soc. 1956. Nat. Hist., 10:311-398. Rattlesnakes, their habits, life histories, and influence on mankind. Univ. California Klauber, 1 plants: a checklist and ecological sum- mary. Texas Agr. Exper. Stat. MP-585 (re- vised ed.), 121 pp. Greene, H. W. The food habits and feeding behavior of New World coral snakes. Master of Arts Thesis, The University of Texas at Arlington. 1973b. Defensive tail display in snakes and am1 973a. 1976. 1980. phisbaenians. J. Herp., 7:43-61. Scale overlap, a directional sign stimulus for prey ingestion by ophiophagous snakes. Z. Tierpsych., 41:113-120. Review of behavioral ecology: an evolutionJ. and Davies, N. 61:168-169. Dietary correlates of the origin and radiation of snakes. American Zool., 23:431-441. A diet-based model of adaptive radiation in ary approach, by Krebs, B.J. 1983. MS. Mamm., snakes. Manuscript in preparation. Greene, H. W. and Burghardt, G. M. Behavior and phylogeny: constriction in an978. cient and modern snakes. Science, 200:741 77. Greene, H. W. and Campbell, J. A. 972. Notes on the use of caudal lures by arboreal green pit vipers. Herpetologica, 28:32-34. 1 Grijs, P. 1898. Beobachtungen an Reptilien in der Gefangenschaft. Elaps fulvius. Zool. Gart., 39:276- Description of a supposed new species of Storeria from Florida, Storeria victa. Science, 19:199. HlGHTON, R. 1956. Systematics and variation of the endemic Florida snake genus Stilosoma. Bull. Florida St. Mus., 1:73-96. 1959. Press, Berkeley. R. and Davies, N. B. J. Behavioural ecology: an evolutionary approach. Sinauer Assoc, Sunderland, Mas- 1978. sachusetts. Lambiris, A. J. 1967. Observations on the feeding behavior of Naja haje and Naja nicricollis. J. Herp. Assoc Africa, (3):36. Lank.es, 1928. J. Micrurus (Elaps) frontalis D. B. im Terrarium. Bl. Aquar. Terrar. Kunde, 39:68-72. 1938. Micrurus {Elaps) lemniscatus Linne. Bl. Aquar. Terrar. Kunde, 49:53-54. Lewis, T. H. 1951. The biology of Eeiolopisma laterale (Say). Amer. Midi. Natur., 45:232-240. Loop, M. S. and Bailey, L. G. 1972. The effects of relative prey size on the ingestion behavior of rodent-eating snakes. Psychon. Sci., 28:167-169. LOVERIDGE, A. 1938. Food of Micrurus fulvius fulvius. Copeia, 1938:201-202. 1944. Cannibalism in the common coral snake. Copeia, 1944:254. R. H. Geographical ecology. Harper and Row, Publ., New York. MacArthur, R. H. and Pianka, E. R. 1966. On optimal use of a patchy environment. Amer. Nat., 100:603-609. 1972. P. 1892. HOLLING, C. Krebs, MacArthur, 279. Hay, O. Maiorana, V. 672. S. Some characteristics of simple types of predation and parasitism. Can. Entomol., 91: Malloy, J. F. Food habits of snakes in an east Texas State Fish Hatchery. Master of Science Thesis, S. F. Austin State University, Nacogdoches. 1971. Hornocker, M. 970. An analysis of mountain lion predation upon Martin, mule deer and biogeography of reptiles and amphibians the Gomez Farias region, Tamaulipas. Mexico. Misc. Publ. Mus. Zool. Univ. Michigan, 101:1-102. Marx, H. and Rabb, G. B. 1972. Phyletic analysis of fifty characters of advanced snakes. Fieldiana: Zool., 63:1-321. Idaho Primitive Area. Wildlife Monogr.. 21:39 pp. Jackson, D. R. and Franz, R. 1981. Ecology of the eastern coral snake (Micrurus fulvius fulvius) in northern peninsular Florida. Herpetologica, 37:213-228. Kardong, K. V. 1975. Prey capture in the cottonmouth snake (Agkistrodon piscivorus). J. Herp., 9:169-175. 1980. C. Behavior of an unobservable species: diet selection by a salamander. Copeia, 1978:664- 1978. 385-398. 1 13:185-203. Natural history notes on some snakes of eastern Texas. Texas J. Sci., 16:210-215. 964. W. Texas J. Sci., J. P. elk in the Evolutionary patterns in advanced snakes. Amer. Zool., 20:269-282. P. S. 1958. A in Matthes, 1860. B. Abhandlungen iiber den Zahnbau und die Lebensweise von Elaps fulvius Cuv. [sic], Elaps tenere B. and G. und Elaps tristis B. VERTEBRATE ECOLOGY AND SYSTEMATICS and G. aus den Vereinigten Staaten von Nord-Amerika und Elaps corallinus Pr. Max. aus Brasilien. Pp. 52-59. //? Drechsler, A. (ed.), Denkschriften der Naturwissenschaftlichen Gesellschaft Isis ze Dresden. Dresden. McDowell, 1970. On the status and relationships of the omon Island elapid snakes. J. Zool., Mertens, R. 1956. Beobachtungen an Korallenschlangen Terrarium. Aquar. Terrar. Z., 9:74-77. J. Sol161: im E. Coral snake preyed upon by bullfrog. Copeia, 1949:288. Minton, S. A. The feeding strike of the timber rattlesnake. 1969. J. Herp., 3:121-124. Mitchell, J. D. The poisonous snakes of Texas with notes 1903. on their habits. Trans. Texas Acad. Sci., 5: 19-48. Mole, R. R. The Trinidad snakes. Proc. Zool. Soc. Lon1924. don, 1924:235-278. Morse, D. H. 1 980. Behavioral mechanisms in ecology. Harvard Univ. Press, Cambridge. 1949. Myers, C. W. 1965. Biology of the ringneck snake, Diadophis punctatus, in Florida. Bull. Florida St. Mus., 10:43-90. Nalleau, G. 966. La biologie 1 i iffe, 980. 1 yellow lipped sea krait (Lalicauda colubrina).}. Herp.. 14:422-424. Rosenberg, H. histochemistry, and emptying the venom glands of some elapid snakes. J. Morph., 123:133-156. Roze, J. A. A checklist of the New World venomous cor1967. al snakes (Elapidae), with descriptions of new Histology, mechanism of forms. Amer. Mus. Novitates, 2287:1-60. Rliick, J. D. 1948. Collecting coral snakes, Micrurus fulvius tenere, in Texas. Herpetologica, 1978. origin of the New World proteroglyphous snakes and its bearing on the study of venom delivery systems in snakes. Doctoral dissertation. University of Kansas. Law- The rence. 980. 1 role of venom delivery strategies in snake evolution. Evolution, 34:1194-1204. The Schmidt, K. 1932. comportement predateur de new coral species Schoener, T. W. 1969. Optimal size and specialization in constant and fluctuating environments; an energy-time approach. Brookhaven Symp. Biol., 22: 1031971. Theory of foraging Syst.. 2:369-404. strategies. Ann. Rev. Ecol. Seib, R. L. 1981. Shaw, C. 1971. Size and shape in a neotropical burrowing colubrid snake, Geophis nasalis, and its prey. American Zool., 21:933. E. The coral snakes, genera Micrurus and MiUnited States and northern coral snake, Micrurus fulvius. Herpetologica. cruroides, of the 13:111-118. Mexico. Pp. 157-172. In Biicherl, W., and Snake eat snake. Florida Wildlife, 21:22-25. Oster, G. and Wilson, E. O. 1978. Caste and ecology in the social insects. Princeton Univ. Press, Princeton, New Jersey. E. R. Buckley, E. (eds.). Venomous Animals and Their Venoms. Volume 2, venomous vertebrates. Academic Press, New York. Shine, R. 1977. Habitats, diet and sympatry in snakes: a study from Australia. Can. J. Zool.. 55:1 1 18-1 128. Convexity, desert lizards and spatial heterogeneity. Ecology, 47:1055-1059. PlCKWELL, G. V. 1972. The venomous sea snakes. Fauna, 4:17-32. SOCHLIREK, Pvke, G. H., Pulliam, H. R. and Charnov, E. L. 1977. Optimal foraging: a selective review of theory' and tests. Q. Rev. Biol., 52:137-154. Strecker, 1966. Quinn, H. R. 1979. Reproduction and growth of the Texas coral snake (Micrurus fulvius tenere). Copeia, 1979: 453-463. 1973. P. Stomach contents of some American snakes, with the description of a of Geophis. Copeia, 1932:6-9. 1968. Rabb, G. 4:215-216. Savitzkv, A. H. 114. et le Vipera aspis au laboratoire et dans la nature. These. Univ. Nancy. P. Fanlac, Paris. Neill, W. T. 1951. Notes on the natural history of certain North American snakes. Publ. Res. Div. Ross Aliens' Reptile Inst., 1:47-60. 1957. Some misconceptions regarding the eastern Pianka, W. and Chiszar, D. A. C. A descriptive analysis of predatory behavior in the 1967. S. B. 145-190. Minton, Raix 161 and Marx, H. Major ecological and geographical patterns 1908. the evolution of colubrid snakes. Evolu- tion, 27:69-83. K. J. The reptiles and batrachians of McLennan County. Texas. Proc. Biol. Soc. Washington. 11:69-84. Telford, 1952. S. R. A herpetological survey in the vicinity of Lake Shipp, Polk County. Florida. Quart. Florida Acad. Sci., 15:175-185. B. in E. Einige Freilandbeobachtungen an Korallenschlangen der USA. Aquar. Terrar. (Berlin), 2:26-28. 1955. J. Tollestrup, K. 1980. Sit-and-wait predators versus active foragdo they exist? Amer. Zool., 20:809. ers: SPECIAL PUBLICATION-MUSEUM OF 162 Vitt, L. 1977. Congdon, J. D. and Dickson, N. A. Adaptive strategies and energetics of tail autotomy in lizards. Ecology, 58:326-337. J., Voris, H. K., Voris, H. H. and Liat, L. B. 1978. The food and feeding behavior of a marine snake, Enhydrina schistosa (Hydrophiidae). Copeia, 1978:134-146. Voris, H. K. and Moffett, M. W. Size and proportion relationships between 1981. the beaked sea snake and its prey. Biotropica, 13:15-19. Wright, A. H. and Wright, A. A. 1957. Handbook of snakes of the United States and Canada. Comstock Publishing Associates. Ithaca, New York. Zegel, J. 1975. 7. ing swallowing (see above, and Frazzetta 1970). My casual observations of captive elapids (Bungarus, Micrurus, Naja, Ophiophagus, alternnesia) indicate that W they have more swallowing prey of a particular relative diameter than do some snakes in other families brids). Note (e.g., difficulty some boids, viperids, Atractaspis, 3. — The colu- taxa in Fig. 3 and their average total cm) are: Tantilla gracilis (19), Vir- lengths (to nearest ginia striatula (22), V. valeriae (22), Storeria occipi- tomaculata (23), 5. tum Diadophis punctatus (30). dekayi (28), Tropidoclonion linea- Cemophora (31), coccinea (44), Opheodrys vernalis (44), Sistrurus mil- Heterodon iaris (45), Notes on collecting and breeding the eastern coral snake, Micrurus fulvius fulvius (L.). Bull. S. E. Herp. Soc, 1(6):9— 10." platyrhinos (68), O. aestivus (69), Agkistrodon contor- — Much is nomic or design principles; in so doing, they usually assume that the animal is behaving optimally and set out to determine, via alternative models, how this is accomplished (Krebs and Davies 1978; Oster and Wilson 1978; Greene 1980). Maiorana (1978) referred to such hypothetical diet parameters as "ideal," a more appropriate label than optimal in view of the procedures used in these studies. Note 2. — The factors affecting gape in snakes are poorly studied, but probably include the elasticity of throat and neck skin, and the length and mobility of certain cranial elements (Gans 96 1 Greene, MS). Circumstantial evidence suggests that coral snakes have relatively restricted gapes: they have fewer scale rows (15) than most snakes (implying less interscalar skin 1 and consequently Lampropeltis triangulum (49), trix (76), L. calligaster (92), Elaphe guttata (99), L. getulus ( 1 07), Coluber constrictor (112), Farancia abacura (120), Masticophis flagellum (130), Crotalus horridus (130), Pituophis melanoleucus (132), and E. ob- The size structure of this assemblage is perhaps even more bimodal in terms of potential prey for Micrurus fulvius (average total length 64 cm) than Fig. 3 suggests; this is because several taxa of intermediate length are either rare in east Texas (C. cocci- soleta (145). of the recent literature on foraging couched in terms of optimality, and some studies even purport to test an hypothesis that animals are feeding optimally. In practice, these approaches test feeding performance against constructs based on eco- Note movements (McDowell 1970), and this is also movements of the entire head dur- vertical suggested bv lateral C. Appendix theory NATURAL HISTORY ; less capacity for stretching), rela- 4—1 newly hatched coral snakes (see Campbell 1 973) to prey odors. Surface wash extracts were prepared by placing mealworms, newborn mice, a ground snake (Sonora semiannulata), or earthworms in a beaker of distilled water at 60°C for three minutes. The prey to water ratio was 3 g/10 ml. Extracts were stored frozen and warmed to room temperature before use. For testing, a sterile cotton swab was dipped in a vial of extract and then slowly moved to within 5 mm of the snout of a snake. Repeated attempts with each extract failed because the hatchling coral snakes always responded to the swabs with rapid crawling and body thrashing. It nevertheless seems likely that Micrurus fulvius exhibits innate preferences for snakes and lizards, because these comprise almost all known natural prey for all sizes of coral mandibular elements than many other snakes (Marx and Rabb 1972; Greene, MS). As in most other elapids, snakes, and because such prey and a centipede are the only kinds that have been accepted by very small, naive coral snakes in captivity (Campbell 1973; Zegel 1975; the maxillae are greatly foreshortened and. at most, slightly mobile. Apparently the palatine and pterygoid bones are mostly restricted to anterior-posterior and see Burghardt 1970, and Arnold 1980, for reviews of the roles of naive preferences and experiential factors in the recognition of food by snakes). tively shorter quadrate bones, and nea, O. vernalis) or proportionately stout for their lengths (A. contort rix, H. platyrhinos, S. miliaris). Note attempted to test the responses of two relatively shorter Vertebrate Ecology and Systematics— A Tribute to Henry S. Fitch Edited by R. A. Seigel. L. E. Hunt, J. L. Knight. L. Malarct and N. L. Zuschlag 4X4 Museum of Natural History. The University of Kansas. Lawrence i 1 The Role of Chemoreception in the Prey Selection of Neonate Reptiles Pennie H. von Achen and James L. Rakestraw Lampropeltis getulus, appear to be less responto this methodology (Brock and Myers 1979). Exploration of saurian chemosensory mechanisms, although scant, has shown innate Introduction sive The role of in the chemoreception prey selec- was the focus of our research on ten species of Kansas snakes and two species of Kansas lizards. Recent investigators, most notably Burghardt 970b. 1971,1 973) have demonstrated innate chemical preferences in certain snakes and lizards for the kinds of prey normally eaten in the wild. However, such innate tion of neonate reptiles predisposition to chemical stimuli in some species Eu nieces (Loop and Scoville 1972; Burghardt of 1973) and Gerrhonotus (Burghardt 1977). The innateness of the behavior implies an evo- ( 1 lutionary origin; thus one might expect a phylogenetic basis for the differential reliance upon feeding preferences are subject to some degree of variation, including geographic variation paral- this sensory system. Our objectives were to de- termine whether the chemoreceptive responses to prey odors that have been shown for some reptiles are of widespread occurrence in squa- leling those in the animals' natural diets (Burg- hardt 1970a; Arnold 1977). Furthermore, dispolymorphism within local populations, even within broods, has been demonstrated, aptinct mates and whether these responses differ among species and among higher taxa regardless of life parently serving to prevent overspecialization, thereby permitting better utilization of available histories. food resources (Arnold 1977; Burghardt 1975; Gove and Burghardt Methods 1975). Elimination of visual and olfactory senses sults in unaltered prey attacks in at least — Twenty-three broods of snakes, Subjects. representing ten species (one hundred and thirtyfour individuals), and four broods of lizards rep- re- some kinds of snakes (Wilde 1938; Burghardt and Hess 1968; Burghardt 1970b). Snakes with vomeronasal nerve lesions resenting two species (thirty-one individuals) were born in captivity to gravid females captured in to respond differentially chemical cues (Halpern and Frumin 1979). Therefore, the primary receptor of this chemical fail central or eastern to were kept itive lizards accidently picked up chemicals with the tongue while drinking, eating, and mating. With increased sensitivity of Jacobson's organ facilitating kept in gallon jars in moist manipulations of the tongue transmits the chemical cues to Jacobson's organ, the number of tongue flicks elicited by an odoriferous object, as well as actual attacks, seem to be reliable measures of the reptile's interest in ity. Preparation. Sheffield et a/. 1 968). (many tongue flicks While these when tested and attacks) others, such wood frass with variety of prey animals were Although most investigators have followed Burghardt's extract preparation technique (1968). Carr and Gregory (1976) suggest that since reptiles presumably respond to odors emanating from the surface of the prey, rubbing a moistened cotton swab over the prey ricines (Burghardt 1967, 1969, 1975; Burghardt 1 —A collected (Table the object (Burghardt 1967). Past works have dealt predominantly with nat- snakes displayed a strong response The females moistened plastic wrap. Larger snakes were housed in wooden cages (30 x 30 x 60 cm) with screen fronts which held small water containers and open boxes full of wood frass. Experimental animals were maintained in a concrete building with natural lighting, temperature, and humid- tongue, the system became proficient at responding to airborne chemicals (Gove 1979). Since the and Hess 968; 1). rition, and subsequently released. Each brood was housed collectively in the container in which they were born. The lizards and small snakes (Storeria dekayi and Diadophis punctatus) were information appears to be the vomeronasal system: the tongue, Jacobson's organ, and associated nerves. This system likely evolved as prim- and concomitant Kansas (Table in individual containers until partu- as 163 2). SPECIAL PUBLICATION-MUSEUM OF 164 Table 1. Data on reptiles at Species Eumeces fasciatus Ophisaurus attenuatus Coluber constrictor Lampropeltis calligaster Diadophis punctatus Storena dekavi Thamnophis sirtalis Thamnophis radix Nerodia sipedon Agkistrodon contortrix Sistrurus catenatus Crotalus viridis time of NATURAL HISTORY testing. Capture site of gravid female (Kansas county) Johnson Age Brood si/e (days) letup. Time (CDT) VERTEBRATE ECOLOGY AND SYSTEMATICS Scoring. — Burghardl's tongue flick-attack score (1967), an arbitrary value system based on the number of tongue flicks and length of attack latencies, was used to calculate a for each species (Table 2). "response profile" This score is based on the assumption that an attack is a more signifi- cant response than any number of tongue flicks, and that a more desirable stimulus leads to an attack with a shorter latency than a less desirable stimulus. The formula for attacking reptiles is represented by: 165 outnumbered two one by spiders stomach contents (Fitch 1954). Perhaps this is because as adults, most grasshoppers are too large for even an adult skink to subdue and ingest. Spiders and orthopterans combined comtion, they are : in prise the bulk of the diet in the wild. Only 6% of the skinks attacked harvestman swabs, compared to 47% that attacked the grasshopper swab. Harvestmcn constitute a minor food source the local population. Ophisaurus attenuatus likewise showed in a strong response to spider (P < 0.015) and orthopteran Score The base = base unit is + unit the (60-attack latency) maximum number of touch- ing tongue flicks given by any individual of the experimental group tested to any of the stimuli in a sixty second trial. An attacking reptile was given a score identical to the base unit for that trial length minus the attack lareptile which did not attack was given tency. a score identical with the number of tongue flicks species plus the A emitted towards the swab. Czaplicki a high test-retest reliability method was used (r = .86). ( 1 975) found when this scoring To eliminate re- sponses other than those elicited by swabs (e.g., exploratory tongue flicking), only those tongue flicks which touched the swabs were used. swabs (grasshopper P < 0.003). (cricket P < 0.003). Orthopterans comprise 58% of their diet, spiders 12.5%, in a food sample of a local population (Fitch, pers. comm.). Although the cricket swab elicited as offered, diets, worms mice are often eaten, but crickets are the most frequently consumed prey (Fitch 963). The mouse swab elicited the strongest response, although that was negligible. No attacks were made on any swab. 1 Lampropeltis calligaster eats mainly small was shown for one more types of prey items over the control swab. Using the Wilcoxon Signed Ranks Test to In all species a preference investigate differences between the control and the most preferred swab (those with the highest tongue flick-attack score for each species), we tention (Table determined that for four species the difference was significant (P < 0.05) (Fig. 1). In addition. Diadophis punctatus showed a significant response (P < 0.03) to a swab other than the one receiving the highest tongue flick-attack score. Although comparison was not possible with the Wilcoxon Signed Ranks Test for Thamnophis a paired /-test showed highly significant results (P < 0.005). Preferred stimuli generally sirtalis. to the prey species in the natural by food samples from the local The responded strongly to grasshoppers and spiders (Table 2). Although grasshoppers received the most atten- 2). No diet of local punctatus attacks were made. populations of Diadophis exclusively of composed almost is earthworms (Fitch 1975). While the earthworm swab did elicit the strongest response, the snail swab received almost as much interest and was significant at P < 0.03. No attacks were made. Storeria dekayi showed an overwhelming preference for the earthworm swab (P < 0.0001), supporting Collins' (1974) assessment that cally they eat primarily a weaker, but to the fasciatus 1978). While the some response as expected. Since snakes have been found in stomach and ringneck scat contents (Fitch 1978). it is somewhat surprising that ringneck swabs received so little atracer did elicit Eumeces reptiles (Fitch surprisingly aroused no more interest than the distilled water swab, the swabs of red sided garter snake and eastern yellow bellied or population. Lizards. — prey swabs. Of the three prey are not included in their natural all mouse swab Results diet as indicated attacks as the spider trials i responses to mammals and corresponded many were characterized by shorter attack latencies and more tongue flicks. Snakes. — Coluber constrictor showed m n imal swab, the spider still lo- earthworms. There was significant (P < 0.01) response snail (Table 2). pond Thamnophis sirtalis preys chiefly on mammals, frogs and worms (Fitch 1965). Fish are not well represented in the diet of the local popula- SPECIAL PUBLICATION-MUSEUM OF 166 Table 2. NATURAL HISTORY Tongue-flick attack (TFA) scores for 12 species of Kansas reptiles. Number of Pre\ swab attacks N= Eumeces fasciatus (P < 0.023, 17) a Distilled water 1 Harvestman (Leiobunum vittatum) Differential grasshopper {Melanoplus differentialis) House spider (Achaeranea tepidariorum) = 14) a Ophisaurus attenuatus (P < 0.015. N Distilled water Differential grasshopper (Melanoplus differentialis) Cricket (Acheta assimilis) Prairie wolf spider (Lycosa rabida) Coluber constrictor (t = N= 1.39, P < 13. 0. 10) b Distilled water Cricket (Acheta assimilis) House mouse (Mus musculus) Earthworm (Allolobophora caliginosa) = Lampropeltis calligaster (P < 0.18, N a 5) Distilled water House mouse (Mus musculus) Red-sided garter snake (Thamnophis sirtalis) Eastern yellow-bellied racer (Coluber constrictor) Prairie ringneck snake (Diadophis punctatus) Pond = N Diadophis punctatus (P < 0.14, Distilled water 17) a snail (Physa hawni) (Tenebrio molitor) Earthworm (Allolobophora caliginosa) Mealworm Storena dekayi (P < 0.0001, Distilled water Pond snail (Physa hawni) Earthworm (Allolobophora Mealworm Thamnophis N= 56 ) a caliginosa) (Tenebrio molitor) sirtalis (t = 3.25, N= 15, P< 0.005) b Distilled water Bullfrog (Rana catesbeiana) Earthworm (Allolobophora caliginosa) Red shiner (Notropus lutrensis) Mealworm (Tenebrio molitor) Thamnophis radix (P < 0.052, N= 8) a Distilled water Earthworm (Allolobophora caliginosa) Red shiner (Notropus lutrensis) Pond snail (Physa hawni) House mouse (Mus musculus) Nerodia sipedon (t = 1.48, N = 5. P < 0.1 5) b Distilled water Red shiner (Notropus lutrensis) Leopard frog (Rana pipiens) Tree frog (Hyla chrysoscelis) Agkistrodon contort nx (r = 1.57. N = 5, /> < 0.10) b Distilled water House mouse (Mus musculus) Cricket frog (Acris crepitans) Prairie ringneck snake (Diadophis punctatus) Western slender glass lizard (Ophisaurus attenuatus) Mean TFA score ± SE VERTEBRATE ECOLOGY AND SYSTEMATICS Table 2. Continued. Number Prev swab Sistrurus catenatus (P < attacks 0.1 10, N= 5) a Distilled water Bullfrog (Rana catesbeiana) Prairie vole (Microtus ochrogaster) White-footed mouse (Peromyscus leucopus) House mouse (Mus musculus) Fence lizard (Sceloporus undulatus) Crotalus 167 viridis (P < 0. 1 10, N= 5) a Distilled water Leopard frog (Rana pipiens) Prairie vole (Microtus ochrogaster) White-footed mouse (Peromyscus leucopus) House mouse (Mus musculus) of Mean TFA t SE score 168 SPECIAL PUBLICATION -MUSEUM OF NATURAL HISTORY MEAN TONGUE FLICK-ATTACK SCORE FOR PREFERRED STIMULUS LESS CONTROL SCORE COLUBER CONSTRICTOR LAMPROPELTIS CALLIGASTER DIADOPHIS PUNCTATUS VERTEBRATE ECOLOGY AND SYSTEMATICS In their test on six families of lizards. Bissinger and Simon (1979) noted that the difference in families might indicate the relative importance of the vomeronasal system. Accordingly, mean responses to preferred prey swabs provide the basis for comparing each species' differential sensitivfrequency of tongue extrusions chemical cues. Fig. 1 Ranked by their levels of sechemical discrimination, the reptiles are interesting pattern. lective second generally arranged into subfamilies (based on Dowling's [1975] classification). The three species with the lowest response Coluber constrictor, Lampropeltis calligaster, and Diadophis punctatus, belong to the subfamily Colubrinae. Coluber constrictor, a snake of open grassland, appears to rely largely on sight (Collins 1974). Movement of nearly any scores. and ingest the prey. Perhaps a sixty trial is insufficient for such snakes. The importance of the vomeronasal system in different shows the difference in the tongue flick-attack scores between the most preferred type of swab (the one receiving the highest tongue flick-attack score) and the distilled water control swab for each of the twelve species. These responses vary among the species and could represent the relative importance of chemoreception in the selection of prey. While only a tentative indication, this does suggest an ity to to constrict 169 to the Crotalinae seems not to be strongly associated with prey detection. Previous work on rat- tlesnakes (Chiszar and Radclifle 1976; Chiszar et al. mal 1 978) has demonstrated that visual or ther- signals are needed to elicit attacks and rel- atively few tongue flicks are emitted prior to Once the strike has occurred, however, tongue flicking is initiated and continues while the snake trails the stricken prey and investigates striking. the is presumably carcass, Tongue flicking again to locate commences the head. after the prey swallowed, perhaps to detect any remaining prey in the vicinity. However, rattlesnakes are able to detect prey solely by chemical cues (Cowles and Phelan 1958). The lack of any attacks by the crotalines that we tested supports the supposition that visual or thermal cues are necessary to elicit a strike, even though detection can be accomplished by odor alone. The fourth snake in an intermediate position a natricine, Nerodia sipedon. Unlike the other natricine tested, N. sipedon exhibited a relatively is low degree of tongue flicking and made no prey small animal stimulates the racer to pursue and attack. Olfaction appears to play a minor role in attacks. Burghardt (1968) has reported similar the finding and capture of prey. erally feeds in the water, volatile Diadophis punctatus is thought to depend on smell for prey detection, and its secretive life under rocks seems to confirm this. However, un- der the conditions of our experiments, the odor of its primary prey evoked little response, al- though a lower scoring swab did elicit a signifi- findings for this species. Because this snake gen- Drummond (1979) suggests that this species re- sponds to and integrates visual and chemical cues. Out of the water attacks can be elicited from experienced snakes by visual stimuli alone. When the snake is submerged, visual and/or mechanstimuli are adequate. Attack frequency in- cant response. ical Lampropeltis calligaster, unlike the natricines usually used in these works, is a constrictor. Oth- creases er investigators using colubrid constrictors have found their responses towards test swabs differ chemical cues alone. somewhat from scores, Storeria dekayi, that of natricines. Myers (1979) were unable Brock and to find any significant difference between the control and prey swabs for ingestively naive L. getulus. However, Wil- when visual ones. The diffuse chemical cues three snakes with the highest response Thamnophis sirtalis, and constrictors that overpower, hold and swallow their prey. Their high response scores corroborate previous findings same ingestion of prey, E. vulpina moved more deliband hesitantly, often taking several hours Thaninophis radix, are terrestrial natricine non- tulus erately accompany However, attack can be induced by liams and Brisbin ( 1978) found that adult L. gehad an innate preference for certain prey extracts despite restricted diets. Burghardt and Abeshaheen (1971), working with another colubrid constrictor, Elaphe vu/pina, found that, in contrast to a garter snake's immediate attack and chemical cues not be as useful as visual or tactile cues. may based on essentially the testing technique (Burghardt 1967, 1969, 1970a, 1971; Burghardt and Hess 1968). Prey common and tongue flick activity was among the highest of all the snake species tested, suggesting a strong dependence on chemosensory methods of prey selection. Sheffield attacks were ct al. (1968) note that prey attacks were always SPECIAL PUBLICATION-MUSEUM OF 170 preceded by at least one tongue flick that actually touched the swab, and we noted only two exceptions among the one hundred sixty-five individuals tested in this study. One E. fasciatus at- NATURAL HISTORY viduals. Moreover, tests to date are based adult food preferences. It is known young prey than do the of some species prefer different adults (Mushinsky and Lotz 1980). known about on that the If more were juvenile diets, utilization of prey tacked a harvestman swab immediately upon introduction, and an O. attenuatus likewise attacked a cricket swab. These attacks were not could be assured. surprising since many lizards are known to reto visual cues, including movement. How- Furthermore, we tentatively conclude that various obscure inhibiting factors altered results spond ever, Chis/ar et Thamnophis do adults. It ually replace al. (1976) noted that juvenile tongue more often than flick their is possible that other senses gradthe dependence on che- some of moreceptian during ontogeny (Burghardt 1969; Burghardt and Pruitt 1975). That would explain how a garter snake could catch swiftly moving prey such as a frog, which would seldom remain motionless to permit close approach and preliminary tongue flicks by the predator. Lizards are generally thought to be less dependent on Jacobson's organ than snakes. Terrestrial lizards, however, usually have better developed olfactory/vomeronasal organs and concomitant decreased vision as compared with arboreal lizards (Gravelle 1980). E. fasciatus exhibited a when tested, among the natricines in response scores. While Loop and Scoville (1972) found no dif- relatively strong chemical preference placing it tongue flicking or prey attack behavior congener, E. inexpectatus, our findings support the conclusion of Burghardt ( 973) that there ferential in a 1 innate chemical recognition of prey. Burghardt also noted, as we did, the very low rate of tongue is flicking in Eumeces compared with snakes in similar tests. Ophisaurus attenuatus displayed the highest of response of all the species tested, both in terms of tongue flicks and attacks. This is con- level sistent with the theory that those lizards showing a lack of elaborate visual communication use Jacobson's organ more frequently (Bissinger and Simon 1979). Furthermore, Gove (1979) dempattern of these onstrates that the tongue flick lizards are more similar to snakes than are most lizards. Other studies, addressing different questions, have used techniques similar to those used here. However, future investigators should be aware of several problematic areas. One involves the arbitrary age at which the hatchlings are tested. Postnatal onset of hunger following absorption of stored yolk material may take a short or relatively long time in different species and indi- items which would generate to varying degrees, maximum responses depending on the species and perhaps on the individual. The gentle momentary handling involved in transferring the animal to the experimental container may have involved psychological stress that resulted in suppression of the normal responses to food far beyond the five-minute adjustment period in some hypersensitive kinds. In the more secretive kinds that normally spend their time in burrows or beneath sheltering objects, the experimental container's lack of the necessary thigmotactic stimuli may have inhibited feeding behavior. Differential responses to may light intensities and to temperatures also have been involved. Finally, most studies have not differentiated between tongue flicks which touched the swab and those which did not To minimize the risk . of confusing tongue flicks not directly stimulated by the odor of the swab, such as exploratory tongue flicks, only those flicks touching the swab were used. This measures response to chemicals of low volatility. Sheffield et al. (1968) showed macromolecules were the attack Thamnophis. Cowles and Phelan that nonvolatile stimulus for (1958) theorized that the external nares, receiving olfactory stimuli, were highly sensitive but of low discrimination, alerting the snake to visual stimuli of movements and initiating lingual air sampling and subsequent specific analysis by Jacobson's organ. In other words, olfaction conveys volatile information from a distance, while Jacobson's organ is most sensitive to proximal compounds of low volatility. Indeed either the tongue or lips must touch the object before an attack is released (Sheffield et al. 1968). Presum- ably then, touching tongue flicks are more significant indicators of interest than non-touching tongue flicks. Once the prey is inside the mouth, gustation mediates which prey are suitable for swallowing (Burghardt 1969). If prey moves on after detection, a snake may trail it utilizing Jacobson's organ. However, Elaphe (Burghardt and Abeshaheen 1971) and Nerodia (Dunbar 1979) have demonstrated discrimination of prey on the VERTEBRATE ECOLOGY AND SYSTEMATICA basis of volatile chemical cues alone. In our tests, the constrictor L. calligaster displayed substantial discriminatory behavior encouragement, information on feeding habits, and critical comments regarding this manuscript. when both touching and non-touching tongue flicks are considered. By contrast, only a weak response was made directly towards the swab. On the other hand, N. sipedon, similarly appraised on the basis of all tongue tlicks made, responded more to the control Li Arnoi vestigation i). 1977. 1978. constraints, we conclude C III I) S. J. Polymorphism and geographic variation Some in of early experience on feeding effects responses in Thamnophis meaningful scoring procedure to use. extrusions and attacks KATURE I feeding behavior of the garter snake Thamnophis elegans. Science, 197:676-678. needed to determine the most With some reservations due I the than to the scented swabs. Clearly more inis 171 the sirtalis. common garter snake. Animal Behav., 26:455- 462. to the foregoing that purposeful tongue Bissinger. B. E. and Simon. C. A. 1979. Comparison of tongue extrusions in representatives of six families of lizards (Reptilia. Lacertilia). J. Herp.. 3(2): 33-1 39. do consistently vary among 1 species, the pattern generally following subfamily groupings. This may well suggest a phyiogenetic relationship of the differential dependence on chemoreception in the selection of prey. 1 Brock, O. G. and Myers, S. 1979. Responses of ingestively naive Lampropeltis getulus (Reptilia. Serpentes. Colubridae) to prey extracts. J. Herp., 13(2):209-212. BURGHARDT, G. M. 1967. Summary Inexperienced young of two lizard and ten snake species from Kansas were presented with cotton swabs scented with body surface odors of various animals including the preferred prey species of each kind of reptile. Distilled water swabs were used as controls. The number of tongue flicks, attacks and attack latencies were recorded. In all cases a preference was shown for one or more prey swabs over the water control swab, generally corresponding to the diet of the local population. This supports earlier findings that suggest innate chemical preferences. On the basis of tongue 1968. 1 1 969. The ( 246-257. 1970b. Chemical perception in reptiles. Pp. 241308. In Johnston, J. W., Moulton. D. G. and Turk. A. (eds.).. Communication by Chem- flicks and attacks, var- in their levels of responsivity to chemical stimuli. to 1971. 1973. 1975. conform with subfamily groupings, the colubrinae show the lowest level The saurian Eumcces ranks among the nat- but Ophisaurus attenuatus responded above all the other species tested in exhibiting the highest effectiveness of this sensory modality . Acknowledgments We would like to thank Nancy Zuschlag for her preliminary data analysis. We also wish to thank Dr. Henry S. Fitch for his suggestions. The ontogeny, evolution, and stimulus conof feeding in humans and reptiles. Pp. 253-275. //; Rare. M. and Mailer. O. (eds.). The Chemical Senses and Nutrition. Academic Press. Inc.. New York. Burghardt, G. M. and Abeshaheen. J. P. 1971. Responses to chemical stimuli of prey in newly hatched snakes of the genus Elaphe. Animal Behav.. 19:486-489. Burghardt. G. M. and Hess. E. H. 1968. Factors influencing the chemical release of prey attack in newborn snakes. Journal of Comp. Physiol. Psych.. 66:289-295. Burghardt. G. M. and Pruitt. C. H. Role of tongue and senses in feeding of naive 1975. 1 ricines, Chemical prey preference polymorphism in newborn garter snakes. Thamnophis sirtalis. Behaviour. 52:202-225. of discrim- inatory behavior, natricines, the highest level for snakes, and crotalines, intermediate between the two. Appleton-Century-Crofts, New York. Chemical-cue preferences of newborn snakes: influence of prenatal maternal experience. Science, 171:921-923. Chemical release of prey attack: extension to naive newly hatched lizards, Eumcces fascial us. Copeia. 1:178-181. ical Signals. prey selection for each species. reptiles tested can be ranked according to Tending Comparative prey-attack studies in newborn snakes of the genus Thamnophis. Behavior. 33:77-114. 970a. Intraspecific geographical variation in chemical food cue preferences of newborn garter snakes Thamnophis sirtalis). Behavior. 36: ious levels of discriminatory behavior were recorded, suggesting the relative importance of chemoreception Chemical-cue preferences of inexperienced snakes: comparative aspects. Science, 157: 717-721. Chemical preference studies on newborn snakes of three sympatric species of Natrix. Copeia. 4:732-737. 977. trol SPECIAL PUBLICATION-MUSEUM OF 172 garter snakes. Psych, and Behav., 14:185-194. Burghardt. G. M.. Wilcoxon, H. C. and Czaplicki, J. A. 1973. Conditioning in garter snakes: aversion to and experienced Carr. palatable prey induced by delayed illness. Animal Learn, and Behav.. 1(4):3 17-320. C. M. and Gregory, P. T. flicks be used to measure niche Canadian J. Zool., 54:1389-1394. Chiszar, D.. Carter, T., Knight, L., Simonsen. L. and Taylor, S. 1976. NATURAL HISTORY 1963. 1965. 1975. Can tongue Investigatory behavior in the plains garter snake (Thamnophis radix) and several additional species. 1978. Chemosensory searching by rattlesnakes is for Collins, 1974. J. 978. released by striking: a repReview, 9(2):54-56. T. Amphibians and reptiles in Kansas. Univ. Kansas Mus. Nat. Hist. Pub. Educ. Ser. 1: 1-283. Cowles, R. B. and Phelan. R. L. Olfaction in rattlesnakes. Copeia, 2:77-83. 1958. Czaplicki, J. A. Habituation of the chemically elicited prey1975. attack response in the diamond-backed water snake, Natrix rhombifera rhombifera. Her- 1979. 1974 New 76. 1975. to chemical cues from prey. J. Chem. Ecol., 1:25-40. Gravelle, K. and Simon. C. A. Field observations on the use of the tongue1980. Jacobson's organ system in two Iguanid lizards, Sceloporusjarrovi and Anolis trinitatis. Copeia, 2:356-359. Halpern, M. and Frumin, N. 979. Roles of the vomeronasal and olfactory systems in prey attack and feeding in adult garter snakes. Physiol, and Behav., 22:11831 1189. Loop, M.S. and Scoville, S. A. 1972. Response of newborn Eumeces inexpectatus Mushinsky, H. R. and Lotz, K. H. 980. Chemoreceptive responses of two sympatric 1 water snakes to extracts of H. M. Stimulus control of amphibious predation in the northern water snake (Nerodia sipedon sipedon). Zeitsch. fur Tierpsychol.. 50:18- gested prey species. 535. Chem. commonly in- Ecol., 6(3):523- A(2):7-12. on chem- preference of the northern water snake. Natrix sipedon sipedon (Reptilia, Serpentes, Colubridae). J. Herp.. 13(2): 165-169. Fitch. H. S. Life history and ecology of the five-lined skink 1954. (Eumeces Mus. Nat. J. Sheffield. L. P., Law. J. H. and Burghardt. G. M. 1968. On the nature of chemical food sign stimuli for newborn snakes. Comm. Behav. Biol., L. Effects of early feeding experience ical 960. Responses of ecologically dissimilar populations of the water snake Natrix sipedon si- pedon York. 44. 1 lizard Gove, D. and Burghardt, G. M. Drummond, 1979. comparative study of snake and 254-256. A provisional classification of snakes. In Dunbar. G. A to prey-object extracts. Herpetologica, 28: Yearbook of Herpetology. HISS Publica- 1979. study of the prairie kingsnake. Trans. Sci. 81(4):353-363. tongue-flicking, with an evolutionary hypothesis. Zeitsch. Fur Tierpsychol., 51:58- petologica. 31:403-409. tions. field Gove, D. Dowling, H. G. 1975. A Kansas Acad. wounded prey lication report. Herp. sirtalis. Hist., 53. 1 Animal Learn, and Behav.. 4(3):273-278. Chiszar, D. and Radcliffe, C. W. 1976. Rate of tongue flicking by rattlesnakes during successive stages of feeding on rodent prey. Bull. Psychon. Soc, 7(5):485-486. Chiszar, D., Radcliffe, C. W. and Smith, H. M. Univ. Kansas Publ., 15:493-564. A demographic study of the ringneck snake {Diadoplus punctatus) in Kansas. Univ. Kansas Mus. Nat. Hist. Misc. Publ., 62:1- Thamnophis Mus. Nat. sizes? 1976. Natural history of the racer Coluber conUniv. Kansas Publ.. Mus. Nat. Hist., 15:351-468. An ecological study of the garter snake, strictor. fasciatus). Hist., Univ. Kansas Publ., 8:1-156. Autecology of the copperhead. Univ. Kansas Publ., Mus. Nat. Hist., 13:85-288. Wilde, W. 1938. S. The role of Jacobson's organ in the feeding reaction of the common garter snake, Tham- nophis sirtalis sirtalis. J. Ex. Zool., 445-465. Williams, P. R. and Brisbin, I. L. 1978. Responses of captive-reared eastern kingsnakes (Lampropeltis getulus) to several prey odor stimuli. Herpetologica, 34:79-83. Vertebrate Ecology and Systematics— A Tribute to Henr> S. Fitch L Knight. L Malaret and N. L. Zuschlag Edited by R. A. Seigcl. L. E. Hunt. 14X4 Museum of" Natural History, The University of Kansas. Lawrence .1 j Ecology of Small Fossorial Australian Snakes of the Genera Neelaps and Simoselaps (Serpentes, Elapidae) Richard Shine Introduction Most of continental Australia in is 1. Within these two genera, at least five arid, but the herpetofauna of these enormous deserts is poorly known. Recent studies have clarified the tax- onomy Table distinct "species-groups" are evident: of Australian desert reptiles (i) Australia. Storr (e.g., 1967, 1979; Greer 1979; Cogger 1979), but the ecology of these animals remains virtually unstudied. The present paper is based on dissections of snakes from museum collections, and describes the general biology and life-histories of several small snake species from the arid zone. These snakes belong to the genera Neelaps (two species) and Simoselaps ( 1 1 species); both Storr (1967) and Cogger ( 1979) suggest that these genera are closely related to each other. All of the Neelaps and Simoselaps species are characterized by small body size (< 50 cm snout- Neelaps bimaculatus and N. calonotus are slender unbanded species of south-western (ii) The Simoselaps "bertholdi group" tholdi, anomala, littoralis. (ber- minima) are all short heavy-bodied snakes with distinct yellow-and-black bands and lacking an upturned rostral (Fig. ). 1 (iii) The Simoselaps "semifasciatus group" {semifasciatus, approximans, incinctus, rohave less distinct bands along the body peri) {incinctus lacks bands). The rostral has a sharply upturned, angular leading edge (Fig. 1). (iv) vent length), bright colouration, and fossorial Five Simoselaps species show a pro- Simoselaps australis and S. fasciolatus may not be closely related to each other. Both nounced upturned edge on the species resemble S. semifasciatus in general shape and colour; 5". australis has a sharp- sumably edged rostral whereas habits. rostral scale, preas an adaptation to burrowing. In this regard, as well as in general appearance and habits, Neelaps and Simoselaps are strikingly con- (v) vergent with small sand-dwelling snakes from other continents (e.g., Chilomeniscus, Chionactis, rant that in Methods I examined all specimens (N = 953) of Neelaps and Simoselaps in the collections of the Western Australian Museum, the South Australian Mu- (1967, 1979) recently has revised the group, describing several new species in the process. Ac- cording to Storr ( 1 967), Neelaps and Simoselaps are not sufficiently distinct to warrant generic seum, the National Museum, the Queensland Museum and the Australian Museum. I took the following data from each specimen: (i) snoutvent length (SVL), measured by running a tape measure along the body; (ii) gut contents; (iii) separation; Storr considers that they belong to the single large genus Vermicella (together with V. it is only doubtfully included genus (Storr 1979). in Africa). Although Neelaps and Simoselaps species may be among the most abundant snakes over most of Australia, they have attracted little study. Storr the Bandy-Bandy, does Simoselaps warro is a rainforest species of northeastern Queensland, and is so aberthis Ficimia, Gyalopion in North America; Pro- symna and Elapsoidea S. fasciolatus not. annulata). Cogger (1979) prefers to recognize the three genera separately. I follow Cogger's (1979) nomenclature in the were: present study, because of my subjective impression that Vermicella {sensu Storr) is too heter- licles reproductive maturity or immaturity (criteria males— large testes or opaque efferent ducts: females— gravid, ogeneous a group. Geographic distributions of the Neelaps and Simoselaps species are given by Storr (1967, 979) and Cogger (1979) and are briefly summarized large oviducts, or ovarian fol- >5 mm); and follicles Growth (iv) diameters of ovarian or oviducal eggs, in mature females. rates were estimated from seasonal dis- tributions of body sizes; this 1 in 173 more detail below. method is explained SPECIAL PUBLICATION-MUSEUM OF 174 Table 1. Sample sizes, body sizes and sexual size dimorphism NATURAL HISTORY in Neelaps and Simoselaps. VERTEBRATE ECOLOGY AND SYSTEMATICA * • ^ ^^^^ 41 +<» 175 SPECIAL PUBLICATION -MUSEUM OF 176 Table 2. Prey items found Pre) items in NATURAL HISTORY stomachs of Neelaps and Simoselaps species. VERTEBRATE ECOLOGY AND SYSTEMATICS 177 I U OO u O, 3- 24 22 20 30 28 26 Fig. 2. 3.99, P> X = 7.73, 2 Fecundity and inferred body-size .20) or S. semifasciatus P> (N = 88, 3d.f., Neither are the seasonal distributions of these two species different from each other (N = .05). 2 205, 3 d.f, x = 3.45, P> .30). Data on feeding activity (the proportion of snakes containing food items) show a different pattern (Fig. 5). In both S. bertholdi and S. semifasciatus, feeding is most common in summer, at 34 32 MEAN SNOUT -VENT LENGTH OF ADULT 00 (cm) hatching in Neelaps and Simoselaps species. in error on this subject. Glauert suggested that most Simoselaps species eat "insects and other small forms of life, including frogs and small is lizards" (1957, p. 40). Kinghorn ( 1 964) suggested that the diet of Neelaps bimaculatus consisted mainly of small insects. that S. australis fed on Mackay (1949) believed slugs, beetle larvae, worms. McPhee (1979) noted and that S. semifas- ciatus probably ate only insects. Worrell (1963) on skinks. and ceases during winter. correctly asserted that S. bertholdi feeds Gow (1976) Discussion on Body Sizes. — The tendency for females to grow males in Neelaps and Simoselaps species is not unexpected. Females are the larger sex in most, but not all, of the small Australian elapids studied to date (females larger in Cacophis, Furina, Drysdalia and Vermicella; males larger than Unechis— Shine 1978a, 1980a, 1980b. 1981a, 1981b). Female size superiority is also the most common situation among snakes in general, and is correlated with the absence of male combat behaviour (Shine 1978b). larger in Food Habits. — Published literature generally insects, suggested that N. bimaculatus feeds and that 5. australis probably does also (as well as feeding credited and 5". on skinks). bertholdi with feeding lizards, on Gow (1976) insects, frogs and recorded captive S. warm feeding on skinks. Storr (1967) speculated that the geographic distribution of Neelaps and Si- moselaps was constrained by competition with lygosomine skinks. Data from the present study suggest that these lizards are food items rather than competitors. The repeated assertion that these snakes feed on invertebrates (especially insects) is not sup- ported by data in Table 2. Lizards are the only SPECIAL PUBLICATION-MUSEUM OF 178 NATURAL HISTORY bertholdi S. 61 CO LU 2- < z IS) 1 6-1 1 1 1 1 1 1 co semifasciatus S. 4- Z 2- .iiiiiii -— —— i 8 10 12 —r— 18 1 14 16 SNOUT-VENT LENGTH (cm) Fig. 3. Body-size distributions of juvenile Simoselaps bertholdi and 'non-growing season" (April through October). S. semifasciatus collected prey type of Neelaps and the Simoselaps "ber- tholdi subdues tholdi group," and squamate eggs are the only " prey taken by the S. "semifasciatus group. Stud- in the same manner on several other small Australian elapids have revealed an analogous situation: published lit- fasciatus ies almost unanimous in suggesting that invertebrates are the main food items, but diserature is sections show that lizards portion of the diet. This rina, is comprise the major true in Cacophis, Fu- Demansia, Unechis and other groups (Shine 1977c, 1980a, 1980c, 1981a, 1981b). The saurophagous Neelaps and Simoselaps species mainly eat fossorial lizards: skinks of the genus Lerista and pygopodids of the genus Apra- However, the heavy-bodied Simoselaps "bertholdi group" also take non-fossorial lizards. The skinks Ctenotus, Menetia and Morethia are surface-active forms; Ctenotus is a very large and sia. robust prey for these small snakes. The inclusion of these prey items in the S. bertholdi diet may be related to the daily activity cycle of the snakes. Waite (1929), Worrell (1963) and Gow (1976) may be active diurnally, the other (nocturnal) Simoselaps for which records are available. Simoselaps ber- The its during the scincid prey by constriction, as do pythons (Bush 1981). specialization of the Simoselaps "semi- group" on squamate eggs was an un- expected finding. known H No other Australian snakes are to feed predominantly upon eggs, although oophagy apparently is common in the large northern colubrid Stegonotus cucullatus (McDowell 1972 found squamate eggs in 6 of 18 stomachs with identifiable food; the rest contained lizards, mice, frogs and orthopterans). Occasional oophagy has been recorded in several other Australian elapids. These include small species such as Cacophis harrietae C. squamulosus (N = sia olivacea (N = 1), (Shine 1980c), (N = 4 eggs), DemanD. psammophis (N = 2) and Drysda/ia coronoides (N = 6) 9) (Shine 1980a), (Shine 1981a). Eggs have also been found in the stomachs of large species: Austrelaps superbus (N = 2), Pseudechis porphyriacus (N = 3) and = 8) (Shine 1977c). HowPseudonaja textilis (N ever, eggs form only a small part of the diet in note that S. bertholdi all unlike squamate eggs were the only food recorded in stomachs of Simoselaps roperi (N = all of these species. In contrast, VERTEBRATE ECOLOGY AND SYSTEMATICS 179 25 n Nee/ops S auitralu 20 <> 10 O x x » JFMAMJJASOND * S < 1 1 1 ! 1 1 1 1 ° o° " O O O 1 1 o JFMAMJJASOND 1 I 20 I 1 I 1 1 1 1 1 1 1 20 —i O S. bertholdi S semifosciohjs 15- 15- O O 10- 10 5- OO o O J Ox ox OX OOO O O x oo O 10, M M o o OO OO o OCXDOOCO ^ O OO O o T" M -T- O J N MONTH Fig. 4. in Monthly variation bertholdi, crosses show diameter of the largest ovarian follicle in mature female Neelaps and Simoselaps: bimaculatus, crosses show N. calonotus; in S. bertholdi graph, circles show S. 5. littoralis. In all graphs, dots show oviducal eggs. in Neelaps graph, circles show A', = 27). were common semifasciatus (N of 4 food items) and present in S. fasciolatus (1 of 4 food items). The lack of 3) and S. in 5. australis (3 recognizable embryos in all these eggs suggests that the oophagous Simoselaps raid nests soon Simoselaps) are saurophagous. lished literature reveals the lation many A survey of pub- same general corre- snakes in general, but there are exceptions. Reptile eggs are an important among several "shovel-nosed" dietary component of after oviposition. species (e.g., However, some oophagous snakes take eggs of embryonic development (e.g., Prostvwja-Broadley 1979; Oligodon— Wall 1921). Snakes that feed primarily on squamate eggs lorhynchus— Klauber 1940; Salva do ra— Blair 1960; Oligodon— Wall 1921), and an occasional component in others (e.g., Rhinocheilus lecontei tessellatus—Sha-w and Campbell 1974; Aspidelaps— Branch 1979; Heterodon — Shaw and Campbell 1974). However, some "shovel-nosed" species do not feed on reptile eggs: Lytorhynchus is saurophagous (Minton 966) and the extensive array of North American desert colubrids with Moll and Legler (1971) found that most predation on turtle nests occurred at this time, probably because predators can locate the nest more easily. Observations by Blair ( 1 960) suggest that the snake Salvadora lineata preys chiefly upon squamate eggs from freshly-laid nests. at all stages may show morphological adaptations to this diet. within Neelaps and upturned rostrals (e.g.. Chionactis, ChilomenisGyalopion) feed only on inverte- cus, ticimia, Simoselaps is the shape of the rostral scale. The "shovel-nosed" species (i.e., those with an upturned angular edge to the rostral) are egg-eaters ly, "semifasciatus group," S. australis), whereas and other the snakes lacking this feature (Neelaps 1979: Phyl- 1 A clear correlate of oophagy (S. Prosymna— Broadley (e.g., Shaw and Campbell 1974). Similarthe upturned rostral is lacking from at least two snake species that feed mainly on squamate brates eggs (Cemophora coccinea— Palmer and Tre1970; Elapsoidea sundevalli— Branch gembo SPECIAL PUBLICATION-MUSEUM OF 180 Table 3. where y = NATURAL HISTORY Fecundity of Neelaps and Simoselaps species: Table gives values to solve the equation y and x = 9 SVL (cm). Regression fit by least squares. clutch size ax + b VERTEBRATE ECOLOGY AND SYSTEMATICS i < CO < - O : O z o O a. O -2o 181 NATURAL HISTORY SPECIAL PUBLICATION-MUSEUM OF 182 and pygopodid lizards are the only prey taken by Neelaps species and by the 57moselaps bertholdi species-group, whereas the parent: scincid Simoselaps semifasciatus species-group feeds ex- on squamate eggs. Oophagy is common 5. australis, and recorded in 5. fascio- Bush, B. Reptiles of the Kalgoorlie-Esperance region. Press, Perth. 46 pp. 1981. Vanguard Cogger, H. G. 1 Reptiles and Amphibians of Australia. A. H. A. W. Reed, Sydney. 608 pp. 979. & clusively also in latus. Oophagous species show adaptations of scalation (upturned edges of the rostral for bur- rowing) and dentition (flat blade-like posterior maxillary teeth for slitting egg-shells). Feeding occurs only in the warmer months of the year, Glauert, tralia. Gow, G. Fecundity and is is spring, with ovulation in summer. sizes 2.5 to 5.3), low (mean clutch correlated with mean adult body size in an interspecific comparison. Body size at hatching also increases with mean adult body size. Females attain larger body sizes than males, and mature at larger sizes. Analysis of body-size distributions suggests that sexual maturity is at- months of age. tained at 20 to 32 Greer. A. encouragement, stimulation, and assistance with identification of prey items and access to published literature. Finally my thanks go to Henry S. Fitch, whose superb studies on squamate ecology have laid the foundation for all his subsequent work Mus., 32(7):321-338. Klauber, Branch. W. R. 1979. The venomous snakes of southern Africa. Part 2. Elapidae and Hydrophidae. The Snake, 11:199-225. Broadley, D. G. 1979. Predation on reptile eggs by African snakes of the genus Prosvmna. Herpetologica, 35: 338-341. Angus & Robert- L. M. Two new subspecies of Phyllorhynchus, the leaf-nosed snake, with notes on the genus. Trans. San Diego Soc. Nat. Hist., 9: 1 95-214. Mac kay, D. R. The Australian coral snake. Proc. Roy. Zool. 949. 1940. 1 Soc. N.S.W., 1949:36-37. S. R. McDowell, 1972. 1979. The species of Stegonotus (Serpentes, Colubridae) in Papua New Guinea. Zool. Meded., 47:6-26. D. R. The Observer's Book of Snakes and Lizards of Australia. Methuen, 157 pp. Minton, 1966. S. A., Jr. A contribution to the herpetology of West Amer. Mus. Nat. Hist., 134: Pakistan. Bull. 27-184. O. and Legler, J. M. The life history of a neotropical slider turtle, Pseudemvs scripta (SchoepfT), in Panama. 1:1-102. Bull. L.A. County Mus. (Sci), Palmer, W. M. and Tregembo, G. Moll, E. 1971. 1 1970. Notes on the natural history of the scarlet snake Cemophora coccinea copei Jan in North Carolina. Herpetologica, 26:300-302. C. E. and Campbell, S. 1974. Snakes of the American west. A. A. Knopf, Shaw, N.Y. 329 pp. Shine, R. 1977a. Reproduction in Australian elapid snakes. I. Testicular cycles and mating seasons. Aust. J. Blair, W. F. 1960. The Rusty Lizard. A population study. Univ. Texas Press, Austin. 185 pp. Australia. son, Sydney. 197 pp. in this field. Literature Cited Robertson, E. Kinghorn, J. R. 1964. The Snakes of McPhee, G. Cogger (Australian Museum). Too all of them, I am grateful. I thank especially Allen Greer for & Eremiascincus, a new generic name for some Australian sand swimming skinks. Rec. Aust. 979. Acknowledgments This study would not have been possible without the full co-operation of the following curators: G. M. Storr (Western Australian Museum), A. Edwards (South Australian Museum), J. Covacevich (Queensland Museum), J. Coventry (National Museum of Victoria), A. E. Greer and H. of the Snakes of Western AusNaturalists Club, Perth. 62 pp. Snakes of Australia. Angus Sydney. 88 pp. in commences in W. A. F. 1976. 1 saurophagous as well as oophagous species. At least five of the thirteen species studied are oviparous. In mature females, vitellogenesis L. A Handbook 1957. Zool., 25:647-653. 1977b. Reproduction in Australian elapid snakes. II. Female reproductive cycles. Aust. J. Zool., 25:655-666. 1 977c. Habitats, diet and sympatry in snakes: a study from Australia. Canad. J. Zool., 55:1118- 1128. 1978a. Growth rates and sexual maturation in six species of Australian elapid snakes. Herpetologica, 34:73-79. 1978b. Sexual size dimorphism and male combat in snakes. Oecologia. 33:269-278. VERTEBRATE ECOLOGY AND SYSTEMATICS Western Australia and the Northern TerRoy. Soc. Western Aust. 50:80-92. Revisionary notes on the genus Vermicella 1980a. Comparative ecology of three Australian snake species of the genus Cacophis (Serpentes: Elapidae). Copeia 1980:831-838. 1980b. Reproduction, feeding and growth in the Australian burrowing snake 'ermicella annulate. J. Herpetol.. 14:71-77. 980c. Ecology of eastern Australian whipsnakes of in ritory. J. 1979. (Serpentes, Elapidae). Rec. West. Aust. Mus., 8:75-79. 1 1 the genus 389. 1981a. Demansia. Venomous snakes J. Waite, 1 E. R. 929. Herpetol., 14:381- The Reptiles and tralia. Amphibians of South AusGovernment Printer, Adelaide. 270 pp. in cold climates: ecology of the Australian genus Drysdalia (Serpentes: Elapidae). Copeia, 1981:14-25. 1981b. Ecology of Australian elapid snakes of the genera Furina and Glyphodon. J. Herpetol., 15:219-224. Storr, G. M. 1967. 183 The genus Vermicella (Serpentes, Elapidae) Wall, W. 1921. Worrell, 1963. F. Ophidia Taprobanica. or the Snakes of Ceylon. H. R. Cottle, Government Printer, Colombo. 581 pp. E. Reptiles of Australia. Sydney. 207 pp. Angus & Robertson, Vertebrate Ecology and Syslematics— A Tribute to Henry S. Fitch Edited by R. A. Seigcl. L. E. Hunt. J. L. Knight. L. Malaret and N. L. Zuschlag c 1984 Museum of Natural History. The University of Kansas. Lawrence Scaphiodontophis (Serpentes: Colubridae): Natural History and Test of a Mimicry-Related Hypothesis Robert W. Henderson Snakes of the sibynophiine colubrid genus Scaphiodontophis Taylor and Smith are relatively rare in collections and are something of a Food and Feeding Behavior. — Both species of Scaphiodontophis are stenophagous. apparently feeding almost exclusively on scincid lizards of two enigmatic charhave extremely long tails and exhibit a high incidence of tail injuries; and one of two basic color patterns exhibited by members of the genus has coral snake-like banding re- the genus Sphenomorphus in nature (Alvarez del Toro 1960; Landy etal. 1966; Scott 1969; Stuart 948; Taylor and Smith 943; pers. observ.), but also taking Gymnophthalmus (Teiidae) and Eumeces (Scincidae) (Alvarez del Toro, in litt.). curiosity since they possess acteristics: all taxa 1 stricted to the anterior portion of the body. In I Note.— Scaphiodontophis taxonomy has long in a chaotic state. However, an unpublished been dissertation by Morgan (1973) dealing with the entire colubrid subfamily Sibynophiinae has ably Morgan recognized two is uncommon in Scaphiodontophis and prey and with incredible usually swallowed alive of time necessary for prey (Scincella lateralis) ingestion by a 41.0 cm SVL 5. annulatus with a follow Morgan's (1973) clas- stopwatch on eight occasions. The watch was sification in this paper. started as soon as the snake grasped the lizard and it was stopped when the lizard's body (tails Natural History — Scaphiodontophis prowling cage during speed (Alvarez del Toro 1960; R. W. Van Devender, in litt.: pers. observ.). I recorded amount monotypic species of Scaphiodontophis: S. annulatus (including S. carpicinctus and S. zeteki listed by Peters and Orejas-Miranda [1970]) and Habitat. its It fed readily on Anolis and on at least one occasion captured the anole from below while the snake was under the leaf litter and the anole was on top of the litter. The captive S. venustissimus constricted an anole on one occasion and Alvarez del Toro (1960) observed constriction of larger prey by S. annulatus from Chiapas. Mexico. Subduing prey by constriction is prob- of Scaphiodontophis and, second, to test a hypothesis related to coral snake mimicry. I S. venustissimus leaf litter-covered floor of the day. The purposes of this paper are first to summarize what is known about the natural history S. venustissimus. observed a captive on the addition, they have peculiar hinged, shovel-like teeth (Savitzky 1981). clarified the situation. 1 were removed immediately prior to feeding) was no longer visible. Mean time of ingestion for six lizards between 40-47 SVL was 7.73 ± 2.09 sec (2.8-16.9), but the four fastest times had a mean of 4.87 ± 0.91 sec (2.8-7.2). Two skinks 52 SVL took 5.2 sec and 20.5 sec for ingestion. One 45 SVL skink which did not have ranges from mm southern Tamaulipas, Mexico to northern Colombia. S. annalatus is primarily a rainforest inhabitant (Alvarez del Toro 1960; Duellman 1965; Martin 1955; Neill and Allen 1959; Stuart mm mm 1935,1 958; Wilson and Meyer 1 982). It has also been taken in dense scrub forest (Duellman 1 965). removed, was grabbed by the tip of the tail. The snake worked its way to the lizard's snout and then swallowed it head first; the entire episode took 10.0 sec. All lizards were swallowed head first. its tail pine savanna and parkland (Henderson and Hoevers 1975), coffee groves (Slevin 1939; Taylor and Smith 1943), banana plantations (Roze 1969) and citrus groves (McCoy 1970). It is a leaf litter species (Alvarez del Toro 1960; Henderson and Hoevers 1975; McCoy 1970; Slevin 1939; Taylor and Smith 1943; J. Wright, in litt.) and may also be subterrestrial (Neill and Allen 1959). Scaphiodontophis venustissimus is found in wet lowlands (Scott 1969), occurs in leaf litter (Taylor 1954) and is also "fossorial" (Scott 1969). — The only display obDefensive Behavior. served in Scaphiodontophis has been tail and body thrashing. S. J. annulatus W. Wright (in in the field in litt.) observed it in northern Guatemala. An S. annulatus "was found in a fallen bush that was overgrown with herbs and grasses along with a considerable 185 amount of leaf litter. Mv attention SPECIAL PUBLICATION-MUSEUM OF 186 NATURAL HISTORY y*~-** >^^s Fig. 1. Scaphiodontophis venustissimus from Limon, Costa Rica. (Photo by R. was called to the snake because of the noise it was making. I heard the thrashing right up to the The banded portion point of close inspection. of the snake was elevated for at least a third of the length of the body and was visible above the bush. The head and neck remained motionless. The posterior more unicolor part of the snake was undulated and thrashing in the bush. The thrashing was not like the tail fluttering prattling) of some snakes, as much of the body moved as well." Likewise, I have observed that S. annulatus is always inoffensive and never offers to bite, but it does have a peculiar response to tactile stimuli: the body is vigorously twitched and both . . . ends of the body are thrashed about. It never failed to startle me! R. W. Van Devender {in litt.) has observed similar behavior in S. venustissimus. This may be comparable to the tail thrashing in Clelia clelia described by Greene (1973). Reproduction.— Alvarez del Toro and Smith (1958) reported a clutch of four S. annulatus eggs laid on 16 June and hatched on 15 August 1956 in Chiapas, Mexico. W. F. Pyburn {in litt.) colfrom beneath a lected three S. annulatus eggs ' "*3«v3k W. Van Devender.) Mexico on 3 August 964. on 14 September and it was opened egg contained a living male snake of 1.4 cm SVL; a second egg was opened on 9 October and it contained a living snake 11.5 cm SVL. The third egg hatched on 12 October and the snake was 16.7 cm SVL and 4.5 g. The number of anterior bands in the "hatchlings" was variable (2 with rotting log in Veracruz, 1 One 1 2 bands, 1 Limon clutch of three eggs; cm SVL A specimen of S. venhad a the gravid snake was 43.5 with 4 bands). ustissimus from Prov., Costa Rica (Carl S. Lieb, in litt.). Test of a Coral Snake Mimicry-Related Hypothesis Smith (1975, 1977) found that wooden dowels painted from end to end with coral snake colors and pattern and presented to naive, laboratory reared individuals of two species of neotropical reptile-eating birds (motmots and kiskadees) caused avoidance and alarm, and the birds would not peck at them. Dowels painted with coral snake colors but in stripes rather than rings, and those VERTEBRATE ECOLOGY AND SYSTEMATICS Fig. 2. Scaphiodontophis annulatus from Honduras. (Photo by painted in rings but not with coral snake colors, did not cause alarm and were attacked by the birds with little when or no hesitation. Likewise, only the end-third of a dowel was painted with coral snake colors and pattern, the birds attacked the dowel but directed their pecks at the unpainted end. Other end-third models, in color and pattern combinations as for the solid models, had pecks directed mostly or entirely at the paintS. venustissimus (Fig. 1) pattern has black bands bordered by yellow bands and red inter- spaces: this pattern typically covers the entire body and the tail. The venter yellowish and marked with small dark spots. length of the venustissimus almost invariably exhibits is S. this pattern. In 5. annulatus (Fig. 2) banding is frequently restricted to the anterior part of the body or to the entire W. body but not the tail. The pattern is of yellow-bordered black bands (range of 2 to 18 triads [Morgan 1973]) with red interspaces. Those portions of the body and/or tail which do not exhibit triads are of a drab grey or brown Porras.) 3 rows of dark spots, giving the impression of stripes. Again, the venter is ground color with unpatterned. Scaphiodontophis has an extremely long tail. Mean tail length expressed as a percentage of SVL in male and female S. annulatus is 86.3 (78.0-96.2) and 69.1 (58.2-92.6), respectively; in venustissimus it is 67.3 (60.1-72.0) in males and 56.7 (52.3-60.2) in females (Morgan 1973). Greene (1973) suggested ed ends. The L. 187 that "the tail of any escaping animal generally trails the body, and thus would be more likely to be grasped by a pursuer than any other part." Assuming that tail is predator inflicted (see below) and since S. annulatus is a living example of Smith's ( 1 975, damage 1977) "end-third" model and .S\ venustissimus a example of her "solid ring" model. I hy- living pothesized that if the coral snake color pattern does confer some selective advantage, then snakes with the S. venustissimus pattern should show a significantly lower incidence of tail injuries than snakes with the annulatus pattern. Methods. — Forty-eight preserved specimens of SPECIAL PUBLICATION-MUSEUM OF IKS Fig. 3. NATURAL HISTORY Scaphiodontophis venustissimus from Palmar Sur. Puntarenas, Costa Rica. (Photo by R. W. Van Devender.) Scaphiodontophis with an S, annulatus pattern, and 29 specimens with the venustissimus pattern, were examined for tail injuries. A few S. annulatus were banded the entire length of their body and tail. These specimens were regarded as having the 5. venustissimus-type pattern. Two specimens I examined had a "dugandi" pattern (Roze 29 (51.7%) with the injuries while 15 of S. ven- ustissimus pattern had suffered tail injuries. Using a chi-square contingency table, the incidence vs. uninjured tails in S. annulatuspatterned and S. venustissimus-patlerned snakes was compared. Differences were found to be not of injured significant (P > .05, x 2 = 0.1494, 1 d.f.). 1969; Henderson 1983) in which the anterior portion of the body is banded, the posterior poris not banded, but the tail is banded (Fig. These specimens were discounted from the statistical treatment. An additional number of snakes was examined but disregarded because coral their tail injuries (Dunn 1954; Hecht and Marien 1956; Savage terior and Vial 1974). tion Discussion 3). appeared recent (i.e., the posend of a caudal vertebra was exposed) and therefore may have been collector-inflicted. Also, some juveniles known to have been hatched in captivity were also discounted. A dissecting microscope was used to examine tails in which damage was not obvious, but with which the possibility existed that a minor injury may have occurred. — Twenty-seven of 48 (56.3%) snakes with the 5. annulatus pattern had sustained tail Results. Scaphiodontophis venustissimus resembles a snake, primarily Micrurus nigrocinctus matic annulatus S. is, however, enighave band- in that individuals frequently ing restricted to the anterior portion of the body. Perhaps, as suggested by Echternacht (1973) mimicry may "in some just be in a developmental stage species of the S. venustissimus and S. annulatus groups. ." A. H. Savitzky (in lift.), however, suggests that "Scaphiodontophis shares a common ancestor with Simophis. If so, the . mimetic pattern . is probably primitive (and. VERTEBRATE ECOLOGY AND SYSTEMATICS Tabu 1. Ratios of tail length to Scaphiodontophis annulatus. SVL and ( ienera species" S. annulatus* Terrestrial Arboreal Arboreal Micrurus terrestrial 1/1 total length in 189 snakes various adaptive zones as compared to male Tail length SVL" ( ienera species" Tail length loial length*** SPECIAL PUBLICATION-MUSEUM OF 190 Do are these data suggest that long-tailed snakes susceptible to predation? Possibly, but more it suggests that such species are more successful at escaping conflicts with predators alternatively than are short-tailed species (i.e., snakes with shorter tails exhibit fewer tail injuries because — fewer escaped with only a broken tail most were killed by the predator). Arboreal snakes with prehensile tails (such as Dipsas catesbyi and Imantodes cenchoa) may show a low incidence of tail breaks because the tail around a branch making it is usually more wrapped difficult for a predator to grab. Also, prehensile tails may be mechanically less susceptible to breaking, and perhaps these snakes are more likely attacked in the head region; injuries thus become more serious and escape less likely. Being nocturnal may also decrease the incidence of predation. The frequency of broken or regenerated tails has been used as an indicator of predation intensity (e.g., Brooks 1967; Parker 1972; in lizards Parker and Pianka 1975; Schall and Pianka 1980; al. 1977), although Schoener (1979) and and Fuentes (1980) have suggested that tail break frequency might be a better indication of predator efficiency. Pianka (1970) found that there was a positive correlation between tail breaks in Cnemidophorus tigris and number of potential predators in northern to southern sam- NATURAL HISTORY drophidion dendrophis and Rhadinaea decihave evolved a tail long enough to sustain piens), several predator-inflicted breaks. If a predator were to grab a Scaphiodontophis near the base tail, that may be the only attack the tail (and the snake) could sustain. Liner ( 960) found that the tail of a Pliocerus elapoides "was given of the 1 off like that of a lizard" plane on the expanded transverse processes. dontophis to disease, but Taylor later (1954) reported that 5. venustissimus apparently breaks deliberately when restrained by it. three occasions he grabbed a snake by the its tail On tail and three times the tail broke. He experienced the same result when attempting to catch another 5. venustissimus; the tail broke off in his hand and the snake escaped. I have observed no indication of He trans- nulatus). (Morgan [1973], however, found no evidence of a fracture plane in the caudal vertebrae of Scaphiodontophis, and found that tail breaks had always occurred between successive vertebrae.) In Wilson's opinion "this grooving of the transverse processes of the caudal vertebrae of Pliocercus and perhaps Scaphiodontophis is a point of sufficient weakness that allows the vertebrae to break samples show tail breakage incidence at 100% (e.g., Clark [1971] for Scincella lateralis). Taylor and Smith (1943) attributed the high incidence of stub tails in Scaphio- on the verse processes of "a few caudal vertebrae" from specimen of S. zeteki nothus (=S. an- the size class it a single Vitt et Some tried to pick ( 1 also found very shallow grooving Jaksic ples. when he 968) examined caudal vertebrae in P. elapoides and described a fracture up caudally. Wilson tail. The when selective the snake advantage of is seized by this adapta- seems obvious. As in lizards, the essential portion to the animal survives while the tail remains behind to occupy a predator." Although the results of the chi-square test do tion not support my hypothesis (i.e., Scaphiodontophis with the venustissimus pattern do not have an obvious selective advantage over those with the S. annulatus pattern), other interpretations of the data are feasible. If predation pressure on leaf litter rainforest snakes is high, as the inci- dence of incomplete tails in Scaphiodontophis suggests, then each of the two color patterns, in conjunction with the easily broken tail, may have selective value. (One does not have to have an advantage over the other.) For 5. venustissimus it is obvious: it mimics a venomous coral snake in color and pattern. Coral snakes and their look- birds (Howell 1957; disease in preserved specimens of Scaphiodontophis and therefore attribute the alikes are preyed high incidence of caudal damage to encounters with predators. I believe that the extremely long tail and the ability to autotomize the tail are anti- mammals (Jackson 1979) despite having aposematic patterns. Since these snakes are going to be preyed upon despite their aposematic colors and pattern, a tail which can be easily autoto- tail predator adaptations. Tail autotomy allows the attacked snake an opportunity to escape a potential predator. Since there is no tail regeneration in snakes (but see Sharma 1980), any portion of the forever. To compensate phis, for this, tail lost is gone Scaphiodonto- and possibly other snake species (e.g., Den- upon by Pough 1964; Skutch 1971; Smith 1969) and mized can only be advantageous. Alternately, Scaphiodontophis annulatus is es- sentially bi-patterned, whereas S. venustissimus is uni-patterned. Since 5. annulatus is bi-patterned, a predator has the option of choosing which pattern to attack: a dull-colored, striped VERTEBRATE ECOLOGY AND SYSTEMATICS Fig. 4. R. Scaphiodontophis anmdatus (41.0 cm SVL) from Honduras 191 in a typical diurnal posture. (Photo by W. Henderson.) banded pattern. I would more often attack the dull, striped pattern and Smith's (1975, 1977) experiments support this; naive birds would attack a dowel if the coral snake pattern was restricted to only one end of it. A predator does not have to make a choice with a uni-patterned snake and it may just as likely grab the snake at mid-body or at the head as at the tail (although I suspect the caudal region is more frequently pattern, or a brightly colored suggest that the predator grabbed than more anterior regions). Thus. S. venustissimus exhibits a high incidence of tail injuries because it has an autotomizeable tail and is able to escape conflicts with would-be predators. In addition, the coral snake pattern, even without benefit of a tail display (Gehlbach 1972) and confuse a predator and cause it tail. S. anmdatus exhibits a high incidence of tail injuries because a predator must make a choice between attacking two patterns and most likely will go for the dull posterior pattern which ends with a tail that is easily autotomizeable. Other leaf litter snakes may be ex- may inhibit to go for the posed to similar predation pressure as Scaphio- dontophis, but because they lack tail autotomy. they are killed by the predator, rather than being able to escape minus only a portion of their tail. In conclusion, I offer three possible functions of observed behavior, incidence of and color pattern in tail damage Scaphiodontophis: 1) Scaphiodontophis {anmdatus and venustissimus) exhibit a high incidence of broken tails because they are adapted to autotomy and are exceptionally long in order to sustain several breaks. Color pattern in both species is potentially confusing and/or inhibitive to potential predators. 2) An alternative interpretation of pattern function in Scaphiodontophis is based on the antipredator strategies of flight and defense, and on the phenomenon of flicker fusion (Jackson et al. 976). Brattstrom (1955) suggested that the coral snake pattern conceals the bearer by a disruptive 1 when it is immobile and, when fleeing, the banded pattern prohibits the predator from shift- effect ing focus rapidly forward to maintain the snake of vision and the snake may therefore in its field escape. Jackson et al. (1976) concurred with SPECIAL PUBLICATION-MUSEUM OF 192 Brattstrom (1955) that a regularly banded pattern compromise between the of disruptive concealment and gener- "may strategies represent a NATURAL HISTORY by thrashing its tail through the leaf litter? Although the moving tail could attract the attention of potential predators (as ation of a deflective illusion during flight." Scaphiodontophis annulatus bears a pattern that can autotomizeable be regularly or irregularly banded anteriorly and tain such attacks. is essentially striped posteriorly. Jackson ciated came et al. According to (1976), a striped pattern more with flight is asso- did Wright), the long The motionless, coral snake- patterned head would attract less attention and is mimetically colored. than defense, and they to the conclusion that it which would be the part of the snake most likely attacked, is adapted to sustail, Summary "The aposematic- while the disruptive-deflective effect might be of value against color-insensitive mam- Snakes of the sibynophiine colubrid genus Scaphiodontophis have remarkably long tails (up to 96% of SVL in males) and they exhibit a high incidence (>50%) of broken tails. Scaphiodon- malian predators. ." Observations of captive and pattern the mimetic functions might be most useful with predators, like birds, that can perceive color greatest . tophis venustissimus has coral snake-like colors . S. annulatus, even in where there is not opprolonged movement, have illus- relatively small enclosures portunity for trated to me Even with a of 5. the confusing effect of the pattern. knowledge of the appearance priori annulatus, it takes several seconds to de- entire length of the body and tail, annulatus usually has the coral snake colors and pattern restricted to the anterior one-half (or but S. less) of the body. Preserved S. venustissimus ex- hibit fewer tail injuries, but not significantly so, than S. annulatus. Assuming that the tail injuries and whether the snake is coming or going. In addition, I have observed captive 5. annulatus on numerous occasions with only the banded part of the body exposed and are predator inflicted, the incidence of injuries in the two species suggests that the coral snake under a cover object (Fig. 4). I beJackson et al. (1976) and as stated by that the patterns of both snakes are confusing and inhibitory to potential predators: in venus- termine where the head the rest of lieve, like is it Pough (1976), that "... a pattern of brightly col- pattern and colors confer no selective advantage to S. venustissimus over S. annulatus. tissimus because it is conclude I a coral snake mimic, and spicuous depending on the light conditions, the visual capacity of the predator involved, and the bi-patterned and a predator must choose which pattern to attack (most likely the non-coral snake-like posterior). behavior of the snake." The ored crossbands can be both cryptic and con- field observations of S. 3) Finally, Wright's annulatus, along with its unusual color pattern, suggests a function of tail thrashing apart from defense. Sphenomorphus apparently the primary prey species of Scaphiodontophis, and it would not be surprising that such extreme stenophagy would give cherriei rise to is in tails it is of both species are autotomizeable and Another and pattern of potentially able to sustain several breaks. potential function of the color Scaphiodontophis is as an anti-predator strategy of flight and defense. Finally, the long tail, at least in 5. annulatus, may be useful in flushing scincid lizard prey from leaf litter substrate. anatomical and Acknowledgments possibly behavioral characteristics adapted to Sphenomorphus predation. Savitzky (1981) has noted a number of anatomical peculiarities in Scaphiodontophis which apparently are adaptations for swallowing hard-bodied prey (i.e., scincid lizards). Sphenomorphus cherriei is, like Sca- phiodontophis, a rainforest, leaf annulatus because litter (Fitch 1973; Stuart 1958; pers. observ.) inhabitant and much of its activity occurs beneath the leaf litter (Fitch 1973). Fitch (1973) noted that a collector could flush Sphenomorphus by "trampling" through the litter. Could Wright (in litt.) have been watching a Scaphiodontophis trying to flush prey For the loan of specimens and/or information on specimens at their respective institutions I thank J. A. Campbell, D. F. Hoffmeister, J. P. Karges, A. G. Kluge, A. Leviton, C. S. Lieb, H. Marx, T. P. Maslin, C. J. McCoy, P. Meylan, W. Pyburn, A. H. Savitzky, H. K. Voris, J. W. Wright, and R. G. Zweifel. For the use of photographs I thank Louis PorF. ras and R. W. Van Devender. Miguel Alvarez del Toro, R. W. Van Devender and J. W. Wright provided useful field observations of Scaphiodontophis. VERTEBRATE ECOLOGY AND SYSTEMATICA H. W. Greene, M. A. Nickerson, J. (Serpentes: Colubridae) from Costa Brenesia, 19/20:359-362. A. Roze, R. A. Sajdak, A. H. Savitzky, and an anonymous reviewer provided many useful comments on an earlier draft of the manuscript. Finally, to Henry Fitch who, for over 15 years has been mentor and source of inspiration and encouragement: Thanks for everything. 193 Henderson. R. W., Binder, M. H. and Sajdak, (Colubridae) on Isla Saona, Republica Dominicana. Amphibia-Reptilia, 2:153-163. Henderson. A 1975. Literature Cited W. and Hoevers, R. checklist and key L. Howell, T. R. Birds of a second-growth rain forest area of Nicaragua. Condor 59:74-1 1. 1957. 1 Jackson, J. F. of some ophidian tail displays on the predatory behavior of grison (Galictis sp.). Copeia, 1979:169-172. 1979. Effects petologica. 14:15-17. Brattstrom, B. H. "mimic" problem and coral snake pro- tective coloration. Evolution, 9:217-219. gosoma 87. Clark, D. R. The 1971. strategy of skink, tail Lvgosoma autotomy laterale. J. in the ground Exp. Zool., 1 76: 295-302. Dixon, J. reptiles of the Upper Amazon Basin, Iquitos Region, Peru II. Crocodilians, turtles and snakes. Contrib. Biol. Geol. Milwaukee DUELLMAN, W. E. 1965. Amphibians and Reptiles from the Yucatan Peninsula, Mexico. Univ. Kansas Mus. Nat. 1979. Nat. Hist. Misc. Publ., 65:1-352. 1954. Liolaemus lizards. J. Herpetol., 14:137-141. Landy, M. J., Langebartel, D. A., Moll, E. O. and Smith, H. M. A collection of snakes from Volcan Tacana. Chiapas, Mexico. J. Ohio Herpetol. Soc. 5: 93-101. coral snake "mimic" problem in field study of Costa Rican lizards. Univ. Sci. Bull., 50:39-126. Coral snake mimicry reconsidered: the strategy of self-mimicry. Forma et Functio. 5: 311-320. Greene, H. W. tail displays by snakes and amphisbaemans. J. Herpetol., 7:143-161. Hecht, M. K. and Marien, D. The coral snake mimic problem: a reinter1956. pretation. J. Morphol., 98:335-356. Defensive Henderson, R. W. 1983. Herpetological records from the An unusually patterned Scaphiodontophis Gomez Far- of southwestern Tamaulipas. Mexico. Copeia, 1955:173-180. ias region C. J. The snake fauna of Middlesex. duras. Morgan, 1973. J. British Hon- Herpetol., 4:135-140. E. C. Snakes of the Subfamily Sibynophiinae. Ph.D. Diss., University Southwestern Louisiana. 205 pp. Myers, C. W. 1967. The pine woods snake, Rhadinaea Jlavilata S. A Kansas Gehlbach, F. R. 1973. P. S. Pan- (410):1-18. 972. Martin, McCoy, ama. Evolution, 8:97-102. Echternacht, A. C. The color pattern of Sonora michoacanensis 1973. (Duges) (Serpentes. Colubridae) and its bearing on the origin of the species. Breviora, Fitch, H. 1973. Liner, E. A. A new subspecies of false coral snake (Pli1960. ocercus elapoides) from San Luis Potosi, Mexico. Southwest Nat., 5:217-220. 1970. The and Campbell, H. W. approach. Amer. Natur.. 110:1029-1053. Jaksic, F. M. and Fuentes, E. R. Correlates of tail losses in twelve species of 1980. The biology of an equatorial herpetofauna in Amazonian Ecuador. Univ. Kansas Mus. E. R. Ill, dorsal pigmentation pattern of snakes as an antipredator strategy: a multivariate 1955. Hist. Publ., 15(2):577-614. Ingram, W. The 1966. R. and Soini, P. Public Mus., (12): 1-91. 1 J. F., 1976. The 1977. Dunn, Jackson, R.. Jr. Population ecology of the ground skink, Lylaterale (Say). Ecol. Monogr., 37:71- 1967. amphibians and America. Milwau- 63. Reptiles de Chiapas. Instituto Zoologico de Estado (Chiapas), Tuxtla Gutierrez. Alvarez del Toro, M. and Smith, H. M. Notulae herpetologicae Chiapsiae II. Her1958. 1960. Brooks, G. G. to the kee Public Mus. Contrib. Biol. Geol., (5):1- Alvarez del Toro, M. The R. A. Ecological relationships of the tree snakes Uromacer catesbyi and U. oxyrhynchus 1981. reptiles of Belize. Central 1955. Rica. 1 974. (Cope). Bull. Florida State Mus., 1 1:47-97. The systematics of Rhadinaea (Colubridae). a genus of New World snakes. Bull. Amer. Mus. Nat. Hist.. 153:1-262. Neill, W. T. and Allen, R. Studies on the amphibians and reptiles of 1959. British Honduras. Publ. Res. Div. Ross Allen's Reptile Inst., 2(1): 1-76. Parker, W. 1972. S. Ecological study of the western whiptail lizard, Cnemidophorus tigris gracilis, in Ari- zona. Herpetologica. 28:360-369. Parker, W. S. and Pianka, E. R. 1975. Comparative ecology of populations of the SPECIAL PUBLICATION-MUSEUM OF 194 lizard Via stansburiana. Copeia 1 975:6 5— 1 632. Peters, J. A. and Orejas-Miranda, B. 1970. Catalogue of the neotropical Squamata: Part 1-347. I. Snakes. Bull. U.S. Nat. Mus., (297): PlANKA, NATURAL HISTORY Smith, N. G. 1969. Smith, S. 1975. Comparative autecology of the lizard Cnemidophorus tigris in different parts of its geo- A coral snake "mimic" eaten by a bird. 1977. Co- peia, 1964:223. 1976. Roze, J. Multiple cryptic effects of crossbanded and ringed patterns of snakes. Copeia, 1 976:834836. A. 1969. Una nueva of Costa Rica. Rev. 295-349. Savitzky, A. H. rus) Biol. Trop., 21(2): in snakes: An adaptation for swallowing hard-bodied prey. Science, 212: 346-349. Schall, J. J. and Pianka, E. R. Evolution of escape behavior diversity. 1980. Amer. Natur., 115:551-566. SCHOENER, T. W. 1979. Scott, N. 1969. Inferring the properties of predation and other injury-producing agents from injury frequencies. Ecology, 60: 1 1 10-1 1 1 5. J. A zoogeographic analysis of the snakes of Costa Rica. Ph.D. Diss., Univ. So. California. Sharma, 1980. 1948. 1958. Taylor, Taylor. 1943. A Skutch, A. A naturalist in Costa Rica. Univ. Florida Press, Gainesville. Slevin, J. R. 1939. Notes on a collection of reptiles and amphibians from Guatemala. Proc. California Acad. Sci., 23(26):393-414. A contribution to a knowledge of the savanna region of central Peten, Guatemala. Misc. Publ. Mus. Zool. Univ. Michigan, (29): 1-56. The amphibians and reptiles of Alta Verapaz, Guatemala. Misc. Publ. Mus. Zool. Univ. Michigan, (69): 1-109. A study of the Uaxactun-Tikal area of northern El Peten, Guatemala. Contr. Lab. Vert. Biol. Univ. Michigan, (75): 1-30. E. H. Further studies on the serpents of Costa Rica. Univ. Kansas Sci. Bull., 36(1 1):673-801. H. and Smith, H. M. review of the American sibynophiine snakes, with a proposal of a new genus. Univ. Kansas Sci. Bull., 29:301-337. E. A Congdon, J. D. and Dickson, N. A. Adaptive strategies and energetics of tail autotomy in lizards. Ecology, 58:326-337. Vitt, L. J. and Hulse, A. C. Observations on feeding habits and tail dis1973. play of the Sonora coral snake, Micruroides euryxanthus. Herpetologica, 29:302-304. Wilson. L. D. A fracture plane in the caudal vertebrae of Pliocercus elapoides (Serpentes: Colubridae). Amphiesma The F. L. C. J., 1968. rare case of autotomy seen in Coral-snake pattern recognition and stimulus generalization by naive great Kiskadees (Aves: Tyrannidae). Nature, 265(5594):535 Vitt, L. 1977. B. D. stolata (Linn., Serpentes: Colubridae). Snake, 12:60. 1971. 1935. 1954. Hinged teeth 187:759- 536. Stuart, coral falsa del genero Scaphio- dontophis (Serpentes: Colubridae) de Colombia. Caldasia, 10(48):355-363. Savage, J. M. and Vial, J. L. 1974. The venomous coral snakes (genus Micru- 1981. Innate recognition of coral snake pattern by 760. H. F. 1964. M. a possible avian predator. Science, graphic range. Ecology, 51:703-720. Pough, coral snakes. Copeia, 1969:402-404. E. R. 1970. Avian predation of J. Herpetol., 1:93. Wilson. L. D. and Meyer, J. R. 1982. The snakes of Honduras. Milwaukee Public Mus. Publ. in Biol, and Geol., 6:1-159. Zug, G. 1979. R., Hedges, S. B. and Sunkel, S. Variation in reproductive parameters of three neotropical snakes, Coniophanes fissidens, Dipsas catesbyi and Imanlodes cenchoa. Smithsonian Contr. Zool., (300): 1-20. Vertebrate Ecology and Systematics— A Tribute to Henry S. Fitch L Knight. L. Malaret and N. L. Zusehlag Edited by R. A. Seigel. L. E. Hunt. l'»N4 Museum of Natural History. The University of Kansas. Laurence .1 Dominance in Snakes Charles C. Carpenter Dominance behavior has been suggested by a number of observers as an explanation for a function of combat rituals in snakes. Other reported may certain species of snakes, few indicate interactions in the field be- types of behavioral actions which by their nature could be interpreted as relating to dominance are: territoriality, rivalry or competition, winning and defeat, submission, pursuit and chase, flight tween individuals. Kennedy 965) and Bennion and Parker 1976) recorded interactions between male Masticophis and Andren (1975) and Volsoe (1944) have observed males showing aggression ( 1 ( or avoidance, strength and weakness and physically overcome. It is not my aim to review all of the related to breeding sites in I 'ipera berus. Combat most easily recognized agonistic interactions between male snakes (Carpenter and rituals are the literature using these terms, but significant references will be cited below. Ferguson 1977). What must be involved in snake interactions which may result in the dominance of one individual? There must be species recognition between conspecifics. I believe the evidence is good that an individual snake recognizes its own Confusion or uncertainty arises from the lack of criteria for measuring or determining dominance in snakes, and what the functions of dominance might be. Recent definitions of dominance infer this behavior should indicate priority for resources such as food, a mate, a territory (Wilson 1975; Brown 1975). Good evidence for priority use of a resource in snakes is scarce or contact— implicating the vomeronasal modality or channel for species, especially after tongue this recognition (Noble and Clausen 1 936; Kubie 1978; Devine 1976). Sex recognition must also occur and this appears to be rapid between et al. wanting. The criteria or behaviors used by observers to judge dominance vary with the group being observed; what may be used for lizards, birds or mammals, may and accomplished by chemical cues (Noble 1937; Froese 1980). perhaps sometimes conspecifics not apply for snakes. The opI shall use for dominance visually. Individual recognition or identification involves the ability of one member of a species to identify an individual conspecific. There is no erational definition behavior interact in the field. But while these studies show limited movement of in snakes is as follows: Dominance is an interaction between two snakes in which one satisfactory evidence that a snake distinguishes individual performs certain actions (physical or otherwise) which ultimately causes the other in- one conspecific individual from another, though I see no reason they could not accomplish this by detecting individual differences in chemical dividual to avoid these actions (subordinate). Interactions between snakes are difficult to observe in the or visual cues. Individual recognition is considered important in establishing social herarchies in groups of birds and mammals. The observa- although the potential is there for certain species patiently observed. In captive snakes the evidence for social dominance is more direct and field, tions of Barker et al. (1979) strongly suggest that individual recognition occurred in the formation of a hierarchy in a captive group of four males easily observed. Social behaviors are known to occur for snakes during the interactions of courtship and mating, and one female Python molurus. combat Combat Rituals.— Over the past few years I have observed many interactions in captivity between conspecific male snakes. I believe I recognize dominance as a result of most of these interactions whether or not actual combat rituals occurred. Numerous descriptions of combat rituals are available from the literature (Carpenter and Ferguson 1977) and I will use aspects of those descriptions which relate to the establish- rituals, and possibly, parental care and could be associated with different types of aggregations such as feeding, communal nesting, denning and other hibernation groupings, and cover concentrations. The spacing of adults of certain species and their tendency to have limited home ranges (Seibert and Hagen 947; and Cope 1947; Fitch 1949; Carpenter 1952; Barbour et al. 1969) suggests that snakes activity or 1 Stickel es SPECIAL PUBLICATION- MUSEUM OF 196 Fig. 1. Dominant male crawling over subordinate male NATURAL HISTORY in submissive posture. Crotalus viridis. ing of physical superiority and those behaviors which indicate a dominant-subordinate relation- ations the area for flight is small so that the dominant can easily reestablish contact. The domi- ship between two snakes. During a combat ritual between two conspecific male snakes, one, or both individuals at- nant may again perform the solicitation display before the subordinate with the latter sometimes tempt to obtain a higher or superior position by assuming a posture with the head and anterior trunk higher in display than his adversary (Crotalus— Carpenters al. 1976; Sistrurus— Carpenter 1 979), or over that of his adversary (hovering) {Lampropeltis— Carpenter and Gillingham 1977; Murphy et al. 1978). The snake in the superior position then attempts to force his adversary quickly looping around him and pushor throwing (topping) him down (Crotalus, ing op. cit.) or by lowering the hovering head and down by anterior trunk down on the anterior region of his adversary {Lampropeltis, op. cit.) forcing the lower snake down and pinning his anterior region These actions are obvious physattempts by one individual to force another into a lower posture. to the substrate. ical When one individual has been forced to a lowand anterior trunk) the superior snake may then return to the solicer or prostrate posture (head itation display (Carpenter et combat ritual al. 1976) of the with the other again rising in re- sponse and repeating attempts at topping. Such may continue for an hour or more with does not respond but takes a low posture, usually coiling against a wall of the chamber, the dominant then crawls over (Fig. 1) and may lie on the suborrising in response. If the subordinate dinate (Fig. 2). If move away, the the subordinate tries to flee or dominant will continue to crawl on top of him. If the subordinate raises his head, the dominant may respond by moving over his head region, as if forcing him to mainover or lie tain his submissive (negative) posture (Fig. 3). Each time the subordinate shows activity, the dominant may crawl away, and when the subordinate again becomes active, the dominant will start to return at which time the subordinate will try to avoid the approach of the dominant by fleeing, only to be pursued by the dominant, the subordinate attempting to climb the sides of the chamber. When contact is again made, between the two males, the subordinate may again attempt to flee or again assume the submissive posture. During many of my attempts to induce combat of male snakes, no tendency to perform a ritual was exhibited, or only one rituals in a variety actions individual might rise to a solicitation posture. In numerous topping bouts. During these bouts it appears that the more aggressive snake keeps his posterior trunk region over some posterior part such instances, the non-responder, or one of the individuals, appears to act subordinate, assuming a submissive posture without overt physical of the other snake. contact, or Often after a combat has been proceeding for sometime one individual attempts to crawl away from the contact with the other, sometimes ritual becoming hyperactive and violently thrashing away. Following these attempts to move away, the superior snake (dominant) follows or pursues the fleeing snake (subordinate). In captive situ- may which appears actively try to avoid the other dominant. From these ac- to be it appears that dominance may occur without contact between two males. tions Observations of Dominance The following brief summaries are from un- published notes on complete videotapes of the VERTEBRATE ECOLOGY AND SYSTEMATICS Fie. 2. Dominant male Crotalus viridis lying interactions of staged encounters between paired Lampropeltis getulus holbwoki.— After initial contact the larger male continued to follow the smaller male, crawling over him, hovering and pushing down on his anterior region. The smaller male tried to escape or avoid the larger male and at times assumed a submissive coil with his head vibrating his tail when pinned by the larger male. The subordinate male burrowed completefiat, beneath the sand. Lampropeltis getulus holbwoki. — The larger of the two males on first contact immediately ly hovered over the smaller male which tried to avoid contact. The larger male persisted in crawling over, hovering and attempting to pin the smaller male which tried to flee. They separated and when the larger male returned and continued his dominance actions, the smaller male at times tail rapidly. The smaller male finally The male hovered and pinned the min and then persisted during the 35 min they were together, intermittantly, in pursuing, crawling over, hovering and pinning the smaller male, until he was removed. Over this time the larger male bit the smaller male times, some bites lasting over 30 sec. During these long bites, the smaller male was shaken vigorously. The larger male (dominant) exhibited no overt actions towards a second female added at this time. The smaller male (subordinate) remained active during the entire episode moving about the chamber, climbing the sides and pushing its rostrum against the glass front of the chamber. male. larger A second, smaller, male, when introduced, was immediately approached rapidly by the larger male and crawled upon, the smaller male trying to flee to a corner where he backed up body loops against the sides of the chamber. After separating, the smaller male fled from approach of the 1 Crotalus atrox. of combat — M\er approximately ritual, the more 10 min aggressive male per- forming solicitation displays appeared dominant when the other male ceased to display, tried to was pursued, and then assumed a sub- escape, him. Sistrurus miliarias. — The initiator made con- with the other male which jerked in response, then the initiator crawled over the other tact which assumed a loose separated and coil with head down. They made contact as the initiator again away pursued by the The initiator crawled over the fleeing male. The submissive male lay coiled and when contacted waved his tail vertically. the other violently thrashed larger male. Lampropeltis calligaster.—A male L. c. calligaster placed with a slightly smaller female showed no courtship actions for a period of 12 min at which time a smaller male was added. Over the next 50 min the larger male continually pursued, bit and crawled over the smaller male, biting him four times and hovering over and pinning him six times before he was removed, but directed no actions towards the female. 1 missive coil with the dominant crawling over burrowed under the sand and escaped. A male L. c. rhombomaculata was then introduced and within 35 sec the larger male had crawled over the smaller male with the smaller male crawling away pursued by the larger slightly smaller on subordinate male Crotalus molossus. smaller male within 2Vi male snakes. vibrated his 197 initiator. Sistrurus catenatus. — On contact one male im- coil, head down. male moved back and forth over the coiled male, the inferior male waved his tail. If the submissive male moved from his coil, he immediately assumed a submissive coil again when contacted by the now dominant male, the former tail waving. The dominant moved away and then back over the subordinate male several mediately assumed a submissive As the superior times. SPECIAL PUBLICATION -MUSEUM OF 198 NATURAL HISTORY in Fig. Submissive (negative) posture of a male Lampropeltis getulus holbrooki. 3. Bothrops godmani. — On first approach, one snake bit the other. As the two males crawled over each other, one rose to a solicitation display. The biter's actions appeared dominant as he crawled over and lay on the other. The subordinate crawled off and was pursued by the dominant which displayed and kept himself higher than the subordinate. The dominant continued his display actions while crawling over the sub- ordinate, with the subordinate finally thrashing violently to escape, only to be crawled upon by the dominant. pursued and Vipera lebentina.— One male, after performing solicitation display persistently, lay upon crawled over and assumed a head down. the other which coil (sub- missive posture) with its Crotalus triseriatus. — Afler nearly one hour of continuous combat ritual between two males (a female also present) with no apparent superior male, one bit the other, and a few seconds later the one bitten first bit and held onto the other assumed a submissive coil in a corner with the lighter male crawling over him. This light male (C viridis) was then matched with a larger male Crotalus molossus. The C. viridis immediately crawled over the C. molossus which coiled with its head down and waved its viridis stayed on top of tail (no rattling). The the molossus, crawling back and forth and at times rising to display. The C. molossus tried to crawl away but the viridis stayed on top of him and persisted in crawling back and forth, the C. molossus responding by vertical tail waving. When placed with a second male C. molossus, the C. viridis immediately crawled over him as he assumed a submissive coil. Then when a male Agkistrodon contortrix was placed with the C. viridis, the latter crawled over this new male, but the A. contortrix did not form a submissive coil and did not appear to respond to the dominance C C C actions of the C. viridis. A male Agkistrodon piscivorus was then chamber with The placed for 3 to 4 sec. After this reciprocal biting episode, in the displaying ceased and both males lay quiet with no further interaction. immediately performed a solicitation display and persisted in displaying for some time finally il- all Crotalus combat v. viridis. — One male initiated the with solicitation displays, the other responded, but soon moved off, pursued by the initiating male. They again displayed intermittantly ritual and the initiator started to crawl over the other male, which tried to retreat (violently) and then assumed a submissive coil with the now the C. viridis. C. viridis response from the two topping attempts, the C viridis appeared dominant, forcing the A. piscivorus down. At one time the A. piscivorus waved liciting a solicitation display A. piscivorus. In C viridis crawled over him. It his tail as the appeared that the A. piscivorus was trying to avoid the persistent actions of the C. viridis. dominant initiating male crawling back and forth over the subordinate. Crotalus viridis. — The lighter male persistently When Published records of species for which domi- contact he nance, dominance-like behavior, submissive be- The darker male havior and territoriality have been stated, sug- followed and crawled over the darker male. the darker male tried to move from was pursued by the male. light Observations by Others VERTEBRATE ECOLOGY AND SYSTEMATICS gested or inferred follow. to A brief reference is made interpretation of each account. my Boidae.— Python molurus (Barker Dominance with linear hierarchy. et al. 1979). Combat ritual with spur gouging and escape. Individual recognition. Sanzinia madagascariensis (Carpenter Dominance-subordination resulting et al. 1978). from combat ritual Colubridae. el al. 1976). Combat obsoleta bairdi (Brecke ritual with strong rivalry. obsoleta (Rigley 1971). Combat ritual suggesting dominance as a result. Coronella aus- Elaphe o. (Andren and Nilson 976). Males bite while fighting. Coluber viridiflavus (Guibe and Saint Girons 1955). Combat ritual with the victor first triaca 1 to mate. Lampropeltis triangulum Flight by subordinate after combat (Shaw 1951). ritual. domination and territoriality suggested from combat ritual. Sistrurus miliarius (Carpen- Social ter 1979). Dominance as a result of combat rit- ual. Viperidae.— Vipera sp. (Prior 1933). TerritoVipera aspis (Naulleau 1970). Territoriality suggested. Vipera berus (Andren riality suggested. and loser, with chasing, combat rituals. Vipera berus (Volsoe 1944). Dominance suggested, with winner of combat ritual pursuing female. Vipera berus (Guibe and 1975). Indicates winner with spur gouging. — Elaphe 199 Lam- in Saint Girons 1955). Territoriality suggested. There are many other descriptions of combat rituals in the literature but these observers did not record the consequences or conclusions of interactions or were possibly not aware that dominance might be occurring. propeltis getulus holbrooki (Carpenter and Gillingham 1977). As a result of a combat ritual one Discussion male exhibited dominance actions, the other male subordinate behaviors. Lampropeltis mexicana alterna (Murphy of combat et al. 1978). Dominance as a ritual. Lampropeltis pyromelana (Martin 1976). Aggression with biting and result (Kennedy 1965). Territoriality and dominance proposed from aggression on mating area. Pseudaspis cana (Fitzsimons 1962). Males fight vigorously with gashing bites. Ptyas mucosus (McCann 1935). chasing. Masticophisf. JJagellum In a recent paper (Carpenter 1977) I discussed the role of different signal channels in communication between snakes, stressing the impor- tance of tactile actions in agonistic and courtship interactions. Tactile and visual communication appear to play the significant roles in determining dominance and subordination in snakes. The chemical senses may also be important. The actions employed in combat rituals and Territoriality suggested in contest for supremacy. Elapidae. — Demansia textilis (Fleay 1937. 1951). Dominance was indicated with the other agonistic interactions involving contact (tactile) signals are tongue flicking, crawling over, "weaker" subordinate retreating, the "stronger" male intimidating rivals. Pseudechis porphyria- down, entwining, spur use and cus (Fleay 1937, 1951). Territorial "right" sug- combat gested from ritual. dorsal crawl, lying on. pinning, topping, pushing biting, and are apparently used in determining dominance. Visual signals are those of vertical or oblique displays, hovering, pursuit, The subordinate snake Crotalidae.— Agkistrodon piscivorus leucosto- and dominance from combat ritual. Crotalus adamanteus (Wagner 1962). During combat ritual the dominant has more stamina. Crotalus atrox (Jenni 1966). Dominance apparent for victory and defeat clearly defined as result of combat ritual. ma (Perry 1978). Suggests territoriality Crotalus atrox (Foree 1949). Territoriality suggested. Crotalus cerastes (Lowe and Norris 1950). Territoriality suggested of combat ritual. and discussed as a result Crotalus horridus atricaudatus (Sutherland 1958). Dominance a possible result of combat ritual. Crotalus lepidus klauberi (Carpenter et al. of combat Dominance in Dominance apparent result Crotalus ruber (Shaw 1948). apparent from the result of rivalry 1976). ritual. combat ritual. Crotalus v. viridis (Thorne 1977). and perhaps approach. signals its submission by avoiding the dominant, thrashing on contact, fleeing (retreat), tail waving, submissive (negative) posture, and sometimes burrowing, which are visual signals. If the subordinate raises his head (visual) or begins to move (tactile or visual) dominant will refits the criterion submissive The posture spond. these are signals to which the for submissive postures in other animals, that offering the lowest or smallest profile. I believe the evidence is strong that in is, many instances the interplay of these agonistic signals results in individual male snakes becoming dominant and their adversaries becoming subordi- nate. is the function of individual dominance male snakes? The resources over which male What in SPECIAL PUBLICATION- MUSEUM OF 200 snakes might compete are a mate, food, and space. The strongest evidence of possible competition for a mate is and localized in Vipera berus. A male having identified a reproductive female and repeatedly chase off, other males (Andren 1975;Volsoe 1944), with possible temporary territoriality. The evidence for competition for food and space is less evident, though combat rituals are noted in the presence of food will fight with, (Shaw 1951; Sutherland 1958). Since many of observations occurred in the absence of a female or food, dominance does occur in the absence of these resources. my We need to know more about mating strategies in how snakes and these may relate to domi- nance. Shine (1978) provided data that "reveals between the occurrence of male a high correlation combat, and sexual dimorphism in which the male is the largest sex" and states "These results strongly support the hypothesis that large male size is an adaptation to intrasexual competition." Dominance a natural consequence of intra- is We need evidence of sexual male competition. a resource reward for dominance. What are the taxonomic relationships of dom- NATURAL HISTORY combat general lack of biting during rituals and dominance-subordinate encounters supports this. Certain of the actions seen performed by the i.e., the dorsal crawl and crawling over, are similar to the actions performed by a male courting a female, and a courted female may as- dominant, sume a submissive posture, or retreat and be chased by a male. The similarity of these subordinate actions by a male to those of a courted female may provide communication signals that lead the dominant male to homosexual action, aligning next to the subordinate male and attempts to tail search and effect intromission, i.e., though chemical signals should direct otherwise. Is dominance related to larger size? This appears to be true for the Lampropeltis getulus holbrooki observed, but more detailed measurements of size and weight are needed to verify this assumption. The observations of Crotalus male dominated larger males of different species (perhaps an artifact of viridis indicate that this captivity). The determination of the existence and dominance as a social factor in nificance of sig- nat- urally occurring populations of snakes will be my inance? Since combat rituals have been observed difficult. in the Boidae, Colubridae, Elapidae, Crotalidae an awareness of this phenomenon of dominance domi- by other herpetologists and stimulate them to watch for this behavior in the field. and Viperidae, nance is likely this spread suggests that to be a phenomenon occurring in The fact that it is recorded groups of snakes. mostly for the larger species of snakes may be due to the difficulty, or lack, of observing the all smaller species. For those species where aggregations are mon and multiple courtships occur (two or I hope that observations will create Summary The existence of dominance-subordinate re- lationships between individual conspecific snakes commore i.e., has been suggested by observers recording combat rituals, mostly from captive encounters. Using an operational definition of dominance in appears that combat rituals do not occur and that dominance is not through various actions over a subordinate which males courting a female some at the natricine colubrids), same time, it likely to occur. there individual recognition between male snakes and is this a necessary attribute of domIs inance (this is suggested in Python molurus by Barker et al. 1979, where a linear hierarchy occurred)? When does the onset of dominance interac- tions occur, turity? i.e., at a certain size, age, sexual Most combat rituals ma- have been observed (presumably sexually mature) males. The establishment of dominance, whether by in large combat pears to dominance actions, apthe ritualistic function of gaining rituals or other fulfill superiority without significant physical harm. The snakes based on one snake exhibiting superiority in turn performs certain actions, I believe the evidence is clear that dominance does occur in certain species. The dominant male performs actions such as displaying higher and attempting to force his op- ponent to remain lower by forcing him down, by topping or pinning and then persistently crawling over or lying on the subordinate male; biting is The subordinate snake shows it submission by avoiding, fleeing, tail waving or assuming a submissive posture. The dominant will pursue the subordinate if it flees and will respond to movements from the submissive posvery infrequent. ture, repeating its dominant actions. VERTEBRATE ECOLOGY AND SYSTEMATICS A series of observations of male encounters dominancesubordinate relationships using the above befor nine species of snakes all indicate obsoleta bairdi (Yarrow). Herpetologica, 32: 389-395. Brown, J. L. 1975. haviors. Literature records for 25 species of snakes suggest the occurrence of dominance-like behavior. The evidence that dominance functions to give priority for a resource has not been adequately demonstrated, and is often observed in the absence of food, a mate, and in a confined space. Dominance and dominance-like behavior have been observed in five different families of snakes (Boidae, Colubridae. Elapidae. Crotalidae. Viperidae). More information will be needed if different mechanisms are used determine determining dominance The dominance in these combat rattlesnakes 1979:638-642. rituals, while subordinate behaviors are likely to be Carpenter, C. C. 1952. Comparative ecology of the common garter snake (Thamnophis s. sirtalis), the ribbon snake (Thamnophis s. sauntus) and Butler's garter snake (Thamnophis butleri) in mixed populations. Ecol. Monog., 22:235-258. Carpenter, C. C. 1977. Communication and displays of snakes. Amer. Zool., 17:217-224. Carpenter, C. C. A combat ritual between two male pygmy 1979. in actions of snakes are likely to be related to the behaviors used in The Evolution of Behavior. W. W. Norton & Company. Inc.. New York. to and other groups. more similar between families. J. wish to thank James B. Murphy and his staff Department of Herpetology at the Dallas Zoo, Dallas, TX and Frank Bryce of the VenI in the Am Laboratory in Cache, Oklahoma for the courtesies extended in the use of their facilities. ritual of the rock rattlesnake dominance. Southwestern Naturalist, 22: 517-524. Carpenter, C. C, Murphy. J. B. and Mitchell, Literature Cited A. with spur use in the Madagascan boa (Sanzmia madagascariensis). Herpetologica, 34:207-212. Combat bouts Devine, M. C. C Social behavior of Vipera berus during the reproductive period. Norwegian J. Zool.. 24: 234-235. and Nilson. G. Hasselsnoken (Cornonella austriaca) — a 1976. utrotningshotad omart! Fauna och Flora, 2: 61-76. Barbour, R. W., Harvey. M. J. and Hardin, J. W. 1 969. Home range, movements, and activity of the Andren. Copeia, B. The combat L. 1975. rnilianus). (Crotalus lepidus). Copeia. 1976:764-789. Carpenter, C. C. and Gillingham, J. C. A combat ritual between two male speckled 1977. kingsnakes (Tampropeltis getulus holbrooki: Colubridae, Serpentes) with indications of 1978. Andren, (Sistrurus Carpenter, C. C. and Ferguson, G. W. 1977. Variation and evolution of stereotyped behavior in reptiles. Pp. 335-554. In Gans. C. and Tinkle, D. W. (eds.). Biology of the Reptilia Vol. 7. Academic Press, London. Carpenter, C. C, Gillingham, J. C. and Murphy, 1976. Acknowledgments 201 C worm snake. Carphophis amoenus amoenus. Ecology. 50:470-476. Barker, D. G., Murphy, J. B. and Smith. K. W. 979. Social behavior in a captive group of Indian pythons. Python molurus (Serpentes, Boieastern 1976. Species discrimination in mate selection by free living male garter snakes and experimental evidence for the role of pheromones. Herpetological Reviews, 1976:(abstract). Fitch, H. S. 1949. Study of snake populations in central California. Amer. Midi. Natur., 41:513-579. Fitzsimons, V. F. M. 1962. Snakes of Southern Africa. MacDonald and Co., Ltd., London. Fleay, D. 1937. Black snakes in combat. Proc. Roy. Soc. N.S. Wales. Aug.. 40-42. 1 dae) with formation of a linear social hierarchy. Copeia, 1979:466-471. Bennion, R. S. and Parker. W. S. Field observations on courtship 1976. and aggressive behavior in desert striped whipsnakes, Masticophis t. taeniatus. Herpetologica, 32: 1976. B. J., An Murphy. 1951. Savage battle between snakes. Walkabout. 17:10-13. FOREE, K. 1 949. Dallas trio witness rare spectacle rattlesnake courtship or death battle. Dallas Morning News, 2:5. Froese, R. D. 30-35. Brecke. Fleay, D. J. B. and Seifert, W. inventory of reproduction and social behavior in captive Baird's ratsnakes, Elaphe 1980. Reptiles. Pp. 39-68. In Roy. M. A. (ed.). Species Identity and Attachment. Garland STPM Press, New York. SPECIAL PUBLICATION -MUSEUM OF 202 Guibe, and Saint Girons, H. J. Espace vital et territorire chez La Nature, 245:358-362. 1955. Jenni, B. 1966. Perry, les reptiles. NATURAL HISTORY J. An observation of "dance behavior" in the western cottonmouth, Agkistrodon piscivorus leucostoma (Reptilia, Serpentes, Viperidae). J. Herpetology, 12:429-430. 1978. Combat dance of the male rattlesnake rarely seen by man. Outdoor Oklahoma, 22:6-7. Kennedy. Prior, H. T. Territorial behavior in the eastern coach- 1965. ample of 492-493. whip. Masticophis flagellum. Anat. Rec. 151: KiBiE, 499 (abstract). Vagvolgyi, A. and Halpern, M. Roles of the vomeronasal and olfactory systems in courtship behavior of male garter snakes. J. Comp. Physiol. Psychol., 92:627- J. L., 1978. "Combat dance" of 1971. gressive behavior and territoriality in snakes. Natur. Hist. Misc., (66): 1-1 3. B. E. Notes on breeding behavior in a captive pair of Sonoran mountain kingsnakes, (Lampropeltis pvromelana). Bull. Maryland Herpetol. Soc, 12:23-24. Murphy, J. B., Tyron, B. W. and Brecre, B. J. 1978. An inventory of reproduction and social be1 976. havior in captive gray-banded kingsnakes, Lampropeltis mexicana alterna (Brown). Herpetologica, 34:84-93. McCann, 1935. Shaw, 1 C. E. The combat "dance" of some crotalid 948. Shaw, C. E. 1951. Male combat in American colubrid snakes with remarks on combat in other colubrid andelapid snakes. Herpetologica, 7:149-168. Shine, R. 1978. Sexual size dimorphism and male combat in snakes. Oecologia, 33:269-277. Stickel, 1 947. W. H. and Cope, J. B. The home ranges and wanderings of snakes. Copeia, 1947:127-136. Sutherland, I. D. W. The "combat dance" of 1958. rat snake {Zamenis mucosus) fighting. Natur. Hist. Soc, 38:409. Bombay Thorne, E. T. 1977. Sybille life, Espace vital et territorire chez Vipera aspis. In Richard, G. (ed.), Territoire et Dominance Vital. Serie Ecologie et Ethologie. No. Masson 1 Noble, G. K. and Clausen, H. J. 936. The aggregation behavior of Storeria dekayi and other snakes with special reference to the sense organs involved. Ecol. Monog., 6: 269-316. Wyoming Wild- Structure and seasonal variation of the male 944. reproductive organs of Vipera bents. Spolia Mus. Zool. Hauniensis (Copenhagen), 5:1- et Cie, Paris. The sense organs involved in the courtship of Storeria, Thamnophis and other snakes. Bull. Amer. Mus. Natur., 73:763-725. Creek snake dance. 41:14. VOLS0E, H. Noble, G. K. 1937. the timber rattle- snakes. Herpetologica, 14:23-24. Male 1. snakes. Herpetologica, 4:137-145. Naulleau, G. 1970. the black rat snake, 1 C. J. remarkable ex- Elapheo. obsoleta. J. Herpetology, 5:65-66. Seibert, H. C. and Hagen, C. W., Jr. 947. Studies on a population of snakes in Illinois. (Crotalus cerastes), with a discussion of ag- Martin, A Copeia, 1946:6-22. and Norris, K. W. Aggressive behavior in male sidewinders, H., Jr. 1950. the adders. reptilian rivalry. Countryside, 9: RlGLEY, L. 641. Lowe, C. J. The dance of 1933. J. P. 157. Wagner, 1962. R. T. Notes on the combat dance in Crotalus ada- rnanteus. Bull. Phildephia Herpetol. Soc, 10: 7-8. 1 Wilson, 1975. E. O. Sociobiology. The New Synthesis. Harvard University Press, Cambridge. Vertebrate Ecology and Syslematics — A Tribute to Henry S Fitch Edited by R. A. Seigel. L. E. Hunt. J. L. Knight. 1 Malaret and N. L. Zuschlag *S4 Museum of Natural History. The UniversitJ of Kansas. awrence < l 1 l An Experimental Study of Variation in Habitat Selection and Occurrence of the Deermouse, Peromyscus maniculatus gracilis John H. Fitch Introduction at Habitat selection has important consequences the levels of both the individual and the pop- ulation. Many organisms must actively select lection has not been extensively studied. Miller (1973) reported that prairie deer mice (Peromyscus maniculatus bairdi) from North Dakota selected a simulated forest habitat corresponding the type of habitat in which to live from a variety of accessible choices. The choice of a particular to their natural habitat; in Michigan, habitat exposes the organism to a specific set of dividuals selected that simulated habitat in favor selective pressures that can profoundly affect its survival and breeding success (Partridge 1978). Variations in habitat selection may lead to modifications level of gene frequencies at the is available, in habitat selection local or regional populations how- among occupying the same The purpose of this study was to investigate the local variations in habitat selection among (Doyle 1975). two populations of the woodland deer mouse. Peromyscus maniculatus gracilis, in relation to observed differences in habitat occurrence. Hab- type does not provide sufficient evidence that individuals are actively choosing that habitat (Klopfer 1969). Habitat occurrence may also be affected by external factors such as predation itat occurrences of populations were first verified and an open lichen-grass habitat 36 kilometers apart by trapping studies. Animals from each site were then allowed to choose between simulated forest and open habitat models under controlled laboratory conditions. These laboratory tests were needed to establish whether in a forest habitat 1965), interspecific competi- Turner 1961). P. biome. population positive correlation between the spatial distribution of a population and a specific habitat tion (Sheppe 1961: on variations where in grassland habitats, in- of forest. Little information ever, A (Kettlewell 1959. m. bairdi occurs only intraspecific competition (Kluyver and Tinbergen 1953). or by internal morphological or physiological constraints (Bursell 1960; Klopfer 1969). Habitat and its relationship to habitat occurrence must therefore be verified experimentally selection variations in habitat occurrence reflected real dif- (Meadows and Campbell ferences in habitat selection or whether they indicated acceptance of suboptimal habitats be- 1972). Experimental studies that test the relationship between habitat selection and habitat occurrence usually allow animals to select a specific habitat or model of the habitat from several choices under controlled conditions. The relationship between habitat selection and occurrence has been experimentally verified and reviewed in vertebrate species offish (Sale 1969; Casterlin and cause of competition. The woodland deer Reynolds 1978). amphibians (Wiens 1970, 1972), reptiles (Kiester et al. 1 975). birds (Klopfer 1 965; Michigan, and as Hilden 1965), and mammals Canada from Pennsylvania north to southern Quebec and from western Manitoba east to Maryland (Hall and Kelson 1959). In Michigan. P. the Upper is widely distributed throughout Peninsula, on many islands in Lake far 1952; This subspecies is generally restricted to woodland habitats and occurs most commonly in up- vary within land deciduous associations of maple, birch, and beech in northern Michigan (Dice 1925; Blair (Harris may m. gracilis south as Missaukee County in the lower peninsula of the state (Burt 1946). Wecker 1963; Fitch 1979). Patterns of habitat selection mouse occurs throughout the northeastern United States and southeastern and among populations of a species or even sub- New York. m. gracilis species. Intrapopulation variation in habitat se- 1941). In lection has been documented experimentally in plankton (Doyle 1975, 1976) and in mice (Fitch to 1979). Interpopulation variation in habitat se- ecotonal areas (Klein 1960). Harris (1952) reported that P. m. gracilis from the Upper Pen203 woodland P. is restricted habitats even in forest-grassland SPECIAL PUBLICATION-MUSEUM OF 204 NATURAL HISTORY miles Fig. 1. General Refuge (see map of Michigan showing the location of the Kingston Plains (see 1) and the Cusino Wildlife 2). insula of Michigan consistently selected simulated forest models in preference to grassland models under controlled laboratory conditions. of hardwoods, and infrequent man-made clearings. Most of the forests in this part of the Upper Peninsula were logged over in the late 800's and 1 now Fitch (1979) discovered a population of P. m. gracilis inhabiting an open lichen-grass habitat are in the by associations of northern white cedar {Thuja occidentalis), black spruce (Picea mariana), and balsam (Abies balsamea). Alder (Alnus rugosa), Labrador tea (Ledum groenlandicurri), and wintergreen (Gaultheria procumbens) are found commonly in this habitat. The upland hardwood forests have associa- Kingston Plains, Alger County, Michigan, approximately 36 kilometers from the Cusino Wildlife Refuge, where Harris (1952) obtained his animals. In the present study, trapping done was both study areas to confirm patterns of habitat occurrences, and animals from both loin calities were then tested for habitat selection. Study Areas Cusino The Cusino Wildlife Refuge is located in the southern part of Alger County, Michigan, approximately 47 kilometers south of Lake Superior (Fig. 1). The region is characterized by lowland swamp conifer habitats, upland associations secondary growth. Lowland swamp conifer habitats are charac- terized tions of striped maple (Acer pensylvanicum), sugar maple (Acer saccharum), and American beech (Fagus grandifolia). Elderberry (Sambucus pubens), bedstraw (Galium sp.), and bracken fern (Pteridium aquilinum) are commonly found in upland forests. Man-made clearings are dominated by blue(Poa nemoralis) and orange hawkweed (Hieracium aurantiacum). grass VERTEBRATE ECOLOGY AND SYSTEMATICA 205 4 \ \ ?$*& <rT*. Fig. 2. habitat in the Kingston Plains characterized by lichen-grass associations in the study grid. and stumps served as shelter areas for P. m. gracilis. Open Partially buried logs Kingston Plains The Kingston Plains is located in the north- eastern corner of Alger County, Michigan, ap- proximately rior (Fig. 1). kilometers south of Lake SupeThis area, approximately 28 square 1 1 kilometers in extent, was once a well-developed upland red and white pine forest (Jenkins 1942). From 1880 was extenmajor fire burned to 1890. the Plains area sively logged, and in 1890 a and subsequent fires desoil the layer and exposed rubiorganic stroyed over the area. This fire con sand (Veatch et al. 1929). Much of the Kingston Plains is still characterized by open, well-drained expanses of old Open habitats are characterized by common associations of hair cap moss (Polytrichum commune) and lichens (Cladonia mitis, Cladina al- and C. rangiferina) interspersed with sparse clumps of hair grass (Deschampsia flex- pestris, ( Vaccinium bracken fern (Pteridium aquilinum). sour- uosa) and Festuca rubra. Blueberry sp.), dock {Rumex sp.) and orange hawkweed (Hieracium aurantiacum) are common in open habitats. Small isolated woodlots have associations of red maple (Acer rubrum). white pine (Pinus strobus), American beech (Fagus grandifolia). and white birch (Betula cerulea). charred tree stumps and occasional snags with vegetational associations of lichens, grasses, and Materials and Methods ferns (Fig. 2). Small, shallow lakes occur in lowland areas, and small isolated enclaves of pine, Determination of Habitat Occurrence maple, and birch are found in some parts of the Kingston Plains. Attempts have been made to Habitat occurrence in P. m. gracilis was determined by setting out transects of Sherman live- reforest small sections with red and white pine, but have met with limited success due to the traps in three habitat types (swamp conifer, up- land hardwood, and open field) at Cusino and marginal soil conditions. two habitat types (woodlot and open lichen-grass SPECIAL PUBLICATION-MUSEUM OF 206 Table 1. NATURAL HISTORY Vegetation used in simulating forest and open habitat models in compartments. Habnal model Approximate surface area or density in compartment Plant species Cladonia mitis, grey lichen Cladina rangifenna and C. Open alpestris, lichen Deschampsia flexuosa, hair grass Pinus strobus.whixe pine (stumps and logs) Pinus strobus, white pine (stumps and Acer rubrum, red maple (leaves) Acer rubrum. red maple (saplings) Forest baited with a mixture of rolled oats and peanut butter. Each habitat type was sampled for four consecutive nights with 50 traps each night, for a total of 200 trap nights. Transects were moved each night in order to sample other areas within each habitat. Trapping was done between 1 5 August and 20 September 1973. Information on size, weight, sex. reproductive condition and estimated age was taken on all captures. Males captured from Cusino upland forest habitat and from Kingston open habitat were marked and returned to the laboratory for habitat selection tests. Several females from these habitats were also marked and returned to the laboratory in the hopes of establishing breeding colonies. and The remaining females were marked released. Determination of Habitat Selection of surface area of surface area of surface area 1 logs) 1 0% 0% 90% of surface area of surface area .43 per sq. .43 per sq. Betula cerulea, white birch (saplings) Pinus strobus, white pine (seedlings) associations) in the Kingston Plains. Traps were set in transect lines at 10-meter intervals and 30% 60% .43 per sq. way was connected m m m to the floor of each com- partment by a sloping, wire mesh causeway. Structural vegetation characteristically found and open habitats during the autumn season was used to create a habitat model in each in forest compartment. Species of plants used in the compartments and their approximate surface areas are listed in Table 1. All species used were either dominant or very common in their respective Ground cover in open habitat comwas dominated partments by associations of lichen, grass, and bracken ferns (Fig. 4). Forest compartments had short red maple, beech, birch, and white pine saplings as well as red maple seedlings and leaf litter (Fig. 3). One small stump and three pieces of logs were placed in both open and forest habitat compartments. Each compartment was provided with a nestbox, running wheel, water bottle, and wire container of food as described by Fitch ( 1979). Water was supplied in each compartment by a 100-ml graduated cylinder that was upended and at- habitats. Apparatus.— The test apparatus used in study to determine habitat selection was the same as that used by Fitch (1979) to evaluate tached to the outside wall so that only the metal drinking tube protruded into each compartment. Food (Purina Mouse Breeder Chow) was placed differences between animals captured from Kingston Plains forest and open habitats. Single animals were placed in one of four pens, each in a Test this measuring 2.44 m x 1.83 m x 1.52 m. Each pen was divided by a plywood partition into two equal compartments each measuring 1.83 m x 1.22 m. A runway (24 cm x 7 cm x 7 cm) mounted 30 cm above the floor connected the Plexiglas compartments. Movements between compartments and total time spent in each compartment were measured by a treadle equipped with a mercury switch installed inside the runway. The run- 6-mm wire mesh container that was suspended in one corner of each compartment. Each pen was illuminated by two 20-watt flu- orescent tubes attached to overhanging crossbeams parallel to the center partition. The rest test room was poorly illuminated in order emphasize habitat cues within the pens. An hours artificial light cycle of 4 hours light and dark was maintained by means of an automatic of the to 1 timer. A faint illumination 1 of 0.01 footcandles was provided during the dark period. All time-related dependent variables were re- VERTEBRATE ECOLOGY AND SYSTEMATICS Fig. 3. Forest habitat compartment. Note runway, wire mesh causeway, and 207 natural forest vegetation. SPECIAL PUBLICATION-MUSEUM OF 208 NATURAL HISTORY i,v*-.-..« Fig. 4. Open habitat compartment. Note runway, wire mesh causeway, and natural vegetation from grass associations in the Kingston Plains. lichen- VERTEBRATE ECOLOGY AND SYSTEMATICA Table Dependent variables used 2. to measure habitat selection simulated forest and open habitat models. for Unit of measurement Variable Initial habitat 209 Habitat compartment choice first entered after release from central runway Percent Percent Percent Percent First night compartment time Second night compartment time Third night compartment time Average night compartment time corded by a 12-volt Esterline Angus Event Recorder in order to analyze both the distribution of time and the total time spent in each compartment. Dependent Variables. — Five different dependent variables were chosen on the basis of pilot measure habitat selection (Table 2). These variables included a time-independent studies to variable, the initial habitat selected initial by each an- was released from a central area. This choice was defined as the first habitat that imal as it each animal actually entered. The other four variables were time-dependent and measured the amount of time that each animal spent in each of two habitat compartments on each of three succeeding nights. Peromyscus are nocturnal; therefore, time spent in each com- partment during the dark hours of the light cycle should more clearly reflect active selection of habitats than that spent during light hours. of of of of time time time time domly strate and upon which the animal placed all four feet tail. Individuals remained in the pens for the foll lowing 3 nights and 2 h days. Nestboxes were checked once daily, during daylight hours, to verify the animal's position. At the end of each test period, the animal sure. was removed from the enclo- each compartment to specific pens to reduce potential pop- room were reduced by orienting habcompartments of the pens in opposite directions. Variations in behavior due to age and sex within the itat were avoided by using males at least 90 days of age as test subjects. Seasonal effects were avoided by using mice captured during the fall effects season. Habitat selection was tested for 12 individuals from Cusino upland forest habitat and 12 individuals from Kingston Plains open lichen-grass habitat. — A Chi square contingenStatistical Analyses. cy test was used to test for differences in the proportions of individuals that initially chose 1969). ment was actually entered. The initial habitat choice was defined as the first compartment sub- in ulation olfactory effects. Potential position effects ing the light-on period. Usually the causeways were explored several times before a compart- 1 spent scent equally in all habitat compartments. Individuals from each habitat were assigned ran- habitat 1 spent in each compartment spent in each compartment Ten animals from each habitat were placed in each pen prior to the tests in order to distribute Experimental Procedure.—The mouse to be was placed in the central runway and plastic transparent doors at either end of the runway were closed. The mouse was allowed to habituate to the runway for 15 minutes after which time the runway doors were pulled up from outside the pen. The animal was then free to enter either compartment via the wire causeways. Tests were usually begun between 600 and 700 hours durtested spent in each compartment compartments that were either from or similar to those natural habitats they were trapped. different in which The four time-dependent variables were recorded in terms of the percentage of time spent in the compartments that modeled open habi- The percentage data were transformed tats. arcsin values in order to distribution for statistical tests to normal (Sokal and Rohlf conform to a One-sample /-tests were used to evaluate the magnitude of habitat selection differences within Cusino and Kingston Plains sample groups for each time-dependent variable. Within these two groups, habitat selection was defined as the statistical difference between the group mean of the percentage time spent in open habitat compartments and a theoretical value of 50% (arcsin = 45), which indicated no specific selection of habitats. was An a priori alpha significance level of .05 each test. set for 210 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 3. Summary of P. m. gracilis captures in relation to trapping effort in various habitat types in the Cusino Wildlife Refuge and the Kingston Plains, Alger Table 4. Chi square analysis of the numbers of individuals that initially chose habitat compartments either different or similar to those natural habitats in County, Michigan. which they were trapped. VERTEBRATE ECOLOGY AND SYSTEMATICS Table Results of a one-sample 5. captured in Cusino mean forest *** ** * P < P < P < .001. .01. .05. of within group means for four time-related variables for animals habitats. Each column lists the arcsin transformation of groupare in relation to significant departures from 50% utilization of open /-test and Kingston open percentages. Probabilities listed habitat compartments. 211 SPECIAL PUBLICATION-MUSEUM OF 212 NATURAL HISTORY variety of habitats on islands than in adjacent mainland areas (Grant 1970; Hatt el al. 1948; that innate preferences for the parental habitat existed in a closely related taxon, Peromyscus Phillips 1964). These island popuwere hypothesized to be under less predation and competition pressure than populations of P. m. gracilis in the nearest mainland maniculatus Ozoga and lations The bairdi. results of these studies make the differ- ences in patterns of habitat selection between charred tree stumps provide shelter, and avian Cusino and Kingston Plains populations espeThese populations are only 36 kilometers apart and yet differ in their patterns of habitat occurrence and habitat selection. Innate patterns of habitat selection may also differ between the two populations. The Kingston Plains has been available for and mammalian predator populations are low. colonization by P. areas. cially interesting. Predation and competition pressures are probably lower for P. m. gracilis in the Kingston Plains than in surrounding regions of continuous forest (Fitch 1979). In open lichen-grass habitats, old Potential competitors such as the meadow vole {Microtus pennsylvanicus) and the red-back vole (Clethrionomys gapped) may restrict the distribution of P. m. gracilis in some areas of northern in. gracilis for approximately 80 years in its present form. At present, conditions such as shelter availability, and lack of predation and competition pressures seem favorable to colonization. An interesting question is: how Michigan (Manville 1949; Ozoga and Verme 1968). Both species, however, were extremely rare in the Kingston Plains, at least at the time was of this study. Kingston Plains selected habitat models in the laboratory that most closely resembled habitats in which they were captured. However, the two habitats were not isolated from one another and In contrast, potential predators such as the redtailed hawk {Buteo jamaicensis) and the great horned owl (Bubo virginianus) were sighted and heard on numerous occasions in the Cusino area. Potential competitors such as Microtus pennsylvanicus and Clethrionomys gapperi were captured frequently in open grassy habitat and swamp conifer habitat, respectively. it colonized? Fitch (1979) has noted that individuals captured from woodlot and open habitats in the there was sufficient movement of individuals between the habitats to indicate that populations were not isolated. Therefore, polymorphic vari- ation in habitat selection seems likely within the Kingston Plains population, with one segment of the population occupying open habitats and the other occupying forest habitats in small wood- Habitat Selection lots. Patterns of habitat occurrence as established by trapping studies were directly related to pat- terns of habitat selection as established by laboratory habitat model selection tests based upon five dependent variables. Group means of indi- viduals captured from Cusino forest and Kingston Plains open habitats were significantly dif- from one another and from no active choice all dependent variables measured. ferent for Patterns of habitat selection established in this study for animals captured from the Cusino forest habitat are consistent with the results re- ported by Harris (1952). Although sample sizes of laboratory-reared offspring were unfortunately too small for statistical analyses, the trend was to select forest habitat models similar to those selected by their parents. Harris (1952) reported similar results and hypothesized that habitat selection might involve an innate preference for Wecker (1963) confirmed the parental habitat. Polymorphic variation may occur less comthe Cusino population. The two wild caught individuals from Cusino that selected open habitat type model compartments might be examples of such variation. Such animals might colonize open habitats in the Cusino monly within area if they encountered the same favorable qual- that existed in the Kingston Plains open habitats. In the Cusino area, however, such anities imals may be less frequent in the population because they are under greater selective pressures from competition and predation when they enter open habitats. Summary Unusual patterns of habitat occurrence were discovered in woodland deermice, Peromyscus maniculatus gracilis, that inhabited ecologically disturbed habitats of the Kingston Plains. Alger VERTEBRATE ECOLOGY AND SYSTEMATICA 213 and occurrence of the Kingston Plains population were compared with those of a population and Peter G. Murphy added much by providing a blend of advice, encouragement, and stimulating criticism. Rollin H. Baker and John A. in. gracilis occurring in forest habitat within the Cusino Wildlife Refuge, 36 kilometers to the south. The following results were obtained: King also furnished necessary scientific equipment, financial support, and laboratory space. I would also like to express gratitude and appre- County, Michigan. Patterns of habitat selection of P. 1 ) 2) Habitat occurrence of P. m. gracilis was tested by live-trapping in three habitat types in the Cusino Wildlife Refuge and two habitat types in the Kingston Plains. Capture rates in Cusino habitats were highest in upland forest and no animals were captured from open grassy fields. In the Kingston were equally high in both small woodlots and in open lichen-grass habitats. Capture rates per 100 trap nights were approximately two times higher in both Plains, capture rates Kingston Plains habitats than in Cusino up- land forest habitats. 3) Significant numbers of individuals from Cu- ciation to my itorial suggestions. with great pleasure that I dedicate this habitat selection and occurrence to an of study It is excellent scientist, friend, Fitch. and father, The substance of this volume Henry S. honor in his bears testimony to the effects which his enthusiasm, originality, and high standards have had upon family, students, and colleagues alike. His studies, sometimes re" have contribferred to as "Fitchian Ecology. characteristic research uted greatly to the eventual integration of the fields of natural history and theoretical ecology. sino forest and Kingston Plains open habitats selected habitat models simulating the habi- from which they had been captured. Habitat selection was measured by one dependent variable independent of time and four time- wife, Sally, for her assistance in preparing the manuscript and for her helpful ed- Literature Cited tats 4) 5) dependent variables. Group means of individuals captured from Cusino forest habitat and from Kingston Plains open habitat were significantly different from no choice of habitats for all time- Blair. W. F. 1941. The small Size of Mamm.. individuals from Kingston Plains open Burt, W. H. open habitats selected habitat models simulating those habitats in which their respective parents were captured. Different patterns of habitat selection and oc- currence were discovered in populations of P. m. gracilis only 36 kilometers apart. A hypothesis concerning the colonization of the Kingston Plains 1960. 1 946. life 23:27-36. E. The effect of temperature on the consump- The Mammals of Michigan. Univ. Michigan Press, Ann Arbor. XV + 288 pp. Casterlin, M. E. and Reynolds, W. W. 1978. Habitat selection by juvenile bluegill sunfish. Lepomis macrochirus. Hydrobiologica, 59(1): 75-79. Dice, L. R. The mammals of Marion Island. Grand Tra1925. verse County. Michigan. Univ. Michigan. Occas. Paper, Mus. Zool.. 160:1-8. Doyle, R. W. 1975. Settlement of planktonic larvae, a theory of habitat selection in varying environments. was presented. Amer. Natur.. 109:113-126. Analysis of habitat loyalty and habitat pref- erence in the settlement behavior of planktonic marine larvae. Amer. Nat.. 110(975): Acknowledgments wish to express my appreciation to the people who aided me directly in this study. Rollin H. Baker, the late James C. Braddock, John A. King. range and notes on the tion of fat during pupal development in Glossina. Bull. Entomol. Res.. 51:583-598. 1976. I home history of the woodland deermouse and eastern chipmunk in northern Michigan. Jour. BlRSELL, li- forest in 10. 1942. dependent variables. Group means of individuals from Cusino forest habitat differed significantly from those of chen-grass habitat for all dependent variables. 6) Laboratory-reared offspring from individuals captured in Cusino forest and Kingston Plains mammal population of a hardnorthern Michigan. Univ. Mich.. Contrib. Lab. Vert. Genetics. 17:1- wood 719-730. Fitch. J. H. 979. Patterns of habitat selection and occurrence 1 in the deermouse. Peromyscus maniculatus SPECIAL PUBLICATION-MUSEUM OF 214 gracilis. Publ. Grant, 1970. Hall, 1 of the Museum, Mich. E. 959. Hatt, R. 1948. Meadows, C, Pope, C. H. life: figs.. 1 map. Habitat selection 2:53-75. Ann. Zool. Fenn., New trial D. and selection for pattern. Science, 148:1290-1296. Kiester, A. R., Gorman, G. D. and Arroyo, D. C. Habitat selection behavior of three species of Anolis lizards. Ecology, 56(l):220-225. 1 entific Publications, 494 pp. Sale, P. F. 969. Pertinent stimuli for habitat selection by the 1 juvenile manini, Acanthurus triostegus sandvicensis. Ecol., 50(4):6 16-623. Sheppe, 1961. W. 1 969. copus noveboracensis and P. maniculatus New York. Ecol. Monogr., 30:387-407. Klopfer, P. H. 1965. Behavioural aspects of habitat selection: a 1969. preliminary report on stereotypy in foliage preferences of birds. Wilson Bull., 77:376381. Habitats and Territories. New York: Basic 1 Turner, 1961. Veatch, 1929. populations in E. R. A. Survival values of different methods of cam- L. R. and Lesh. F. R. of Alger County, Michigan. Bureau of Chemistrv and Soils, Series 1 929, no. J. O., Schoenmann, Soil survey 32. Wecker, S. C. The 1963. role of early experience in habitat se- by the prairie deermouse Peromyscus maniculatus bairdi. Ecol. Monogr., 33:307- lection J. 325. A. 1970. Effects of early experience on substrate pattern selection in Rana aurora tadpoles. Co- 1972. Anuran habitat selection: early experience and substrate selection in Rana cascadac tadpoles. Anim. Behav.. 20(2):2 18-220. titmice. Archives Neerlandaises de Zoologie, mammal San Francisco: Co., 776 pp. ouflage as shown in a model population. Proc. Zool. London, 136:273-284. Territory and the regulation of density in 10:265-289. Manville, R. H. 1949. A study of small F. J. Biometry. The principles and practice of sta- W. H. Freeman and WlENS, Books, Inc.. 17 pp. Kluyver, H. N. and Tinbergen, L. 1953. Systematic and ecological relations of Peromyscus oreas and Peromyscus maniculatus. Trans. Amer. Phil. Soc, 105:421-446. tistics in biological research. Ecological relationships of Peromyscus leugracilis in central L. J. Small mammals of conifer swamp deerlands in northern Michigan. Mich. Acad. Sci.. Arts & Letters, Papers, 53:37-49. Partridge, L. 978. Habitat selection. In Krebs, J. R. and Davis, N. B. (eds.). Behavioral Ecology: An Evolutionary Approach. Oxford: Blackwell Sci968. J. 1960. and Verme, J. J. Sokal, R. R. and Rohlf, 183:918-921. Klein, 305-348. Ozoga, aspects of the genetic control of indusmelanism in the Lepidoptera. Nature, Insect survival Michigan. Mich. State Univ., Publ. Mus.. Biol. Series, 2(6): 1 in birds. F. Cassel). Mammals of Beaver Island, 1964. An B. Mammalogists (with J. and Phillips, C. J. J. J. A study of the land vertebrates of the islands of eastern Lake Michigan. Cranbrook Inst. Sci., Bull. 27, xi + 179 pp.. Island Kettlewell, H. 1975. I. Ozoga, L. Jenkins, B. C. 1942. Unpublished Pittman-Robertson quarterly reports. Project 6-R, July 15, 1942. Cusino Wildlife Experiment Station files. 1965. J. Ronald Press. 2 vols. van Tyne, J., Stuart, and Grobman, A. B. T., experimental study of habitat selection by prairie and forest races of the deermouse Peromyscus maniculatus. Contr. Lab. Vertebr. Biol., Univ. Michigan, 56:1-53. Hilden. O. 1959. and Campbell, Habitat selection by aquatic invertebrates. Adv. Mar. Biol., 10:271-382. Miller, C. A. 1973. Behavioral habitat selection in Peromyscus and Microtus. Paper presented at 53rd Annual Meeting of the American Society of Harris, V. T. 1965. P. S. 1972. Experimental studies of competitive interaction in a two-species system. The behavior of Micro! us. Clethrionomys and Peromyscus species. Anim. Behav., 18:411-426. R. and Kelson, K. R. The Mammals of North America. New York: 43 1952. northern Michigan. Univ. Michigan. Misc. Publ., Mus. Zool., 73:1-83. State Univ., Biol. Series, 5(6):443-484. P. R. NATURAL HISTORY peia, 1970:543-548. Part IV Systematics and Biogeography Vertebrate Ecology and Systematics — A Tribute to Henry S. Fitch Edited by R. A. Seigel, L. E. Hunt. J. L. Knight. L. Malaret and N. L. Zuschlag 1484 Museum of Natural History. The University of Kansas. Lawrence i Herpetogeography Sierra in the Mazatlan-Durango Region of the Madre Occidental, Mexico Robert G. Webb am- Interest in studying the distribution of tane pastures cut by arroyos (crossing two of them at Rio Chico and Mimbres) to the broad, irreg- phibians and reptiles along a transect in southern Sinaloa and adjacent Durango, Mexico, began in forms the crest of the Sierra Madre. Here, the general elevation of the undulating transect route is about 2438 m (8000 ft) with peaks near 2804 m (9200 ft). The highway then drops approximately 2347 m (7700 ft) in 104.5 road km (65 mi) over an airline distance of about 35 km (22 mi) in descending the steep, ular plateau that June of 955 when I first crossed the Sierra Madre Occidental. At that time the rough, narrow, unimproved road from Villa Union, Sinaloa to Ciudad Durango, Durango, used mostly by busses 1 and was trucks, season. pleted virtually impassable in the rainy Now, a paved road (Highway in November 1960, provides 40), com- mostly west-facing slopes to the coastal lowlands of Sinaloa. The highway in the initial stages of this descent winds along the upper slopes of bar- for year- round transportation along a scenic route; especially impressive is the rough barranca country that straddles the border of the two Mexican states. The paved highway nity to collect rancas that mostly have a southern exposure. These south-facing slopes are cut by spectacular barrancas and canyons and show the most rugged topography of the transect. Just before El Palmito, the highway crosses the Durango-Sinaloa state line, which also marks the juncture of the Central and Mountain Time Zones. Farther west at Loberas, where the Pacific Ocean may be seen on clear days some 96.5 km (60 mi) away, the highway crosses a ridge and begins its switchback route on westerly exposed slopes. The Tropic of Cancer (23°27T5") intersects Highway 40 about affords the opportu- amphibians and reptiles in what would otherwise be itats. The highway in a region of relatively inaccessible habalso crosses the Sierra Madre highest elevation its and thus tra- maximum diversity of habitats. The purpose of this report is to record the known kinds of amphibians and reptiles, to ascertain verses a the kinds of distributional patterns along the transect based on the occurrence of each species each major faunal region, and to relate these patterns to major herpetofaunal assemblages. in 0.7 km east of Santa Rita. to Santa Lucia, less so to transect of the Sierra Madre Occidental The is oriented in a generally northeast-southwest direction in southwestern Durango and southern ( 1 1 1 west to may (198.5 mi) km (14 mi) northtotal route of 320 km 22.5 be traveled in six hours under From Durango on the Mesa del Norte of the Mexican Plateau at an elevation of about 1905 ft) the highway ascends through west of the Continental Di- drainage the year. Climates vary from an arid-tropical in the coastal lowlands of Sinaloa to a cool-temperate on the Mexican Plateau. In both places (vicinities Physiography and Climate (6250 is is westward into the Ocean. The eastern slopes of the Sierra Madre are drained by the Rio del Tunal, a large tributary of the Rio Mezquital. Most of the high plateau of the Sierra Madre is drained by tributaries of the Rio de Acaponeta (to the south) and the Rio del Presidio (to the north). Watercourses on the western slopes drain into the Rio del Baluartes or the Rio del Presidio. Large tributaries intersecting Highway 40 generally have all some water throughout normal driving conditions. m rapid which Pacific and coincides with Highway 40 that 84 mi) meanders for approximately 296 km between Durango, Durango and Villa Union, Sikm south of Villa Union naloa (Fig. 1). About Highway 40 joins the coastal Highway 5, which some Mazatlan. The entire area vide so that Sinaloa, in turn continues is after the highway traverses gently rolling foothills to the relatively flat coastal lowlands (Fig. 2). Description of Transect The The descent Chupaderos, mon- of Mazatlan and Durango), the afternoons, highest in 217 is rainfall, usually in heaviest and temperatures are summer and fall months. The driest SPECIAL PUBLICATION-MUSEUM OF 218 NATURAL HISTORY Topographic map showing transect across Sierra Madre Occidental and spatial relationships of mentioned in text. The numbered localities, identified in gazetteer and arranged west to east, are: 1, Mazatlan. 2, Villa Union. 3, Concordia. 4, Chupaderos. 5, Panuco. 6, Copala. 7, Santa Lucia. 8, Potrerillos. 9, Santa Rita and El Batel. 10, Loberas. 11, El Palmito. 12, Revolcaderos. 13, El Espinazo. 14, Los Bancos. 15, Buenos Aires and Puerto Buenos Aires. 16, La Ciudad. 17, Las Adjuntas. 18, El Mil Diez and El Salto. 19. Estacion and Hacienda Coyotes. 20, Llano Grande. 21, Navios. 22, Rancho Santa Barbara. 23, Mimbres. 24, Rio Chico. 25, Metates. 26, Tapias and Durango. Fig. 1. localities months are generally March through May. Mamore rain (annual average about 86.4 zatlan has cm or 34 in.) and higher temperatures (annual average about 24°C or 75°F) than Durango (48.3 cm or 19 in., and 17°C or 63°F). At Durango about 83% of the total rain into October, in.), August (10.2 cm most of (9.1 or 4.0 it falls from mid-June in July (12.5 cm or 4.9 cm in.). or 3.6 in.), and September The lowest average monthly in.). Occasional west coast tropical cyclones account for deluge rainfall in the Mazatlan-Villa Union area — e.g., 32.0 cm (12.6 in.) of rain fell in 24 hours on 12 September 1968 at Siqueros 30 km NE Mazatlan; Schmidt 1976:22). Near (ca. Mazatlan, the lowest average monthly temperis 19°C (67°F) in January, February, and March, whereas the highest temperatures are 26 ature to 27°C (79 to 8 1°F) from June through October. 1°C (53 to 54°F) in December and January, whereas the highest are 20 moisture-laden, westerly winds sweep inland from the Pacific Ocean and precipitate 22°C (69 to 72°F) from May through August. In winter, cold northerly winds may drop tem- that are often peratures below freezing. At Mazatlan, about 86% of the annual rainfall occurs in the months of the Sierra July through October, most of it in August (24.4 cm or 9.6 in.) and September (27.2 cm or 10.7 orographic precipitation (1976:20) notes that the temperatures are 10 to 1 to Warm, most rain on the highest parts of the Sierra Madre shrouded in clouds and where hail storms are not infrequent. This highest part of Madre provides for the extremes of in Sinaloa. mean annual Schmidt precipi- VERTEBRATE ECOLOGY AND SYSTEMATICS M 219 SPECIAL PUBLICATION-MUSEUM OF 220 go), Crossin (1967, Mixed Boreal-Tropical in Si- Smith (1971, Tropical-Deciduous), and Hardy and McDiarmid (1969, Sinaloa). Hardy and McDiarmid 969) utilized Holdridge's classification and terminology of bioclimates in their naloa). ( 1 herpetofaunal study of Sinaloa, recognizing in the transect area (from east to west) the Lower Montane Dry Forest, Subtropical Dry Forest, Tropical Dry Forest, and Tropical Semiarid Forest. The Lower Montane Dry Forest corresponds Mixed Boreal-Tropical, the Subtropical Dry Forest to the Tropical-Deciduous (here considered somewhat more extensive), and the to the Tropical Semiarid Forest to the Thorn-Scrub. The Tropical Dry Forest, not recognized, is here considered to be a transitional zone between the Thorn-Scrub and Tropical-Deciduous. NATURAL HISTORY manzanita (Arctostaphylos), with grasses common, and sotol, maguey (Agave), and prickly pears in some places. Bare ground, rarely exposed, is covered with grasses, pine needles, oak leaves, and loose rock. The terrain is hilly, rocky, and dissected by numerous canyons. Level areas are extensively cultivated, mostly in corn, and grazed by livestock. The highway dips into two canyons having tributaries of the Rio Mezquital, the Rio Chico and Rio Mimbres. Riparian flora, best developed along the Rio Chico, is principally of large willows (Salix), alder (Alnus), buttonbush (Cephalanthus), smartweed (Polygonum), a small sedge (Eleocharis), and patches of water lily (Nymphaea). This woodland, merging at higher elevations with the Pine-Oak, extends for about 48 km (30 mi) between elevations of about 2103 m (6900 and 7400 ft), the most marked change seemingly about 9 or 10 km (6 mi) west of Mimbres. and 2255 Mesquite-Grassland The western part of the (Fig. 3) Mesa del Norte of the Mexican Plateau in Durango is climatically a grassland of mixed and short grasses, especially grasses {Bouteloua). This grassland has been modified by agrarian development and grazing of livestock, permitting an invasion of Pine- Oak (Fig. 4) On grama shrubby components. In most places the vegetation consists of a low grassy cover, often sparse with bare soil exposed, with scattered herbs, mesquite (Prosopis), huizache (Acacia farnesiana), prickly pear (Opuntia), and occasionally juniper cholla. The relatively level terrain is inter- and rupted by scattered low hills pings. Foothills of the Sierra with rock outcrop- Madre have a rather open scrub cover of catclaw (Acacia), leatherplant (Jatropha), ly some grasses, occasional prick- pears and sotol (Dasyliriori), and a large tree- yucca ( The Yucca). eastern terminus of the transect at at is in the an elevation Mesquite-Grassland Durango, of approximately 1905 (6250 ft). Immediately after leaving the city westward, the highway rises through the yucca-foothill zone, which extends m for about 8 2103 m km (6900 mi) to an elevation of about where a rocky landscape, hav- (5 ft), ing thin dark soils largely concealed by a cover of grasses and scrub oaks, is transitional to the Pine-Oak through a montane savanna or woodland. This transitional, open wooded area consists tall pines, scrub oaks, juniper (Ju- of scattered niperus), pinon pine (Pinus cembroides), and is the plateau-like crest of the Sierra Madre a forest of pines, principally Chihuahua pine (Pinus leiophylla), Durango pine (P. durangenand white pine (P. strobifonnis), and several sis) large oaks. The gently rolling terrain, often with rock outcroppings, has an open understory of grasses and herbs and scattered manzanita, ju- and large madronos (Arbutus). On drier generally at the lowest elevations or on south nipers, sites, or east-facing slopes, oaks are more abundant than pines, whereas moist, deep, protected canyons often support fir (Abies religiosa) and Douglas fir (Pseudotsuga mucronata). Many swift, cold, clear-water streams (small trout and water ouzel observed) drain the plateau. Forested areas are interspersed with extensive meadowy Herbs include various grasses, a yellow areas. aster-like composite, buttercups, violets, geraniums, a white-flowered smartweed, and small euphorbs and mints. Rocky slopes moist from seepage support mosses, ferns, and in some places columbine Much of the area is grazed (cattle), lumbered, and cultivated (mostly corn and po(Aquilegia). tatoes). The Pine-Oak extends for about 1 16 km (72 mi) along the highway at a general elevation of 2438 (8000 ft) where the maximum elevation is about 2804 (9200 ft) between Las Adjuntas and La Ciudad; a few mountain peaks rise some m m VERTEBRATE ECOLOGY AND SYSTEMATICS 221 % Fig. Mesquite-Grassland. Top, 3. Bottom, foothill tree-yucca habitat. 4 Lovelace, Jr.). ca. 182 airline km km N Durango W Tapias, (3 km E La Zarca). Durango (24 July 1973). Durango (22 July 1973. both photographs by Richard C. SPECIAL PUBLICATION- MUSEUM OF 222 NATURAL HISTORY * r -V j' ~ >' , - '- XT" ' - ' < - Fig. 4. :» v#" MP Pine-Oak, 10 road km SW El Salto, Durango (both photographs m 305 (1000 ft) above this general level. A few kilometers west of Buenos Aires the Pine-Oak merges with the Mixed Boreal-Tropical region near 2408 m (7900 ft). ical-Deciduous. 1 1 July 1970 by author). The Mixed Boreal-Tropical cov- mountainous terrain at the highest elevations in large barrancas and canyons, and is best developed on south-facing slopes. Steep ers rugged, boulder-strewn hillsides with rock outcrops, in- Mixed Boreal-Tropical This habitat is unique, is (Fig. 5) relatively sharply de- and is somewhat transitional between the Pine-Oak and, at lower elevations, the Troplimited, terrupted by small, relatively level areas, are covered in most places with a tall pine-oak woodland and often a dense understory of herbs, shrubs, and thick tangles of Pinus oocarpa, vines. P. teocote, Common and P. pines are lumholtzi (the VERTEBRATE ECOLOGY AND SYSTEMATICS "pino triste," an indicator species for this re- Oaks include Quercus macrophylla gion). and Q. viminea; other large glandulosa), and in some places magnolia {Magnolia shiedeana) and hop-hornbeam {Ostrya virginiana). The under(broadleaf). Q.fulva, trees are story, madrono {Arbutus dense in places, includes the large Tithonia calva. Rhus terebinthifolia, and Cer- herbs of the genus the tree-like Bocconia arborea, thick cocarpus macrophyllus, Stevia (spp.). shrubs tall brambles of Rubus, and some poison ivy {Toxicodendron). Open hillsides may have bracken fern {Pteridium) and scattered magueys and small prickly pears. Secluded moist areas may harbor begonias {Begonia) and a tropical bamboo palm (Chamaedorea). Orchids, ferns, lichens, and mosses are common, and many are epiphytic common 223 is most lush in the shaded and narrow mountainous ravines and arroyos and in the larger canyon bottoms. Trees and shrubs include hillsides, morning-glory tree {Ipomoea arborescens), guavas {Psidium), sugar apple {Annona squamosa) pricklenut (Guazuma ulmifolia). coleto {Oreopanax peltatum), sandboxtree {Hura polyandra). the large eardrop tree {Enterolobium cyclocarpum), the Acacia-hke Lysiloma divaricata, large figs {Ficus), as well as the genera Brosunum, Ceiba, Haematoxylwn, Bursera, and Acacia. There is also some bamboo and, in broader valleys, bananas and papayas. Fresh-water crabs {Pseu- dothelphusa) occur in cascading rocky streams. The Tropical-Deciduous occurs for about 77 km (48 mi) along the highway between elevations of about 1798 and 122 (5900 and 400 ft). m {Psit- Thorny acacias become increasingly abundant at lower elevations and with the advent of organ- tacanthus, usually on oaks). Along with mosquitoes, biting black flies {Simulium) are a pipe cactus indicate the transition to the coastal lowland Thorn-Scrub; this rather broad transi- nuisance in the rainy season. The Mixed BorealTropical, recognized elsewhere in Durango (Webb tional with bromeliads (most exserta and T. are Tillandsia benthamiana) and mistletoe zone seems to extend from near Chupa- deros to the vicinity of Concordia. and Baker 962), has some resemblance to a cloud 1 Thorn- Scrub forest. This region occurs for approximately 5 1 km (32 mi) along Highway 40 between elevations of 2408 (7900) and 1798 (5900 ft). Some 8 or 9 m km (5-6 mi) west of Buenos Aires the transition from the Pine-Oak is observed as the highway slowly descends on southerly facing slopes of large barrancas; about 13 or 14 km (8 mi) west of Buenos Aires, the vegetation has a tropical aspect with mosses, ferns, and dripping water on the sheer rock walls of the roadcuts. (9 About 1 5 km mi) west of El Palmito the highway crosses a saddle at Loberas onto the uppermost western slopes of the Sierra Madre and into a transitional zone with the Tropical-Deciduous. Tropical- Deciduous (Fig. 6) The Tropical-Deciduous covers most of the west-facing slopes of the Sierra Madre. At the highest elevations pines and oaks are common on the exposed tops of hills, but at lower ele- vations (ca. 1069 m or 3500 ft, and 4 km below Santa Lucia) pines are replaced by oaks, which in turn are mostly absent below 884 (2900 ft). The oak woodland consists of both deciduous m and evergreen species of Quercus. The probable climax vegetation, modified by clearing on many The (Fig. 7) vegetation of the Thorn-Scrub forms dense about 7 to 9 (25-30 ft) in m thickets, averaging and covers the coastal plain that is some 32 to 40 km (20-25 mi) wide. The relatively level terrain becomes increasingly more hilly inland height, with extensive rock outcrops in some places. The plant cover consists principally of species of Acacymbispina), Mimosa, Cassia, Caesalpinia, and Bursera, and the guamuchil {Pithecollobium sonorae). The organ-pipe cactus cia (mostly A. is scattered and bromeliads and prickly pear, as well as some of the plants of the TropicalDeciduous, are of occasional occurrence. Along the coast, water hyacinth mats are common in the rainy season in roadside sloughs. Large coconut palms {Cocos nucifera) occur near the beach {Pachycereus pecten-arboriginum) characteristic. Terrestrial and a mangrove {Rhizophora mangle) fringes coastal areas. Much of this habitat on either side of the coast highway between Villa Union and Mazatlan is now being cleared for various purposes. Gazetteer The place-names listed below, arranged al- phabetically by states, are those mentioned in 224 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS 225 remarks, including approximate road distance from other localities, elevation, and faunal re- Las Adjuntas: Small village 17.7 km (1 mi) west El Salto, 2515 (8250 ft). Pine-Oak (17). Llano Grande: Large ejido 22.2 km (13.8 mi) gion. Most localities are indicated by signposts. All place-names along the transect route are not go, mentioned below. Elevations are approximate; they may vary many meters (depending on air km the text. Each locality pressure) is followed by descriptive when recorded same place at the 1 m and 70.3 km (43.7 mi) west Duran2408 m (7900 ft), Pine-Oak (20). Los Bancos: Small village observed about east El Salto 1 south of highway, 7 ical (14). Metates: Small ejido on east brim of Arroyo Rio Chico, 20.9 km (13 mi) west Durango and 4.0 km (2.5 mi) east Rio Chico, 2195 m (7200 1 Durango Mesquite-Grassland/Pine-Oak transition (25). village in Barranca de los (9 mi) west Rio Chico and 30.9 km (19.2 mi) east Llano Grande. 2225 m Buenos Aires: Small settlement 4.8 km (3 mi) west La Ciudad and 3.2 km (2.2 mi) east Puerto Buenos Aires, 2591 m (8500 ft), Pine-Oak (15). Coyotes, Estacion: Lumber town about 2 km offhighway, 4 km (2.5 mi) east Hacienda Coyotes and 12 km (7.3 mi) west Llano Grande, 2408 m (7900 ft), Pine-Oak (19). ft), Mimbres: Small Mimbres, 14.5 km (7300 ft), Mesquite-Grassland/Pine-Oak transi- tion (23). Navios: Small village about 11.3 west Rancho Santa Barbara and Coyotes, Hacienda: Ranch 7 km (4 mi) east El Salto, 2454 (8050 ft), Pine-Oak (19). Durango (Ciudad): Capital of state and eastern east Llano Grande, km 198 mi) from Maand 92.5 km (57.5 mi) east El Salto; mileage from highway at Parque Guadiana on west view, 3.2 m m ft), km Espinazo (Espinazo del Mesquite-Grass- (Devil's Backbone) connecting two ranges with drops of several hundred meters on either side; roadside stop and scenic view where east Revolcaderos, 2377 m 2256 monument (7800 real-Tropical (13). El Mil Diez: Small village, 2 at 1.2 ft), km (0.8 Pine-Oak km ft), (12.3 mi) Mixed Bo- km north highway mi) west El Salto, 2515 m (8250 (18). El Salto: Large lumber town about 93 km (58 mi) west Durango and 95 km (59 mi) east El Palmito, Sinaloa, 2469 m (8100 ft), Pine-Oak (18). La Ciudad (=Ciudad): Old lumber camp-town km (16.4 mi) west Las Adjuntas and 4.8 km (3 mi) east Buenos Aires, 2484 m (8150 ft), Pine-Oak (16). ft), (7 ( 1 1 mi) mi) Pine-Oak mi) east El Espinazo, 2560 m (8400 ft). (15). m (7400 ft), Mesquite-Grassland/Pine-Oak transition (22). commemorates dedication of completion of highway on 30 November 1960; 13 km (8 mi) west Puerto Buenos Aires and 20 (8000 km km Rancho Santa Barbara (formerly Weicher Ranch): Cattle ranch 1.7 km (1.1 mi) west Mimbres and 29 km (18 mi) east Llano Grande. Ridge Diablo): (8 Pine-Oak land (26). El m 7.7 Puerto Buenos Aires: Roadside stop for scenic km (2.2 mi) west Buenos Aires and 13 ( (6250 2438 1 (21). zatlan side of city, 1905 (4.2 mi) west Puerto Buenos Aires and about 6 km (3.3 mi) east El Espinazo, 2286 m (7500 ft). Mixed Boreal-Trop- at Each entry terminates with a number in parentheses, which indicates its geoand 2). graphic position on the maps (Figs. different times. terminus of transect, 320 km Revolcaderos: Small village 40.5 west La Ciudad and 10.9 Palmito, Sinaloa, 2042 m km km (25.2 mi) (6.7 mi) east El (6700 ft), Mixed Bo- real-Tropical (12). Rio Chico: Small settlement in arroyo, 4.0 km (2.5 mi) west Metates and 14.5 km (9 mi) east Mimbres, 1981 m (6500 ft), Mesquite-Grassland/Pine-Oak transition (24). Tapias: Small suburb of Durango, 3 km (1.9 mi) west Parque Guadiana, 1905 m (6250 ft), Mesquite-Grassland (26). Weicher Ranch: See Rancho Santa Barbara. Sinaloa 26.3 Fig. 5. Mixed Boreal-Tropical. Richard C. Lovelace, Jr.). 13 road km SW Chupaderos: Small village-truck stop across Rio Chupaderos, 5.3 El Palmito. Sinaloa (both km at bridge (3.3 mi) west photographs 13 July 1973 by 226 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS 227 Fig. 7. Thorn-Scrub. Top, leeward beachside thicket, 2 km N Mazatlan, Sinaloa (note startled ctenosaur atop organ-pipe cactus); habitat now destroyed (photograph 9 August 1957 by author). Bottom, 5 km E Villa Union, Sinaloa (photograph 14 July 1973 by Richard C. Lovelace, Jr.). Fig. 6. Tropical-Deciduous. Top, panoramic view looking west showing Highway 40 and Santa Lucia. Sinaloa. Bottom, arroyo habitat with small creek. 2 km E Santa Lucia, Sinaloa (both photographs 13 July 1973 by Richard C. Lovelace. Jr.). SPECIAL PUBLICATION-MUSEUM OF 228 km turnofF to Copala and 17.5 Concordia, 244 m (800 ft). (10.9 mi) east Tropical-Deciduous/ Thorn-Scrub transition (4). Concordia: Large town 20.7 km (12.9 mi) east junction highways 40 and 15, 122 m (400 ft), Tropical-Deciduous/Thorn-Scrub transition (3). Copala: Mining town (church observed in arroyo at Km signpost 70) reached by dirt road 2 km (1.2 mi) from turnoff at roadside truck stop (Copalita, mileages therefrom), 18 km (11.2 mi) west Santa Lucia and 5.3 km (3.3 mi) east Chu- m paderos, 579 (1900 Tropical-Deciduous ft), (6). Small village 3.5 El Batel: Potrerillos m 1646 and (5400 3.4 km (2.1 km mi) east mi) west Loberas, (2.2 Tropical-Deciduous (9). El Palmito: Large village 1.2 km (0.8 mi) west Durango-Sinaloa state line and 14. 1 km (8.8 mi) east Loberas, 1935 m (6350 ft). Mixed Borealft), Tropical (11). Loberas: Roadside stop for scenic view westward (microondas station, and small group of casitas 0.8 km (2.1 mi to the east, erected in 1970's), 3.4 mi) east El Batel, 1922 m (6300 ft). Mixed Boreal-Tropical/Tropical-Deciduous transition (10). Mazatlan: Seaport-tourist resort on small penand western terminus of transect, about insula 22.5 (50 km ft), (14 mi) northwest Thorn-Scrub ). Villa Union, 15 m ( 1 Panuco: Mining settlement 10 km (6 mi) by dirt road off highway at km signpost 70, 1 Km . 1 Copala and 16.9 km (10.5 mi) west Santa Lucia, 640 m (2100 ft), Tropical-Deciduous (5). (0.7 mi) east NATURAL HISTORY Villa Union (formerly Presidio): Large town on south side of Rio del Presidio, about 22.5 km (14 mi) southeast Mazatlan and 20.7 km (12.9 mi) from Concordia; mileage from junction of km south of town, highways 40 and 15 about 30 m (100 ft), Thorn-Scrub (2). 1 Composition of Herpetofauna This section documents the occurrence of the kinds of amphibians and reptiles in the five herpetofaunal regions along the transect. Introduced species (Gehyra mutilata, Ramphotyphlops braminus), the estuarine crocodile (Crocodylus acutus), sea turtles, and the sea snake {Pelamis pla- turus) are not included. Species will doubtless be added, especially in the Sinaloan tropical habitats. Known ranges probably will be extended northward (e.g., Eumeces parvulus) or southward Syrrhophus (e.g., decurtatus). The some excluded interorbitalis, Phyllorhynchus hiatus in geographic range of species will perhaps be rectified to the north and where records of occurrence south are not now available for the transect area belli, Lowe, Jones, and Wright 1968; Terrapene nelsoni, Smith and Smith 1980; Tantilla bocourti, McDiarmid, Copp. and Breedlove 1976; Trimorphodon tau, McDiarmid and Scott 1970, but see subsequent discussion of Pseudoewycea (e.g., distribution patterns-barranca corridors). Assignment of some species to faunal regions will probably collection be altered pending further data of Ctenosaura pectinata, and sev- (e.g., eral snakes, especially ybelis aeneus, Boa constrictor which are here and Ox- restricted to the mi) east Santa Lucia and 2.4 km (1.5 mi) west Santa Rita, 1615m (5300 ft), Tropical- Thorn-Scrub, probably occur in the adjacent Tropical-Deciduous). Taxa are assigned to a particular faunal region based on their overall dis- Deciduous tribution; several species (e.g., Potrerillos: km Highway construction village 8.5 (5.3 (8). Same as Villa footnote); name re- Presidio (Presidio de Mazatlan): Union (see Conant 1969:89, tained for railroad stop, Estacion Presidio, about 3 km south of Villa Union, and for Rio del Presi- dio (formerly Rio Mazatlan). Santa Lucia: Small village and truck stop, 18 km (11.2 mi) east Copala and 29.6 km ( 1 8.4 mi) west El Palmito, ciduous (7). 1 100 m (3600 ft), Tropical-De- mi) east Potrerillos, 1676 Deciduous (9). m (5500 ft), Tropical- kelloggi, B. represented by many localities in, and assigned to, the Thorn-Scrub penetrate eastward to only the Chupaderos-Copala region, which is barely into but near the transition to the Tropical-De- ciduous. No attempt abundance of species Some Santa Rita: Rancho and restaurant-bus stop, 1.1 km (0.7 mi) west El Batel and 2.4 km (1.5 Bufo marinus, B. marmoreus, Pachymedusa dacnicolor, Smilisca baudini, Coniophanes lateritius) is made to indicate relative in particular faunal regions. however, are represented by only Hylactophryne tarahumaraensis, Sceloporus clarki, S. nelsoni, Eumeces brevirostris, Pituophis deppei, and Crota- one lus species, locality in a region (e.g., molossus in Mixed Boreal-Tropical; VERTEBRATE ECOLOGY AND SYSTEMATICS Sceloporus jarrovi, Gyalopion quadrangularis, and Crotalus lepidus in Tropical-Deciduous) and seem to be of rare occurrence, not to mention the enigmatic Anolis utowanae in the Thorn- Some species, excluded for various reasons, are discussed below. Scrub. — Phrynohyas venulosa (Laurenti). This large arboreal tree frog is reported from Presidio, Si- Boulenger (1882:327), Giinther (1901[1885-1902]:272),andGadow(1905:207). The record represents the northernmost on the west coast of Mexico if the locality is correct. The vicinity of Villa Union has been relatively well explored in recent years but no specimens have become available since. Rana pipiens complex.— The taxonomic status and distribution of ranid frogs of the R. pipiens complex along the transect is unknown. In Sinaloa two species, R. magnaocularis and R. forreri, are sympatric at Concordia (Frost and Bagnara 1976:335). Frogs from the Pine-Oak region in Durango seem to represent the recently described R. chiricahuensis (Platz and Mecham 1979). Another taxon, presumably a subspecies naloa by of R. berlandien, occurs in the Mesquite-Grassland of Durango. — Lepidochelys olivacea (Eschscholtz). The ridmay be the most abundant species of sea dark brown and pale yellow spots on pale brown dorsal surfaces; these spots (brown and yellow alternating) are mostly in parallel rows on the back and tend to form bands on the tail. Yellow spots on the side of head tend to form pre- and postocular stripes. This juvenile pattern becomes obliterated with increase in size with the largest individuals mostly uniform pale brown or with evidence of indistinct yellow spots. Urosaurus needs verification. Coluber from the favored the dispersal of this species. A hatchling 20 SVL, later destroyed) was active (night (ca. mm of 22 August) among window fixtures on the second floor of the Hotel Belmar. Both young and adults were captured on 7-8 June. Young geckos, about 25 mm SVL, have contrasting patterns of state. this species — Wilson The status of this species in Du- of snake in Sinaloa based on the of "Mazatlan." The only other few known records of occurrence in Mexico are from the locality of Michoacan and Oaxaca. Mazatlan another place of that name in may Guerrero comments concerning type-loof Sphaerodactylus torquatus by Taylor 1947:304-305). or Oaxaca (see cality Accounts of The amphibians and Species reptiles considered in the subsequent analysis of distributional patterns consist of 145 taxa. Discussion of them has been deferred for inclusion in the terminal Appendix. near Acaponeta, Nayarit. Gehyra mutilata (Wiegmann). — This intro- duced lizard is abundant at night on the walls of beachfront establishments in Mazatlan. Increased urbanization along the beach north of Mazatlan as observed in the years since 1955 has Jan. rango requires further study. Geagras redimitus Cope. — Hardy and McDiarmid (1969:162) discuss the occurrence of (probably Lepidochelys) are caught in the vicinity of Mazatlan in July, August, and September; individuals float at the surface and are relatively Beach and Isla de la Piedras) in May and early June, but adults are not captured then. The largest rookery in the general area is said to be south oaxaca constrictor (1966) records one specimen of this snake from Coyotes, Durango. No other specimens are known states easy to catch. Flesh and eggs are used locally for food. Some turtles nest near Mazatlan (North (Boulenger).— species on the west coast of Mexico. Occurrence of the species in the Mazatlan-Villa Union area refer to fishermen say sea turtles lateralis the nearest locality as 36 miles north Mazatlan. These two localities are the southernmost for the of Mazatlan. Carapaces and/ or skulls were found on the beach north of Mazatlan on 12 August, 6 and 8 June, and 23 July (different years). Local ornatus Hardy and McDiarmid (1969:141-142) discuss a questionable record for Presidio and mention ley turtle in the vicinity 229 Distribution of Herpetofauna al The herpetofauna considered for distributionpurposes consists of 145 taxa— 2 salamanders (1.4%), 35 frogs (24.1%), 5 turtles (3.4%). 33 lizards (22.8%), and 70 kinds of snakes (48.3%). The assignment of these taxa to faunal regions allows for the discussion of the herpetofauna of each region and the distributional patterns along the transect. Representation of higher taxa in each of the five regions is shown in Table 1. The total number of taxa is greatest in the Thorn-Scrub SPECIAL PUBLICATION-MUSEUM OF 230 Table regions 1 . Frequency (number and percentage) of higher taxa of amphibians and (MG, Mesquite-Grassland; PO, Pine-Oak; MBT, Mixed Thorn-Scrub). Taxa NATURAL HISTORY reptiles in the five herpetofaunal Boreal-Tropical; TD, Tropical-Deciduous; TS, VERTEBRATE ECOLOGY AND SYSTEMATICS Boreal-Tropical, Tropical-Deciduous, and Thorn-Scrub; one of these (Masticophis mentovarius) that seems established in the MesquiteGrassland is discussed below under Barranca Corridors. Sixteen taxa (2 frogs. 2 turtles, 3 lizards. 9 snakes) occur only in the Tropical-De- ciduous and Thorn-Scrub. These variable distributional patterns are depicted in Fig. 2. An unusual distributional pattern, not influenced by the east-west course of the transect, is the occurrence of four species in only the Mesquite-Grassland and Thorn-Scrub. This pattern, represented by Scaphiopus couchi, Bufo punc- Hypsiglena torquata, and Arizona elegans (two subspecies), is explained as southern attenuations of geographic ranges of Nearctic species on either side of the Sierra Madre. Barranca Corridors. — The Sierra Madre Occidental is cut by many large barrancas and artatus, royos that provide corridors for the dispersal of tropical species eastward (all drainage westward to Pacific Ocean): the most notable involved drainage is that of the Rio Mezquital with headwaters draining the vicinity of Ciudad Durango. Documentation of tropical species far to the east barrancas has been previously noted by Baker (1962) and by Crossin et al. (1973). Tropical species may extend into or very near non-tropical areas. in these to east-west dispersal, 231 I have compared only ad- jacent faunal regions along the transect. The numbers of taxa that bridge the four transitional zones between adjacent faunal regions are 15(1 salamander. 5 frogs. 2 turtles. 3 lizards. 4 snakes). Mesquite-Grassland/ Pine-Oak; 1 1 (3 frogs. 2 liz- Pine-Oak/Mixed Boreal-Tropical; 21 (5 frogs, 8 lizards, 8 snakes). Mixed Boreal-Tropical/Tropical-Deciduous; and 24 (2 ards, 6 snakes), frogs, 2 turtles, 8 lizards, 12 snakes), Tropical- Deciduous/Thorn-Scrub. The fewest number of shared taxa (11) suggests the most pronounced faunal break between the Pine-Oak and Mixed Boreal-Tropical. All 1 1 taxa that bridge the tran- zone between those two regions extend their ranges west from the Pine-Oak (Fig. 2); this transition zone is thus most effective as a barrier sition to the eastern dispersal of tropical species. Ek- man's (total (in Udvardy 1969:274) formula A + B compared regions)/C (shared taxa of two which the highest numerical value indicates the greatest faunal change, also marks the most abrupt transition between the Pine-Oak and taxa), in Mixed Boreal-Tropical with a value of 6.73 (Mesquite-Grassland/Pine-Oak. 4.67; Mixed Boreal-Tropical/Tropical- Deciduous, 3.67; Webb and Tropical-Deciduous/Thorn-Scrub, 4.87). A slightly different manipulation of the numbers of total taxa in. and shared taxa between, Dispersal of tropical species eastward in barranca corridors is exemplified by: 1) Anolis neb- each region marks the Mixed Boreal-Tropical as having the greatest discrepancy between percentages of shared taxa with adjacent regions (i.e., 65.6% of the taxa in this region is shared with ulosus and Dryadophis cliftoni near the brims, and Geophis dugesi in the more mesic bottoms, of large canyons near the Pine-Oak locality of Llano Grande, 2) a locality of 9.7 miles west the Tropical-Deciduous and only 34.4% is shared with the Pine-Oak, a difference of 3 1 .2%). In the Durango for Trimorphodon tau (Univ. New Mexico 22790) in Mesquite-Grassland, 3) the locality of "ca. 10 mi SW Durango" for Elaphe Tropical-Deciduous the discrepancy percentage of shared taxa with adjacent faunal regions is 6.7% (53.3% shared with Thorn-Scrub, 46.6% triaspis (Dowling 1960:76), 4) the occurrence of two specimens of Masticophis mentovarius from 6 miles southeast Durango and 7 miles northeast Durango (Johnson 1977:300). and 5) the doubt- with Mixed Boreal-Tropical), and in the PineOak is 10.2% (31.4% shared with Mixed Boreal- continuity of populations of Hylactophryne august i (disjunct along transect route, see species account) provided by rocky barranca habitats (an break less intervening locality for the species from such a habitat is 6 miles southeast Llano Grande. UTEP). Tropical and 41.6% with Mesquite-Grassland). These data indicate the greatest east-west faunal in the Mixed Boreal-Tropical, with ical regions. Although various formulas (primarily to admagnitude of the two compared regions) have been proposed to indi- just for the differences in cate degree of faunal resemblance, Faunal Assemblages In an effort to determine the degree to which each of the four transition zones acts as a barrier the herpetofauna mostly aligned to the western trop- all show the (as does Ekman's formula, see above). For example, the values based on the formulas of Jaccard,C/N, + N 2 - C x 100, and Simpson, C/N, x 100 (in Udvardy 1969:273). same general trend SPECIAL PUBLICATION-MUSEUM OF 232 and of Duellman (1965:677), 2C/N, + N 2 (here modified to avoid decimal fractions) x 100 when applied, respectively, to the four faunal transitions along the transect are: Mesquite-Grassland/ NATURAL HISTORY Neotropical Herpetofauna.— Taxa of tropical occur not only in the Pacific coastal affinities Thorn-Scrub, Tropical-Deciduous, and Mixed Boreal-Tropical, but also in the Mesquite-Grassland that is composed mostly of Nearctic species. and 40.6; Pine-Oak/Mixed Boreal-Tropical, 19.6, 34.4, and 32.8; Mixed Bo- The real-Tropical/Tropical-Deciduous, 37.5, 65.6, transect represent and blages. Pine-Oak, 25.5, 44. 1 , and Tropical-Deciduous/Thorn-Scrub, 25.8, 53.3, and 41.0. Lower values indicate fewer taxa in common to the two areas. These data indicate highest resemblance between the Tropical-Deciduous and Mixed Boreal-Tropical, about the same degree of resemblance between Mesquite-Grassland and Pine-Oak as between ThornScrub and Tropical-Deciduous, and the least resemblance between Pine-Oak and Mixed Boreal54.6; tropical species in these two two regions of the different tropical assem- The Pacific coastal assemblage comprises about 75 (80.6%) of the total of 93 taxa in the three tropical regions. Fifteen of these that seem to reach their northernmost extent of range in the transect area include seven frogs (Eleutherodactylus hobartsmithi, Tomodactylus nitidus, T. Syrrhophus teretistes, Hyla bistincta, H. smaragdina, Gastrophryne usta), four lizards saxatilus, Tropical. The distributional data, as well as the abrupt climatic change personally experienced in winter bulleri, S. heterolepis, S. utiformis, colimensis), and four snakes (Dryadophis c/iftoni, D. melanolomus, Rhadinaea hes- on Physiography and Climate) emphasizes the distinction between the Pine-Oak peria, Leptodeira maculata). (see section (Sceloporus Eumeces Of the 34 taxa in the Mesquite-Grassland, 4 1.8%) are judged to have tropical affinities with the Mesa Central, the southern tropical highland and Mixed Boreal-Tropical. This transition corresponds to that between two major herpetofaunal assemblages, the classic Neotropical and Nearctic zoogeographical realms. Northern Nearctic and southern Neotropical species overlap on either side of the Sierra Madre along the of the Mexican Plateau. The four taxa include one frog (Bufo occidentalis), one turtle (Kinosternon integrum subsp.), one lizard (Sceloporus spinosus), and one snake (Pituophis deppei). east-west trending transect. Nearctic Herpetofauna.— Aside from that of Acknowledgments the Pine-Oak and most of the herpetofauna of the Mesquite-Grassland, some species of Nearc- have extended their ranges far the south on the Pacific side of the Sierra tic affinities to Madre also into tropical habitats. Of the 93 taxa in the three tropical regions along the transect, 18 (19.4%) have Nearctic and all but one of them (Gyalopion recorded only once from the adjacent TropicalDeciduous) are restricted to the coastal ThornScrub. These 18 taxa, some of which seem to be near their southernmost extent of range (marked affinities, with asterisk), consist of four frogs (Scaphiopus Bufo kelloggi, *Bufo punctatus, *Gastro- couchi, phryne olivacea), five lizards (*Coleonyx varie- gatus, Callisaurus draconoides, Holbrookia ele- gans, Sceloporus clarki, Eumeces callicepha/us), and nine snakes (^Arizona elegans, Gyalopion quadrangularis, Rhinocheilus lecontei, Phyllorhynchus browni, Pituophis tnelanoleucus, Salvadora deserticola, Sonora aemula, Tantilla yaquia, *Micruroides euryxanthus). Some of these ( 1 Field work was financed by grants from the Bache Fund of the National Academy of Sciences (1961, Grant 463), the National Science Foundation (1962, part of Grant G-23042 to William W. Milstead), and the Penrose Fund of the American Philosophical Society (1964, Grant 3542). I acknowledge the authorities of the Direccion General de la Fauna Silvestre, Mexico, D.F. for issuing scientific collecting permits. Of the many companions, I am most grateful to Rollin H. Baker and his wife Mary, J. Keever Greer, Leslie C. Drew, and Rudolph A. Scheibner. Field work through the years has been enhanced by the cooperation and hospitality extended by many local residents, especially Rodolfo Corrales and Fidel Gutierrez of Ciudad helpful field Durango. Summary species are represented by tropically adapted In studying the distribution of 145 kinds of amphibians and reptiles across the Sierra Madre subspecies. Occidental from Cd. Durango, Durango to Ma- VERTEBRATE ECOLOGY AND SYSTEMATICS zatlan, Sinaloa, five herpetofaunal regions are recognized (from east to west)— Mesquite-Grassland. Pine-Oak, Mixed Boreal-Tropical, Tropi- cal-Deciduous, and Thorn-Scrub. An enumeration of localities places each of the 145 taxa in one or more region and provides for a Crossin, R. in the coastal Sinaloan Thorn-Scrub, and snakes are the most abundant component in each region. Each region contains endemic taxa with est the highest percentage in the Thorn-Scrub. The most abrupt faunal break is between the Pine- Boreal-Tropical, which also marks the transition between the Nearctic and Oak and Mixed Neotropical zoogeographical realms. New state records include Diadophis punctatus and Pituofor 1 R. H. Biotic relationships in the 1973. Dixon, J. 1969. 1 960. 1 Zoologica. 45:53-80. Duellman, W. A 1965. A biogeographic account of the herpetofauna taxonomic study of the middle American snake, Pituophis deppei. Univ. Kansas Publ. Mus. Nat. Hist., 10:599-610. of Michoacan. Mexico. Univ. Kansas Publ. Mus. Nat. Hist., 15:627-709. Hylid frogs of middle America. Monogr. Mus. Nat. Hist. Univ. Kansas. (1): 1-753. E. R. reptiles of the Mexican expedition of 1934. Proc. Acad. Nat. Sci. Philadelphia. 88:471-477. S. and Bagnara, J. T. A new species of leopard frog (Rana pipiens complex) from northwestern Mexico. Co- The amphibians and D. Storeria storerioides in western Mexico. Herpetologica, 16:63-66. Frost, The life history and systematics of Ambystoma rosaceum. Copeia, 1961:371-377. Armstrong, B. L. and Murphy, J. B. The natural history of Mexican rattlesnakes. 1979. Univ. Kansas Publ. Mus. Nat. Hist., Spec. J. 976. 1 peia. 1 J. K. of the Mexican state of Durango. Publ. Mus. Michigan State Univ., Biol. Ser., 2:25-154. Boulenger, G. A. 1882. Description of a new genus and species of frogs of the family Hylidae. Ann. Mag. Nat. 1 962. Mammals (5)10:326-328' Catalogue of the lizards in the British Museum (Natural History). Vol. II. Taylor & Hist., 1885. Francis, 1887. London. Catalogue of the lizards in the British Museum (Natural History). Vol. III. Taylor & The 905. London. and Fugler, C. M. Chrapliwy. 1955. Amphibians and reptiles collected in the summer of Systematics of the mexicana species group of the colubrid genus Lampropeltis, with an hypothesis [of] mimicry. Breviora, (466): 1- 1982. 36. Gehlbach, Mexico 1 Chrapliwy. Williams, K. and Smith, H. M. 1961. Noteworthy records of amphibians from Mexico. Herpetologica. 17:85-90. P. S., 1963. 1969. R. Semiaquatic snakes of the genus Thamnophis from the isolated drainage system of the Rio Nazas and adjacent areas in Mexico. Copeia, 1963:473-499. A review of the water snakes of the genus Natrix in Mexico. Bull. Amer. Mus. Nat. Hist.. 142:1-140. 1-270, pis. 1-31. C. L. G. Reptilia and Batrachia. Biologia Centrali-Americana. Dulau and Co.. London. 1885-1902. 1: 121-128. Conant, F. R. Herpetology of the Zuni Mountains region, northwestern New Mexico. Proc. U.S. Nat. Mus., 116:243-332. Gloyd. H. K. The rattlesnakes, genera Sistrurus and Cro1940. talus. Chicago Acad. Sci. Spec. Publ.. (4):i1965. vii, in Mexican amphibians and London. 2:191-244. Garstka, W. R. Gunther, A. 1953. Herpetologica. distribution of reptiles. Proc. Zool. Soc. Francis. P. S. 1976:332-338. Gadow, H. Publ., (5): 1-88. Baker. R. H. and Greer, E. 1960. Dunn, 1961. Rio Eumeces brevirostris group. Contnb. Sci. Los Angeles County Mus.. ( 68): 1—30. Dowling, H. G. A taxonomic study of the ratsnakes. genus 1960. Elaphe Fitzinger. VII. The triaspis section. 1936. J. del the 1970. Anderson, Canon Mezquital. Durango, Mexico. Southwestern Nat., 18:187-200. R. Taxonomic review of the Mexican skinks of Durango. Literature Cited tufted jay. Proc. Western Foundation Vert. ZooL :265-299. Crossin, R. S.. Sot it. O. H., Webb, R. G. and Baker, phis deppei for Sinaloa. and Dryadophis cliftoni and Sahadora bairdi S. The breeding biology of the 1967. distri- butional analysis of the herpetofauna along the transect route. The total number of taxa is high- 233 Hardy, L. M. A systematic revision of the colubrid snake J. Herpetol.. 9:107-132. Hardy, L. M. and McDiarmid. R. W. The amphibians and reptiles of Sinaloa, 1969. Mexico. Univ. Kansas Publ. Mus. Nat. Hist.. 1975. genus Gyalopion. Johnson, 1977. 18:39-252. D. J. The taxonomy and distribution of the neo- whipsnake Masticophis mentovarius (Reptilia, Serpentes. Colubridac). J. Herpetropical tol.. 11:287-309. SPECIAL PUBLICATION-MUSEUM OF 234 L. M. The subspecies of Klauber, 1949. Smith, H. M. the ridge-nosed rattle- snake, Crotalus willardi. Trans. San Diego Soc. Nat. Hist., 11:121-140. Langebartel, D. A. A new 1959. lizard (Sceloporus) from the Sierra Madre Occidental of Mexico. Herpetologica, 15:25-27. and Wright, J. W. A new plethodontid salamander from Sonora, Mexico. Contrib. Sci. Los Angeles County Mus., (140): 1-11. Lowe, C. H., Jones, C. 1968. McCoy, C. J., J. Jr. Notes on snakes from northern Mexico. Southwestern Nat., 9:46-48. 1964. McDiARMID, W. R. Variation, distribution and systematic status of the black-headed snake Tantilla yaquia 1968. Smith. Bull. So. California Acad. Sci., 67: 159-177. McDiarmid, R. W., Copp, J. T., and Breedlove, D. E. 1976. Notes on the herpetofauna of western Mexico: new records from Sinaloa and the Tres Marias Islands. Contrib. Sci. Nat. Hist. Mus. Los Angeles County, (275): 1-1 McDiarmid, R. W. and Scott, N. 1970. J., 7. Jr. Geographic variation and systematic status of Mexican lyre snakes of the Trimorphodon Mus. Los Angeles County, ( 1 79): 1 — 1939b. The Mexican and Central American lizards of the genus Sceloporus. Zool. Ser. Field Mus. Nat. Hist., 26:1-397. 1942. Mexican herpetological miscellany. Proc. U.S. Nat. Mus., 92:349-395. Smith, H. M. and Chrapliwy, P. S. 1958. New and noteworthy Mexican herptiles from the Lidicker collection. Herpetologica, 13: 267-271. Smith, H. M. and Smith, R. B. 1980. Synopsis of the herpetofauna of Mexico. Volume Vermont Platz, J. E. 1979. Herpe- J. S. new species of leopard frog {Rana pipiens complex) from Arizona. Copeia, 1979:383-390. Reeve, W. 1952. and Mecham, Rana J. chiricahuensis, a L. Taxonomy and distribution of the horned Phrynosoma. Univ. Kansas Sci. 34:817-960. lizards genus Bull., RlEMER, W. 1955. Mexican Mamm., Robinson, M. D. 1979. Systematics of skinks of the Eumeces brevirostris species group in western Mexico. Contrib. Sci. Nat. Hist. Mus. Los Angeles County, 968. 1972. from western Mexico. Occas. Pap. Mus. Zool. Louisiana State Univ., (35): 1-1 Schmidt. R. H., 1976. A geographical survey of Sinaloa. Southwestern Stud., Texas Western Press, Univ. Texas El Paso, (50): 1-77. 1044. 55:45-65). A new subspecies of Conopsis nasus from Chihuahua, Mexico. Herpetologica, 17:13J. R. and Harris, H. S., Jr. subspecies of Crotalus lepidus from western Mexico. Great Basin Nat., 32:16- A new H. review of the Mexican forms of the lizard genus Sphaerodactvlus. Univ. Kansas Sci. Bull., 31:299-309. Thomas, R. A. and Dixon, J. R. A re-evaluation of the Sceloporus scalaris 1976. group (Sauria: Iguanidae). Southwestern Nat., 20:523-536. Thompson, F. G. E. 1947. A 1957. A new Mexican gartersnake (genus Thamnophis) with notes on related forms. Occas. Pap. Mus. Zool. Univ. Michigan, (584): 110. Tihen, J. 1949. 1954. A. A review of the lizard genus Barisia. Univ. Kansas Sci. Bull., 33:217-256. Gerrhonotine lizards recently added to the American Museum collection, with further revisions of the genus Abronia. Amer. Mus. Novit., (1687):l-26. Udvardv, M. D. 1969. 2. Jr. + 24. Taylor, (3 19): 1-1 3. A new natricine snake of the genus Adelophis (1979), pp. xviii 18. Rossman, D. A. and Blanev, R. M. 1 Mexican turtles. BibIII. John Johnson, Tanner, W. W., Dixon, the distribution of certain toads. Herpetologica, 11:17-23. to Tanner, W. W. J. Comments on Guide Addendum Smith, R. B. 1971. Seasonal activity and other aspects of the ecology of terrestrial vertebrates in a neotropical monsoon environment. Unpubl. M.S. Thesis, Northern Arizona University, Flagstaff, 127 pp. (not seen, referenced from Bateman, G. C. and Vaughn. T. A. 1974. Nightly activities of mormoopid bats. J. 1 microhylid genus Gastrophrvne. toL, 6:111-137. VI. liographic 1961. 43. Myers, C. W. 974. The systematics of Rhadmaea (Colubridae), a genus of New World snakes. Bull. Amer. Mus. Nat. Hist., 153:1-262. Nelson, C. E. 1972. Systematic studies of the North American The lizards of the torquatus group of the genus Sceloporus Wiegmann, 1828. Univ. Kansas Sci. Bull. (1936), 24:539-693. 1939a. An annotated list of the Mexican amphibians and reptiles in the Carnegie Museum. Ann. Carnegie Mus., 27:31 1-320. 1938. tan group (Colubridae). Contrib. Sci. Nat. Hist. NATURAL HISTORY F. Dynamic zoogeography. Van Nostrand Reinhold Co., New York. Webb, R. G. 1 960. Notes on some amphibians and reptiles northern Mexico. Trans. Kansas Acad. 63:289-298. from Sci., VERTEBRATE ECOLOGY AND SYSTEMATICS 1 966. Resurrected names for Mexican garter snakes. Thamnophis cyrtopsis (Kennicott). Tulane Stud. Zool.. 13:55-70. 1 967. 1968. Variation and distribution of the iguanid lizard Sceloporus bulleri, and the description of a related new species. Copeia. 1967:202213. The Mexican skink Eumeces lynxe (Squa- 1969. mata. Scincidae). Publ. Mus. Michigan State Univ., Biol. Sen. 4:1-28. Variation, status, and relationship of the 1972. iguanid lizard Sceloporus shannonorum. Herpetologica. 25:300-307. Resurrection of Bufo mexicanus Brocchi for a highland toad in western Mexico. Herpetologica, 28:1-6. 235 KU — Museum of Natural History. University of Kansas MCZ — Museum of Comparative Zoology. Harvard University MSU — The Museum, Michigan State Universiu — University of Illinois Museum of Natural UIMNH History UMMZ— Museum of Zoology. University of Michigan USNM — National Museum UTEP— Laboratory of Natural History Environmental Biology. University of Texas at El Paso The reference for Hardy and McDiarmid (1969), often cited for Sinaloan localities, [page number]"; is abbreviated to some localities cited by them "H-M: mod- are for more precise orientation in faunal regions (stated mileage presumably by road). Supplementary data are provided for some Sinaloan species. Data for species in Durango are minimal (usually only localities) ified A 1976. review of the garter snake Thamnophis elegans in Mexico. Contrib. Sci. Nat. Hist. 1977. Comments on Mus. Los Angeles County. (284): 1-1 3. snakes of the genus Geophis (Colubridae) from the Mexican states of Durango and Sinaloa. Southwestern Nat., 21: to contemplated publication of more detailed information elsewhere. owing 543-559. Class Webb, R. G. and Baker. R. H. 1962. Terrestrial vertebrates of the Pueblo Nuevo area of southwestern Durango. Mexico. Amer. Midi. Nat., 68:325-333. Welbourn, W. C, Jr. and Loomis, R. B. 970. Three new species of Hannemania (Acarina, Trombiculidae) from amphibians of western 1 Mexico. 65-73. Wilson. L. D. 1 966. Bull. So. California Acad. Sci.. 69: Family Ambystomatidae Ambystoma rosaceum Taylor. Durango: Vicinity El La Ciudad (Anderson 1961): 1.6 km Buenos Aires (Welbourn and Loomis 1970:69. 71). W Salto to near Pine-Oak. Ambvstoma tigrinum subsp. Durango: Navios. mi (AMNH); 4 mi NE Navios (UTEP): 13 mi N Durango (MCZ). The taxonomic status of these sal1 The range of the Rio Grande racer in Mexico and the status of Coluber oaxaca (Jan). Her- S Navios petologica, 22:42-47. amanders in Durango land and Pine-Oak. ZWEIFEL, R. G. 1954a. A new frog of the genus Rana from western Mexico with a key to thq Mexican species of the genus. Bull. So. California Acad. Sci.. 53: 131-141. 954b. Notes on the distribution of some reptiles in western Mexico. Herpetologica. 10:145-149. A survey of the frogs of the augusti group, 1956. 1 genus Eleutherodactylus. it.. (1813): 1-35. 1967. Amphibia Order Caudata Amer. Mus. NovAmer. Eleutherodactylus augusti. Cat. phib. Rept:4 1.1-4 1.4. Am- is uncertain. Mesquite-Grass- Order Anura Family Pelobatidae Scaphiopus couchi Baird. Durango: Vicinity Durango (Chrapliwy Williams, and Smith 1 96 1 :86). Sinaloa: Mazatlan-Villa Union area east to 4.7 mi NE Concor, dia (H-M:71-72). Recently were active in daytime on mi metamorphosed 1 1 toadlets July about temporary N Mazatlan: 26 (UTEP) ranged in mm. MesquiteGrassland and Thorn-Scrub. Scaphiopus multiplicatus Cope. Durango: Several specimens within seven-mile radius Durango. 10 mi Metates. 15 mi E Coyotes (MSU): 6 mi SE Llano Grande (UTEP). Mesquite-Grassland and Pine-Oak. rain pools 1 length from 9 to 19. averaging 13.8 Appendix This appendix provides accounts of the 45 taxa that formed the data base for the analysis of distributional patterns. Accounts are brief, the primary intent being 1 W only to provide locality records that indicate occurrence in one or more of the faunal regions. Place-names are explained in the gazetteer and geographically oriented in Figs. 1 and 2. Localities are documented by either literature citations or by museum acronyms, the latter indicating one or more specimens in the follow- ing institutions: Family Leptodactylidae Eleutherodactylus hobartsmithi (Taylor). Vicinity 1 Chupaderos [24.8 mi E jet Sinaloa: hwys 40-15 = ca. mi E Chupaderos]. Santa Lucia (H-M:73). Tropical- Deciduous. AMNH — American (specimens Eleutherodactylus occidentals Taylor. Sinaloa: 7.1 9.8 mi E Concordia. 15.7 mi E ConSanta cordia (H-M:74): 5 mi Copala. 2 mi geles Lucia (MSU). The two Museum of Natural History CSULB— California State University Long Beach now presumably in Los AnCounty Museum Natural History) mi E Concordia. SW MSU SW specimens were active at SPECIAL PUBLICATION-MUSEUM OF 236 night on dirt roads in the rainy season (30 July and 1 August 1960). Food in the stomach of the Copala specimen (identified by Dr. George W. Byers, Department of Entomology, University of Kansas) consisted of: Neuroptera, Myrmeliontidae (1 larva); Diptera, Tipulidae, Limonia; Lepidoptera, Noctuidae (1 larva); Phalangida (2 specimens); and many earthworm fragments. Tropical-Deciduous. Eleutherodactylus vocalis Taylor. Sinaloa: Vicinity Concordia locality close to Chu= ca. mi E Chupaderos; 24.8 mi E jet hwys 40-15 paderos] east to Santa Lucia and Potrerillos (H-M:75). Chupaderos [MCZ 1 Tropical-Deciduous. Hylactophryne augusti cactorum (Taylor). Sinaloa: Vicinity Santa Lucia, 6-7 mi NE Concordia (H-M:72). Frogs of this species (KU), in company with individuals (KU) of the more frequently observed Eleutherodactylus vocalis (Webb 1960:289), were obtained at night as they perched on boulders wet from splashing water of a cascading stream near Santa Lucia. Tropical- Deciduous. Hylactophryne augusti latrans (Cope). Durango: 2.5 Tapias (AMNH). Durangan specimens, tentatively assigned to H. a. latrans, may represent intergrades with H. a. cactorum (Zweifel 1967:41.3). Mes- mi W quite-Grassland. Hylactophryne tarahumaraensis Taylor. Durango: Las Adjuntas mi Las Adjuntas, 6 mi (Zweifel 1956:29); 5.5 mi SW El Salto (KU); 6 mi SW El Salto (UTEP); 2 mi E El Espinazo (CSULB). PineOak and Mixed Boreal-Tropical. NW 10 WSW Leptodactylus melanonotus (Hallowell). Sinaloa: Mazatlan-Villa Union area, 11 and 12 mi NE Concordia (H-M:78, as L. occidentalis). Thorn-Scrub. Syrrhophus teretistes Duellman. Sinaloa: Several localities extending from 3.4 mi NE Concordia to mi NE W Revolcaderos (MSU); 49 mi NE Concordia, Si- SW naloa [=ca. 1.5 mi Revolcaderos] (H-M:78). Sinaloa: Vicinity Santa Lucia east to Santa Rita and El Batel [47.2 mi NE Villa Union = W mi NE El Batel] Palmito (UTEP). Mixed 1.4 El (H-M:78); 8 (road) mi Boreal-Tropical and Tropical-Deciduous. Tomodactylus saxatilus Webb. Durango: 0.5 mi Revolcaderos (MSU); 23.5 km SW Buenos Aires [=7.5 W mi 7 1 ). NE mi NE Concordia and Copala (H-M: Thorn-Scrub. Bufo marinus (Linnaeus). Sinaloa: Mazatlan-Villa area east to 4.7 80). W mi Union area east to 2 mi ENE Copala and Panuco (H-M:81). Thorn-Scrub. Bufo marmoreus Wiegmann. Sinaloa: Mazatlan and Villa Union east to vicinity Chupaderos [26 mi NE Villa Union = 2 mi NE Chupaderos] (H-M:82). Thorn1 Scrub. Bufo mazatlanensis Taylor. Sinaloa: Mazatlan-Villa Santa Lucia and 27.2 mi east to 0.6 mi NE Concordia [=1.8 mi E Santa Lucia] (H-M:83-84). Tropical-Deciduous and Thorn-Scrub. Bufo microscaphus mexicanus Brocchi. Durango: Several localities vicinity El Salto and Las Adjuntas (Webb 1972:5); 9 mi E El Espinazo [=1 mi E Puerto Buenos Aires] (CSULB). Pine-Oak. DuBufo occidentalis Camerano. Durango: 3 mi mi La rango (UTEP), 10 mi SW El Salto (KU), Revolcaderos (MSU), 4 Ciudad (AMNH), 0.5 mi mi E El Palmito, Sinaloa (H-M:85). Sinaloa: Localities extending from 2.2 km NE Santa Lucia to 2.6 km SW El Palmito [47.2 mi NE Villa Union = 1.4 mi NE El W Union area 1 W 1 W W (H-M:85). Mesquite-Grassland, Pine-Oak, Mixed Boreal-Tropical, and Tropical-Deciduous. Bufo punctatus Baird and Girard. Durango: Durango Batel] W 2.5 mi Tapias, Rio Chico (UTEP). Simi E Mazatlan, about 3 mi SE Mazatlan (H-M: (AMNH); naloa: 2 86). Some literature records for Sinaloa attributed to Riemer by Hardy and McDiarmid (1969:86) seem to be in error. Riemer (1955:22) is only geographically orienting place-names in Sinaloa and other states. Mesquite-Grassland and Thorn-Scrub. Family Hylidae vicin- El Villa Union = 1.4 mi ity El Batel [47.2 mi Batel] (H-M:78, as S. modestus). Tropical-Deciduous. Tomodactylus nitidus petersi Duellman. Durango: 0.5 NE NATURAL HISTORY Revolcaderos] (Welbourn and Loomis 1970: El Palmito (H-M:79). Mixed Sinaloa: 8 (road) mi W Boreal-Tropical. Family Bufonidae Bufo cognatus Say. Durango: Several specimens within eight-mile radius Durango (AMNH. MSU, UTEP). Mesquite-Grassland. Bufo compactilis Wiegmann. Durango: "near Durango, NE city" (UMMZ); 2 mi NE Coyotes, 9.7 mi NE El Salto, 10 mi Metates (Webb 1972: 1-2). Mes- W quite-Grassland and Pine-Oak. Bufo debilis insidior Girard. Durango: 5 mi S Durango (MSU). Mesquite-Grassland. Bufo kelloggi Taylor. Sinaloa: Mazatlan- Villa Union Hyla arenicolorCope. Durango: Localities extending from Cerro de Mercado [=ca. 3 km N Durango] to 5 El Espinazo (Duellman 1970:698). Sinaloa: 44 km mi NE Villa Union, 47.2 mi NE Villa Union [both localities near El Batel] (H-M:88); 6.4 km SE Santa Lucia (Welbourn and Loomis 1970:68); 8 (road) mi W W Palmito (UTEP). Mesquite-Grassland, Pine-Oak, Mixed Boreal-Tropical, and Tropical-Deciduous. El Espinazo Hyla bistincta Cope. Durango: 5 km (Duellman 1970:698). Sinaloa: 1.6 km E Santa Lucia El W (Duellman 1970:699). Mixed Boreal-Tropical and Tropical-Deciduous. Hyla eximia Baird. Durango: Localities vicinity Durango west to 53 km SW El Salto and 14 km E El Espinazo [both localities near Puerto Buenos Aires] (Duellman 1970:702). Mesquite-Grassland and PineOak. Hyla smaragdina Taylor. Sinaloa: Localities extending from Copala east to Potrerillos [27.2 mi E Concordia = 1.8 mi E Santa Lucia] (H-M:89, Duellman 1970:712). Tropical-Deciduous. Hyla smithi Boulenger. Sinaloa: Mazatlan-Villa Union area east to 0.5 km S Santa Lucia and 1 1 mi NE Copala [=ca. 0.2 mi S Santa Lucia] (H-M:90). Tropical-Deciduous and Thorn-Scrub. Pachymedusa dacnicolor (Cope). Sinaloa: MazatlanVilla Union area east to 3.2 km SW Copala and 12 mi VERTEBRATE ECOLOGY AND SYSTEMATICA NE Concordia [=ca. 2 mi Thorn-Scrub. SW Copala] (H-M:92-93). Union area east to 4.7 Mesquite-Grassland and Pine-Oak. Fragmentary data a captive female (Rio del Presidio. in on eggs deposited by Pternohyla fodiens Boulenger. Sinaloa: MazatlanVilla 237 mi NE Concordia (H-M: Thorn-Scrub. Smilisca baudini (Dumeril and Bibron). Sinaloa: Mazatlan-Villa Union area east to 6.5 mi E Concordia [=2.3 mi NE Copala] (H-M:95-96). Thorn-Scrub. 94). 1 Sinaloa) are: 9 eggs (total), deposited 6 October (5). 8 October (2), and October (2 eggs); weights (taken 6-13 October) ranged from 3.74 to 5.30, averaging 4.2 gm; measurements (taken 3 October) of length ranged from 25.0 to 27.1, averaging 25.6 mm. and of width from 15.4 to 16.8, averaging 16.3 mm. 1 1 Triprion spatulatus spatulatus (Giinther). Sinaloa: Union area east to vicinity Localities in Mazatlan-Villa Concordia (H-M:88). Thorn-Scrub. 1 Family Emydidae Pseudemvs scripta ornata (Gray). Sinaloa: MazatlanUnion area (H-M: 106. Smith and Smith 1980: Villa Family Microhylidae 518). Thorn-Scrub. Gastrophryne olivacea (Hallowell). Sinaloa: Mazatlan-Villa Union area east to 5 km SW Concordia (H-M: 98). Thorn-Scrub. Gastrophryne usta (Cope). Sinaloa: Venadillo [near Mazatlan], east to 4 km Presidio and NE Concordia mi N Mazatlan. and 4.5 mi NE Concordia (H-M:99); 9 mi SE Villa Union (Nelson 972: 131). Thorn-Scrub. Hypopachus variolosus (Cope). Sinaloa: MazatlanVilla Union area east to 4.5 mi NE Concordia and 9 km NE Concordia (H-M: 100, as H. o. oxyrrhinus). 1 Rhinoclemmys pulcherrima rogerbarboun (Ernst). mi Concordia. 7.5 mi E Concordia. Santa Lucia (H-M: 107. Smith and Smith 1980:397). Thorn-Scrub and TropSinaloa: Mazatlan, Presidio de Mazatlan. 9 ical-Deciduous. Order Squamata— Suborder Sauria 1 1 Thorn-Scrub. Rana pustulosa KU SW El Coleonyx variegatus fasciatus (Boulenger). Sinaloa: mi S Presidio, 7.4 mi S jet hwys 15 and 40 (H-M: 110). Thorn-Scrub. Phyllodactylus tuberculosus saxatilis Dixon. Sinaloa: Mazatlan-Villa Union area east to near Santa Lucia Boulenger. Sinaloa: Several localities km in error for 2. 1 km E Santa vicinity Santa Lucia (2 1 4463 1), 7.2 Lucia for mi Family Gekkonidae 10 Family Ranidae 14 Batel. 10 W mi Santa Rita (H-M: 103); mi mi NE El Batel [=ca. 1 W H-M: 103): cave below (west) Copalita [see Copala] (LACM). The above El Palmito] (Zweifel 1954a: 13 1-1 32. 14). Over the relatively flat terrain in the immediate vicinity of Mazatlan and Villa Union, these geckos occur under bridges and in road culverts. None could be found on likely rocky hillsides of oceanfront uplifts near Mazatlan. Tropical-Deciduous and Thorn- (H-M:l Scrub. records of occurrence combine those listed by Hardy and McDiarmid (1969:103) for R. pustulosa and R. sinaloae. W The two names are synonyms, the large ho- lotype of/?, pustulosa being indistinguishable from large females of R. sinaloae. However, some frogs from the Santa Lucia area (e.g., MCZ 32591-94) show features of Rana tarahumarae. For purposes of this report only one taxon is recognized. Mixed Boreal-Tropical and Tropical-Deciduous. Class Reptilia Order Testudines Family Kinosternidae Kinosternon hirtipes murrayi Glass and Hartweg. Durango: Several localities vicinity Durango extending west to 6 mi ENE El Salto (Smith and Smith 1980: 146-147). Mesquite-Grassland and Pine-Oak. Kinosternon integrum subspp. Durango: Localities vicinity Durango extending west to Hacienda Coyotes (Smith and Smith 1980:122). Sinaloa: Mazatlan-Villa Union area east to vicinity Santa Lucia (H-M: 104-1 05, Smith and Smith 1980:130). In this report two sub— Sinaloan specspecies of A', integrum are recognized imens having coarse yellow-blotched head patterns in Thorn-Scrub and Tropical-Deciduous, and Durangan specimens that lack contrasting blotched head patterns Family Iguanidae W Anolis nebulosus (Wiegmann). Durango: 0.5 mi Revolcaderos (MSU, UTEP). Sinaloa: Mazatlan-Villa Union area east to 5 km SW El Palmito (H-M:l 151 16). Gravid females were obtained on 22 and 29 June near Santa Lucia. Thorn-Scrub, Tropical-Deciduous, and Mixed Boreal-Tropical. Anolis utowanae Barbour. Sinaloa: about 10 mi Mazatlan (H-M:l 16). Thorn-Scrub. N Callisaurus draconoides bogerti Martin del Campo. Sinaloa: Several localities in immediate vicinity Ma- Individuals seem mostly restricted low sand dunes. Two gravid females were obtained on June 27; hatchlings and adult males and females were captured on 22 August (different years). Progressive urbanization along the beach- zatlan (H-M:l 19). to the leeward side of front north of Mazatlan, as witnessed in years since 1955, has eradicated suitable habitat for Callisaurus. Thorn-Scrub. Ctenosaura pectinata (Wiegmann). Sinaloa: Mazatlan-Villa Union area east to near Copala [1 mi S. 26 mi E Villa Union] (H-M: 1 24); 3 mi NE Copala (MSU). A low-flying hawk clutching a sizeable ctenosaur in its and when startled, released the preConcordia. 30 June sumed prey unharmed (ca. 4 mi 1961). The record of Ctenosaura from La Ciudad. Du- talons, alighted, rango is in error (see W Conant 1969:86). Thorn-Scrub. SPECIAL PUBLICATION-MUSEUM OF 238 Holbrookia approximates subsp. Durango: Durango, Rio Chico (AMNH); 4 mi E Durango (KU). The nomenclature is in accord with the as yet unpublished data of Ralph W. Axtell. Mesquite-Grassland. Holbrookia elegans elegans Bocourt. Sinaloa: Ma- Union area east to 10 km NE Villa Union and 9.4 mi NE Villa Union (H-M: 126-127). North of Mazatlan, individuals do not occur in the sand-dune zatlan-Villa habitat with Callisaurus draconoidcs, but are found a few hundred meters inland in sparsely vegetated, open fields with a sandy substrate. The specific status of H. elegans anticipates ultimate documentation by Ralph W. Axtell. Thorn-Scrub. Iguana iguana (Linnaeus). Sinaloa: Mazatlan-Villa Union area (H-M: 127-1 28). Just north of Mazatlan, a few hundred meters inland from the beach small iguanas rested at night on branches in a thorn-scrub thicket where the terrain was partly inundated by heavy rains (11 August 1957). Thorn-Scrub. Phrvnosoma douglassi brachycercum Smith. Durango: 5 mi N Durango (Reeve 1 952:9 8, KU). MesquiteGrassland. Phrvnosoma orbiculare bradti Horowitz. Durango: Coyotes (Smith 1 939a:3 1 5), El Salto (Smith 1 942:36 1 ), 1 10 mi E El Salto (Reeve 1952:940), lenger 1885:242). Pine-Oak. La Ciudad (Bou- W Sceloporus bulleri Boulenger. Durango: 0.5 mi 1 967:206), 1 .8 mi NE El Palmito, Sinaloa (AMNH). Sinaloa: Localities extending from vicinity Santa Lucia to near El Palmito [37 mi E Con- Revolcaderos (Webb mi E Loberas] (Webb 1967:206; H-M: 129); ca. 14 mi NNE Copala [ca. 2.5 mi SW Santa Lucia] (MCZ). Mixed Boreal-Tropical and Tropical- cordia^ NATURAL HISTORY Pine-Oak, Grassland, Mixed Boreal-Tropical, and Tropical-Deciduous. Sceloporus nelsoni Cochran. Sinaloa: Mazatlan-Villa Union area east to 5 km SW El Palmito (H-M: 138). Mixed Boreal-Tropical, Tropical-Deciduous, and Thorn-Scrub. Sceloporus poinsetti macrolepis Smith and Chrapliwy. Durango: Durango, Coyotes, La Ciudad (Smith Durango (Smith and 1938:614); El Salto, 10 mi Chrapliwy 1958:268); 4 mi SW Coyotes (Chrapliwy and Fugler 1955:124); Buenos Aires (AMNH). Mesquite-Grassland and Pine-Oak. W Sceloporus scalaris Wiegmann. Durango: 9.9 mi W comm.); localities exLa Ciudad] tending from vicinity Neveria [=4 mi Metates (Thomas and Dixon 1976:535). to 10 mi Mesquite-Grassland and Pine-Oak. Sceloporus spinosus spinosus Wiegmann. Durango: Durango (Smith 1939b:93), 4 mi E Durango (KU), 2.5 mi Tapias (UTEP). Mesquite-Grassland. Sceloporus utiformis Cope. Sinaloa: Vicinity Villa Union east to Santa Lucia area (H-M: 140). TropicalDeciduous and Thorn-Scrub. Urosaurus bicarinatus tuberculatus (Schmidt). Si- Durango (Ernest A. Liner, pers. W W W Union area east to Santa Lucia (H-M: 141). The report of this species from La Ciudad. Durango is in error (see Conant 1969:86). TropicalDeciduous and Thorn-Scrub. naloa: Mazatlan-Villa Family Scincidae ca. 2 Deciduous. Sceloporus clarki boulengeri Stejneger. Sinaloa: MaUnion area east to 5 km SW El Palmito (H-M: 32-1 34). A gravid female (seven eggs) was obzatlan-Villa 1 SW Mixed Boreal-Tropical, Tropical-Deciduous, and Thorn- tained on 2 August 1960 (5 mi Eumeces W Tanner. Durango: El Salto [=near Los Localities extending from 33 mi Bancos] east to Coyotes (Dixon 1969:14); localities ex- km SW El Salto to 16 km E Llano Grande (Robinson 1979:11). Pine-Oak and Mixed Botending from 24 real-Tropical. Copala). Scrub. brevirostris bilineatus N 1 Eumeces callicephalus Bocourt. Sinaloa: 5 mi (8 km) Mazatlan, Presidio (H-M: 143, Robinson 1979:12); mi N Mazatlan (MSU, UTEP). Four of these skinks Sceloporus grammicus microlepidotus Wiegmann. Las Adjuntas, 2 mi E El Salto Durango: 7 mi (Chrapliwy and Fugler 1955:124); La Ciudad (Boulenger 1885:233, Gunther 1 890[1 885-1 902]:72); Rancho Santa Barbara (MSU); Buenos Aires (AMNH). Pine- (one adult, 12 July 1963, MSU; two adults and one hatchling, 23 July 1965, UTEP) were found in loose loamy soil among half-buried rocks, bricks, pieces of tile, and decaying palm fronds in a palm savanna. The SVL, have blue-brown three adults, 67. 72, and 73 Oak. tails Sceloporus heterolepis shannonorum Langebartel. Revolcaderos (Webb 1969:302, Durango: 0.5 mi neck. SW W km SW El Palmito, 8 mi (13 km) W km NE Santa Lucia [=ca. 2.3 mi E Loberas] (H-M: 39, Webb 969:307); 37 miles by road 307). Sinaloa: 5 El Palmito, 19.2 1 1 from Concordia [=2 mi E Loberas or 6.8 mi W El Palmito, type-locality] (Langebartel 1959:25). Mixed Boreal-Tropical. Sceloporus horridus albiventns Smith. Sinaloa: near Mazatlan (H-M: 134). Thorn-Scrub. Sceloporus jarrovi jarrovi Cope. Durango: 25 mi SE Durango (UTEP); 24 mi N Durango (UIMNH); 20 mi Metates (UTEP); El Durango (AMNH); 10 mi Salto (Dunn 1936:473); La Ciudad (Boulenger 1885: W W SW La Ciu224, Gunther 1890(1885-1 902]:69); 6 mi Redad, 17 mi NE El Batel, Sinaloa [=ca. 0.5 mi El volcaderos] (Zweifel 1954b: 145). Sinaloa: 4.7 mi Palmito (UTEP), 10 mi NE El Batel (Zweifel 1954b: 1 145), 1 km NE W W Santa Lucia (H-M: 135). Mesquite- mm and two have The indistinct pale stripes (colors in hatchling of 27 mm on head and life) is black dorsally (head orange-brown) with dark blue tail, and pale orange head striping (ventrolateral stripe white on lip, yellow on neck). Midbody scale rows are 26, 26, 28, and 28. The record of E . callicephalus from La Ciudad, Durango (Boulenger 1887:378) is in error (see discussion by Conant 1969:86). Thorn-Scrub. Eumeces colimensis Taylor. Sinaloa: 1.5-1.6 km E Santa Lucia (H-M: 144, Robinson 1979:1 1). The only known Sinaloan specimen was foraging in early afternoon among rocks and dense broadleaf shrub-cover adjacent to a rocky, cascading stream. Another skink, believed to be this species, that later escaped was found in the morning of 5 August 1960 climbing up the side of our tent that was pitched in a level, dense herb5 mi SW Copala (just below ChupadeThe small specimen was about 45 mm SVL, had covered area, ros). tail, whitish venter, broad white lateral stripes on anterior half of body, and a blackish head and back; a blue VERTEBRATE ECOLOGY AND SYSTEMATICS the pale head stripes were reddish. Tropical-Deciduous. Eumeces lynxe belli (Gray). Durango: 30 mi E El Rancho Santa Barbara (Webb 1968:22). Pine- Salto. Oak. 239 W Durango City [=ca. 7 mi W Rancho Santa Barbara]. 5 mi E El Salto (Tanner 1961:17); 0.5 mi W Revolcaderos (MSU, UTEP). Sinaloa: 19.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas], 37 mi E Concordia [=ca. 2 mi E Loberas] (H-M: 157). Pine-Oak and Mixed Boreal-Tropical. Family Teiidae Cnemidophorus costatus subspp. Sinaloa: MazatlanUnion area east to 5 km SW El Palmito (H-M: 147, 148). Two subspecies, C. c. huico and C. c. ma- Villa zatlanensis, intergrade in the transect area (not differinfluenced by faunal regions) and are not entially recognized in this study. Mixed Boreal-Tropical. Tropical-Deciduous, and Thorn-Scrub. Cnemidophorus scalaris scalaris Cope. Durango: 5 mi S Durango (MSU); 2.5 mi Tapias, Rio Chico (UTEP). Mesquite-Grassland. W W DuDiadophis punctatus subsp. Durango: 32 mi Rancho Santa Barbara] (McCoy rango [=ca. 6.5 mi Los Coyotes (Gehlbach 1965: 1964:47); 20 mi 307); 1.6 mi E El Palmito, Sinaloa (UTEP). Sinaloa: mi El Palmito (UTEP). The two UTEP specimens were found DOR. The Sinaloan specimen, the first recorded from that state, is badly mashed. Gehlbach (1965:305) regarded Durango ringneck snakes as in- W NW 1 W tergrades (D. p. dugesi x D. p. regalis). Pine-Oak and Mixed Boreal-Tropical. ReDryadophis cliftoni Hardy. Durango: 0.5 mi W W volcaderos (MSU), ca. km Los Bancos (AMNH). Sinaloa: 8 road mi SW El Palmito (UTEP); 19.2 km NE Santa Lucia, 1 km NE Santa Lucia. 1.1 mi Santa 1 1 W Family Anguidae Barisia imbricata ciliaris (Smith). Durango: Coyotes (Tihen 1949:245). 10 mi E El Salto (Tihen 1954:12), 15 km NE El Salto (MCZ), 10 mi SW El Salto (KU). 6 mi SE Llano Grande (UTEP). Pine-Oak. Elgaria kingi ferruginea (Webb). Durango: 3.2 road mi NE El Palmito, Sinaloa (UTEP). Sinaloa: 1 km NE Santa Lucia (H-M.T51). Mixed Boreal-Tropical and Tropical-Deciduous. Gerrhonotus liocephalus liocephalus Wiegmann. Sinaloa: 7.2 mi E Santa Lucia. 19.2 km NE Santa Lucia. 5 mi SW El Palmito (H-M: 151). Mixed Boreal-Tropical Family Helodermatidae Thorn-Scrub. Serpentes Elaphe triaspis intermedia (Boettger). Sinaloa: MaUnion area east to Santa Lucia (H-M: 1 62); 2.5 mi NE Santa Lucia (MSU). Tropical-Deciduous and Thorn-Scrub. Geophis dugesi dugesi Bocourt. Durango: .8 mi NE El Palmito, Sinaloa (AMNH). Sinaloa: 19.5 mi SW Buenos Aires, Durango [=ca. 0.4 mi E El Palmito] (Fort Worth Museum of Science and History): Loberas at 1 1 Mixed Boreal-Tropical. Gyalopion quadrangulans (Gunther). Sinaloa: Mazatlan-Villa Union area, and 3.2 km SW Santa Lucia constrictor imperator Daudin. Sinaloa: MazaUnion area east to 10 mi Concordia 156). Thorn-Scrub. 1977:551). Family Boidae (H-M: Villa 1 77 (AMNH); 5 km SW El Palmito. 9.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas] (H-M: 163. Webb Leptotyphlops humilis dugesi (Bocourt). Sinaloa: Mazatlan, Presidio (H-M: 156). Thorn-Scrub. NE tlan- Villa stuarti Smith. Sinaloa: 8 Union (H-M: 159). Thorn-Scrub. Drymarchon corais rubidus Smith. Sinaloa: Mazatlan-Villa Union area east to 2.2 km NE Santa Lucia (H-M: 160). Tropical-Deciduous and Thorn-Scrub. Drymobius margaritiferus fistulosus Smith. Sinaloa: Mazatlan, 5 mi N Mazatlan, Presidio (H-M: 161). Km Squamata — Suborder Family Leptotyphlopidae Boa Dryadophis melanolomus km N 1 Heloderma horridum horridum (Wiegmann). Sinaloa: Mazatlan area east to 2 mi ENE Copala (H-M: Order Tropical-Deciduous. zatlan-Villa and Tropical-Deciduous. 153). (H-M: 158). The Durango specimens are the first recorded from that state. Mixed Boreal-Tropical and Rita (H-M:168-169. Hardy 1975:116). Tropical-Deciduous and Thorn-Scrub. Heterodon nasicus kennerlyi Kennicott. Durango: 9 mi NE Durango (UIMNH), 29 km N Durango (Dunn 1936:476). Mesquite-Grassland. Hvpsiglena torquata (Gunther). Durango: Durango (AMNH), 16 mi N Durango (Zweifel 1954b: 147). 2.5 mi Tapias (UTEP). Sinaloa: Mazatlan-Villa Union W Thorn-Scrub. area east to 2.7 km NE Chupaderos (H-M: 170-1 71). Nomenclature follows Hardy and McDiarmid (1969: Family Colubridae Mesquite-Grassland and Thorn-Scrub. Imantodes gemmistratus latistratus (Cope). Sinaloa: 170). E Adelophis foxi Rossman and Blaney. Durango: Va mi El Mil Diez (Rossman and Blaney 1968). Pine-Oak. Arizona elegans expolita KJauber. Durango: 4.4 mi ESE Durango (UMMZ). Mesquite-Grassland. N N 3 Arizona elegans noctivaga Klauber. Sinaloa: Mazatlan (H-M: 156). Thorn-Scrub. 1.1 mi Coniophanes lateritius lateritius Cope. Sinaloa: 8 km Villa Union and about 30 mi NE Villa Union [=ca. mi NE Copala] (H-M: 157). Thorn-Scrub. Conopsis nasus nasus Gunther. Durango: 32.5 mi Several localities vicinity Mazatlan. 2.2 km NE Santa Lucia (H-M: 1 72-173). Tropical-Deciduous and ThornScrub. Lampropeltis getulus splendida (Baird and Girard). Durango: 5.1 mi ESE Durango (UMMZ). MesquiteGrassland. Lampropeltis mexicana (Garman). Durango: Mimbres (MCZ); Rio Chico, Rancho Santa Barbara (Garstka 1982:31). Pine-Oak. SPECIAL PUBLICATION -MUSEUM OF 240 Lampropeltis triangulum sinaloae Williams. SinaMazatlan-Villa Union area east to 6 km SW Concordia (H-M: 175, as L. t. nelsoni). Thorn-Scrub. Leptodeira maculata Hallowell. Sinaloa: MazatlanVilla Union area east to Santa Lucia (H-M: 76). Tropical-Deciduous and Thorn-Scrub. loa: 1 Leptodeira punctata (Peters). Sinaloa: Several records in Mazatlan-Villa Union area (H-M: 177-1 79). Leptodeira septentrionalis polysticta Giinther. SinaThree specimens from north of Mazatlan (the near- loa: 29 km, H-M: 179). Thorn-Scrub. Leptodeira splendida ephippiala Smith and Tanner. Sinaloa: About 10 km SW Concordia, 12.3 km SW Santa Lucia, 2.4 km NE Santa Lucia, 14 mi SW El est, Batel, Presidio (H-M: 180). Tropical-Deciduous and Leptophis diplotropis (Giinther). Sinaloa: MazatlanVilla Union area east to 9.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas] and 10.6 mi E Santa Lucia [=ca. 1 1 mi E Loberas] (H-M: 182). Mixed Boreal-Tropical, Tropical-Deciduous, and Thorn-Scrub. Masticophis bilineatus Jan. Sinaloa: Mazatlan-Villa east to 34 mi E Villa Union [=ca. 4 mi SW Union area Santa Lucia] (H-M: 183). Tropical-Deciduous and Thorn-Scrub. Masticophis flagellum linear ulus Smith. Durango: about 10 km SSE Durango (UTEP). Mesquite-Grassland. Masticophis mentovarius striolatus (Mertens). Sinaloa: Mazatlan-Villa Union area, 14 km E Concordia. Santa Lucia, 19.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas] (H-M: 186). Mixed Boreal-Tropical, Tropical-Deciduous, and Thorn-Scrub. Masticophis taeniatus girardi Stejneger and Barbour. W Durango (AMNH). Mesquite-Grass- land. Nerodia valida valida (Kennicott). Sinaloa: Maza- Union area east to 24.8 mi E Villa SW Thorn-Scrub. Oxybelis aeneus auratus 88). (Bell). Sinaloa: Mazatlan- Union area, 4.4 mi SW Concordia, 20 mi E Villa Union [=ca. 4 mi Chupaderos] (H-M: 189). Thorn- W Scrub. Phyllorhynchus browni Stejneger. Sinaloa: 10 km N Mazatlan (H-M: 191). Thorn-Scrub. Pituophis deppei deppei (Dumeril). Durango: CoyDuotes, Llano Grande (Duellman 1960:605); 3 mi El rango (UTEP). Sinaloa: 4 km E Loberas (12 km Palmito) (AMNH). The record for Sinaloa seems to be the first for that state. Mesquite-Grassland, Pine-Oak, and Mixed Boreal-Tropical. Pituophis melanoleucus afftnis Hallowell. Sinaloa: Mazatlan-Villa Union area east to 5 mi SW Concordia (H-M: 192). Thorn-Scrub. W W Pseudoficimia frontalis (Cope). Sinaloa: Localities from 12.8 to 18.3 mi N Mazatlan, Presidio, 4 mi NE Concordia (H-M: 194). Thorn-Scrub. Rhadinaea hespena Bailey. Sinaloa: Santa Lucia, 12.3 km SW Union area east to (H-M: 195-1 96). Thorn-Scrub. zatlan-Villa 1.4 mi E Concordia W La Salvadora bairdi Jan. Durango: 24 road mi mi E Revolcaderos] (Univ. Arizona, Charles M. Bogert, pers. comm.). Sinaloa: 2.2 km NE Santa Lucia, 19.2 km NE Santa Lucia (H-M: 198); 9 mi El Palmito (MSU). The MSU specimen was found DOR, as well as another badly mashed specimen 1 W SW El Palmito, Sinaloa) that was not saved. The Durango locality is the first in the state for this species. Mixed Boreal-Tropical and Tropical-Decidu- (from 4 mi Santa Lucia, 19.2 km NE Santa Lucia [=ca. mi E Loberas] (H-MT94-195; Myers 1974:243); 2 km E Loberas, Km 175 (Myers 1974:243). Mixed Boreal-Tropical and Tropical-Deciduous. 2.3 1 Salvadora deserticola Schmidt. Sinaloa: 9 mi N Mami S Villa Union (H-M: 199). Thorn-Scrub. Salvadora grahamiae grahamiae lineata Schmidt. zatlan, 10 Durango: 2.5 mi W. Tapias, 10 mi E El Salto (AMNH); 15 mi ENE El Salto (MCZ). Mesquite-Grassland and Pine-Oak. Sonora aemula (Cope). Sinaloa: 40 mi S Mazatlan (McDiarmid, Copp, and Breedlove 1976:12). ThornScrub. Storeria storerioides (Cope). Durango: Salto (Anderson 1960:63). Sinaloa: 19.2 Lucia [=ca. 2.3 mi E Loberas], 9.6 mi La Ciudad, El km NE SW Santa El Palmito (H-M:201). Pine-Oak and Mixed Boreal-Tropical. Svmpholis lippiens Cope. Sinaloa: 9 mi N Mazatlan, 10.8 mi N Mazatlan, 13.3 mi SE Rio Presidio (H-M: 202). Thorn-Scrub. Tantilla calamarina Cope. Sinaloa: Mazatlan, 29 km N Mazatlan (H-M:203). Thorn-Scrub. Tantilla wilcoxi wilcoxi Stejneger. Durango: 1 5 km Tapias (MSU), Rio Durango (MCZ), 2.5 mi W WSW Chico (UTEP). Mesquite-Grassland. Union [=ca. 1 mi E Chupaderos] (H-M: 187-188); Chupaderos on Rio Chupaderos, 5 mi Copala (Conant 1 969: Villa 1 Pine-Oak. Rhinocheilus lecontei antonii Duges. Sinaloa: Ma- ous. Thorn-Scrub. tlan-Villa Rhadinaea laureata (Giinther). Durango: Coyotes, lOmiEElSalto, 10 mi SW El Salto (Myers 1974:244); mi S Navios (AMNH); 6 mi SE Llano Grande (UTEP). Ciudad [=ca. Thorn-Scrub. Durango: 8 mi NATURAL HISTORY Tantilla vaquia Smith. Sinaloa: 5.8 mi N Mazatlan. N Mazatlan (H-M:203, McDiarmid 1968:176). Thorn-Scrub. Thamnophis cyrtopsis collaris (Jan). Durango: 1.6 km E Sinaloa-Durango state line (Webb 1966:62). Si16 mi naloa: Mazatlan-Villa Union area east to 5 km SW El Palmito (H-M:205-206). Mixed Boreal-Tropical, Tropical-Deciduous, and Thorn-Scrub. Thamnophis cyrtopsis cyrtopsis (Kennicott). DuranMetates, Rio Chico go: 12 mi N Durango, 10 mi W (Webb 1966:59). Mesquite-Grassland. Thamnophis cyrtopsis pulchrilatus Cope. Durango: 2 mi NE El Salto, 3 mi E El Salto, Hacienda Coyotes (Webb 1966:66). Pine-Oak. Thamnophis elegans errans Smith. Durango: Several W extending from Hacienda Coyotes to mi Buenos Aires (Webb 1976:12). Pine-Oak. Thamnophis eques mega/ops (Kennicott). Durango: Vicinity Durango west to 33 mi ENE El Salto [=Mimbres] and Rancho Santa Barbara (Conant 1963: localities 1 487). Mesquite-Grassland. virgatenuis Conant. Durango: from near Coyotes west to 3 mi E Las Adjuntas (Conant 1963:490). Pine-Oak. Thamnophis melanogaster canescens Smith. Durango: Durango, 6 mi E Durango, 10 mi N Durango, Rio Chico, Mimbres, Coyotes, Hacienda Coyotes, 6 mi SW Thamnophis eques Localities extending VERTEBRATE ECOLOGY AND SYSTEMATICS Mil Diez (Conant 1963:481-482). Mesquite-Grassland and Pine-Oak. Thamnophis nigronuchalis Thompson. Durango: 5.6 El Salto, S side El W El Salto (Thompson 1957:1). 6 mi SW El Salto mi (UTEP). Two snakes from Coyotes and from 33 mi ENE El Salto [=Mimbres], although referred to Thamnophis rufipunctatus (Thompson 1957:9; Conant 1963: 480), were discussed by Conant (1963:481) as having features of T. nigronuchalis. Pending further study, T. rufipunctatus is excluded from consideration in this report. Pine-Oak. Trimorphodon biscutatus biscutatus (Dumeril, Bibron, and Dumeril). Sinaloa: Mazatlan-Villa Union area east to 4.8 km NE Santa Lucia (H-M:208. as T. lambda paucimaculata). Tropical-Deciduous and Thorn-Scrub. Tropidodipsas annulifera Boulenger. Sinaloa: Ma- zatlan area east to Santa Lucia (H-M:209-210). Tropical-Deciduous and Thorn-Scrub. Tropidodipsas philippi (Jan). Sinaloa: 31.6 mi N Mazatlan (H-M:210). Thorn-Scrub. Family Elapidae Micruroides euryxanthus neglectus Roze. Sinaloa: 16.3 mi Mazatlan, 20 mi N Mazatlan (H-M: 210-211). Thorn-Scrub. Micrurus distans distans (Kennicott). Sinaloa: 9.9 mi N Mazatlan, 11.6 mi N Mazatlan. 9.1 mi NE Concordia, 6.5 km SW Concordia (H-M:2 11-21 2). Thorn- NNW Scrub. 241 W (H-M:2 14-2 5); 4 km S Santa Lucia. 5 mi Concordia (Armstrong and Murphy 1979:6): 5 mi El El Palmito (UTEP). Mixed BoPalmito, 8 road mi cia 1 W W real-Tropical, Tropical-Deciduous, and Thorn-Scrub. Crotalus lepidus k/auberi Gloyd. Durango: Rancho Santa Barbara (MSU), Coyotes (Gloyd 1940:1 12). Pine- Oak. Crotalus lepidus maculosus Tanner, Dixon and Har- W La Ciudad and 16 mi SW La Durango: 15 mi Ciudad [both ca. 2-3 mi El Espinazo], km Los mi La Ciudad [=ca. 2 mi E El Espinazo] Bancos, (Tanner, Dixon, and Harris 1972:16-17). Sinaloa: 5 km SE El Palmito, 19.2 km NE Santa Lucia. 7 and 9 mi NE El Batel (H-M:2 6); mi Durango-Sinaloa state line, 4.8 mi E Santa Rita, 12.5 mi El Palmito [=ca. 0.5 mi E Potrerillos] (Tanner, Dixon, and Harris 972: 6-17). Mixed Boreal-Tropical and Tropical-Deris. 1 1 W W 1 1 1 W 1 W W 1 ciduous. Crotalus molossus nigrescens Gloyd. Durango: Coy(Gloyd 1940:164): 16 km Durango. 8.3 km E Coyotes, Los Bancos (Armstrong and Murphy 1979:33). Mesquite-Grassland. Pine-Oak. and Mixed W otes, El Salto Boreal-Tropical. Crotalus pricei pricei Van Denburgh. Durango: Las ENE El Salto. Llano Grande, Los Bancos (Armstrong and Murphy 1979: 38). Pine-Oak and Mixed Boreal-Tropical. Adjuntas, near Coyotes, 14 mi Crotalus scutulatus scutulatus (Kennicott). Durango: W mi S Durango (MSU), 10 mi Durango (AMNH). 2.5 mi Tapias (UTEP). Mesquite-Grassland. Crotalus stejnegeri Dunn. Sinaloa: 2.2 km NE Santa Lucia (H-M:217); between 10 and 15 mi NE Concormi E Concordia (McDiarmid, Copp. and Breeddia. 5 W 1 Family Viperidae Agkistrodon bilineatus bilineatus (Giinther). Sinaloa: mi N Mazatlan. Mazatlan, Presidio (H-M:213); 4 mi SE Villa Union (UTEP). Thorn-Scrub. Crotalus basiliscus basiliscus (Cope). Sinaloa: Mazatlan-Villa Union area east to 19.2 km NE Santa Luc7.5 love 1976:14). Tropical-Deciduous. Crotalus willardi meridionalis Klauber. Durango. Coyotes and Weicher Ranch (Klauber 1949:133): near Llano Grande (Armstrong and Murphv 1979:65). PineOak. Vertebrate Ecology and Systematics — A Tribute to Henry S. Fitch Edited by R A. Seigel, L. E. Hunt. J. Knight. L. Malaret and N. L. Zuschlag 1984 Museum of Natural History. The University of Kansas. Lawrence I i Systematic Review of the Percid Fish, Etheostoma lepidum* Alice Echelle, F. Anthony and Clark Hubbs Until recently, the greenthroat darter, Etheo- A. Echelle, 67. 3064; (F) UNM 53; (G) UNM 49. Colorado stoma lepidum (Baird and Girard). was known only from south and central Texas in east-flowing drainages of the Edwards Plateau (Strawn 955a, 1957). However. Hubbs and Echelle (1972) reported that Roster's (1957) "Etheostoma sp." in the Pecos River drainage of New Mexico is E. TNHC 2435. 3096; (I) TNHC 207 3 2 Guadalupe River drainage: (J) TNHC 6116. 2977. Nueces River drainage: (R) TNHC 3225. 3057; (L) TNHC 5282, 3105, 5645. E. grahami: Rio Grande drainage: (M) TU 27708; (N) TNHC 3264. 36 5; (O) TNHC 3475. lepidum. This extended the known range of the species well to the north and west and to the opposite side of the High Plains Divide from the 3536. 1 previously recognized distribution. Hubbs and Echelle (1972) noted that the New Mexico population had declined since the time of W. ter's collections in the disjunct occurrence J. 1 Counts and Measurements. — Fin ray and scale counts and body measurements follow Hubbs and Lagler (1958) except as follows: number of transverse scale rows is counted from anal fin origin to base of first dorsal fin: two counts were minimum at above are Strawn's (1955a, 1961) descriptions of fin color and five meristic characters of Texas populations. Hubbs (1967) described variation in survival of offspring from intra- and Texas populations, and Hubbs and Delco (1960) interspecific crosses involving several in egg . scales variation in dorsal some 1 made population prompted the present review of the species. Other studies of geographic variation in morphological characters of E. lep- described 1 . The New Mexico idum 1 declining status of the 1940's and 1950's. and Ros- River drainage: (H) aspects of geographic variation complements of Texas populations. above the depth of caudal pedunclewith the scale row lateral line begins lateral line and includes the median dorsal scale, scales below lateral line begins with the scale row below the lateral line and ends with the median ventral scale; caudal fin length is from caudal base to tip of middle ray; pectoral and pelvic fin lengths and heights of first and second dorsal fins are lengths of longest rays. Scalation in various areas was coded as fol- = no lows: scales: 1 = one to several imbedded or exposed scales covering less than one-half the area; 2 = partially unsealed, but scales covering — Museum abbrevia- the area; 3 = completely For the nape and belly, a score of meant no scales on midline from, respectively, origin of dorsal fin to head and origin of anal fin to base UNM = Univer- of pelvics. more than one-half scaled. Materials and Methods Collections examined. tions in the following list are: sity of New Mexico Collection of Vertebrates; = Texas Natural History Collection of the Texas Memorial Museum, University of Texas at Austin; OSU = Museum Collection of Fishes, Oklahoma State University; TU = Tulane Uni- TNHC versity. Collections used for counts and mea- surements are as follows (letters in parentheses refer to localities as given in Tables 1-5; complete locality data available from authors): E. lepidum: Pecos River drainage: (A) 1 1 342. 1 1 343. 52. 63; (D) 1 1 OSU UNM 50. 55; (C) UNM 57. 59; (E) UNM 65, 66, 344; (B) UNM 51, * Dedicated to Henry S. Fitch and Virginia R. Fitch, esteemed parents and friends. Etheostoma (Oligocephalus) lepidum (Baird and Girard) Figs. 1-2. Tables 1-5 — Types and Nomenclature. Baird's and Girard's (1853) original description of the species as Boleosoma lepida, was based on specimens collected from the Rio Leona, a tributary of the Nueces River, at Uvalde. Texas, by J. H. Clark during the first United States and Mexican Boundary Survey. Existing type specimens include one syntype at the University of Michigan Museum of Zoology (UMMZ 86335) and six syntypes at the U.S. National Museum of Natural History 243 (USNM 744). Following article 74 SPECIAL PUBLICATION-MUSEUM OF 244 NATURAL HISTORY m * >• ' <* it*. Fig. 1. Mature adults of Etheostoma lepidum from Sago Spring, upper end of Unit 4, Bitter Lake National New Mexico. OSU 11342, 2 June 1971. Top, male, 45.1 mm SL. Bottom, female, Wildlife Refuge, Roswell, 37.7 SL. mm of the International Code of Zoological we clature UMMZ idum, number USNM 86335, and under a new catalog 223024), five specimens of (USNM 744. The sixth retains the original rard ( 1 Nomen- designate as paralectotypes of E. lep- specimen of USNM 744 number as the lectotype. Gi- 859a) referred to E. lepidum from the up- per Nueces (Leona River) as Poecilichthys lepidus and, in a paper published in the same volume (Girard 1859b), described Oligocephalus leonena new species, also from Leona River. Two sis as syntypes bearing the latter name are at the Mu- (1966) noted that the syntypes have "well-developed ctenoid scales on the opercle, a character of E. grahami . . . ." They note, however, that although badly dried, the body appears more elongate than in either E. lepidum or E. grahami. A junior synonym of E. lepidum, Etheostoma was described by Evermann and Kendall (1894), from Comal Springs of the Guadalupe River drainage as a result of a mistaken lepidogenys, comparison (Hubbs, Kuehne, and Ball 1 953) with E. spectabile rather than E. lepidum. The two syntypes for the latter description are deposited Museum seum of Comparative Zoology (MCZ 24580) with at the information that they were collected in the Nueces River by J. H. Clark and sent to (USNM 44840). Hubbs, Kuehne, and Ball (1953) MCZ USNM Evermann and Kendall (1894) regarded O. leonensis a junior synonym of E. lepidum. However, Collette and Knapp from in 1853. the National of Natural History and Hubbs and Echelle (1972) used the trinomial, E. lepidum lepidogenys in reference to, respectively, Guadalupe River populations and "the Guadalupe-Colorado river stocks." VERTEBRATE ECOLOGY AND SYSTEMATICS 33- w- 30 100 27 245 SPECIAL PUBLICATION-MUSEUM OF 246 breeding tubercles (Collette 1965), a character which sets these three species apart spectabile. from E. — A member of the subgenus OliDiagnosis. gocephalus characterized by the following combination of traits: no breeding tubercles in males; branchiostegal membranes, breast, and pelvic and anal fins blue green in breeding males; first dorsal fin in breeding males with two red bands and 1- 2 blue to blue green bands (always one on distal margin); nape and breast largely naked; opercle naked; checks unsealed or partially completely scaled; so, rarely moderately arched, line scales 44-60, pored lateral line incomplete, total lateral 19-42. scales, — Counts are presented in Tables Measurements are shown in Table 5. General aspects of body form and pigmentation are shown in Fig. 1. Strawn (1961) presented data on variation in five counts (lateral line scales, anal soft rays, and rays in first dorsal, second dorsal and pectoral fins). In the following acDescription. 1-4. count, ranges for Strawn's (1961) data are presented in brackets. Fin rays of spinous dorsal 7-12 [4-1 1], modally 9 in all populations except two tributaries of Colorado River where mode = 10. Soft dorsal most frequent mode =11. Anal soft rays 5-8 [4-9], rarely 5. Anal spines 1 or 2; in New Mexico populations, 2 more comfin rays 9-13 [8-14], mon than in 1; Texas, 2 is almost the exclusive count. Pectoral rays, 10-13 [9-14], usually 1 1 or 12. Pelvics have one spine and 5, rarely 4 or 6 (in New Mexico) soft rays. Lateral line scales 44- 60 [45-67], usually 48-55; pored lateral line scales 19-42, usually 29-36. Transverse scale rows 1219. modally 14-16. Caudal peduncle scales 4-6 = (modally 5) above lateral line and 4-7 (mode 6 in all populations except one with 5) below NATURAL HISTORY from New Mexico except those from Blue Spring where 12 of 55 (22%) had 1-5 small, imbedded, nonoverlapping scales adjacent to the posterior and/or ventral border of eye; Texas samples generally as described for Blue Spring, but specimens from San Saba and Guadalupe rivers more fully specimens with cheeks comwith covered pletely large overlapping scales. Moore's ( 1 968) comment that the cheeks are fully scaled in E. lepidum may have been based on such specimens, but this is uncommon for the species. Nape naked to lightly scaled (posterior region) except in Blue Spring and in Guadalupe River where occasionally more than half-covered with scales, rarely (two specimens from Blue scaled, occasional Spring) completely covered. Breast completely in New Mexico samples except in Blue Spring where 47% (26 of 55) had light scalation near bases of pectoral fins; largely naked in Texas naked samples, but specimens often have scales similar to those described for Blue Spring. Preopercular pores 5-7. modally 6. Mandib- pores 2modally 6 with canal interrupted and 4 pores anteriorly, 2 posteriorly. Supratemporal pores generally 1-3, modally 2. Supratemporal canal ular pores 2-7. modally 4. Infraorbital 8, usually interrupted middorsally. but occasionwhen uninterrupted, an extra ally uninterrupted; pore occurs middorsally. Coloration. Breeding Males. — Head dark brown dorsally followed posteriorly by 8-10 dark brown to olive brown blotches; 10 to 12 dark greenish brown lateral bars on body, darker and more complete on caudal peduncle; abdomen whitish gray ventrally with immaculate genital papilla; sides of abdomen dark red to reddish orange or rust colored; similar pigmentation extends dorsally between lateral bars and well above lateral line. Suborbital bar pronounced, extends below angle of jaw. lateral line. downward from Belly usually completely scaled, but occasional specimens with small naked area anteriorly or on the midventral line and rarely h to more than Vi naked. Variation in codes for anterior squamation is given in Table 4. Opercles completely Ventrolateral sides of head, branchiostegal membranes and breast blue green. Pectoral fins with l naked. (The comment by Moore 968, that specimens from San Saba River have scaled opercles 1 Other areas of anterior squamation generally naked or lightly covered, but with specimens from Blue Spring. New Mexico, and San Saba and Guadalupe rivers, Texas, more heavily scaled. Cheeks completely naked in all specimens is incorrect.) orbit to well 6-7 variably distinct vertical rows of dusky spots on rays, usually a reddish to rust-colored spot in axil. Pelvic fins dark blue green to blackish memore brightly colored individuals, sially with, in scattered red, orange, or rust-colored spots laterally. Anal fin blue green, darker at base, often with 2-3 horizontal rows composed of 2-3 small. orange or rust-colored interradial spots. red, fin with 5-7 vertical rows composed of dusky to reddish brown spots on rays. First dor- Caudal VERTEBRATE ECOLOGY AND SYSTEMATICA Table numbers 1 Species . Caudal peduncle scale counts in E. lepidum and E. grahami. Locality and Methods. as listed in Materials letters refer to 247 museum collection SPECIAL PUBLICATION-MUSEUM OF 248 Table 2. Lateral line scale counts in E. lepidum NATURAL HISTORY and E. grahami. Locality letters as in Table 1. VERTEBRATE ECOLOGY AND SYSTEMATICS Table 3. Species and Fin ray counts in E. lepidum and E. grahami. Locality letters as in Table 249 1. SPECIAL PUBLICATION-MUSEUM OF 250 Table 4. Codes for anterior Locality letters as in Table squamation in E. NATURAL HISTORY lepidum and E. grahami. See Methods for description Cheek Nape Opercle Species and Breast locality E. lepidum New Mexico A 15 0.0 15 0.0 11 38 0.0 37 0.3 20 C 11 0.0 11 0.0 8 D 26 43 23 20 0.0 26 55 23 20 0.0 26 B E F G 12 0.2 0.0 0.0 0.3 34 0.0 7 0.0 22 0.0 12 35 27 29 0.0 32 22 3 0.0 0.0 3 17 40 42 0.0 24 16 0.0 6 36 1 1 Texas H I 31 4 4 16 2 1.2 18 6 1.6 J E. 0.1 K 13 27 0.7 L 29 13 0.3 10 3 0.2 11 2.8 12 16 6 0.3 16 2.6 21 2 8 1.2 15 3.0 10 5 grahami M N O of codes. 1. 1 4 12 VERTEBRATE ECOLOGY AND SYSTEMATICA 251 mouths of its major tributaries has resulted from extensive habitat alterations which include National Wildlife Refuge— see Fig. 2) are closely the associated with dense vegetation. In Texas the species occurs most abundantly in vegetated riffles when in sympatry with E. spectabile, but a channel straightening, brush removal, irrigation diversions, and pollution from oil fields, munic- toward gravel riffles occurs in the Nueces River drainage where E. spectabile is absent (Hubbs. Kuehne, and Ball 1953). E. lepidum occurs abundantly in most springfed habitats of the Edwards Plateau, and also does well in certain reservoirs of the area. Distribution and abundance of Texas populations has not been shift substantially altered by ever, human activities. How- pumping of water from underground aqui- and cattle feedlots. The populations at Blue Spring and Bitter Lake National Wildlife ipalities, Refuge are well protected by the present ownseem in no immediate danger of erships and elimination. Strawn (1955b) noted that greenthroat darters spawn repeatedly when kept at temperatures ranging from ... [16] (and probably lower) "will " to [23°C] Hubbs (1961. 1967) described eventually have adverse effects as springs diminish in flow. The abundance of E. lepidum has declined in New Mexico. The following account is based on developmental temperature tolerances (7-29°C) of the South Concho River and Nueces River 957) noted that, populations. Hubbs and Strawn Roster's early collections and a compilaassembled by J. E. Sublette with relatively uniform temperatures, the breeding season is 10 to 2 months long with depressed fers will W. J. tion of other records New Mexico of Eastern never abundant in the University. Although Pecos River proper (Ros- comm.). several large collections (32330 specimens) were made in the 1940's and early 950"s from the Pecos River and the mouths of large tributaries. Black River and Rio Felix; as late as 1961 a collection of 60 specimens (Arizona State University #0936) was taken from ter, pers. 1 Cottonwood Creek. N of Artesia, Eddy County. Large populations are now known only from three Blue Spring, a 4 km spring and spring localities: 1 ) run of the Black River drainage. 8 km E of Whites Eddy County. 2) small springs and waterfowl management ponds on the Bitter Lake National Wildlife Refuge near Roswell. Chaves County, and 3) Pecos River (Mike Hatch, pers. comm.) at Boiling Springs (Major Johnson Spring). 9.6 km S of Lakewood. Eddy County. Since 1961. and excepting a small, uncatalogued collection by a party from New Mexico State University which was taken at Carlsbad in 1966 City. comm.). the species has been taken from only three additional locations, and each consisted of single specimens. However, one of the three collections (Eastern New Mexico University #01 5.02) was made from the Rio Penasco. 1.7 km S. 29.2 km EMayhill. Chaves County, an area sufficiently isolated from presently (D. Jester, pers. known major populations that it is difficult to explain the single specimen as a stray individual from known areas of dense concentration. Thus, permanent population may occur somewhere the Rio Penasco drainage. The virtual elimination of E. lepidum from the Pecos River and a in ( in the Guadalupe River, 1 at a springfed locality 1 breeding more in mid-summer, while at a locality with variable water temperatures, breeding oc- November through April. The New Mexico populations seem to respond similarly. Based on dissection of females larger than 35 mm curred from SL. ripe eggs were present in all collections ex- amined from Blue Spring (collections made in April. May. October, and November) and in a June collection from Sago Spring on the Bitter Lake National Wildlife Refuge, but in an August collection of 3 females from Dragonfly Spring on the Refuge none had ripe eggs. Roster's collections from the Pecos River and the mouths of larger tributaries include two February collections in which 7 of 8 females were ripe, one July of 22 females was ripe, collection in which only and six August collections in which, excluding a collection from the mouth of Black River, only of 24 was ripe: in the excluded collection. 6 of 14 females were ripe. Thus. New Mexico pop1 1 ulations apparently experience depressed breeding activity in the summer. Zoogeography.—The New Mexico population of E. lepidum represents a disjunct occurrence of the species (Fig. 1). Elsewhere, the species is restricted to limestone springs and associated waters of the Edwards Plateau where its range coincides well with the Balconian Province as defined by Blair (1950). The similar E. grahami occurs in the lower Pecos and elsewhere in the Rio Grande drainage (Fig. 1). At present E. grahami is isolated from E. lepidum by a 300 km segment o\" the Pecos River (from Malaga. New Mexico to Sheffield. 252 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Means and, in parentheses, standard deviations of standard length (SL) and proportional measure5. ments, as thousandths of SL, for E. lepidum and E. grahami.* Table VERTEBRATE ECOLOGY AND SYSTEMATICA Table 5. Continued. 253 SPECIAL PUBLICATION-MUSEUM OF 254 widely disjunct population in the middle Pecos River drainage of southeastern New Mexico has declined noticeably since the 1950's; at present, large populations occur only in Blue Spring, near Whites City, and at the Bitter Wildlife Refuge, near Roswell. lepidum and E. grahami diverged Coast. In late Pliocene or early Pleistocene the lower Pecos River (presently occupied by E. gra- hami) eroded headward and presumably captured the middle to upper Pecos from the Colorado River or another drainage of central Texas. The latter Baird, 1 S. F. event brought E. lepidum into the Pe- cos River and created the present disjunct distribution of the species. At present E. lepidum and E. grahami are separated by km segment of the Pecos River. a saline, Blair, W. F. The biotic provinces of Texas. Texas 1950. 2:93-117. Collette, Collette, to Sublette obscure and Mike Hatch museum records, for locality records Department, and the U.S. Fish and Wildlife Ser- Fink (MCZ) for information regarding type specimens in their care, to Royal D. Suttkus (TU) and Robert F. Martin (TNHC) L. for loans of specimens, and to Warren Bounds, Larry Kline, and Delbert Boggs for asand permission to collect on property under their control. Reeve M. Bailey and an sistance anonymous reviewer provided valuable comments on the manuscript. Part of this work was done at The University of Oklahoma Biological Station. Literature Cited Bailey, R. 1955. and Knapp, L. W. Nat. Mus., 119:1-88. Distler, D. A. Distribution and variation of Etheostoma 1968. spectabile (Agassiz) (Percidae. Teleostei). Univ. Kansas Sci. Bull., 48:143-208. Echelle, A. A. and Echelle, A. F. 1978. The Pecos River pupfish. Cyprinodon pecosensisn. sp. (Cyprinodontidae), with com- ments on its evolutionary origin. Copeia, 1978:569-584. S. and Underhill, How to Know J. C. the Freshwater Fishes. Wrn. C. Brown, Dubuque, Iowa. Evermann. B. W. and Kendall, W. C. 1894. The fishes of Texas and the Rio Grande Ba- vice for collecting permits, to Reeve M. Bailey (UMMZ), Bruce B. Collette and Susan Karnella William B. B. (Pisces, Percidae, Etheostomatini). Proc. U.S. Eddy, in New Mexico, to the New Mexico Fish and Game Department, the Texas Parks and Wildlife (USNM) and B. B. Catalog of type specimens of the darters 1966. sin, loan of valuable early collections of E. lepidum from New Mexico, to James E. Sublette for his efforts at ferreting out J. Sci., Systematic significance of breeding tubercles in fishes of the family Percidae. Proc. U.S. Nat. Mus., 117:567-614. 1965. Koster for the J. C Boundary Survey, under Lt. Col. Jas. D. Graham. Proc. Acad. Nat. Sci. Philadelphia, 6:387-390. 300 Acknowledgments are indebted to William and Girard, ican 1978. We Mus. Zool., Description of new species of fishes collected by Mr. John H. Clark, on the U.S. and Mex- 853. in allopatry, Grande and the more northern drainages of the Gulf the latter in the ancestral Rio Misc. Publ. Univ. Michigan, 93:1-44. Lake National breeding condition declines in summer. It is hypothesized that, in Pliocene times, E. in family Percidae. The New Mexico population conforms with published reports on breeding season in Texas populations; ripe females occur during most months of the year, but former NATURAL HISTORY M. and Gosline, W. A. Variation and systematic significance of vertebral counts in the American fishes of the considered chiefly with reference to their geographic distribution. Bull. U.S. Fish Comm., 1892, 12:57-126. Girard, C. 1859a. Ichthyological notices, 5-27. Proc. Acad. Nat. Sci. Philadelphia, 11:56-68. 1859b. Ichthyological notices, 28-40. Proc. Acad. Nat. Sci. Philadelphia, 11:100-104. Hubbs, C. L. and Lagler, K. F. 1958. Guide to the Fishes of the Great Lakes Region. Cranbrook Inst. Sci. Bull., 26:1-213. Hubbs, C. 1961. Developmental temperature tolerances of four etheostomatine fishes occurring in Texas. Copeia, 1961:195-198. 967. Geographic variations in survival of hybrids between etheostomatine fishes. Bull. Texas 1 Mem. Mus., 13:1-72. Hubbs, C. and Delco, E. A. 960. Geographic variations in egg complement of E. lepidum. Texas J. Sci., 12:3-7. Hubbs, C. and Echelle, A. A. 1972. Endangered nongame fishes of the upper Rio Grande basin. Pp. 147-167. In Symposium on Rare and Endangered Wildlife of the Southwestern United States. New Mexico Dept. Game and Fish, Santa Fe. Hubbs, C, Kuehne, R. A. and Ball, J. C. 1953. The fishes of the upper Guadalupe River, Texas. Texas J. Sci., 5:216-244. Hubbs, C. and Strawn, K. 1957. The effects of light and temperature on the 1 VERTEBRATE ECOLOGY AND SYSTEMATICS Etheofecundity of the greenthroat darter, el 602. 1955a. the Vertebrate Animals of the northeastern United States, Inclusive of Marine Species. World Book, N.Y. Jordan D. S., Evermann, B. W. and Clark. H. W. Check-list of the Fishes and Fish-like Ver1930. tebrates of North America north of the Northern Boundary of Venezuela and Co- lombia. Rep. U.S. Fish Comm., Hill Book United States. Co., N.Y. , luscan faunas, east-central New Mexico. New Mexico Bur. Mines and Min. Res. Memoir 955b. 1957. MOORE, G. A. II. Pp. 22-165. In Blair, W. F., 1. Aquar. J., 26:408-412. A method of breeding and raising three Texas darters. 2. Aquar. J.. 27:12-14, 17. 31. influence of environment on the meriscounts of the fishes, Etheostoma grahami and E. lepidum. Unpubl. Ph.D. dissertation, The tic Univ. Texas, Austin. 1961. A comparison of meristic means and variances of wild and laboratory-raised samples of the fishes, Etheostoma grahami and E. lepidum (Percidae). Texas J. Sci., 13:127159. Thomas, 1972. 30. A method of breeding and raising three Texas darters, 1 1928. KOSTER W. J. 1957. Guide to the Fishes of New Mexico. Univ. New Mexico Press, Albuquerque. Leonar d, A. B. and Frve, J. C. Pliocene and Pleistocene deposits and mol1975 Fishes. Pt. Vertebrates of the Strawn, K. Jordan D. S. Manual of 1929. 1968. ah, McGraw stoma lepidum (Girard). Ecology, 38:596- 255 R. G. The geomorphic evolution of the Pecos River svstem. Baylor Geological Studies, Bull. 22. Vertebrate Ecology and Systemaucs— A Tribute to Henry S. Fitch Edited by R. A. Seigel, L. E. Hunt. J. L. Knight. L. Malaret and N. L. Zuschlag c 1984 Museum of Natural History- The University of Kansas, Lawrence A New Anolis fitchi, Species of the Anolis aequatorialis Group from Ecuador and Colombia Ernest The Williams and William E. aequatorialis group of the iguanid lizard is characteristic of cloud forests on genus Anolis the slopes of the Andes in northwestern South America. Vast areas of these forests still remain and the Anolis aequatorialis group, inaccessible, like many other elements of the cloud forest fau- remained poorly known. The only published field observations on any of these lizards na, has are those of Henry S. Fitch. Therefore, ticularly appropriate to associate his a handsome new species in the it is name par- with eulaemus sub1 section of the group. Anolis fitchi new species Frontispiece m tained on 1 9 October 1971 by William E. Duell- man. Paratypes. — ECUADOR: Provincia Napo: KU KU 142864, 142866, same data as holotype; 142867-69. Rio Azuela at Quito-Lago Agrio road, 1740 m, William E. Duellman and Joseph October 1971: MCZ 158324, same locality, Kenneth Miyata, 24 February km upstream 1979; KU 164162, Rio Salado, from Rio Coca, 1410 m, William E. Duellman, 7 October 1974; Ku 164163-65, same locality, William E. Duellman and Alan H. Savitzky, 18 T. Collins, 20-21 1 KU 178960, same locality David C. Cannatella, 18 July 1977; KU 178961, same March 1975; Duellman AMNH unknown; 28900, "Volcan Sumaco," Carlos Olalla, January, 1924; 214869, "upper Rio Napo," Jorge Olalla, date unknown; USNM USNM 214870, La Alegria on Rio Chingual, ±3 Sibundoy, ±20 km N La Bonita, 1930 m. James A. Peters, 24 June 1962. COLOMBIA: km N KU 169823-26, Departamento de Putumayo: km of El Pepino, 1440 m, William E. Duellman, 27-29 September 1974. W 10.3 Diagnosis.— Anolis fitchi is a member of the eulaemus subgroup of the Anolis aequatorialis species group; i.e., it has the moderate size and narrow toe lamellae characteristic of all members of the group but has the subdigital pad under the phalanx projecting above the proximal end of I, rather than continuous with the latter ("Norops condition" as understood by Boulenger 1885). Anolisfitchi is similar to A. eulaemus Boulenger but differs in having the dewlap on the male with dark skin and large scales in single or double lines (rather than light skin and minute phalanx — KXJ 142865, an adult male, from 16.5 km (by road) north-northeast of Santa Rosa, Provincia Napo, Ecuador, 1 700 elevation, obHolotype. E. Martha C. Lynch, 17 July 1977; MCZ 124350-51, "Loreto region." collector and date locality, and in the presence of a moderate-sized mottled or spotted dewlap in females (female dewlap rudimentary with dark skin scales in multiple lines) in eulaemus). Anolis fitchi similar also to A. the scales around the interparietal slightly larger than those on the nape (those scales smaller, hardly distinguishable from nape scales in A. ventrimaculatus) and in the presence of the mod- dewlap in females (no trace of a dewlap in female A. ventrimaculatus). Description (Counts for Holotype in Parentheses).— Anterior head scales small, multicarinate. erate tuberculate. or wrinkled; snout between By a lapsus, Williams (1976) used the name aequatorialis group in a table and eulaemus group in a key on the following page. The intention was to use is ventrimaculatus Boulenger but differs by having 1 1-18 (16) scales across second canthals; some scales within shallow frontal depression larger than those immediately anterior to depression; 6-9 1 the oldest name in each group as the nominate form. Anolis aequatorialis Werner 1894 antedates A. eulae- mus Boulenger 908, and hence is the appropriate name for the whole defined series. However, A. eulaemus is the earliest-named member of one of the two quite 1 distinct subgroups, which therefore are called the A. and the A. eulaemus subgroup. aequatorialis subgroup (9) scales bordering rostral posteriorly; 10 or 11 between supranasals dorsally; anterior naabove, or just behind, suture between rostral scales sal and first supralabial (Fig. 1). Supraorbital semiby 1-3 (3) scales; no differen- circles separated tiated supraocular disc, but some scales slightly enlarged, keeled; one moderately elongate supraciliary followed by one or two shorter scales. 257 SPECIAL PUBLICATION-MUSEUM OF 258 Fig. first Dorsal view of head of holotype of Anolis fitchi. 7-10 (8) rows of loreals, Temporal and supratemporgranular; no differentiated double in- and second uppermost al scales 1. largest; longest. tertemporal line of enlarged scales; scales in depression surrounding interparietal distinctly but variably enlarged, posterior and posterolater- ones grading abruptly into dorsals and supratemporals; interparietal smaller than ear, separated from supraorbital semicircles by 3-6 (4) scales. Suboculars separated from supralabials al by one row of 8middle of eye (Fig. 2). Mental semidivided, wider than long, posteriorly in contact in an approximately transverse line with 5-8 (6) scales between infralascales or narrowly in contact; 11 (11) supralabials from rostral to not clearly differentiated; median throat scales small, swollen, grading into bials; sublabials much longer scales laterally. Dewlap large in male, extending to middle of belly; scales in closely packed single rows, sep- arated by naked skin; lateral scales larger than ventrals; dewlap in NATURAL HISTORY females extending just pos- sions narrow; 21-24 (22) lamellae under pha- langes II and III of fourth toe. Tail compressed but without dorsal crest; verticils not distinct; two enlarged middorsal rows of scales; postanals weakly enlarged sometimes not evident. in males, Color in Preservative. — Dorsum pale brown with broad dark brown middorsal blotches confluent with or narrowly separated from broad diagonal marks on flanks, or flanks dark brown with many round pale spots. Head pale brown above; limbs pale brown above with broad dark brown transverse bars. Tail brown blotches becoming pale brown with broad indistinct posteriorly. Venter dull tan, flecked or not with dark brown; throat dark with lighter spots or transverse Male dewlap dark brown with tan scales. Female dewlap blotched black on brown. In both sexes a more or less conspicuous complex light, streaks. often white, blotch containing black spots or oblique streaking just above dewlap. Color in Life (See Frontispiece). Dorsum — ol- of axilla. to four middorsal rows of body scales ive-green to tan with dark brown markings, with or without yellowish tan flecks and/or round spots slightly enlarged, keeled, swollen, subimbricate; laterally; often a tan vertebral stripe in females; granules swollen, pointed, juxtaposed; ventrals larger than dorsals, imbricate or sub- venter brown to yellowish green; male dewlap dark brown with yellowish tan to yellowish green scales; female dewlap mottled or spotted; iris dull terior to level Two lateral imbricate, smooth, tending to be in transverse rows. Some scales on limbs multicarinate; on dorsum of hand large, multicarinate; larger scales supradigital scales multicarinate; digital expan- bluish gray; tongue pinkish gray. Males and females may differ sharply in color. This is emphasized by descriptions of a male and female paratype from the same Ecuadorian lo- VERTEBRATE ECOLOGY AND SYSTEMATICS 259 <^g*2 Fig. 2. Lateral view of head of holotype of Anolis fitchi. KU 1 78960, <5: "Green with reddish-brown Hint of yellow along lateral surfaces. markings. Venter yellow. Dewlap brown with green scales at base, becoming yellow laterally. Tongue cream, iris gray." KU 178961, 2: "Dorsal stripe cream with some reddish-brown infusion; laterally dark cality: brown, then bright lime green. Dewlap scales dull orange at edge, yellow toward throat, marbled with black, belly dirty cream with gray-brown spots. Tongue dark gray. Iris blue-green." (Field notes, John D. Lynch). Colombian specimens may differ slightly; KU 1 69823, 6: "Dorsum pale brown transverse markings. Venter pale brown with dark brown flecks. Dewlap brown with dull yellowish green stripes. Iris, tongue, and lining of mouth blue." KU 169826, 2: "Dorsum green with brown flecks and dorsal blotches. Dewlap greenish white with brown green with dark proximally and orange bars distally. Iris W. E. Duellman). Measurements ofHolotype (mm). — Snout-vent flecks pale blue." (Field notes, length 88; tail length 221; head length 24; head width 12.5. Distribution and Ecology. — Most specimens of A. fitchi have been collected in cloud forest at on the eastern slope elevations of 1410-1930 m (Fig. 3). Loreto is at 550 m. The specimen from "Volcan Sumaco" collected by Carlos Olalla most likely came from the vicinity of the village of San Jose Viejo (Peters 1955:345; Paynter and Traylor 1977:110). The imprecise locality, "upper Rio Napo," presumably is less than 500 m. Extensive collections assembled by Duellman and field associates at several localities on the lowerAndean slopes and in the upper Am- of the Andes azon Basin in northern Provincia Napo (Cordi150 m; Bermejo, 720 m; Puerto llera del Due, Libre, 570 m; Santa Cecilia. 340 m; Lago Agrio. 340 m) do not contain examples of A. fitchi. We have adopted a restricted concept of A. fitchi: only specimens from Provincia Napo, Ecuador and adjacent Departamento de Putumayo. 1 Colombia, are included in the type series. Specimens from farther south (Provincias Pastaza. Tungurahua, and Morona-Santiago. Ecuador, and Departamento Amazonas, Peru) obviously are However, these specimens appear to differ in coloration and may represent more than one taxon. The present samples are close to A. fitchi. inadequate to make a decision at this time. The range of unequivocal A. fitchi extends into southern Colombia, at least into Putumayo. To 260 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 600 M. VERTEBRATE ECOLOGY AND SYSTEMATICS Table 1. Scale characters in the Anolis eulaemus subgroup. 261 262 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. Dewlap and body pattern differences in the eulaemus subgroup (pattern differences emphasized because color in life of eulaemus is unknown). in preservative Dewlap 95 A. eulaemus Large, skin dull, scales lighter, minute, in multiple lines (5-6) separated by naked skin. Rudimentary, represented by folds of skin that are emphasized by intervening dark pigment. A. vent ri'macular us Large, skin dark or light, scales light, large, in single or double lines separated by naked skin. None. A. Large, skin dark at base or all light, scales light, large, in multiple lines (3-4) separated by naked skin. None. Large, skin dark, scales light, large, in single or double rows separated by Moderate, skin mottled or spotted, scales light, large, in single or double rows separated by naked skin. gemmosus A.fitchi naked skin. BODY PATTERN Side of A. eulaemus Neck Pattern dull, vague; a poorly bounded black blotch just above the dewlap. A dark area above the rudimentary dewlap bounded dorsally by an arc of light pigment arising from the ear and then descending to the shoulder. 1 A. ventrimaculatus A narrow light line from labials arching over the upper margin of the ear and Uniform dark, rarely some vague light spots. continuing to the shoulder; above this a black blotch. A. gemmosus A light line above A.fitchi A from labials not arching no black blotch above it. ear, light blotch containing black spots or oblique streaks contiguous with the base of the dewlap. Faint bluish tinge on side of neck, no well-defined pattern. A black blotch in front of shoulder just a light blotch containing black spots. behind Throat A. eulaemus Gray, lighter lateralh. Uniformly dark. A. ventrimaculatus Llniform dark or very weakly vermiculate (juveniles may show bold ver- Boldly vermiculate, dark on A. Nearly uniform, light. miculation). gemmosus at most shades of purhave dark spots Weakly on ple (juveniles may on a light ground). A. tiic/11 Some light streaks or spots on a light to strongly vermiculate, dark light. Light marks or spots on dark back- ground. ground. Dorsum A. eulaemus "Purplish brown above with rather indistinct transverse bars on back and round lighter spots on sides." Boulenger 1908. large A. ventrimaculatus Middorsum uniform dark or crossed by narrow dark bars containing light spots. Flanks boldly spotted with lighter. Oblique narrow white bars with somewhat irregular margins meeting in forward pointing angles middorsally and separated by wide areas of uniform brown. A longitudinal dorsal light zone bound- ed by dorsolateral lightbands or a narrow middorsal light line or a series of middorsal multiply-shaped figures. Flanks with strong to weak or absent spotting or vermiculation. VERTEBRATE ECOLOGY AND SYSTEMATICS Table 2. 263 Continued. Dewlap 99 <3<5 A. gemmosus A Dark transverse bands widest dorsally separated by areas with bold and irregular spotting. Transverse bands tapering on flanks which become entirely spotted or pale dorsum and middorsal light zone with dark mar- gins or a narrow middorsal light line or short transverse bands not extending onto flanks. Flanks patternless or with some dark spotting. flanks nearly uniform. A. fitchi Broad dark transverse bands Flanks more or vermiculate. less A dorsally. middorsal boldly spotted or light zone with dark mar- gins or dark transverse bands narrowed in center (butterfly pattern). Flanks may be obscurely vermiculate or spotted with darker. Venter eulaemus A. A. ventrimaculatus Venter obscurely but densely vermiculate. dark on light. Sides of venter pepper and Densely and finely spotted (juveniles Light with ing also vermiculate). A. gemmosus more uniform brown salt becom- in center. weak dark vermiculations or spotting. Belly with spotting encroaching from flanks or nearlv uniform bluish. Venter immaculate or more or less densely but obscurely vermiculate Belly darkish, edge invaded by dark Belly with dark spots or markings, and spotted. A. fitchi most prominent spots. 1 All comments on the female of eulaemus are based only on AMNH 1 laterally. 18980. Anolis eulaemus was described from a unique male type (BMNH 1946.8.13.31) from Pavas (near San Antonio). Departamento Valle, Colombia, and A. ventrimaculatus from two syntypes— an adult female and a juvenile from the Rio San Juan, Intendencia Choco, Colombia. We maculatus, the lateral scales are large, as in A. fitchi, larger than the ventrals, and in single series here designate the adult female syntype (BMNH 946.8. 13.5) as the lectotype; it is uncertain that Rico, Risaralda (formerly Caldas). This specimen makes more plausible Boulenger's inexact the faded juvenile is the same species. Recently collected material from cloud forests locality Departamento Valle, Colombia, including material from the vicinity of San Antonio, provides an excellent match for the female syntype of A. ventrimaculatus, which lacks any trace of a dewlap. Males from the vicinity of Lago Cal- of the Rio San Juan, and M. G. Palmer, who collected both syntypes of A. ventrimaculatus and 1 in the ima, Departamento Valle, recently collected in numbers along with females of A. ventrimaculatus, agree with these females and not with the type of A. eulaemus in the small size of the scales surrounding the interparietal and differ sharply squamation of the dewA. eulaemus has minute lateral The of lap. type scales on the dewlap that are smaller than the ventrals and are crowded in multiple series of rows that are widely separated by naked skin. On the contrary, in males referred to A. ventrifrom A. eulaemus in the in rows that are closely packed. The new series has permitted recognition as A. ventrimaculatus of a specimen (Instituto La Salle 109) collected by Niceforo Maria "Rio San Juan" at Pueblo for the syntypes of A. ventrimaculatus. Pueblo Rico is near the source the type of A. eulaemus, is reported by Boulenger (191 1), in the same paper in which he described A. ventrimaculatus, to have obtained Leptognathus (=Dipsas) sancti-joannis at "Pueblo Rico, slopes of San Juan River, Colombian Choco. 5200 feet." It is possible that Niceforo Maria's specimen is topotypic or near-topotypic, but Boulenger's careful avoidance of precision leaves the question open. On the basis of the new collections, A. ventri- maculatus seems to be common, but A. eulaemus remains rare in collections. No material has been A collected recently near the type locality. single 1 1 0495) was collected at Lago Calmale (AMNH SPECIAL PUBLICATION -MUSEUM OF 264 ima by Stephen C. Ayala in 1974. None has been obtained in more recent collections from that area. Four other males are known: two from Penas Blancas (where A. ventrimaculatus also occurs), one from the "Farallones de Cali, Pi- chinde," and one from "region alta cerca al Lago Calima." The probable female of the species is AMNH 118980 from "mounrepresented by tains above the north side of Lago Calima ( 700 m)."- All of these localities are on the Pacific 1 versant of the Cordillera Occidental in Departamento Valle, Colombia. The female has no well- developed dewlap, but the area is indicated by longitudinal throat folds. A specimen (BMNH 1910.7.11.4) from "Siato, near Pueblo Rico, Choco," collected by Palmer, indicates that A. eulaemus occurs somewhere near the probable type locality of A. ventrimaculatus. Thus the two species appear to be broadly sympatric; whether they are ever synotopic, like A. aequatoriahs and gemmosus, is unknown. The dewlap of male A. ventrimaculatus has two color morphs — one with dark brown skin covered by yellow lines of scales and one with orange skin covered by lighter lines of scales and A. with a dark blotch at its base. gemmosus from Anolis Andes the western slope of Ecuador, the third previously described member of the eulaemus subgroup of the aequatoriahs species group has not previously the in been associated with this group. Williams ( 1976) punctatus group in error. The of dilations digital gemmosus are narrow as in the aequatoriahs group, not wide as in the puncplaced it in the tatus series ever, A. (cf. the key in Williams 1976). gemmosus (maximum <? size 66 How- mm) is smaller than any other member of the group, and this fact contributed to the failure of Williams to recognize its true relationships. It is interesting member of the group co-occurs that the smallest with the A. aequatoriahs, some individuals of which reach the maximum size for the group. Size differences of this magnitude imply a par- NATURAL HISTORY which Fitch el al. (1976) with their restricted sample (five) of A. aequatoriahs could not demonstrate. titioning of resources The second smallest member of the aequato- mm riahs group (A. ventrimaculatus, 80 maxisnout-vent length) is a parallel case; its mum sympatry with A. eulaemus ( 1 00 mm snout-vent length) again a case of co-existence of two related species that differ significantly in size. is Anolis gemmosus is morphologically somewhat intermediate between A. eulaemus and A. ventrimaculatus: male dewlap it has the lateral scales of the in multiple lines as in A. eulaemus but as large or larger than ventrals as in A. ventrimaculatus. Anolis gemmosus shares with A. ventrimaculatus a condition that has been considered very rare in A nolis— absence of the pa- Frequently in A. ventrimaculatus, less commonly in A. gemmosus, there may be no rietal eye. an interparietal by size, popresence of a central spot or win- scale recognizable as sition, or the dow. gemmosus resembles Anolis A. ventrimacula- dewlap coloration, but as Fitch et al. (1976:121) reported, and K. Miyata (pers. comm.) confirms, there appear to be no discrete morphs, either within or between populations. tus in variable Fitch et al. commented: "The highly variable dewlap did not seem to comprise well-defined classes but tended to form a continuum between extremes. At one extreme were dewlap with little contrast, dull yellowish-green on the basal area, shading to dull greenish-yellow on the outer part. The more contrasting and colorful type of dewlap was similar in having a dull greenish-yellow on the outer part, but the basal part was bluish-green with six narrow sharply defined white stripes diverging from a center on the anterior basal portion. The stripes had bright blue edges proximally at their origins, but distally the blue changed to green and the stripes themselves became suffused with the yellow background and finally blended into it and blended with their brighter colored edges." (pers. comm.) mentioned some vari"The dewlap skin is basically a pale yel- Miyata 2 That this is a female of A. eulaemus must remain ations: lowish-green. In many specimens there is a disorange or yellow wash along the edge as well, and in some specimens there is also a dark somewhat doubtful, for it was not taken in association with any male. However, there is available something lacking for any of the males, color in life: "Green with tinct broad blackish gray diagonal bars on sides. Throat and venter light brown. Small dewlap brown with black stripes. Iris brown. Tongue gray, throat lining unpigmented." (C. W. Myers, in litt.) blue area anterobasally, and, in some, a white area posteriorly. The extent and intensity of these colors is quite variable and in some individuals VERTEBRATE ECOLOGY AND SYSTEMATICS the colors grade into each other while in others they remain as discrete patches. The dewlap scales are arranged in distinct longitudinal rows; these scales are normally white, but in many specimens they are green and, in a few individuals, they are yellow. In some specimens the anterior scales are white and the posterior scales either green or yellow, or sometimes all three colors. These scales, if they are white or yellow, are bordered by either green or blue scales, the latter only on the anterobasal dewlap skin." if there is blue the Recent (Simpson 1979). Application of the albumin clock hypothesis (Wilson et al. 977) to immunological data on frogs of the genus Gastrotheca inhabiting cloud forests on opposite sides of the Andes shows a divergence time of 2-3 1 million years (Scanlan et al. 1980). If the time of speciation events in Gastrotheca are indicative of that for other groups, such as anoles, it might be assumed that populations of cloud forest inhabitants were continuous across the Andes in and subsequently were fragmented by the orogenies and climatic and vegetational shifts in the Pleistocene documented in the late Pliocene, Interpopulational variation in dewlap color is, as Fitch et al. stated, ordinarily rare or minor in anoles. It is presumably another indication of the the palynological record (van der affinity of A. gemmosus and A. ventrimaculatus that they show such variability. However, A. fit- chi known not is to show comparable Difference in A. fitchi is 265 Hammen 1 974). Acknowledgments variation. primarily sex dimor- phism, although the extent of blotching able in the female dewlap. is vari- Body color in A. gemmosus also is highly variable (Fitch et al. 1976; also see Table 2 for pattern repertoire). Anolis andianus Boulenger (1885) from Milligalli, Provincia Pichincha, Ecuador, appears to be well within this variation, both in scales and color, and we here formally synony- We are W. Myers, Amerof Natural History (AMNH), W. Ronald Heyer, National Museum of Natural History (USNM), Alice G. C. Grandison, British ican indebted to Charles Museum Museum (Natural History) (BMNH), and Stephen C. Ayala, Universidad del Valle. Colombia, for making specimens available for our use. Kenneth Miyata has added some useful comments and observations. The Museum of Natural His- Provincia Carchi, Ecuador, very close to the northern border of the country; the species very probably extends northward into at least the De- of Kansas, is abbreviated KU; of Comparative Zoology. Harvard University, MCZ. Duellman's fieldwork was partly supported by grants from the National Science Foundation (DEB 74-01998) and the Na- partamento Narino, Colombia. Anolis fitchi in Provincia Napo, Ecuador, ob- cially mize it with A. Anolis gemmosus O'Shaughnessy ( 1875). gemmosus is known from Maldonado in viously is a close relative of the three species from the western slopes of the Andes. Scale counts for all four taxa overlap (Table 1). Differences in coloration are noted in Table there that males different 2; it is and females of these taxa show differences and resem- interspecific any one species is most Museum tional 304). We are espeGeographic Society grateful to Linda Trueb for the color illustration. Literature Cited Boulenger, G. A. 1885. Catalogue of lizards in the British Museum (Natural History). Edition 2. London, xiii + 497 pp. 1 908. closely related to any other. The presence of related species in cloud forests on the opposite sides of the Andes in Ecuador and southern Colombia is now known to be a common distribution pattern in lizards, snakes, and frogs (Duellman 1979). The differentiation of cloud forest inhabitants on either 1911. rather Andes is most likely a late Cenozoic phenomenon. The final elevation of the northern Andes occurred in the late Pliocene and up to side of the ( 1 conspicuous blances. Although the unity of the group is confirmed by the cross-affinities, it is not easy to say that tory, University the Descriptions of new batrachians and reptiles discovered by Mr. M. G. Palmer in southwestern Colombia. Ann. Mag. Nat. Hist., ser. 8. 2:515-522. Descriptions of new reptiles from the Andes of South America preserved in the British Museum. Ann. Mag. Nat. Hist., ser. 8, 7:1925. Blrt, C. 1931. and Burt, M. D. South American lizards E. in the collection of American Museum of Natural History. Bull. Amer. Mus. Nat. Hist.. 61:1 17-395.' DlELLMAN, W. E. 1979. The herpetofauna of the Andes: Patterns of the SPECIAL PUBLICATION-MUSEUM OF 266 distribution, origin, differentiation, and present communities. Pp. 371-459. In Duellman, W. E. (ed.). The South American Herpetofauna: Dispersal. Kansas, Lynch, J. Origin, Evolution, and Its Monogr. Mus. Nat. Hist. Univ. 1-485. (7): D. A taxonomic and distributional synopsis Amazonian frogs of the genus Eleutherodactvlus. American Mus. Novit., 2696: 1980. of the 1-24. Lynch, 1980. J. D. and Duellman, W. E. The Eleutherodactylus of the Amazonian slopes of the Ecuadorian Andes (Anura: Leptodactylidae). Univ. Kansas Misc. Pub., 69:1-86. Mus. Nat. Sci. Bull., 51:91-128. Museum collection, with de- scription of new Hist., ser. 4, 15:270-281. species. 1 A., Jr. Ann. Mag. Nat. and Traylor, M. A., Jr. Ornithological gazeteer of Ecuador. Harvard University, Cambridge. 151 pp. 977. A. Herpetological type localities in Ecuador. Rev. Ecuatoriana Entomol. Parasitol., 2:335352. Scanlan, B. E., Maxson, L. R. and Duellman, W. E. Albumin evolution in marsupial frogs (Hy1980. Peters, 1955. J. lidae: Gastrotheca). Evolution, 34:222-229. Simpson, B. B. 1979. Quaternary biogeography of the high montane regions of South America. Pp. 157-188. In Duellman, W. E. (ed.). The South Amer- ican Herpetofauna: Dispersal. Kansas, O'Shaughnessy, A. W. E. List and revision of the species of Anolidae 1875. in the British Paynter, R. Hist. Fitch, H. S., Echelle, A. F. and Echelle, A. A. Field observations on rare and little known 1976. mainland anoles. Univ. Kansas NATURAL HISTORY Its Origin, Evolution, and Hist. Univ. Monogr. Mus. Nat. (7): 1-485. VAN der Hammen, T. 1974. The Pleistocene changes of vegetation and climate in tropical South America. J. Bio- geogr., 1:3-26. Williams, E. E. South American anoles: The species groups. Pap. Avuls. Zool. Sao Paulo, 29:259-268. Wilson, A. C, Carlson, S. S. and White, T. J. Biochemical evolution. Ann. Rev. Bio1977. chem., 46:573-639. 1976. Index to Scientific Names 265 48 257. 261-264 andianus abacura, Farancia Abies balsamea *'2 eulaemus f?0 religiosa 223 '"" pensylvanicum rubrum saccharum Achaearanea tepidariorum gryllus acntus, Crocodylus adamanteus, Crotalus Adelophis foxi aemula, Sonora aeneus, Oxybelis aeneus auratus, Oxybelis aequatonalis. Anolis ->nd U4 Sceloporus horridus Alligator mississipiensis Allolobophora caliginosa Alnus rugosa Cladina alpestris. Lampwpeltis mexicana Amblyrhynchus alterna, cnstatus Ambystoma rosaceum. tigrinum.... americana. Ulmus americanus Bufo Fraxinus amoenus, Carphophis andianus, Anolis 239 ' ' - 78 220 223 223 220 223 220 230. 236 glandulosa Arctostaphvlos arenicolor, 76 204. 205 Arbutus )4 Hyla Aristida Arizona 198 iyo 1 99 238 84 66 220 204 w 205. 206 199 Aspidelaps 199 r- aspis. I ipera assimilis. Acheta 199 atricaudatus, Crotalus horridus 197, 199 164-166. 168. 170. 171 58 atrox, Crotalus attenuatus, Ophisaurus... 235 235 ^ JJ 89 126 °v 148 _ 265 230-232 239 239 52 1 119 52. 1 elegans elegans expolita elegans noctivaga arnyi. Diadophis punctatus 1 augusti, Hylactophryne august i cactorum, Hylactophryne.... augusti latrans. Hylactophryne aurantiacum, Hieracium auratus, Oxybelis aeneus Austrelaps superbus australis, Simoselaps , 230.231 236 204, 205 240 1 78 173. 174. 176 177. 179. 180. 182 1" austnaca, Coronella avenaceum, Sorghastrum B 43 443 scoparius Pnmus '" Elaphe obsoleta 233.240 Sa/vadora Annona squamosa annulata, 238 173. 174. 176. 180 Aquilegia aquilinum. Ptendium arborea, Bocconia arborescens, Ipomoea 228 240 257. 260, 261, 264 hallii I 240 repens -32, 240 2 173. 174. 176. 180 Holbrookia Simoselaps Aprasia 199 1 229, 237 257, 261-265 approximans —° Andropogon angustifolia. 260, 261 1 Aparallactus capensis 24 1 piscivorus leucostoma parilis antonii, Rhinocheilus lecontei bilineatus bilineatus 36, 58, 153. 164. 166-168. 198 contortrix albiventris, 26 anomala. Simoselaps 105-1 12. 148, 152. 154, 155 Opheodrys 240 Pituophis melanoleucus 220 piscivorus 260-265 231. 237 ventrimaculatus 1 l • mints nebulosus utowanae 205.206 204 aestivus, Agkistrodon 257-263, 265 punctatus 66 166 Acheta assimilis 89, 90. 92. 93. 98. 100. 101. 126 Acris 90. 92. 93. 95-98. 101. 150. 166 89. crepitans 89 Agave fronlispiece, gemmosus 220. 223 farnesiana Acer affinis. 261 fasciatus ->ru U4 fitchi Acacia. cvmbispina ' carolinensis ermicella annulatus, Scaphiodontophis annulifera, Tropidodipsas Anolis. aequatonalis 223J ^- bairdii, Pcromyscus maniculatus balsamea. Abies Bansia imbricata alians basiliscus, Crotalus basiliscus ' 185-192 241 .... basi/iscus basiliscus, Crotalus.... 156.185.264 257.260.261.264 baudim. Smilisca 267 .... 203. -1- 204 -39 - 241 24^ 228. 23 SPECIAL PUBLICATION-MUSEUM OF 268 223 Begonia NATURAL HISTORY Bungarus caeruleus, 239 228 Pseudoeurycea 223 benthamiana, Tillandsia 229 berlandieri, Rana 173-182 bertholdi, Simoselaps 58, 60, 68, 195, 199, 200 berus, Vipera 205, 206 Betula cerulea 238 bicarinatus tuberculatus, Urosaurus Caesalpinia calamarina, Tantilla bilineatus calligaster Eumeces lynxe 24 1 238 Agkistrodon bilineatus Eumeces brevirostris 240 Masticophis 24 1 bilineatus bilineatus, Agkistrodon bimaculatus, Neelaps.... 173, 174, 176, 177, 179, 180 Trimorphodon biscutatus biscutatus biscutatus, Trimorphodon bistincta, Hyla „ Blattella germanica biscustatus, 24 24 232, 236 1 1 138 Boa 228 239 223 228 237 243 constrictor constrictor imperator Bocconia arborea bocourti, Tantilla bogerti, Callisaurus draconoides Boleosoma lepida Bombina 124, 125, 127, 128 Bothrops godmani boulengeri, Sceloporus clarki Bouteloua brachycercum, Phrynosoma douglassi Brachylophus fasciatus Phrynosoma orbiculare braminus, Ramphotyphlops brevirostris, Eumeces brevirostris bilineatus, Eumeces Brosimum bradti, browni, Phyllorhynchus Bubo Bufo americanus 1 compact His debilis insidior marinus marmoreus 228, 232, 123, 125, 228, 228, mazatlanensis microscaphus mexicanus occidentalis punctatus woodhousei Sceloporus Bungarus caeruleus Bursera bulleri, Buteo jamaicensis butleri. Thamnophis 230, 232, 231, 232, 1 26, Lampropeltis... La mpropeltis 152, 164-166, 168, 169, 171, 197 1 97 calligaster rhombomaculata, Lampropeltis Callisaurus 137, 237 draconoides draconoides bogerti calonotus, Neelaps calva, Tithonia 232, 238 capensis, Aparallactus carinata, Cyclura carolinensis, 1 amoenus 46, 52, carpicinctus, Scaphiodontophis catesbvi, Rana Dipsas Ceiba coccinea cenchoa, Imantodes occidentalis cerastes, Crotalus cherriei, Sphenomorphus Chionactis Rana Chrvsemvs harrietae Hyla Barisia imbricata squamulosus 138 98 1 92 173,179,181 229 80 66 239 1 Cladma alpestris clamitans, 178 178 199 223 205, 206 223 80, 83, 85 picta.. chrysoscelis, ciliaris, 180 179 189, 190 173, 179, 181 Chilomeniscus chiricahuensis, 220 220 43 Cephalanthus 212 177, 178 1 melanoleucus 41,51 123-129, 131, 166, 167 189, 190 223 Cemophora 223 36 1 58, 164, 166-168, 197 cembroides, Pinus rangiferina Cladonia mitis 1 1 185 223 Cassia catenatus, Sistrurus Chaulognathus pennsylvanicus Chelydra serpentina 5 48 148 vermis Chamaedorea 1 57 117, 119 A nolis 236 236 236 236 236 236 236 236 236 236 28 199 24 1 138 Carphophis 1 1 223 cana, Pseudaspis canescens, Thamnophis melanogaster caninus, Dermestes clarki, Rana Sceloporus clarki boulengeri, Sceloporus Clelia clelia Cacophis 237 173, 174, 176, 179, 180 126 232, 238 197 197 calligaster calligaster, Lampropeltis Cercocarpus macrophyllus cerulea, Betula 24- 28 cognatus kelloggi 9 1 calligaster catesbeiana, 1 5 1 Eumeces catenifer, Pituophis 212 virginianus callicephalus, 198 1 1 223 240 66 232, 238 caliginosa, Allolobophora 238 220 238 238 228 228, 230 238 223 232, 240 236 cactorum, Hylactophryne augusti belli 205, 206 205, 206 205, 206 124-129, 131 228, 232 238 86 1 clelia, Clelia 186 Clethrionomys gapperi cliftoni, Dryadophis climacophora, Elaphe 2 1 2 231-233, 239 51, 52 VERTEBRATE ECOLOGY AND SYSTEMATICS 137, 141. 142, 144. 156 Cnemidophorus costatus costatus huico costatus mazatlanensis exsanguis 239 239 239 137-144 52 269 cerastes 199 horridus horridus atricaudatus 58,61 99 229, 230 57, 1 lepidus lepidus klauberi lepidus maculosus 199, 241 scalaris scalaris 239 tigris 190 79 pricei 24 230 24 230 pricei pricei 241 223 236 ruber scutulatus scutulatus 1 99 24 1 241 198 gularis 1 murinus coccinea, 141, 142, 144 Cemophora Cocos nucifera cognatus, Bufo Coleonyx variegatus fasciatus colimensis, Eumeces Thamnophis 232 237 232, 238 230, 240 cvrtopsis 14-16, 18, 20, 22, 24, 25. 27, 29-33, 51 Coluber 35-37, 58-60, 148, 152, 164166, 168, 169 constrictor flaviventris 13, 33-38, 52 constrictor mormon 13, 15-20, 22, 24-38, 51, 52 constrictor. constrictor 197, 198, 228, 1 stejnegeri variegatus collaris, 1 1 molossus molossus nigrescens 13, oaxaca viridiflavus commune, Polytrichium compactilis, Bufo 229 lateritius lateritius Conolophus subcristatus 196-198, 200 viridis lutosus 30 viridis 4! oreganus 1 willardi meridionalis pectinata similis 1 1 78 78 223 117, 121 117, 119 carinata 121 cvrtopsis Thamnophis Thamnophis 117, 119 223 cvrtopsis Thamnophis Thamnophis pulchrilatus, Thamnophis cvrtopsis collaris, cvrtopsis cvrtopsis, 230 240 230. 240 240 230, 240 228 Coluber 35-37, 58-60. 148. 152, 164-166, 168, 169 constrictor flaviventris. Coluber 13, 33-38, 52 13, constrictor imperator, Boa constrictor mormon. Coluber constrictor oaxaca. Coluber contortrix, Agkistrodon corais rubidus, Drymarchon coronata, Tantilla 229 36, 58, 153, 164, 166-168, 198 239 1 52 99 78 coronuta stejnegeri, Cyclura 117, 119 239 costatus, Cnemidophorus costatus huico, Cnemidophorus 239 costatus mazatlanensis, Cnemidophorus 239 couchi, Scaphiopus 231, 232, 235 Coronella austriaca 1 coronoides, Drysdalia 1 crepitans, Acris 89, 90, 92, 93, 95-98, 101, 150, 166 Amblyrhynchus Crocodylus acutus Crotalus adamanteus atrox basiliscus basiliscus D 239 13, 15-20, 22, 24-38, 51, 52 cristatus, 76, 1 cyclocarpum, Enterolobium Cyclura cvrtopsis 1 117, 119. 121 coronuta stejnegeri cymbispina, Acacia constrictor 98, 1 Ctenosaura 228 239 19 99 24 1 1 7, 237 228, 237 viridis viridis cucullatus, Stegonotus 230 239 Boa 41 viridis helleri 205 236 1 Conopsis nasus nasus nasus 36, 46, 57, 58, 61, 68, 164, 167, 168, Ctenotus 189 lateritius viridis 199 Coniophanes fissidens triseriatus 1 19 228 33, 196 199 58, 197, 199 24 1 dacnicolor, dactyloides, Pachymedusa 228, 236 43 Tripsacum 220 236 Dasylirion debilis insidior, Bufo decipiens, Rhadinaea decurtatus. Phyllorhynchus dekavi, Storeria 189, 190 228 58, 148. 149, 152-155, 163-166. 168. 169 delicatissima. Iguana Demansia olivacea psammophis reticulata textilis Dendrophidion dendrophis deppei Pituophis Pituophis deppei deppei deppei, Pituophis dendrophis, Dendrophidion Dermestes caninus Deschampsia jlexuosa 1 1 9 178 178 178 57 199 189, 190 228, 230, 232, 233 240 240 189. 190 138 205, 206 SPECIAL PUBLICATION-MUSEUM OF 270 deserticola Pituophis melanoleucus 30, 41, 44, 48. 49, 51, 52 232, 240 154 Salvadora Diadophis 34,46, 111, 148, 152, 163-166, 168, 169, 230, 233, 239 52 punctatus arnyi 239 punctatus dugesi 239 punctatus regalis 1 66 differentialis, Melanoplus punctatus.... diplotropis, Leptophis catesbyi 1 Dipsosaurus Micrurus distans distans distans, Micrurus 89, Eleutherodactylus hobartsmithi occidentalis peruvianus vocalis Elgaria kingi ferruginea 153 Enhydrina Enterolobium cyclocarpum ephippiata, Leptodeira splendida 90 263 eques megalops, Thamnophis eques virgatenuis, Thamnophis errans, Thamnophis elegans 1 1 17 241 241 Lysiloma Do/omedes sexpunctatus 223 98 238 232, 238 237 douglassi brachycercum, Phrynosoma draconoides, Callisaurus draconoides bogerti, Callisaurus 239 220 260 232, 235 235 260 236 239 eques, distans, divaricata, elegans noctivaga, Arizona Eleocharis 240 Dipsas sancti-joannis NATURAL HISTORY Drvadophis Thamnophis erythrogasier, Nerodia 1 Etheostoma grahami lepidogenys lepidum lepidum lepidogenys pottsi spectabile 1 spectabile pulchellum 231, 232, 233, 239 cliftoni melanolomus melanolomus stuarti Drymarchon corais rubidus Drymobius margaritiferus ftstulosus 232 239 239 239 Drysdalia coronoides 177 1 78 eulaemus, Anolis Eumeces 152, 154, 163, 170, 171, 185 228, 230 brevirostris 238 brevirostris bilineatus 232, 238 232, 238 callicephalus colimensis 148, 149, 151, 152, 154, fasciatus 164-166, 168, 170 1 52, 1 70 dugesi 239 231 239 239 239 Diadophis punctatus Geophis Geophis dugesi Leptotyphlops humilis dugesi dugesi, Geophis dulcis, Leptotyphlops durangensis, Pinus 148, 150, 152 220 inexpectatus 51, 154, 170 Elaphe climacophora 51,52 148, 152, 154 tetragrammus euryxanthus, Micruroides euryxanthus neglectus, Micruroides 189, 232 241 eximia, Hyla Arizona elegans exsanguis, Cnemidophorus exserta, Tillandsia 1 extenuatum, Stilosoma 236 239 37- 44 223 1 1 52 152 guttata obsoleta 51, 52, 58, 148, 152 99 99 36, 51, 52 231 239 169 190 obsoleta bairdi obsoleta obsoleta 1 1 quadrivirgata triaspis triaspis 239 228 lynxe belli parvulus expolita, E 223 240 230 240 240 240 48 243 243-245, 247-254 244 243-254 244 245 244-246, 25 245 257, 261-264 intermedia vulpina elapoides, Pliocercus 1 Elapsoidea 73 Thamnophis elegans elegans, Holbrookia elegans errans, Thamnophis elegans expolita, Arizona 1 52, A nolis 26 1 Brachylophus Coleonvx variegatus Eumeces 1 168, 170 Simoselaps Fest uca rubra 176 Ficimia 238 240 239 173, 174, 176, 179, 180, 182 239 205 ferruginea, Elgaria kingi 238 58, 60, 69, 71, 72, 167 19 237 148, 149, 151, 152, 154, 164-166, 230-232 232, 238 52 fasciatus fasciolatus, elegans Lerista 1 220 farnesiana. Acacia 179 sundevalli Arizona Holbrookia Holbrookia elegans 204, 205 Fagus grandifolia Farancia abacura 154, 173, 179, 181 olivacea Ficus ftssidens, ftstulosus, Coniophanes Drymobius margaritiferus 152 223 1 89 239 VERTEBRATE ECOLOGY AND SYSTEMATICS fitchi, Anolis frontispiece, flagellum, Masticophis Jlagellum Jlagellum flagellum, Masticophis 199 Jlagellum lineatulus, Masticophis Jlavescens, Perognathus Rhadinaea Jlavilata, 1 Coluber constrictor flaviventris, 13, 33-38, 52 205, 206 Deschampsia Jloridana, Rhineura Jlexuosa, 152 43 floridanus, Sylvilagus 237 229 239 fodiens, Pternohyla forreri, Rana foxi, Adelophis 89 Fraxinus americanus frontalis Micrurus 153 Pseudoficimia 240 223 Quercus fulvius, Micrurus Furina H 257-263, 265 1 99 240 43 89 fulva, 147, 150-158, 189 177, 178 271 43 223 Andropogon hallii, Haematoxylum hammondi, Scaphiopus 130 178 166 4 harrietae, Cacophis hawni, Physa helleri, Crotalus viridis 1 Heloderma horridum horridum hespena, Rhadinaea Heterodon 179 52 nasicus 239 nasicus kennerlyi platyrhinos heterolepis, Sceloporus 52, 148 shannonorum, Sceloporus Hieracium aurantiacum heterolepis hirtipes murrayi, Kinosternon Sauromalus Sigmodon 204 212 Gastrophryne olivacea 232, 237 usta 232,237 265 204 229 228, 229 239 260-265 Gastrotheca Gaultheria procumbens Geagras redimitus Gehyra mutilata gemmistratus latistratus, Imantodes gemmosus, Anolis Geophis Holbrookia approximans elegans elegans elegans horridum, Heloderma horridum horridum horridum, Heloderma horridus, Crotalus horridus albiventris, Sceloporus horridus atricaudatus, Crotalus 43 232, 235 197-200 137 238 232, 238 238 239 239 57, 58, 61 Hura polyandra 238 99 239 239 223 Hvdrophis Hyla 153 125 huico, 1 Cnemidophorus costatus humilis dugesi, Leptotyphlops dugesi dugesi dugesi 231 germanica, Blaltella Gerrhonotus 138 163 arenicolor 239 19 1 hobartsmithi, Eleutherodactylus Galium 232 238 204, 205 237 hispidus holbrooki, Lampropeltis getulus gapperi, Clethrionomys 239 232, 240 239 chrysoscelis 230, 236 232, 236 166 getulus, Lampropeltis 163, 169 getulus holbrooki, Lampropeltis getulus splendida, Lampropeltis 197-200 239 eximia smaragdina 232, 236 liocephalus liocephalus gibbosus, girardi. Lepomis 1 3 1 240 223 Masticophis taeniatus glandulosa. Arbutus Gleditsia triacanthos 89 198 godmani, Bothrops Peromvscus maniculatus 203-205, 210-213 148, 152 Tantilla grahami, Etheostoma grahamiae, Salvadora grahamiae lineata, Salvadora 243-245, 247-254 236 smithi Hylactophryne 230,231 236 236 228, 230, 236 august i augusti cactorum august i latrans tarahumaraensis groenlandicum. 240 238 204, 205 204 89 223 Ledum gryllus, Acris Guazuma ulmifolia 152 Cnemidophorus Elaphe Gyalopion quadrangularis Gymnophthalmus 237 237 231, 239 oxyrrhinus oxyrrhinus variolosus Hypsiglena torquata 152 grammicus microlepidotus, Sceloporus grandifolia, Fagus guttata, 236 Hypopachus gracilis gularis, bistincta 1 52 181,232 229, 232, 239 173, 179, 185 I Iguana delicatissima Imantodes cenchoa gemmistratus imperator. ciliaris, Boa 17 19 115-121, 238 115-121, 238 iguana iguana. Iguana imbricata 1 1 189, latistratus Barisia constrictor 90 239 239 239 1 SPECIAL PUBLICATION -MUSEUM OF 272 incinctus, Simoselaps inexpectatus, insidior, Bufo Eumeces 173, 174 lepida, 152, 170 Lepidochelys 236 olivacea lepidogenys Ipomoea arborescens 239 228 223 jamaicensis, Buteo 212 interorbitalis, Boleosoma 230, 232, 237 debilis integrum, Kinosternon intermedia, Elaphe triaspis NATURAL HISTORY Syrrhophus 243 229 229 Etheostoma Etheostoma lepidum lepidum, Etheostoma lepidum lepidogenys, Etheostoma 244 244 243-254 244 lepidus Crotalus 229, 230 244 Poecilichthys jarrovi Sceloporus Sceloporus jarrovi jarrovi jarrovi, Sceloporus Jatropha Juniperus 137-144, 229, 230 238 238 220 220 lepidus klauberi, Crotalus 199, 241 lepidus maculosus, Crotalus 241 Lepomis gibbosus 1 3 1 Leptodactylus melanonotus 236 236 occidentalis Leptodeira maculata Bufo kennerlyi, Heterodon nasicus kelloggi, 228, 232, 236 239 239 kingi ferruginea, Elgaria Kinosternon 237 murrayi integrum hirtipes klauberi, Crotalus lepidus 230, 232, 237 199, 241 232, 240 240 240 240 263 240 punctata septentrionalis polysticta splendida ephippiata Leptognathus sancti-joannis Leptophis diplotropis Leptotyphlops 154 dulcis 148, 150, 152 humilis dugesi 239 Lerista 1 elegans Xenopus lambda paucimaculata, Trimorphodon laevis, 1 Lampropeltis calligaster calligaster calligaster calligaster 27 24 1 51, 154, 196 152, 164-166, 168, 169, 171, 197 rhombomaculata 97 leucostoma, Agkistrodon piscivorus 1 97 Lim nodynastes 1 Limonia 236 163, 169 getulus holbrooki 1 mexicana mexicana alterna pyromelana triangulum triangulum nelsoni triangulum sinaloae praepedita leucopus, Peromyscus 97-200 239 239 1 99 199 51, 52, 199 240 240 Lerista Salvadora Salvadora grahamiae lineatum, Tropidoclonion lineatulus, Masticophis flagellum liocephalus, Gerrhonotus liocephalus liocephalus liocephalus, Gerrhonotus lippiens, 148-150, 152, 154, 155, 185, 190 Urosaurus ornatus 229 189 lateristriga, Rhadinaea littoralis Scincella lateritius Coniophanes Imantodes gemmistratus latrans, Hylactophryne augusti laureata, Rhadinaea 228 239 239 239 236 240 lebentina, Vipera 198 lateritius lateritius lateritius, latistratus, lecontei, Rhinocheilus lecontei antonii, Rhinocheilus lecontei tesselatus, Rhinocheilus Ledum groenlandicum 232 240 Sympholis Simoselaps lumholtzi, Pinus 30 lutrensis, Notropis Lycosa rabida lynxe belli, Eumeces 166 leonensis, Oligocephalus 244 1 53 1 66 1 66 239 223 Lysiloma divaricata Lytorhynchus 179 M macrophylla, Quercus macrophyllus, Cercocarpus maculata, Leptodeira maculosus, Crotalus lepidus madagascariensis, Sanzinia 220 240 239 239 240 222 204 Pinus lemniscatus, Micrurus 240 148, 152 lutosus, Crotalus viridis macrolepis, Sceloporus poinsetti leiophylla, 176 179 77, 82 173, 174, 176, 179, 180 Malaclemys terrapin 179 Leiobunum vittatum 25 lineata lateralis Coniophanes Coniophanes 1 1 getulus getulus splendida lineata picturata 78 176 176 176 176 43, 167 199 76, 238 223 223 232, 240 24 1 1 99 VERTEBRATE ECOLOGY AND SYSTEMATICS magnaocularis, Rana 229 223 77, 79-85 77, 82 77, 82 77, 82 77, 81-85 77, 82 223 203, 212 203-205, 210-213 Magnolia shicdcana Malaclemys terrapin terrapin littoralis terrapin pileata terrapin tequesta terrapin terrapin.. mangle. Rhizophora maniculatus bairdii, Peromyscus maniculatus gracilis, Peromyscus marcianus, Thamnophis 1 52 marmoreus, Bufo 239 204 123, 125, 228, 236 228, 236 Masticophis 33, 51, 60, 68, 195 margaritiferus fistulosus, mariana, Picea Drymobius marinus, Bufo 240 1 99 240 bilineatus flagellum flagellum flagellum lineatulus mentovarius mentovarius striolatus taeniatus 231 240 20, 34, 52, 58, 60, 61, 111 miliarias, Sistrurus minis, A nolis 236 239 Bufo Cnemidophorus costatus 93 megacystis, Zeugorchis 240 43 megalops, Thamnophis eques megalotis, Reithrodontomys melanogaster canescens, Thamnophis 241 melanoleucus, Pituophis... 34, 43-46, 48, 49, 58, 232 melanoleucus affinis, Pituophis 240 melanoleucus catenifer, Pituophis 41. 51 melanoleucus deserticola, Pituophis 30, 41, 44, 48,49, 51, 52 melanoleucus sayi, Pituophis... 41, 44, 48, 50, 52, 54 232 239 236 melanolomus, Dryadophis melanolomus stuarti, Dryadophis melanonotus, Leptodactylus Melanoplus differentialis Menetia mentovarius, Masticophis mentovarius striolatus, Masticophis meridionalis, Crotalus willardi 166 176, 178 23 1 1 Cladonia modeslus, Syrrhophus molilor, Tenebrio molossus, Crotalus molossus nigrescens, Crotalus lemniscatus nigrocinclus 230 241 1 mormon. Coluber 1 195, 199, Morethia constrictor 76, 200 1 78 5-20, 22, 24-38, 51. 52 1 mucosus, Ptyas mucronata, Pseudotsuga multiplicand. Scaphiopus murinus, Cnemidophorus murrayi, Kinosternon hirtipes Mus musculus 3, 1 1 99 220 235 141. 142, 144 237 166, 167 musculus, Mus muticus, Trionyx 166. 167 Gehvra 228, 229 mutilata, 85 N Heterodon nasicus kennerlyi, Heterodon nasus Neelaps bimaculatus 46, 52 231, 237 173, 174, 176-182 173, 174. 176, 177, 179. 180 173, 174. 176, 179. 180 calonotus neglectus, Micruroides euryxanthus 241 nelsoni Lampropeltis triangulum Sceloporus Terrapene Nerodia 1 230 239 239 Conopsis Conopsis nasus nasus nasus, Conopsis Natrix tigrina nebulosus, Anolis 241 236 238 98 236 43 43, 67 212 52 239 nasicus, 240 199 ochrogaster pennsylvanicus Micruroides fulvius 1 197, 198, 228, molurus. Python nemoralis, Poa Neoseps revnoldsi 239 mexicana, Lampropeltis mexicana alterna, Lampropeltis mexicanus, Bufo microscaphus microlepidotus, Sceloporus grammicus Micropterus salmoides microscaphus mexicanus, Bufo Microtus frontalis 1 26 84 205, 206 236 38, 66 mississipiensis, Alligator mitis, 99 223 173, 174, 176 minima, Simoselaps mazatlanesis distans distans 97, 1 Mimosa 240 30 taeniatus girardi taeniatus taeniatus euryxanthus euryxanthus neglectus Micrurus 273 240 228, 238 228 204 152 51. 150. 154. 170 148 148 rhombifera 34-36. 51. 52. 98. 164, 166-169. 171 sipedon erythrogaster valida valida nigra, Salix nigrescens, Crotalus molossus nigriceps, Tantilla nigrocinclus, Micrurus nigronuchalis, Thamnophis 240 43 241 148 1 88 24 1 nitidus, 232 236 239 151, 153. 156, 158 Norops 257. 261 241 153 147, 150-158. 189 153 188 Notechis scutatus 1 nothus, Scaphiodontophis zeteki 1 90 1 66 1 89, 232 24 1 Tomodactylus nitidus petersi, Tomodactylus noctivaga, Arizona elegans Notropis lutrensis nucifera, Cocos Nymphaea 1 1 223 220 NATURAL HISTORY SPECIAL PUBLICATION-MUSEUM OF 274 O 229 oaxaca, Coluber constrictor obesus, 117, 119 Sauromalus obsoleta 51, 52, 58, 148, 152 Elaphe Elaphe obsoleta obsoleta bairdi, Elaphe obsoleta obsoleta, Elaphe 99 199 199 pensylvanicum, Acer Perognathus jlavescens Peromyscus Bufo Cephalanthus Eleutherodactylus Leptodactvlus Thuja 58, 152 occipitomaculata, Storeria 43, 167 243, 245, 246 ochrogaster, Microtus Oligocephalus 244 179-181 leonensis Oligodon Demansia 1 Ficimia 152 78 232, 237 Gastrophryne Lepidochelys oocarpa, Pinus 229 222 Opheodrvs 105-112, 148, 152, 154, 155 41, 109, 110 152 164-166, 168, 170, 171 Opuntia orbiculare bradti, Phrynosoma 220 238 oreganus, Crotalus viridis 41 Oreopanax peltatum ornata, Pseudemys scripta 223 237 229 223 ornatus lateralis, Urosaurus Ostrya virginiana Oxybelis aeneus aeneus auratus Hypopachus oxyrrhinus oxyrrhinus oxyrrhinus, Hypopachus oxyrrhinus, Pachycereus pecten-arboriginum Pachymedusa dacnicolor Panicum virgatum parietalis, parilis, Thamnophis sirtalis Anolis Eumeces paucimaculata, Tnmorphodon lambda parvulus, pecten-arboriginum, Pachycereus pectinata Ctenosaura Spartina Pelamis platurus Pelobates peltatum, Oreopanax pennsylvanicus Chaulognathus Microtus 228 240 237 237 223 228, 236 43 43 52 260, 26 1 228 24 1 223 24 Tropidodipsas Phrynosoma douglassi brachycercum orbiculare bradti Phvllodactylus tuberculosus saxatilis 238 238 237 Phyllophaga Phvllorhynchus browni 1 232, 240 228 1 66 204 decurtatus Physa hawni Picea mariana picta, 80, 83, 85 1 7 6 Chrysemys picturata, Lerista pileata, 77, 82 Malaclemys terrapin Pinus cembroides leiophylla lumholtzi oocarpa strobiformis strobus teocote pipiens, Rana 43 153 228 124, l>b, i28 223 38 179 220 220 220 222 222 220 205, 206 222 123, 126, 128, 166, 167, 229 98 199 piscivorus, Agkistrodon 1 piscivorus leucostoma, Agkistrodon 223 Pithecollobium sonorae 33, 5 Pituophis deppei 1 228, 230, 232, 233 240 deppei deppei melanoleucus 34, 43-46, 48, 49, 58, 232 240 melanoleucus afflnis melanoleucus catenifer 41,51 melanoleucus deserticola 30, 41, 44, 48, 49, 51, 52 melanoleucus sayi 41, 44, 48, 50, 52, 54 1 52 plamceps, Tantilla 228 platurus, Pelamis 52, 148 platyrhinos, Heterodon 190 Pliocercus 190 elapoides 204 Poa nemoralis Poecilichthys lepidus poinsetti macrolepis, Sceloporus polyandra, Hura Polygonum polysticta, Leptodeira septentrionalis 228, 237 1 229 Phrynohyas venulosa durangensis attenuatus 1 260 236 Tomodactylus nitidus petersi, olivacea Ophisaurus 203, 212 203-205, 2 1 0-2 3 peruvianas, Eleutherodactylus philippi, 230, 232, 236 43 235 236 204 vernalis 43, 167 leucopus maniculatus bairdii maniculatus gracilis 1 occidentalis aestivus 204 43 127, 209 Polytrichum Populus commune porphyriacus, Pseudechis pott si, Etheostoma praepedita, Lerista pretiosa, Rana 244 238 223 220 240 205 89 57, 111, 178, 199 245 1 76 97 pricei 1 38 212 Crotalus Crotalus pricei 230 24 1 VERTEBRATE ECOLOGY AND SYSTEMATICS pricei pricei, Crotalus 24 procumbens, Gaulthcria 204 220 Prosopis 79-1 8 52, 98, 148, 152 Prosymna 1 proximus, Thamnophis 73, 1 1 43 Prunus angustifolia psammophis, Demansia Pseudaspis cana 1 78 1 99 57, 111, 178, 199 Pseudechis porphyriacus 80 79 237 228 240 Pseudemys scripta scripta ornata Pseudoeurycea belli Pseudoficimia frontalis Pseudonaja text His Pseudothelphusa Pseudotsuga mucronata Psidium 178 Psittacanthus Pteridium aquilinum Pternohyla fodiens Ptyas mucosus pubens, Sambucus pulchellum, Etheosloma spectabile pulcherrima rogerbarbouri, Rhinoclemmys pulchrilatus, 1 Thamnophis cyrtopsis punctata, Leptodeira punctatus Anolis 223 220 223 223 223 204, 205 237 199 204 245 237 230, 240 240 264 231, 232, 236 Bufo Diadophis 34,46, 111, 148, 152, 163-166. 168, 169, 230. 233, 239 52 punctatus aryni, Diadophis 239 punctatus dugesi, Diadophis 239 punctatus regalis, Diadophis Rana sphenocephala 126 tarahumarae 237 rangiferina, Cladina redimitus, Geagras regalis, Diadophis punctatus relict a, Tantilla 57 152 36 Neoseps Rhabdophis tigrinus reynoldsi, Rhadinaea 1 Quercus ' fulva macrophylla velutina viminea lateristriga 189 laureata 240 Rhineura floridana 1 232 240 lecontei lecontei antonii lecontei tesselatus 1 Rhinoclemmys pulcherrima rogerbarbouri Rhizophora mangle rhombifera, Nerodia Rhus 1 197 Rhinoclemmys pulcherrima 223 237 terebinthifolia rogerbarbouri, 41, 164, 166-169 Ramphotvphlops braminus Rana berlandieri catesbeiana chiricahuensis clamitans forreri magnaocularis pipiens pretiosa 228 124-128 229 Simoselaps rosaceum, Ambystoma ruber, Crotalus roperi, Drymarchon corals 123-129. 131. 166. 167 229 124-129. 131 229 229 123, 126, 128, 166, 167, 229 97 173, 174, 176. 178. 180 235 199 239 205 Tantilla 1 52 205. 206 rubrum, Acer Thamnophis rugosa, Alnus Rumex saccharum, Acer Salix bairdi desert icc'.a ahamiae grahamiae 79 237 223 48 calligasier rhombomaculata, Lampropeltis Salvadora 166 rabida, Lycosa radix, Thamnophis 52 Rhinocheilus nigra salmoides, Micropterus R 89 232, 240 hesperia rufipunctatus, 36, 51, 52 223 223 223 89 223 90 1 1 flavilata Rubus 229, 232, 239 89. 189 decipiens rubra Festuca quadrangularis. Gyalopion quadrivirgata, Elaphe 76 1 Demansia reticulata, 1 99 195, 199, 200 52 1 220 43 Reithrodontomys megalotis repens, Aprasia 152 1 229 239 Abies religiosa, pygmaea, Umbra pvromelana, Lampropeltis Python molurus 126 205, 206 tigrina pygaea, Seminatrix 1 125, 128, 129 temporaria rubidus, 3 237 237 pustulosa sinaloae 237 pustulosa, 275 223 24 1 204 205 204 89,220 43 98 1 54, 1 79 233. 240 232, 240 1 gi lineata lineata Sambucus pubens 52 240 179 204 sancti-joannis Dipsas Leptognathus Sanzinia madagascariensis sartorii, Tropidodipsas sauritus, Thamnophis 263 263 99 1 52 52 1 SPECIAL PUBLICATION-MUSEUM OF 276 Sauromalus 117, 121 1 1 9 hispidus 117, 119 9 obesus variits 1 1 Phyilodactylus tuberculosus Tomodactylus tnelanoleucus .... 237 232, 236 41, 44, 48, 50, 52, 54 173, 174, 176, 179, 180, 182 173, 1 74 173, 174, 176, 179, 180 littoralis 173, 174, 176 173, 174, 176, 178, 180 roperi 173-182 semifasciatus warm 173, 174, 176, 177, 180 223 Simulium scalaris Cnemidophorus scalaris Sceloporns Cnemidophorus Scaphwdontophis anmdatus scalaris scalaris, 239 238 239 185, 187, 189-192 185-192 1 carpicinctus 85 185-192 venustissimus 185 190 zeteki zeteki nothus Scaphiopus 231, 232, 235 couchi hammondi 130 235 nudtiplicatus Sceloporus bulled 1 52, 1 56 232,238 228,232 clarki clarki boulengeri grammicus microlepidotus heterolepis heterolepis shannonorum horridus albiventris jarrovi jarrovi jarrovi nelsoni poinsetti macrolepis scalaris spinosus spinosus spinosus undulatus 238 238 232 238 238 137-144, 229, 230 238 228, 238 238 238 232 238 112, 152, 167 232, 238 utiformis Scincella 154 148-150, 152, 154, 155, 185, 190 43 scoparius, Andropogon 79 scripta. Pseudemys lateralis scripta ornata, 237 Pseudemys scutatus, Notechis 1 scutulatus, Crotalus scutulatus scutulatus scutulatus, Crotalus semiannulata, Sonora senufasciatus, Simoselaps Seminatrix pygaea septentrionalis polysticta, Leptodeira serpentina, Chelydra sexpunctatus, Dolomedes shannonorum, Sceloporus heterolepis shiedeana. Magnolia Sigmodon hispidus similis, 173-182 bertholdi fasciolatus incinctus minima saxatilis sayi, Pituophis NATURAL HISTORY Ctenosaura Simophis Simoselaps anomala approximans australis 1 1 24 1 24 1 148, 152, 153 1 73-1 82 155 152 240 98 98 238 223 43 43 117, 119, 121 188 sinaloae Rana sipedon, Nerodia 34-36, 51, 52, 98, 164, 166-169, 171 sirtalis 58-70, 72, 98, 164-169 52 52 sirtalis parietalis, Thamnophis 52 sirtalis sirtalis, Thamnophis 196 Sistrurus catenatus 58, 164, 167, 168, 197 1 97, 1 99 miliarius Thamnophis Thamnophis sirtalis 232, 236 228, 237 smaragdina, Hyla Smilisca baudini 236 Hyla Sonora aemula semiannulata smithi, 154 232, 240 148, 152, 153 223 43 43 237 237 sonorae, Pithecollobium Sorghastrum avenaceum Spartina peclinaia spatulatus, Triprion spatulatus spatulatus spatulatus. Triprion Etheostoma Etheostoma spectabile, spectabile pulchellum, Sphaerodactylus torqualus sphenocephala, Rana Sphenomorphus 244-246, 251 245 229 1 26 1 85 192 cherriei spinosus Sceloporus Sceloporus spinosus spinosus spinosus, Sceloporus splendida, Lampropeltis getulus splendida ephippiata, Leptodeira squamosa, Annona squamulosus, Cacophis 232 238 238 239 240 223 78 180 178 1 Stegonotus cucullatus stejnegeri 241 Crotalus 117, 119 Cyclura coronuta 223 Stevia StUosoma extenuatum Storeria dekavi occipitomaculata storerioides 173, 174, 176-181 storerioides, Storeria 173, 174, 176, 180 striatula, Virginia 173, 174, 176, 180 173. 174, 176, 177, 179, 180, 182 240 237 Lampropeltis triangulum striolatus, 155 152 154, 155 58, 148, 149, 152-155, 163-166, 168. 169 58, 1 52 230, 240 230, 240 52, 148, 149, 153, 155 Masticophis mentovarius strobiformis, Pinus 240 220 VERTEBRATE ECOLOGY AND SYSTEMATICS 205, 206 strobus, Pinus Dryadophis mclanolomus subcristatus, Conolophus sundevalli, Elapsoidea 239 stuarti, I 1 9 179 178 superbus, Austrelaps 43 240 Syhilagus Jloridanus Sympholis lippiens Syrrhophus 277 maraanus melanogaster canescens nigronuchalis proximus radix rufipunctatus sauntus sirtalis 228 236 232, 236 mterorbitalis modestus teretistes sirtalis parietalis sirtalis sirtalis Thuja occidentalis 152 24 241 52,98, 148, 152 41, 164, 166-169 241 52 58-70, 72,98, 164-169 52 52 204 1 tigrina Natrix 46,52 Rana taeniatus tigrinum, 20, 34, 52, 58, 60, 61. 68, 111 Masticophis 30 Masticophis taeniatus 240 taeniatus girardi, Masticophis 30 taeniatus taeniatus, Masticophis tigrinus, Tantilla 148, bocourti calamarina coronata 152-155 228 240 152 148, 152 gracilis 148 152 152 152 nigriceps planiceps relicta rubra wilcoxi wilcoxi yaquia tarahumarae, Rana tarahumaraensis, Hylactophryne tau, 228, 231 Trimorphodon temporaria, Rana Tenebrio molitor teocote, 240 232, 240 237 228, 230, 236 125, 128, 129 138, 166 222 Pinus tepidariorum, Achaearanea 66 77, 81-85 223 232, 236 228 1 Malaclemys terrapin terebinthifolia, Rhus tequesta, Syrrhophus Terrapene nelsoni teretistes, Malaclemys Malaclemys terrapin terrapin littoral is, Malaclemys terrapin pileata, Malaclemys terrapin tequesta, Malaclemys terrapin terrapin, Malaclemys butleri 223 223 223 Tithonia calva Tomodactylus nitidus nitidus petersi saxatilis torquata, Hypsiglena iorquatus, Sphaerodactylus Toxicodendron triacanthos, Gleditsia triangulum, Lampropeltis triangulum nelsoni, Lampropeltis triangidum sinaloae, Lampropeltis triaspis, 232 236 232, 236 231, 239 229 223 89 51, 52, 199 240 240 231 Elaphe triaspis intermedia, 239 Elaphe Trimorphodon 24 24 biscutatus biscutatus lambda paucimaculata 77, crytopsis cyrtopsis cyrtopsis pulchrilatus 1 Trionyx muticus 85 43 237 43 Triplasis Triprion spatulatus spatulatus 198 Crotalus triseriatus, Triturus 124-128 77, 82 Tropidoclonion lineatum 1 81-85 77, 82 1 79 148, 152, 154 199 178 46, 51, 57, 60, 62, 152, 154, 170 36 cyrtopsis collaris 1 228, 231 tau 48, Tropidodipsas annulifera philippi sartorii 230 230, 240 240 230, 240 58, 60, 69, 71, 72, 167 240 230 240 240 154 1 52 155 241 241 1 52 238 237 tuberculatus, Urosaurus bicarinatus cyrtopsis elegans elegans errans eques eques megalops eques virgatenuis benthamiana exserta 77, 82 textilis Thamnophis Tillandsia 77,82 77, 82 Rhinocheilus lecontei tetragrammus, Eumeces tesselatus, Pseudonaja Rhabdophis Cnemidophorus Tripsacum dactyloides terrapin Demansia tigris, 126 235 36 190 Ambystoma tuberculosus saxatilis, Phyllodactylus U ulmi folia, Gauzuma 223 89 Ulmus americana Umbra pygmaea undulatus, Sceloporus Unechis Uromacer 1 3 1 112. 152. 167 1 77. 78 189 1 Urosaurus bicarinatus tuberculatus ornatus lateralis 238 229 usta. NATURAL HISTORY SPECIAL PUBLICATION-MUSEUM OF 278 232, 237 232, 238 229, 237 Gastrophryne utiformis, Sceloporus utowanae, Anolis viridiflavus. Coluber 199 viridis Crotalus 36, 46, 57, 58, 61, 68, 164, 168, 196-198, Crotalus V 205 Vaccinium... 4 viridis lutosus. 30 1 53 viridis Nerodia valida valida valida. Nerodia 240 240 232 237 237 viridis Coleonyx variegatus fasciatus, Coleonyx variolosus, Hypopachus varius, Sauromalus variegatus, velutina, 1 1 66 169 Elaphe W 9 257, 261-265 venulosa, Phrynohyas venustissimus, Scaphiodontophis 229 185-192 Vermicella annular a 173, 177 173 warro, Simoselaps wilcoxi, Tanti/la wilcoxi wilcoxi wilcoxi, Tantilla 173, 174, 176, 177, 180 willardi mendionalis, Crotalus woodhousei, Bufo 240 240 241 126, 128 46, 52, 111 41, 109, 110 vermis, Carphophis Opheodrys viminea, Quercus vernalis, Vipera 223 Xenopus 60, 199 laevis 1 27 1 27 199 aspis bents lebentina 58,60,68, 195, 199,200 198 240 43 Thamnophis eques virgatum. Panicum virgatenuis, 1 'irginia yaquia, Tantilla Yucca 232, 240 220 54, 155 52, 148, 149, 153, 155 striatula valeriae virginiana, Ostrya virginianus. 1 236 Eleutherodactylus vulpina, 1 108, 199 Leiobunum vittatum, vocalis, 4 1 89 Quercus ventrimaculatus, Anolis I Crotalus Crotalus oreganus, Crotalus viridis, Crotalus viridis helleri, valeriae, Virginia valida, 200 198, 199 viridis Bubo 153 zeteki, 223 zeteki nothus, Scaphiodontophis 212 Zeugorchis megacystis Scaphiodontophis - 185 190 93 AVAILABLE SPECIAL PUBLICATIONS MUSEUM OF NATURAL HISTORY, UNIVERSITY OF KANSAS 1. Catalogue of publications in herpetology published by the University of Kansas Museum of Natural History. By Linda Trueb. Pp. 1-15. December 1976. 2. Catalogue of publications in mammalogy published by the University of Kansas Museum of Natural History. By Robert S. Hoffmann. Pp. 1-19. 15 $0.25. 3. 5. February 1977. $0.25. Maintenance of rattlesnakes in captivity. By James B. Murphy and Barry L. Armstrong. Pp. 1-40. 29 December 1978. $3.00. The natural history of Mexican rattlesnakes. By Barry L. Armstrong and James 7. 8. 9. B. Murphy. Pp. 1-88. 14 December 1979. $6.00. the Pennsylvanian of Kansas. By Robert R. A diapsid reptile from Reisz. Pp. 1-74. 18 February 1981. $5.00. 1 982 Catalog of publications of the University of Kansas Museum of Natural History. Pp. 1-28. November 1982. $1.00 or free with orders. The ecological impact of man on the South Florida herpetofauna. By Larry David Wilson and Louis Porras. Pp. 1-89. 8 August 1983. $7.00. Date Due QL640.5 \ crtcbrati Harvard V48 1984 ccologi MCZ and systematic Library \KI5464 I I II II 2044 062 369 020'

Log In

Vertebrate ecology and systematics: a tribute to Henry S. Fitch

Vertebrate ecology and systematics: a tribute to Henry S. Fitch

Vertebrate ecology and systematics: a tribute to Henry S. Fitch

Vertebrate ecology and systematics: a tribute to Henry S. Fitch

Vertebrate ecology and systematics: a tribute to Henry S. Fitch

Related Papers

RELATED PAPERS