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 E. Hunt James L. 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 Secretary, Museum of Natural History, University of Kansas, Law- rence, Kansas 66045. HARVARD UNIVERSITY H Library of the Museum of Comparative Zoology Front cover: The head ofan adult Osage Copperhead (Agkistrodon contortrixphaeo- gaster) from Douglas County, Kansas. Drawing © 1984 by Linda Dryden. 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 E. Hunt James L. Knight Luis Malaret Nancy L. Zuschlag Museum ofNatural History Department of Systematics and Ecology The University ofKansas Laurence. Kansas 66045 University of Kansas Laurence 1984 University of Kansas Publications Museum of Natural History Editor: Joseph T. Collins MU5, COMR ZOOL LIBRARY JUL U HARVARD UNIVERSITV Special Publication No. 10 pp. i-viii; 1-278; 79 figures 86 tables; 2 appendices Published 21 June 1984 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. INTRODUCTION Henry S. Fitch in Perspective William E. Duellman 3 The Published Contributions of Henry S. Fitch Virginia R. Fitch 5 PART II. REPRODUCTIVE BIOLOGY AND POPULATION DYNAMICS Growth, Reproduction and Demography of the Racer, Coluber constrictor mormon, in Northern Utah William S. Brown and William S. Parker 13 Growth of Bullsnakes {Pituophis melanoleucus sayi) on a Sand Prairie in South Central Kansas Dwight R. Piatt 41 Communal Denning 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 <• 89 Female Reproduction in an Arkansas Population of Rough Green Snakes (Opheodrys aes- tivus) Michael V. Plummer 105 Clutch Size in Iguana iguana in Central Panama A. Stanley Rand 115 Are Anuran Amphibians Heavy Metal Accumulators? Russell J. Hall and Bernard M. Mulhern 123 PART III. 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 (Ser- pentes, Elapidae) Richard Shine 173 Scaphwdontophis (Serpentes, Colubridae): Natural History and Test of a Mimicry-Related Hypothesis Robert W. Henderson 185 Dominance in Snakes Charles C. Carpenter 195 An Experimental Study of Variation in Habitat Selection and Occurrence of the Deermouse, Peromyscus maniculatus gracilis John H. Fitch 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 Anolis fttchi, a New Species of the Anolis aequatorialis Group from Ecuador and Colombia Ernest E. Williams and William E. Duellman 257 INDEX TO SCIENTIFIC NAMES 267 IV Preface This volume is the result of a symposium en- titled. "Perspectives in Fitchian Ecology," held on 9 August 1 980 in conjunction with the annual meetings of the Society for the Study of Am- 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 after 32 years with the Department of System- atics and Ecology at the University of Kansas. Sixteen papers were presented in two sessions during the symposium and. aside from a few additions, the organizational format of this vol- ume closely follows that of the symposium. Manuscripts were submitted and accepted in late 1980 and 1981. but authors were given an op- portunity to update their contributions in early 1 983. In organizing the symposium we were sur- prised by the breadth of research conducted by the participants. Because of Fitch's influence on his past and present students and colleagues, this volume is not restricted to herpetological con- tributions. Thus, the topical emphasis of this vol- ume reflects Fitch's own research interests. The following is a breakdown by subject of the papers contained in this volume versus Fitch's pub- lished papers: ecology (this volume: 78%, Fitch: (73%); systematics and biogeography (17% vs. 19%): conservation (5% vs. 5%); and by taxo- nomic emphasis: squamates (this volume: 73%; Fitch: 62%); other amphibians and reptiles (14% vs. 7%); other vertebrates (13% vs. 23%). We wish to thank Max A. Nickerson of the Milwaukee Public Museum and Al Williams of the University ofWisconsin-Milwaukee and their respective staffs for logistical support in arrang- ing and conducting the symposium. A special note of thanks is extended to Virginia Fitch and other members of the Fitch family for assistance in the development of the symposium. The or- ganizational advice and encouragement of Wil- liam E. Duellman, Curator. Division of Herpe- tology; Philip S. Humphrey, Director, Museum of Natural History; and Richard F. Johnston. Chairman. Department of Systematics and Ecol- ogy, is greatly appreciated. Joseph T. Collins, Editor. Museum Publica- tions, deserves special recognition for his helpful advice and continued patience in answering our many questions concerning the development and execution of the symposium and this volume. The cheerful and patient assistance of Rose Etta Kurtz was invaluable. Finally, we are most grateful to the following persons for reviewing the manuscripts appearing in this volume: Robert D. Aldridge. Stevan J. Arnold, Reeve M. Bailey. Royce E. Ballinger. Thomas J. Berger. William S. Brown. Gordon M. Burghardt, Janalee P. Caldwell, Jonathan A. Campbell, David C. Cannatella. David K. Chisz- ar, Martha L. Crump, Arthur E. Dunham. Don- ald G. Dunlap. Henry S. Fitch. Darrell Frost. J. Whitfield Gibbons. Peter Gray, Harry W. Greene. Wendy Gorman. Harold Heatwole. James E. Huheey, John B. Iverson, Keith V. Kardong. Pe- ter Klopfer, Carl Lieb, Harvey B. Lillywhite. John D. Lynch. Richard Mayden. Roy W. Mc- Diarmid, 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, Vir- ginia Polytechnic Institute and State Univer- sity, 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, Museum of Natural His- tory, University of Kansas. Lawrence, Kansas 66045 Alice F. Echelle. School of Biological Sciences, Oklahoma State University, Stillwater, Okla- homa 74078 Anthony A. Echelle, School of Biological Sci- ences, Oklahoma State University, Stillwater, Oklahoma 74078 John H. Fitch, Massachusetts Audubon Society, Lincoln, Massachusetts 01773 Virginia R. Fitch, University of Kansas Natural History Reservation, Lawrence, Kansas 66044 Harry W. Greene, Museum of Vertebrate Zo- ology, University of California, Berkeley, Cal- ifornia 94720 Patrick T. Gregory, Department of Biology, University of Victoria, Victoria, British Co- lumbia, V8W 2Y2 Russell J. Hall, U.S. Fish and Wildlife Service, Patuxent Wildlife Research Center, Laurel, Maryland 20811 Robert V\ . Henderson. Section of Vertebrate Zo- ology, Milwaukee Public Museum, Milwau- kee, Wisconsin 53233 Clark Hubbs, Department of Zoology, Univer- sity of Texas, Austin. Texas 78712 Bernard M. Mulhern. U.S. Fish and Wildlife Service, Patuxent Wildlife Research Center. Laurel, Maryland 208 1 1 William S. Parker, Department of Biological 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 L. Rakestraw, Museum of Natural His- tory, University of Kansas, Lawrence. Kansas 66045 A. Stanley Rand. Smithsonian Tropical Re- search Institute, APO Miami. Florida 34002 Richard A. Seigel, Savannah River Ecology Lab- oratory, 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 Sci- ences, University of Texas, El Paso. Texas 79968 Ernest E. Williams, Museum of Comparative Zoology, Harvard University, Cambridge. Massachusetts 02138 ft i, . --.. i. c XS a t_ 00 O oX a 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 a o X 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 I it< h Edited b> R. A. Seigcl. L. E. Hunt. J I Knight. I Malaret and N. L. Zuschlag i I "t>84 Museum of Natural History, The University ot Kansas. Lawrence Henry S. Fitch in Perspective William E. Duellman Who is Henry Sheldon Fitch? This quiet, mod- est, unassuming man made his first entrance into the world of biologists by publishing on Oregon birds in the Condor in 1933. Yes, birds! Most of us think of Henry S. Fitch as a herpetologist. Yet, of his 1 50 published papers in the past 50 years, only about two-thirds of them deal with am- phibians 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 ecol- ogist, but 25 of his papers are on systematics and include his classic work on alligator lizards pub- lished in 1934 and his highly perceptive study of western garter snakes published in 1940 (doc- toral 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 history of reptiles. From his earliest papers on reptiles, he has provided extensive field obser- vations. In 1 948, he entered a "naturalist's heav- 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 densities, movements, food, growth rates, hiber- nation, and reproduction— all substantiated with massive quantities of data. Through his efforts this square mile is better known herpetologically than any other in the world. His studies on the natural history of reptiles are classics. Outstanding examples are the thor- ough study of the five-lined skink ( 1954) and the exhaustive study of the copperhead ( 1 960). More recently he has worked on the interactions of behavior and ecology, communities of anoles. and populations and conservation of iguanas. 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 rep- tiles (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- erature research are reflected in his syntheses. These traits combined with dogged determina- tion to learn all there is to know about his sub- jects of study, his continued productivity, and 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 com- monly are narrow specialists. Henry Fitch doesn't fit into a modern pigeon hole. He is a naturalist in the broadest sense of the word. His breadth of knowledge is matched by very few of his con- temporaries and scarcely imagined by most of his younger colleagues. An analogy can be drawn 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 to find them sometimes morassed or abandoned 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 com- monly 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- fully guided the research of many graduate stu- dents, and has collaborated with a diversity of colleagues. A major factor in his remarkable and successful career has been a collaborator, assis- tant, caretaker and charming lady— Virginia R. Fitch. Few 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 history is well worth emulating. Our knowledge of animals in nature would be far greater if there were many more biologists in the world who followed in the footsteps of Henry S. Fitch. Vertebrate Ecolog> and Systematics—A Tnbulc to Henr\ S I itch Edited by R. A. Seigcl. L. E. Hunt. J. I. Knight. L. Malarel and N. L, Zuschlag < 1984 Museum of Natural Hislorj I he 1 Iniversity of Kansas. Lawrence The Published Contributions of Henry S. Fitch Virginia R. Fitch Beginning with his first published paper in 1933. the writings of Henry S. Fitch have en- compassed a wide range of subjects and disci- plines, 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- ductive biology. To date, he has produced 150 papers, all of which appear in the following list. Fitch's published works include as their subjects mammals (19 papers), birds (12), vertebrates in general (5), spiders (3). vegetation and habitats (4), and mollusks ( 1 ), as well as five book reviews, 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 the impressive extent of the knowledge and breadth of interest of Henry S. Fitch. 1933. Bird notes from southwestern Oregon. Condor. 35:167-168 (with J. O. Steven- son). 1934. New alligator lizards from the Pacific Coast. Copeia. 1934:6-7. 1934. A shift of specific names in the genus Ger- rhonotus. Copeia, 1934:172-173. 1935. An abnormal pattern in a gopher snake. Copeia, 1935:144-146. 1935. Natural history of the alligator lizards. Trans. Acad. Sci. St. Louis. 29:1-38. 1936. Amphibians and reptiles of the Rogue River Basin, Oregon. Amer. Midi. Nat.. 17:634-652. 1938. Ranaboylii in Oregon. Copeia, 1938:148. 1938. An older name for Triturus similans Twitty. Copeia, 1938:148-149. 1938. A systematic account of the alligator liz- ards ( Gerrhonotus) in the western United States and lower California. Amer. Midi. Nat., 20:381-424. 1939. Desert reptiles in Lassen County. Cali- fornia. Herpetologica. 1:151-152. 1939. Leptodeira in northern California. Her- petologica. 1:152-153. 1940. A biogeographical study of the Ordi- noides artenkreis of garter snakes (genus Thamnophis). Univ. California Publ. Zool.. 44:1-150. 1940. A field study of the growth and behavior of the fence lizard. Univ. California Publ. Zool., 44:151-172. 1940. Some observations on horned owl nests. Condor. 42:73-75. 1941. The feeding habits of California garter snakes. California Fish and Game, 27:2- 32. 1941. Geographic variation in garter snakes of the species Thamnophis sirtalis in the Pa- cific Coast region ofNorth America. Amer. Mid!. Nat.. 26:570-592. 1942. Interrelations of rodents and other wild- life of the range. Univ. California Agr. Exp. Sta. Bull.. 663:96-129 (with E. E. Horn). 1946. Observations on Cooper's hawk nesting and predation. California Fish and Game. 32:144-154 (with B. Glading and V. 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 hawk. Condor. 48:205-237 (with F. Swenson and D. F. Tillotson). 1946. Trapping the California ground squirrel. Jour. Mammal.. 27:220-224 (with E. E. Horn). 1947. The California Ground Squirrel by J. M. Linsdale (Book review). Jour. Mamm.. 28: 191-192. 1947. A field study of a rattlesnake population. California Fish and Game. 33:103-123 (with B. Glading). 1947. Variation in the skinks (Reptilia: Lacer- tilia)oftheSkiltonianusgroup. Univ. Cal- ifornia Publ. Zool.. 48:169-220 (with T. L. Rodgers). 1947. Predation by owls in the Sierran foothills of California. Condor. 49:137-151. 1947. Ecology of a cottontail rabbit (Sylvilagus auduboni) population in central Califor- nia. California Fish and Game. 33:159- 184. 1947. Rattlesnakes on the range. Pacific Stock- man. 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 squirrels mean destroyed forage. SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Western Livestock Jour.. Oct.:37, 109, 1955. 1 10. 1 12. 1948. Further remarks concerning Thamnophis ordinoidcs and its relatives. Copeia, 1948: 1955. 121-126. 1 948. Habits and economic relationships of the Tulare kangaroo rat. Jour. Mamm., 29:5- 1955. 35. 1948. Ecology of the California ground squirrel on grazing lands. Amer. Midi. Nat., 39: 1956. 513-596. 1948. A study of coyote relationships on cattle range. Jour. Wildlife Management, 12:73- 1956. 78. 1949. Sparrow adopts kingbirds. Auk, 66:368- 369. 1956. 1949. Outline for ecological life history studies of reptiles. Ecology, 30:520-532. 1956. 1949. Use of California annual-plant forage by range rodents. Ecology, 30:306-321 (with J. R. Bentley). 1949. Study of snake populations in central Cal- 1956. ifornia. Amer. Midi. Nat., 41:513-579. 1949. Road counts of snakes in western Loui- siana. Herpetologica, 5:87-90. 1950. A new style live-trap for small mammals. 1956. Jour. Mamm., 31:364-365. 1951. Remarks concerning the systematics of the collared lizard (Crotaphytus collaris), with a description of a new subspecies. Trans. 1956. Kansas Acad. Sci., 54:548-559 (with W. Tanner). 1951. A simplified type of funnel trap for rep- tiles. Herpetologica, 7:77-80. 19 56. 1952. The armadillo in the southeastern United States. Jour. Mamm., 33:21-37 (with P. Goodrum and C. Newman). 1952. The University of Kansas Natural His- 1957. tory Reservation. Univ. Kansas Mus. Nat. Hist. Misc. Publ., no. 4:1-38. 1952. (Book review) Ecological Animal Geog- raphy by Hesse, Allee and Schmidt. Wil- 1957. son Bull., 64. 1953. Ecology of the opossum on a natural area in northeastern Kansas. Univ. Kansas 1958. Publ., Mus. Nat. Hist., 7:309-338 (with L. L. Sandidge). 1953. (Book review) Natural Communities by 1958. L. R. Dice. Wilson Bull.. 65:121-123. 1954. Seasonal acceptance of bait by small mammals. Jour. Mamm., 35:39-47. 1954. Life history and ecology of the five-lined 1959. skink, Eumeces fasciatus. Univ. Kansas Publ., Mus. Nat. Hist., 8:1-156. 1959. Habits and adaptations of the Great Plains skink (Eumeces obsoletus). Ecol. Mono- gr., 25:59-83. Observations on the summer tanager in northeastern Kansas. Wilson Bull., 67:45- 54 (with V. R. Fitch). The coyote on a natural area in north- eastern Kansas. Trans. Kansas Acad. Sci., 58:211-221 (with R. L. Packard). A field study of the Kansas ant-eating frog, Gastrophryne olivacea. Univ. Kansas Publ., Mus. Nat. Hist., 8:275-306. An ecological study of the collared lizard (Crotaphytus collaris). Univ. Kansas Publ., Mus. Nat. Hist., 8:213-274. A ten-year old skink? Herpetologica, 12: 328. Early sexual maturity and longevity under natural conditions in the Great Plains nar- row-mouthed frog. Herpetologica, 12: 281-282. Temperature responses in free living am- phibians and reptiles of northeastern Kansas. Univ. Kansas Publ., Mus. Nat. Hist., 8:417-476. The forest habitat of the University of Kansas Natural History Reservation. Univ. Kansas Publ., Mus. Nat. Hist., 10: 77-127 (with R. L. McGregor). The molluscan record ofsuccession on the University of Kansas Natural History Re- servation. Trans. Kansas Acad. Sci., 59: 442-454 (with D. H. Lokke). Ecological observations on the woodrat, Neotoma jloridana. Univ. Kansas Publ., Mus. Nat. Hist., 8:499-533 (with D. G. Rainey). Aspects ofreproduction and development in the prairie vole (Microtus ochrogaster). Univ. Kansas Publ., Mus. Nat. Hist., 10: 129-161. Observations on hibernation and nests of the collared lizard. Crotaphytus collaris. Copeia, no. 4:305-307 (with J. M. Legler). Natural history of the six-lined racerun- ner (Cnemidophorus sexlineatus). Univ. Kansas Publ., Mus. Nat. Hist., 1 1:1 1-62. Home ranges, territories, and seasonal movements of vertebrates of the Natural History Reservation. Univ. Kansas Publ., Mus. Nat. Hist., 11:63-326. A patternless phase of the copperhead. Herpetologica, 15:21-24. Aspects of needed research on North VERTEBRATE ECOLOGY AND SYSTEMATICA American grasslands. Trans. Kansas Acad. 1 967. Sci.. 62:175-183 (with 5 other authors). 1 960. Criteria for determining sex and breeding maturity in snakes. Herpetologica, 16:49— 51. 1968. 1960. Autecology of the copperhead. Univ. Kansas Publ., Mus. Nat. Hist., 1 3:85-288. 1 969. 1960. (Book review). The Rusty Lizard, a Pop- ulation Study by W. Frank Blair. Copeia, 1960:386-387. 1961. Occurrence of the garter snake Thamno- phis sirtalis in the Great Plains and Rocky Mountains. Univ. Kansas Publ.. Mus. Nat. Hist., 13:289-308 (with T. P. Maslin). 1970. 196 1 . An older name for Thamnophis cyrtopsis (Kennicott). Copeia, 1961:112 (with W. W. Milstead). 1961. The snake as a source of living sperma- tozoa in the laboratory. Turtox News, 39: 247. 1961. Longevity and age-size groups in some 1970. common snakes. Pp. 396-414 in Verte- brate Speciation: A University of Texas symposium. Univ. Texas Press. 1970. 1963. Natural history of the racer Coluber con- strictor. Univ. Kansas Publ.. Mus. Nat. Hist.. 15:351-468. 1963. Observations on the Mississippi kite in southwestern Kansas. Univ. Kansas Publ.. 1 970. Mus. Nat. Hist.. 12:503-519. 1963. Natural history of the black rat snake (Elaphe o. obsoleta) in Kansas. Copeia. 1970. 1963:649-658. 1963. Spiders of the University of Kansas Nat- ural History' Reservation and Rockefeller Experimental Tract. Univ. Kansas Mus. 1971. Nat. Hist., Misc. Publ., no. 38:1-202. 1965. The University of Kansas Natural His- tory Reservation in 1965. Univ. Kansas Mus. Nat. Hist., Misc. Publ.. no. 42:1- 1971. 60. 1965. An ecological study of the garter snake, Thamnophis sirtalis. Univ. Kansas Publ.. 1971. Mus. Nat. Hist., 15:493-564. 1965. Breeding cycle in the ground skink, Ly- gosoma laterale. Univ. Kansas Publ., Mus. 1971. Nat. Hist., 15:565-575 (with H. W. Greene). 1966. Spiders from Meade County, Kansas. Trans. Kansas Acad. Sci.. 69:1 1-22 (with 1971. V. R. Fitch). 1967. Preliminary experiments on physical tol- erances of the eggs of lizards and snakes. 1971. Ecology, 48:160-165 (with A. V. Fitch). Ecological studies of lizards on the Uni- versity of Kansas Natural History Reser- vation. Pp. 30-44 in Lizard Ecology, a symposium. Univ. Missouri Press. Temperature and behavior ofsome equa- torial lizards. Herpetologica 24:35-38. Biotelemetric studies of small vertebrate behavior. Pp. 44-45 in Engineering Re- search, Center for Research, Inc.. Engi- neering Sci. Div., Univ. Kansas. Vol. Ill (1967-1968), E. D. Bevan, Ed. (with H. W. Shirer. W. K. Legler and D. D. Pip- pin.) Data acquisition systems for the study of vertebrate ecology. Pp. 19-20 in Re- search, Vol. IV, The Univ. Kansas, Cen- ter for Research, Inc.. P. McMillan. Ed. (with H. W. Shirer, K. Armitage. W. K. Legler, D. D. Pippitt. J. D. Pauley and J. F. Downhower.) Reproductive cycles in lizards and snakes. Univ. Kansas Mus. Nat. Hist., Misc. Publ., 52:1-247. Comparison from radiotracking of move- ments and denning habits of the raccoon, striped skunk, and opossum in north- eastern Kansas. Jour. Mamm., 5 1(3):49 1- 503 (with H. W. Shirer). A radiotelemetric study of spatial rela- tionships in the opossum. Amer. Midi. Nat. 84(1): 170-1 86 (with H. W. Shirer). Natural history of the milk snake (Lani- propeltis triangulum) in northeastern Kansas. Herpetologica 26(4):387-396 (with R. R. Fleet). Ecological notes on some common lizards of southern Mexico and Central America. The Southwestern Naturalist, 15:398-399 (with A. V. Fitch and C. W. Fitch). A radiotelemetric study of spatial rela- tionships in some common snakes. Co- peia. 1971:118-128 (with H. W. Shirer). (Book review). A Complete Field Guide to Nests in the United States by R. Head- strom. Jour. Wildlife Mgt., 35:188-189. A comparative analysis of aggressive dis- play in nine species ofCosta Rican Anolis. Herpetologica, 27:271-288 (with A. A. Echelle and A. F. Echelle). A new anole from Costa Rica. Herpeto- logica. 27:354-362 (with A. A. Echelle and A. F. Echelle). Further observations on the demography of the Great Plains skink {Eumeces ob- SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY soletus). Trans. Kansas Acad. Sci.. 74:93- 98 (with R. J. Hall). 1972. Radio tracking of wild animals in their natural habitat. Pp. 35-36 in Research. Univ. Kansas Center for Research, Inc.. 1975. Vol. V, P. Nicholas, Ed. (with R. Hoff- mann, H. W. Shirer. L. A. Gold. R. L. Lattis, C. B. Rideout, and R. C. Waltner). 1972. Ecology ofAnolis tropidolepis in Costa Ri- 1975. can cloud forest. Herpetologica, 28( 1 ): 1 0- 21. 1972. Variation in the Central American igua- 1976. nid lizard, Anolis cupreus, with the de- scription of a new subspecies. Occas. Pa- pers. Mus. Nat. Hist., Univ. Kansas, no. 8:1-20 (with A. A. Echelle and A. F. 1976. Echelle). 1972. Observations offish-eating and mainte- nance behavior in two species of Basilis- 1 976. ens. Copeia, 1972:387-389 (with A. A. Echelle and A. F. Echelle). 1973. Observations on the population ecology 1976. of the Central American iguanid lizard, Anolis cupreus. Caribbean Jour. Sci., 13(3- 4):215-229. 1973. Population structure and survivorship in 1976. some Costa Rican lizards. Occas. Papers, Mus. Nat. Hist., Univ. Kansas, no. 18:1- 41. 1973. A field study ofCosta Rican lizards. Univ. 1976. Kansas Sci. Bull., 50(2):39-126. 1973. A new anole (Reptiha: Iguanidae) from southern Veracruz, Mexico. Jour. Herp. 1977. 7(2): 125-1 28 (with R. W. Henderson). 1973. Road counts of hawks in Kansas. Kansas Ornith. Soc. Bull., 24(4):33-35 (with H. A. Stephens and R. O. Bare). 1977. 1973. Yellow-billed cuckoo nesting at Univer- sity of Kansas Natural History Reserva- tion. Kansas Ornith. Soc. Bull., 24(2): 1 2- 15 (with P. von Achen). 1977. 1974. Observations on the food and nesting of the broad-winged hawk (Buteo platypte- rus) in northeastern Kansas. Condor. 76(3):331-333. 1974. Food habits of Basiliscus basiliscus in 1977. Costa Rica. Jour. Herp.. 8(3):260-262 (with R. R. Fleet). 1 975. A preliminary ecological study of the soft- shelled turtle, Trionyx muticus, in the 1978. Kansas River. Israel Jour. Zoology, 24: 28-42 (with M. V. Plummer). 1975. A comparative study of the structural and climatic habitats of Anolis sericeus (Rep- 1978. tilia: Iguanidae) and its syntopic conge- ners at four localities in southern Mexico. Herpetologica, 31:459-471 (with R. W. Henderson). A demographic study ofthe ringneck snake (Diadophis punctatus) in Kansas. Univ. Kansas Mus. Nat. Hist., Misc. Publ.. 62: 1-53. Sympatry and interrelationships in Costa Rican anoles. Occas. Papers, Univ. Kan- sas Mus. Nat. Hist., 40:1-60. A field study of the rock anoles (Reptilia, Lacertilia, Iguanidae) ofsouthern Mexico. Jour. Herp., 10:303-311 (with R. W. Henderson). A new pholcid spider from northeastern Kansas. Bull. Kansas Ent. Soc. (with E. O. Maughan). Sexual size differences in the mainland anoles. Occas. Papers. Univ. Kansas Mus. Nat. Hist., no. 50:1-21. Field observations on rare or little known mainland anoles. Univ. Kansas Sci. Bull.. 51:91-128 (with A. F. Echelle and A. A. Echelle). A new anole (Reptilia: Iguanidae) from Great Corn Island. Caribbean Nicaragua. Contr. Biol. & Geol., Milwaukee Pub. Mus., no. 9:1-8 (with R. W. Henderson). Dragons for dinner. Wildlife Omnnibus, International Wildlife, 6(6): 1 7 (with R. W. Henderson). Age and sex differences in the ctenosaur. {Ctenosaura similis). Contr. Biol. & Geol., Milwaukee Pub. Mus., no. 1 1:1-1 1 (with R. W. Henderson). Age and sex differences, reproduction and conservation of Iguana iguana. Contr. Biol. & Geol., Milwaukee Pub. Mus., 13: 1-21 (with R. W. Henderson). Spatial relations and seasonality in the skinks, Eumeces fasciatus and Scincella laterale in northeastern Kansas. Herpe- tologica, 33:303-313 (with P. von Ach- en). Structure, movements and reproduction in three Costa Rican bat communites. Oc- cas. Papers Mus. Nat. Hist., Univ. Kan- sas, 69:1-28 (with R. K. LaVal). Inter- and intraspecific allometry in a dis- play organ: the dewlap of Anolis (Igua- nidae) species. Copeia, 1978(2):245-250 (with A. F. Echelle and A. A. Echelle). Behavioral evidence for species status of VERTEBRATE ECOLOGY AND SYSTEMATICS Anolis uniformis (Cope). Hcrpetologica 34(2):205-207 (with A. F. Echcllc and A. A. Echelle). 1978. Dragons: 25c/lb. Animal Kingdom, Feb./ March: 12-17 (with R. W. Henderson). 1978. A field study of the red-tailed hawk 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- loporus. Univ. Kansas Sci. Bull., 51(13): 441-461. 1978. Ecology and exploitation of Ctenosaura similis. Univ. Kansas Sci. Bull.. 51(15): 483-500 (with R. W. Henderson). 1 978. Two new anoles (Reptilia: Iguanidae) from Oaxaca, with comments on other Mexi- can species. Contr. Biol. & Geol.. Mil- waukee Pub. Mus.. 20:1-15. 1978. A field study of the prairie kingsnake (Lampropeltis calligaster). Trans. Kansas Acad. Sci.. 81:353-363. 1978. The plight of the iguana. LORE, Milwau- kee Pub. Mus., 28(3):2-9 (with R. W. Henderson). 1978. A 20-year record of succession on reseed- ed fields of tallgrass prairie on the Rocke- feller Experimental Tract. Univ. Kansas Mus. Nat. Hist.. Spec. Publ., 4: 1-1 5 (with E. R. Hall). 1979. Notes on the behavior and ecology of Ctenosaura similis (Reptilia: Iguanidae) at Belize City, Belize. Brenesia, 16:69-80 (with R. W. Henderson). 1980. Remarks concerning certain western gar- ter snakes of the Thamnophis elegans complex. Trans. Kansas Acad. Sci.. 83: 106-113. 1 980. Reproductive strategies of reptiles. Pp. 25- 31 in Reproductive Biology and Diseases of Captive Reptiles (J. B. Murphy and J. T. Collins, eds.), SSAR Cont. Herpetol.. 1:1-277. 1981. Sexual size differences in reptiles. Univ. Kansas Mus. Nat. Hist. Misc. Publ.. 70: 1-72. 1981. Coluber mormon, a species distinct from C. constrictor. Trans. Kansas Acad. Sci. 84:196-203 (with W. S. Brown and W. S. Parker). 1981. Thamnophis sirtalis. Cat. American Amph. Rept.. 270.1-270.4. 1983. 1983. 1982. Reproductive cycles in tropical reptiles. Occas. Papers. Mus. Nat. Hist.. Univ. of Kansas 96:1-53. 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. 13. 1983. Exploitation of iguanas in Central Amer- ica. Pp. 397^ 1 7 in Iguanas of the World: Their Behavior, Ecology, and Evolution (G. M. Burghardt and A. S. Rand, eds.) Noyes Press (with R. W. Henderson and D. M. Hilhs). Thamnophis elegans. Cat. American Amph. Rept., 320.1-320.4. Ctenosaura similis (Garrobo. Iguana Ne- gra. Ctenosaur). Pp. 394-396 in Costa Ri- can Natural History. (D. H. Janzen, ed.) Univ. Chicago Press (with J. Hackforth- Jones). 1983. Sphenomorphus cherriei (Escincela Par- da, Skink). Pp. 422-425 in Costa Rican Natural History. (D. H. Janzen, ed.) Univ. Chicago Press. 1983. Ecological succession in vegetation and small mammal population on a natural area of northeastern Kansas. Proc. Sev- enth North American Prairie Conf., Au- gust 1 980. Pp. 117-121 (with W. D. Ket- tle). In press. The Anolis dewlap: Interspecific vari- ability and morphological associations with habitat. Copeia (with D. M. Hillis). In press. Succession in small mammals on a nat- ural area in northeastern Kansas. Occas. Papers, Mus. Nat. Hist., Univ. Kansas (with V. R. Fitch and W. D. Kettle). In press. Geographic variation in clutch size and litter size in North American reptiles. Univ. Kansas Mus. Nat. Hist. Misc. Publ. In press. Ecological patterns of relative clutch mass in snakes. Oecologia (with R. A. Sei- gel). In press. Thamnophis couchi. Cat. American Amph. Rept. In press. Intergradation of Osage and broad- banded 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 S. Parker Introduction Considerable interest has developed recently in comparative life history and demographic studies because the data point up a number of evolutionary strategies taken by separate inter- and intraspecific populations. To date, the data have been effective mostly in illustrating the se- lection and adaptive basis for the life histories of lizards, birds, and mammals among the ver- tebrates (Stearns 1976; Hutchinson 1978). Rare- ly have data on snakes perfused the general lit- erature even though a number of sound field studies of snake populations have been com- pleted (Blanchard et al. 1979; Branson and Baker 1974; Brown 1973; Carpenter 1952; Clark 1970, 1974; Clark and Fleet 1976; Feaver 1977; Fitch 1949, 1960, 1963, 1965, 1975; Gregory 1977; Hall 1969; Parker and Brown 1974, 1980; Piatt 1969; Prestt 1971; Spellerberg and Phelps 1977; Stewart 1968; Tinkle 1957, 1960; Viitanen 1967). Coluber constrictor (Serpentes, Colubridae) is known to occur from Guatemala to southern Canada (Conant 1975; Etheridge 1952; Stebbins 1966). The species is polytypic, with 10 de- scribed subspecies (Wilson 1970). In the United States, eight of the nine recognized geographic races occur east of the Rocky Mountains (Auf- fenberg 1955; Fitch 1963; Wilson 1970). C. c. mormon occurs west of the Continental Divide. This subspecies has been recorded in most of the states in the western third ofthe U.S. (Auffenberg 1955; Stebbins 1966; Wilson 1978:218.1). An extensive range hiatus in the Rocky Mountains, lack of intergradation, and differences in mor- phology and ecology between the midwestern subspecies, C. c. flaviventris, and C. c. mormon may warrant elevation of the latter taxon to species rank (Fitch, Brown and Parker 1981). Aside from several brief reports on various as- pects of the biology of C. c. mormon (see review of the literature in Fitch 1963), no comprehen- sive ecological study ofthis wide-ranging western form has been conducted. Fitch's (1963) study in Kansas of C. c. flaviventris is, to date, the most extensive ecological investigation of any popu- lation of Coluber constrictor. The present study focuses on the biology of the Western yellow-bellied racer. Coluber con- strictor mormon Baird and Girard, hereinafter called simply "racer." Our approach has been empirical and autecological and has concentrat- ed on one large population of this snake at a single locality in northern Utah over a four-year period. This paper treats growth, maturity, re- production, population structure, and demog- raphy of the racer. Widespread and abundant in North America, 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. Methods Snakes were captured in autumn 1 969 through spring 1973 at a communal denning area in a desert shrub habitat located 4 km W of Grants- ville, Tooele County, Utah (40°36'N, 1 12°32'W, elevation 1580 m), ca. 58 km WSW of Salt Lake City. This area is our primary study locality (area M) where all long-term mark and recapture field work was conducted. We recorded a total of 1 694 captures of 1046 racers at this site. Originally studied by Woodbury and his co- workers in the 1940's (Woodbury et al., 1951), the "main den" (den M) was later sampled in the mid-1960's by Hirth and King (1968) and again in the early 1970's by us. We discovered other actively used dens near den M; these were considered part of a discrete group which we called "M complex." A separate series of newly- discovered dens located 0.8 km to the south was designated "S complex" (Parker and Brown 1 973; Brown and Parker 1976a). The technique we used to capture snakes was to intercept them with a screen wire fence erected aound their hibernaculum. As the dens were sin- 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 13 14 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY covering it with soil from a perimeter trench. Captures occurred almost daily in favorable weather as snakes attempted to enter a den in autumn and leave it in spring. The chronology of sampling Coluber and other snakes at area M is summarized in Parker and Brown ( 1 980). Our results pertain to the four-year period 1 969- 1 972 by sampling dens each autumn and spring from autumn 1969 through spring 1973. Data pre- sented for a given calendar year were derived from sampling in the autumn of that year (den M only) and the spring of the next (den M and other dens). Individuals were processed in the laboratory and most were released within 24 h after capture. Each snake was permanently marked by clipping ventral scutes (Brown and Parker 1 976b). Snout- vent length (SVL = distance from tip of snout to posterior edge of anal scute) and tail length to the nearest 0.5 cm (snakes > 1 year old) or to the nearest mm (hatchlings and juveniles) and live body weight to the nearest 0. 1 g (all snakes) were recorded for each individual at all captures. Re- productive condition of males was determined by obtaining cloacal smears and examining them microscopically for the presence ofspermatozoa. 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 repro- ductive data. These localities are designated as area SLC, vicinity of Salt Lake City, Salt Lake 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 area RB and provided data on body weight changes. Some other females killed for exami- nation of reproductive tracts were from area M. These snakes included several casualties from our marked population and a few others taken > 2 km from the study dens and beyond the max- imum dispersal limits of racers from area M. In most years at area M hatching occurred around mid-August. Juveniles normally arrived at den M in early October at an average age of 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 and spring, snakes placed in a designated year class differed from their actual age by about 7 months. No growth occurred during hibernation so we assigned an equivalent age to autumn and spring-captured snakes as follows: hatchling (age 0),juvenile(1.5and8.5 months), 1 -year-old (13.5 and 20.5 months), 2-year-old (25.5 and 32.5 months), and so on. The simpler age designation in years corresponds to the number of full 5- month growing seasons which a snake had been through and facilitates analysis of age-specific aspects of the life history. Racers > 1 year old were sexed visually by the relatively thick (males) or thin (females) tail base. Juveniles lacked external sexual differences and those in 1972-1973 were sexed using a blunt probe to detect presence (males) or absence (fe- males) of hemipenial sacs. In earlier sample pe- riods juveniles were not sexed and numbers of male and female juveniles were apportioned as- suming a 1:1 sex ratio. Some of these juveniles were sexed later by recapturing them as marked 1 -year-olds after they had attained sufficient dis- cernible sexual dimorphism as yearlings. 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 ini- tially as juveniles. Snakes in all sampling periods that compared closely to these values were as- signed as 1 -year-olds. Records for these initial juveniles and 1 -year-olds that were later recap- tured were then used to determine preliminary length and weight characteristics for 2- and 3-year- olds. 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 in- dividuals were aged through 6 years and a few through 7 years. Lacking prior captures made some error possible in assigning ages of 4 and 5 years to snakes early in the study, but our method of comparing sizes to known-age statistical val- ues was consistent and uniformly applied over all ages. We tended to be conservative in cases involving a size intermediate between two ages, e.g., if the snake was between the two- and three- 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 differ significantly from year to year so both the spring-to-spring and autumn-to-autumn inter- vals used for determining annual growth rates are considered equivalent. Proportional annual increases or decreases in SVL or weight were calculated as the amount increased or decreased during the year divided by the initial size at the beginning of the year. For example, if a 1 -year- old male increased from 31.9 to 48.9 g (an ab- solute increase of 17.0 g/yr), the proportional increase would be 17.0 -5- 31.9 = 0.533/yr, or 53%. Survival rates were measured over two major periods in the annual cycle of Coluber at area M: (1) the winter period of hibernation and (2) the full year. Like growth rate calculations, annual survival rates include one intervening winter pe- riod and were calculated from spring-to-spring or autumn-to-autumn capture records. Population size estimates based on capture- recapture were calculated using the Jolly-Seber stochastic method following Caughley (1977) and Krebs (1978). Eight censuses at den M provided data for the Jolly-Seber analysis over three years (1970-1972). Snakes recaptured following their movement to a different den ofM complex were included in the tabulations as were den M in- dividuals that were experimentally displaced from that den in autumn 1971 (cf. Brown and Parker 1976a). Thus, bias due to these factors was elim- inated. Population sizes were calculated sepa- rately for juveniles (both sexes combined) and for yearling and older (> 1 year inclusive) males and females. Statistical methods in this paper follow Sokal and Rohlf ( 1 969) and Woolf ( 1 968). Mean values are followed by ± one standard error ofthe mean (SE) with the extremes in parentheses. Results Sexual Dimorphism. — Weights of 73 male and 72 female snakes > 1 year old collected during the autumns of 1969-1972 were regressed on snout-vent lengths (Fig. 1). There was a highly significant (<0.01) greater in males (26.76 ± 0.11%. range 23.9- 28.9%, N = 73) than in females (25.07 ± 0.09%. range 23.1-27.0%. N = 72). Although statisti- cally significant, this distinction could not be used in visual sex determination. Size of Snakes of Known Age.—Snout-vent lengths and weights of 1236 Coluber of known 16 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 1. Sizes of Coluber constrictor mormon of known age, 1969-1972. Ages: H = hatchling (not sexed), J = juvenile, 1-7 = years (see text for method of designating age). Except for hatchlings, all measurements were recorded in spring at emergence from hibernation. Mean ± 1 SE, sample size and extremes in parentheses. VERTEBRATE ECOLOGY AND SYSTEMATICS 17 AGE ( yr ) Fig. 2. Growth in snout-vent length of Coluber constrictor mormon, 1969-1972. Data for hatchlings (H) and juveniles (J) include both 66 and 99 (sexes combined). Except for hatchlings. all records pertain to spring only. Horizontal lines = sample mean; solid rectangles = 95% confidence limits for population mean: open rectangles = ± 1 standard deviation (SD); vertical line = range. Means of 66 connected by dashed line, 99 by solid line. Sample sizes indicated above each bar diagram. involved a decrease in weight during the interval (Table 2). Proportions of weight loss records for 1 78 males (95.5%) and 1 55 females (93.5%) were 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 significantly (/ = 3.5, P < 0.0 1 ) more weight than 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 differ- ences in weight loss in males (absolute F = 1.15, P > 0.05; relative F= 0.30, P > 0.05) and in females (absolute F= 0.66, P > 0.05; relative F= 2.49, P> 0.05). Annual Age-specific Growth. — Absolute and relative yearly rates of increase in snout-vent length and weight are summarized in Tables 3- 6. As no recapture records were available to mea- sure growth in the season of hatching directly, growth calculations were based on differences be- tween means of hatchling (mid-August) and ju- venile (October) sizes. Young racers increased 16.0% in SVL and 36.3% in weight during this 1.5-month interval. Weight increase during the first year was rapid. Males increased an average of225% and females 223% of initial juvenile weights (3.2-fold in- creases). One-year-old females nearly doubled their weight again in their second year (mean proportional increase 82%), achieving a growth rate 1.2 times greater than 1 -year-old males. By 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 18 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 100 80 a< 60 H I O UJ $ 40 20 H -i 1 r- 2 3 4 AGE ( yr ) T" 5 -r- 6 Fig. 3. Growth in weight of Coluber constrictor mormon, 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 relative growth rates in males were less than cor- responding rates in females at all ages. Unequal growth rates between years are in- dicated (Tables 3-6). We compared all age-spe- cific absolute rates of growth recorded in 1970 and 1971 against those in 1972. Five age inter- vals from 1-2 yr to 5-6 yr for each sex were tested. Significant (P < 0.05) between-years dif- ferences appeared in seven of 10 Mests of SVL increases and in eight of 10 Mests of weight in- creases. In particular. Coluber that were <5 years old grew significantly faster in 1970 and 1971 than in 1972. Annual Variation in Weight Changes. — To in- clude records of individuals of unknown age not analyzed above, proportions of all large snakes (males mostly >6 years old, females >4 years old) that increased in weight each year were com- pared to each year's rainfall (Fig. 4). A greater 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 oftotal annual rainfall in the 5-month activity period (May-Sept.) each year were 40% (1970). 39% (1971), and 19% (1972). Amounts of weight gained and lost are shown in Table 7. Individual weight gains were signif- icantly greater in 1970-1971 than in 1972 in males (t = 4.75, P < 0.00 1 ) and females (/ = 3.40, P < 0.0 1 ). Individual weight losses were not sig- nificantly greater in 1972 than in 1970-1971 in males (t = 0.30, 0.80 > P > 0.70) and females (t = 1.84, 0.10 > P > 0.05). Female Reproductive Cycle.— Females con- tained enlarged preovulatory oocytes in late April. May, and early June (Brown, unpubl. data). At other times of the year ovaries were small and contained no enlarging oocytes. Available evi- dence indicates production of a single clutch of eggs/2 per year in northern Utah. Clutch Size.— Clutch size was determined from VERTEBRATE ECOLOGY AND SYSTEMATICA 19 Table 2. Winter weight losses in Coluber constrictor mormon > 1 year old. Absolute loss is difference in body weight between last autumn capture and first spring capture at den M; relative loss is percentage of autumn weight. Mean ± 1 SE. sample size and extremes in parentheses. Year Absolute weight loss (g snake) Relali\e weight loss l".,i " 1969-1970 1970-1971 1971-1972 1972-1973 All snakes 4.04 ± 0.56(15) (0.7-8.6) 3.81 ± 0.34(39) (0.2-9.5) 4.72 ± 0.34(56) (0.3-12.2) 4.10 ± 0.36(60) (0.1-15.1) 4.23 ± 0.20(170) (0.1-15.1) 5.11 ± 0.75(18) (1.6-12.5) 6.09 ± 0.72(35) (0.1-20.7) 5.25 ± 0.33 (49) (0.9-10.4) 5.11 ± 0.57(43) (0.3-15.6) 5.39 ± 0.28(145) (0.1-20.7) 6.59 ± 0.80(15) (0.9-11.4) 7.25 ± 0.68(39) (0.3-20.1) 7.54 ± 0.42(56) (1.0-21.3) 7.56 ± 0.55(60) (0.2-21.9) 7.40 ± 0.29(170) (0.2-21.9) 9.17 ± 0.74(18) (3.2-14.3) 7.16 ± 0.63(35) (0.2-16.9) 6.61 ± 0.38(49) (1.7-14.2) 7.53 ± 0.61 (43) (0.5-18.6) 7.33 ± 0.29(145) (0.2-18.6) a sample of 43 reproductive females (Fig. 5). At area M the mean number of eggs/2 was 5.78 ± 0.24 (4-8), mode 5 (N = 18). For these females, a significant (r = 0.53. 0.05 > P > 0.01) linear correlation existed between body size and clutch size; SVL explained 28% of the variation in clutch size. Size of .Eggs. — Measurements of 54 eggs in nine clutches were recorded after oviposition in the laboratory. Eggs averaged 37.78 ± 0.75 (29.2- 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) g. Data on egg size were not recorded in ten additional lab-deposited clutches. Eggs in three of these were weighed indirectly by dividing the female's ovi- positional 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). Analysis of variance (model I) 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 1 ). A model II analysis of the components of vari- ance showed that relatively more of the total variation occurred among clutches (73-86%) than among eggs within clutches ( 1 4-27%). The small- est clutch (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 correlation between female size (SVL) and the mean weight of eggs in her clutch (r = 0.19. P > 0.05; N = 9). Incubation and Hatching. — Between 27 June and 1 July 1971 seven gravid females were col- lected in area RB. These females oviposited be- tween 8-25 July after 9-28 days in an environ- mental chamber maintained at 29°C. Three gravid females from area M collected between 27 June- 3 July oviposited in the laboratory between 12- 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 M 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 of incubation at 29°C. We followed three gravid females with im- planted radio transmitters at area M in 1972 (Brown and Parker 1976a). Two of these females oviposited on 21 and 23 June. Eggs of one clutch were excavated 36 days later and were lab-in- cubated at 29°C an additional 12-13 days; hatch- ing occurred on 11-12 August after 48 and 49 days incubation. At the second field site a hatch- ling was captured by fencing on 10 August. 50 days after oviposition. The third site was exca- vated on 6 August, 4 1 days after oviposition. and one freshly-hatched egg was recovered. In 1971. timing of reproduction between areas RB and M (located 65 km apart) was similar. If most females at area M had oviposited between 5-15 July 1971 and between 20-30 June 1972, with a probable natural incubation period of 45- 50 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- males oviposited in the laboratory. A total of 20 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 3. Age-specific growth in snout-vent length of 199 66 Coluber constrictor mormon during 3 years. Mean ± 1 SE, sample size and extremes in parentheses. Aee VERTEBRATE ECOLOGY AND SYSTEMATICA 21 Table 3. Continued. Proportional increase scar 1970 + 1971 1972 1970-lv": .589 ± (.560 .178 ± (.032 .090 ± (.029 .053 ± (.000 .042 ± (.000 .027 ± (.000 029(2) -.618) 009 (37) -.293) 006 (43) -.160) 009(16) -.127) 009 (8) -.085) 009 (6) -.064) .502 ± (.404 .124 ± (.054 .041 ± (.010 .030 ± (.009 .024 ± (.000 .019 ± (.000 059(5) -.727) 010(16) -.187) 004(16) -.064) .003(19) -.057) .003(21) -.057) 005 (10) -.053) .527 ± .045(7) (.404-. 72 7) .162 ± .008(53) (.032-.293) .077 ± .005 (59) (.010-. 160) .040 ± .004(29) (.000-.085) .029 ± .004 (29) (.000-.085) .022 ± .005(16) (.000-.064) was determined, changes in weight prior to ovi- position were recorded. Ten females captured in spring at emergence and released between 26 April-20 May were later recaptured between 4- 26 June (Table 8). These animals were gravid as determined by palpation of enlarged ovarian oo- cytes or by subsequent oviposition in the field or laboratory. Recaptures occurred 1 7-47 (mean 3 1 ) days after release during which time these fe- males had increased by an average of 32.6% of their initial body weights. Absolute increases av- eraged 30.1 (14.6-48.8) g/snake during the pre- reproductive interval; the mean rate of weight gain was 0.97 ± 0.08 g/day. After ovipositing in the laboratory between 1- 14 July, parturient females were released in the field in July 1971 (3 22) and 1972 (2 22) (Table 9). One female oviposited in the laboratory both years. The four females were later recaptured in August and early September, 31-53 days after release. These spent females recovered an av- erage of 53% (24.7-71.1%) of their parturient weights (Table 9). Mean postreproductive weight recovery was 0.92 (0.45-1.38) g/day. For some area M females, additional weight records were obtained in a following year as they were again recaptured emerging from hiberna- tion. One female (No. 4. Table 9), weighed 1 14.0 g on 5 Sept. 1971 after reproduction. 100.5 g in spring 1972. and 109.2 g in spring 1973. Data showing very similar weights in the spring fol- lowing a known reproductive year as in the spring preceding that year are available for three fe- males in Table 8. Table 4. Continued. 1970 + 1971 Proportional increase year 19" .644 ± .041 (3) 22 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 5. Age-specific growth in weight of 179 6$ Coluber constrictor mormon during 3 years. Mean ± 1 SE. sample si/e and extremes in parentheses. VERTEBRATE ECOLOGY AND SYSTEMATICS 23 TABLE 5. Continued. Proportional increase- year 1970 + 1971 19" 1970-1972 2.153 ± (1.958- .647 ± (.137- .283 ± (.124- .171 ± (.003- .136 ± (.030- .099 ± (.019- .195(2) -2.348) .035(37) -1.081) .016(43) -.579) .021 (16) -.409) .022(8) -.223) .018(9) -.188) 1.744 ± (1.316- .384 ± (.076- .110 ± (.004- .058 ± (.002- .062 ± (.010- .046 ± (.038- .177(5) -2.363) .047(16) -.756) .025(13) -.314) .014(12) -.157) .016(15) -.232) .004(3) -.053) 1.882 - (1.316- .568 ± (.076- .243 ± (.004- .123 ± (.002- .088 ± (.010- .086 ± (.019- 147(7) 2.363) 033(53) 1.081) 017(56) .579) 017(28) .409) 015(23) .232) 015(12) .188) diately 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 weight was 62.7 ± 2.73 (43. 1-77.0) g. Mean RCM was 43.8 ± 1.03% (37.9-49.2%). Females were weighed an average of 6 (1-13) days prior to oviposition during which time some weight loss would be expected through dehydration (al- though 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 be- tween weighing interval and percentage weight loss (r = 0.30. P > 0.05). Estimates ofPopulation Size. — Total numbers of all individual racers captured at the various dens between 1969 and 1972 (Table 10) consti- tute a direct census which was influenced by (1) the snakes' fidelity to the several communal hibernacula and (2) the effectiveness of our en- circling fences in capturing them. We believe both possible sources of error were minimal, assuring a high reliability of our direct counts of individ- uals captured. Nonetheless, each year some un- marked snakes of all ages were caught (see "Age Structure"). Among all snakes one year old or older in 1971, 26-29% at dens M, 1 , and 5 were captures of unmarked individuals (Table 1 1). In 1972 at dens M and S3, new captures comprised 22% and 33% of the samples, respectively. Mark-recapture population estimates using the Jolly-Seber method resulted in population esti- mates for males and females > 1 year old only slightly higher than the actual number of snakes Table 6. Continued. Proportional increase >car 1970 + 1971 1972 19^0-1972 2.653 ± (.022 1.000 ± (.399 .442 ± (.159 .264 ± (.010 .160 ± (.049 .106 ± (.010 .384(3) -3.348) .038(39) -1.510) .028(38) -.873) .047 (20) -.789) .019(13) -.297) .027(8) -.247) 2.055 ± (1.156- .675 ± (.281- .219 ± (.052- .098 ± (.001- .078 ± (.003- .081 ± (.037- .252(7) -3.313) .076(10) -1.061) .037(18) -.754) .021 (13) -.285) .022 (8) -.219) .044 (2) -.124) 2.235 ± (1.156- .934 ± (.281- .370 ± (.052- .199 ± (.001- .129 ± (.003- .101 ± (.010- .219(10) -3.348) .038 (49) -1.510) .026 (56) -.873) .032 (33) -.789) .017(21) -.297) .022(10) -.247) 24 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY _ 40 E o w 30 2 20 2 10 o 1.0 w .8 ct o ? .6 O .4 o CL o q: .2 - ^ 84 78 970 97 972 Fics. 4. Annual proportions of Coluber constrictor mormon that increased in weight in three successive years (1970-1972) compared to yearly rainfall. Upper histogram shows total annual rainfall (unshaded), May- Sept, total (stippled), and June-Aug. total (hatched) recorded at Grantsville, Utah. Weight change records (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. caught at den M in 1970 and 1971 (Table 12). The disparity was greater for juveniles, reflecting the greater difficulty of capturing them and their higher mortality rate. These factors lowered re- capture proportions and tended to raise the es- 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 dis- persal distances recorded by Brown and Parker (1976a) (1.6 km from M complex, 1.8 km from S complex) as radii of circular areas, and assum- ing uniform dispersal in all directions from each den complex, areas occupied by the Coluber pop- ulations 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 weighing 29.728 kg were recorded at M complex, and in spring 1972, 271 Coluber weighing 15.795 kg were recorded at den S-3 in S complex (Table 10). Population and biomass densities at M and S complexes in 1971 were 0.66 and 0.27 snakes/ ha and 37 and 16 g/ha, respectively. Population census data were adjusted by calculated differ- ence factors (Table 12) to estimate total popu- lations. Adjusted population densities were 0.79 snakes/ha at M complex and 0.32 snakes/ha at S complex. Adjusted biomass densities were 39.8 g/ha at M complex and 16.7 g/ha at S complex (Table 13). The two den complexes are located ca. 875 m apart. Thus, a 600-ha region of overlap, encom- passing 60% of the S dispersal area and 75% of the M, could contain additive densities. The overlap densities were 0.78 snakes/ha and 39.6 g/ha. These are the most representative estimates of these parameters under the conditions ofsam- pling and assumptions employed in the calcu- lations. Population Changes. — Population changes during our sampling are shown in Table 14. The racer population increased by 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 1 97 1 were not sustained during 1 972 when the populations declined by 22.2% (den M) and 20.3% (den S3). Sex Ratio. — For each den and sampling pe- riod, 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 pe- riods, males outnumbered females (Table 10). Sex ratios were never significantly different from a 1 : 1 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 fe- males in this sample (x 2 = 0.24, 0.70 > P > VERTEBRATE ECOLOGY AND SYSTEMATICS 25 Table 7. Annual absolute weight changes of 350 Coluber constrictor mormon in three successive years. Data are for 179 36 > 50.0 g and 171 29 > 60.0 g that gained or lost weight during a yearly interval. Mean ± 1 SE. sample size and extremes in parentheses. 26 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 90- 6 o80 I I- o z UJ 70 z > i I- Z> O z 60- 50 o o D A A A -IT 4 Li -10 co UJ o r- ^t— —i— —r 6 7 8 9 CLUTCH SIZE o 5 oo 10 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) are records 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 M females and is described by the equation Y = -0.56 + 0.10X. VERTEBRATE ECOLOGY AND SYSTEMATICS 27 Table 8. Prereproductive (late spring-early summer) weight increases in 10 gravid female Coluber constric- tor mormon at area M, 1971 and 1972. O = ovarian oocytes, E = oviducal eggs. Last two snakes were re- captured prior to being tracked by telemetry. 28 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 10. Numbers, live-weight biomass, and sex ratios of Coluber constrictor mormon captured at five hibernacula in M complex (dens M, 1,2, 3, 5) 1969-1972 and at one hibernaculum in S complex (den S3) in 1972. Number of snakes at den M is total different individuals for both autumn and spring sampling; for all other hibernacula totals are different individuals in spring only. Weights are for animals in spring unless only autumn capture was recorded. Males and females include all ages > 1 year old, juveniles (J) include both sexes < 1 year old (see text). VERTEBRATE ECOLOGY AND SYSTEMATICA 29 Table 12. Comparison of sampled and estimated population sizes of Coluber constrictor mormon at den M in 1 970 and 1971. Difference factor is a proportion calculated by dividing the Jolly-Seber population es- timate by the actual number of snakes caught. Table 13. Population and biomass densities of Col- uber 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 31 FEMALE 95(2) 32 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY FEMALE c (161) -i—i—i — i — i — i — i — i — i — i — " — i < ' " (380) -i—i—i—i — i — i — i — i — i — i — i — <~ (187) "I 1 1 1 1 1 1 1 1 1 1 1 1 ' ' MALE 6+ 5 4 3 2 1 J 6+ 5 4 3 2 1 J 6* 5 4 3 2 1 J i7•2^^^^^^^^x^*•'•'•'^'^'•'•'''^'^'J''-*-'-'-** (209) pyyKTgggg?W5'WW^T?TC??WWTCTCOTWB J (419) 971 r i i — i — i —i—<—> ' i ' (199) 1 '970 i — i — i —S—i—i—i—i—i—i—i — ' — i i i" 1 VERTEBRATE ECOLOGY AND SYSTEMATICS 33 Table 16. Overwintering survivorship of Coluber constrictor mormon > 1 year old estimated by recap- ture proportions at den M. Table 17. Annual survivorship of Coluber constric- tor mormon estimated by recapture proportions at three dens(M, 1,5) in 1970 and 1971 (years combined) and 34 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY ^ .8 a* *" .7 I- 2 .6 a: id W .4 .3 •'/ O A^ / A' / // / VERTEBRATE ECOLOGY AND SYSTEMATICS 35 Table 19. Schedule of age-specific survivorship and fecundity of female Coluber constrictor mormon, x = age (years): Px = age-specific survival rate; lx= survi- vorship to age x: m v = number of female eggs produced each year by a female of age x; R„ = net reproductive rate. See text for assumptions and for adjustment fac- tors of m, schedule. Px lx Unad- justed ITU Adjusted m. I vm x 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 .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 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 1.84 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 = .023 .188 .178 .160 .128 .102 .085 .069 .055 .045 .038 .029 .023 .017 1.187 accrue to females during their increased move- ments in search of oviposition sites (Brown and Parker 1976a; Parker and Brown 1972. 1980) so it seems reasonable to suppose that a higher mor- tality may occur during the shorter prerepro- ductive phase than during the longer postrepro- ductive phase. In both a "good" and a "bad" year (1971 and 1972. respectively), three-year- old females had lower survival rates than either of the adjacent age classes. As most females ma- tured 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% ofthe total). In each sex, mortality was higher in the season of most active repro- ductive behavior, i.e., spring mating activity in males, summer gestation and parturition in fe- males (Feaver 1977). Of the total annual mor- tality in N. sipedon, 32% occurred over the win- ter. This value is almost identical to our data (33% of the annual mortality was overwintering 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 in Utah (present study). H = hatchling, J = juvenile, numeral = years. 36 SPECIAL PUBLICATION-MUSEUM OF 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 parameter Kansas Utah Population density* Body size** Growth rate Reproduction Sexual maturity Clutch size Egg size Hatchling size Relative clutch massf Demography Age distribution Relative contribution to R by female of age x Juvenile survivorship Adult female survivorship Generation time higher (5.0/ha) larger (66 123 g) (25 155 g) faster 1 year (83) 3 years (22) larger (1 1.7 eggs) smaller (5.7 g) smaller (4.2 g) lower (.40) younger (72% 1-3 yr) (28% 4+ yr) higher over ages 2-6; peak at age 3 (19.3%) higher (31 %/yr) lower (62%/yr) shorter (5.1 yr) lower (0.8/ha) smaller (66 56 g) (92 69 g) slower 1 year (<3<5) 3 years (22) smaller (5.8 eggs) larger (7.8 g) larger (6.0 g) higher (.62) older (52% 1-3 yr) (48% 4+ yr) lower over ages 2-6; peak at age 3 (15.8%) lower (23%/yr) higher (79%/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 sex; Fitch 1963) and C. c. 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 5 ages (2-6+ years). somewhat higher relative contribution to R be- tween ages 2-6 years (Fig. 1 2). The distributions indicate that 3-year-old females contribute the most to R in each population. Utah racers have higher adult survival rates than do Kansas racers. Life tables developed for each population show that Utah racers have a somewhat longer esti- mated generation time, suggesting a less frequent turnover of the population. An overall summary ofthe major life history comparisons is presented in Table 21. It is apparent that there are several prominent reproductive and demographic differences be- tween racers in 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 grow larger than males. Feaver (1977) placed C. constrictor, Rhabdophis tigrinus, Thamnophis butleri, and Nerodia sipedon in this group as con- trasted with Crotalus viridis, Agkistrodon con- tortrix, and Elaphe quadrivirgata in which males grow to the larger size. There are several impor- tant reproductive and behavioral differences be- tween the two groups of snakes (cf. Shine 1978); generally the last group of species tends to show late maturity, high adult survivorship, small clutches, and large young as contrasted to the first species group which shows the opposite trends. Viewed at this level, 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 strategies, it is less capable of showing differences in an intraspecific comparison. C. c. flaviventris and C. c. mormon each seems to possess some "K" and some "r" attributes (cf. Pianka 1970; Stearns 1976). Our data on survivorship of C. c. mormon show that there were considerable between-years effects on survival in adult females and lesser effects in adult 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- VERTEBRATE ECOLOGY AND SYSTEMATICS 37 3 4 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 non- gravid 9 body weight. Mean clutch weight was calcu- lated from clutch size and mean weight of eggs for C. c.flaviventris in Kansas (K; Fitch 1963) and C. c. mor- mon in Utah (U; present study). tality variable), a stable environment should fa- vor such traits as fewer young, longer life span, smaller reproductive effort, and slower devel- 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, crickets) are eaten by C. c. mormon almost ex- clusively, whereas C. c. flaviventris takes a mod- KANSAS n ' 242 38 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY these data with those of a population of C. c. flaviventris in Kansas studied by Fitch (1963). Racers were captured at their dens with en- circling screen fences each autumn and spring between 1969-1973. Snakes were measured, weighed, sexed, and permanently marked by scale clipping. The age of each individual was deter- mined by comparing its size to confidence in- tervals for length and weight ofrecaptured known- age snakes. A total of 1046 racers was captured 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 (.v = 28.4 g) weighed significantly more than males (x = 27.0 g) and females con- tinued to be significantly larger in both snout- 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 in 1970 and 1971 than in 1972. Females produced a single clutch per year av- eraging 5.8 eggs. Eggs averaged 38 x 18 mm and 7.8 g. Mean clutch weight/female body weight ratio 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 hatching did not differ significantly from 1:1. Weight increases in prereproductive females in early summer averaged ca. 1 g/day as did post- reproductive weight recovery in late summer. Among randomly-collected females in early summer, 88% were gravid or parturient. In 1971, sampling at six dens yielded 528 rac- ers. The largest number recorded at a single den in one season (spring 1972) was 271 snakes. Us- ing maximum dispersal distances and assuming a 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 M increased from 139 to 189 individuals ( 1 8%/yr) over two successive favorable years (1970, 1971) and declined to 147 individuals (21% decrease) in an unfavorable year (1972). In all samples of snakes >1 year old, males comprised 53% and females 47% of the popu- lation. Age structure favored younger (<5 years) animals which comprised 62-76% of the popu- lation in different years. Large proportions of 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. 1 972 was a poor year for recruitment of juveniles (4.6% of the population). Overwintering survival rates averaged 93% in both sexes. Annual survivorship in juveniles was 23%. First year survival (egg to age 1) was esti- mated to be 1 7%. Adult survivorship in favor- able 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 contrast, literature reports for 1 1 species of col- ubrids indicate an average survivorship of ca. 50% per year. The 21% annual mortality in C. c. mormon may be partitioned into overwinter- ing (7%), prereproductive, and postreproductive mortality. We suggest that prereproductive mor- tality is higher in females from exposure to ad- ditional risks associated with egg laying. A life table for C. c. mormon calculated using the combined female survival rate in 1970 and 1971 showed a net reproductive rate (R ) of 1 . 1 87, a value indicating an increasing population. 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 in the following ways: ( 1 ) lower growth rates and smaller adult size; (2) lower age-specific fecun- dity; (3) larger eggs and hatchlings; (4) higher clutch weight/female body weight ratio; (5) lower juvenile survivorship and higher adult survivor- 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 re- source levels, environmental stability, and pre- dation. we caution against simplistic interpre- tations in contrasting the two populations. Acknowledgments At the University of Utah our studies were supported by American Museum of Natural His- VERTEBRATE ECOLOGY AND SYSTEMATICS 39 tory (Theodore Roosevelt Memorial Fund) grants, a Biomedical Sciences Support Grant FR- 070902, a Graduate Research Fellowship to Brown, and an NDEA Fellowship to Parker. At Skidmore College, Denton W. Crocker, Chair- man, Biology Department, provided an oppor- tunity for portions of this work to be completed. Eric J. Weller, Dean of the Faculty at Skidmore College, allocated financial support to allow Brown's participation in the 1980 annual her- petology meetings and provided funds for manu- script preparation through a Mellon Foundation Grant for faculty development. For assisting us in 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, Fred- erick B. Turner, Stephen C. Stearns, and an anon- ymous reviewer for critically reading the manu- script and suggesting improvements. Brown 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. Literature Cited AUFFENBERG, W. 1955. A reconsideration of the racer. Coluber con- strictor, in eastern United States. Tulane Stud. 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. Herpe- tol., 13:377-402. Branson, B. A. and Baker, E. C 1974. An ecological study of the queen snake. Re- gina septemvittata (Say) in Kentucky. Tu- lane Stud. Zool. Bot., 18:153-171. Brown, W. S. 1973. Ecology of the racer. Coluber constrictor mormon (Serpentes, Colubridae). in a cold temperate desert in northern Utah. Ph.D. Thesis. Univ. Utah, Salt Lake City. 208 p. Brown, W. S. and Parker, W. S. 1976a. Movement ecology of Coluber constrictor near communal hibernacula. Copeia, 1976: 225-242. 1976b. A ventral scale clipping system for perma- nently marking snakes (Reptilia, Serpentes). J. Herpetol.. 10:247-249. 1982. Niche dimensions and resource partitioning in a Great Basin desert snake communitv. Pp. 59-81. In Scott. N. J.. Jr. (Ed.). Herpe- tological Communities. U.S. Fish and Wild- life Service. Wildl. Res. Rep. 13. 239 p. 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 populations. Ecol. Monogr., 22:235-258. Caughley, G. 1977. Analysis ofvertebrate populations. John Wi- ley & Sons. N.Y. 234 p. Clark, D. R.. Jr. 1970. Ecological study of the worm snake. Car- phophis vermis (Kennicott). Univ. Kansas Publ. Mus. Nat. Hist.. 19:85-194. 1974. The western ribbon snake {Thamnophis proximus): ecology of a Texas population. Herpetologica, 30:372-379. Clark, D. R.. Jr. and Fleet, R. R. 1976. The rough earth snake (Virginia striatula): ecology of a Texas population. Southwest. Nat., 20:467-478. Conant, R. 1975. A field guide to reptiles and amphibians of eastern and central North America. 2nd ed. Houghton Mifflin Co.. Boston. Deevev, E. S. 1947. Life tables for natural populations of ani- mals. Q. Rev. Biol., 22:283-314. Etheridge, R. E. 1952. The southern range of the racer Coluber con- strictor stejnegerianus (Cope), with remarks on the Guatemalan species Coluber orten- burgeri Stuart. Copeia, 1952:189-190. Feaver, P. E. 1977. The demography of a Michigan population of Natrix sipedon with discussions of ophid- ian growth and reproduction. Ph.D. Thesis. Univ. Michigan. Ann Arbor. 131 p. Fitch, H. S. 1949. Study of snake populations in central Cali- fornia. Am. Midi. Nat., 41:513-579. 1 960. Autecology of the copperhead. Univ. Kansas Publ. Mus. Nat. Hist.. 13:85-288. 1963. Natural history of the racer. Coluber con- strictor. Univ. Kansas Publ. Mus. Nat. Hist., 15:351-468. 1965. An ecological study of the garter snake. Thamnophis sirtalis. Univ. Kansas Publ. Mus. Nat. Hist.. 15:493-564. 1975. 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: 1 96- 203. Gregory, P. T. 1977. Life-history parameters of the red-sided 40 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY garter snake ( Thamnophis sirtalis parietalis) in an extreme environment, the Interlake re- gion of Manitoba. Natl. Mus. Canada Publ. Zool.. 13:1-44. Hall, R. J. 1969. Ecological observations on Graham's wa- tersnake {Regina grahami Baird and Gi- rard). Am. Midi. Nat.. 81:156-163. Hirth, H. F. 1 966. Weight changes and mortality of three species of snakes during hibernation. Herpetologica, 22:8-12. Hirth, H. F. and King, A. C. 1 968. Biomass densitites of snakes in the cold des- ert of Utah. Herpetologica, 24:333-335. Hl'TC hinson, G. E. 1 978. An introduction to population ecology. Yale Univ. Press, New Haven, Conn. 260 p. Krebs, C. J. 1978. Ecology: the experimental analysis of distri- bution and abundance. 2nd ed. Harper and Row, New York. 678 p. Parker, W. S. and Brown, W. S. 1972. Telemetric study of movements and ovi- position of two female Masticophis t. tae- niatus. Copeia, 1972:892-895. 1 973. Species composition and population changes in two complexes of snake hibernacula in northern Utah. Herpetologica, 29:319-326. 1 974. Mortality and weight changes of Great Basin rattlesnakes (Crotalus viridis) at a hibernac- ulum in northern Utah. Herpetologica, 30: 234-239. 1 980. Comparative ecology oftwo colubrid snakes, Masticophis t. taeniatus and Pituophis mel- anoleucus deserticola, in northern LHah. Milwaukee Public Mus. Publ. Biol. Geol. No. 7:1-104. PlANKA, E. R. 1970. On r and K selection. Am. Nat., 104:592- 597. Platt, D. R. 1969. Natural history of the hognose snakes Het- erodon platyrhinos and Heterodon nasicus. Univ. Kansas Publ. Mus. Nat. Hist.. 1 8:253- 420. Prestt, I. 1971. An ecological study of the viper, Viperabe- rus, in southern Britain. J. Zool. (London), 164:373-418. Shine, R. 1 978. Sexual size dimorphism and male combat in snakes. Oecologia (Bed.), 33:269-277. Sokal, R. R. and Rohlf, F. J. 1969. Biometry: the principles and practice of sta- tistics in biological research. W. H. Freeman Co., San Francisco. 776 p. Spellerberg, I. F. and Phelps, T. E. 1977. Biology, general ecology and behaviour of the snake, Coronella austriaca Laurenti. Biol. J. Linn. Soc. (London), 9:133-164. Stearns, S. C. 1976. Life history tactics: a review of the ideas. Q. Rev. Biol., 51:3-47. Stebbins, R. C. 1966. A field guide to western reptiles and am- phibians. Houghton Mifflin Co.. Boston. 279 p. Stewart, G. R. 1968. Some observations on the natural history of two Oregon garter snakes (genus Thamno- phis). J. Herpetol., 2:71-86. Tinkle, D. W. 1957. Ecology, maturation, and reproduction of Thamnophis sauritus proximus. Ecology, 38: 69-77. 1 960. A population of Opheodrys aestivus (Reptil- ia: Squamata). Copeia, 1960:29-34. Turner, F. B. 1977. The dynamics of populations of squamates, crocodilians and rhynchocephalians. Pp. 157-264. In Gans, C, and Tinkle, D. W. (Eds.), Biology of the Reptilia, Vol. 7. Aca- demic Press, New York. Vial, J. L., Berger, T. J. and McWilliams, W. T., Jr. 1977. Quantitative demography of copperheads, Agkistrodon contortrix (Serpentes, Viperi- dae). Res. Popul. Ecol., 18:223-234. Viitanen, P. 1967. Hibernation and seasonal movements of the 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 para- dox. Am. Nat., 112:595-608. Wilbur, H. M., Tinkle, D. W. and Collins, J. P. 1974. Environmental certainty, trophic level, and resource availability in life history evolu- tion. Am. Nat., 108:805-817. Wilson, L. D. 1970. 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- petologica, 7:1-52. Woolf, C. M. 1 968. Principles ofbiometry. D. Van Nostrand Co., Inc. Princeton, N.J. 359 p. Vertebrate Ecology and Systematica—A Tribute to Henry S Fitch Edited by R. A. Seigel. L. E. Hunt. J. L. Knight. L. Malaret and N. L. Zuschlag i 1 984 Museum of Natural History. The University of Kansas. Lawrence Growth of Bullsnakes (Pituophis melanoleucus sayi) on a Sand Prairie in South Central Kansas Dwight R. Platt Introduction Growth rates of snakes in natural populations have been studied for fifty years. Blanchard and Finster (1933) presented limited data on growth rates ofrecaptured garter snakes and water snakes. KJauber (1937) derived a growth curve for the southern pacific rattlesnake (Crotalus viridis hel- leri = C. v. oreganus) by analyzing a collection of preserved specimens and pointed out that the growth of captive snakes may be distorted. Sei- bert 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. 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 in the northern pacific rattlesnake (Cro- talus viridis oreganus) has been widely cited. He and his students have provided many reports on growth rates of snakes in natural populations (Clark 1970, 1974; Clark and Fleet 1976; Fitch 1960, 1963a, 1963b. 1965, 1975; Fitch and Fleet 1970; Platt 1969). Other notable studies on growth rates in free-living populations of colu- brid snakes include those by 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 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 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 individual variability in growth rates. Two methods have been used to study growth ofsnakes in natural populations: 1 ) summarizing growth records from marked, released and re- captured individuals; and 2) determining size at different ages, usually up to one year old, by an analysis of size frequencies in population sam- 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. The 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 ofsnakes living in the same general environment: 3) to compare the two methods (above) of determin- ing growth rates. This paper reports results based on 709 cap- tures of 471 bullsnakes (Pituophis melanoleucus sayi) and on nine young which were hatched in the laboratory. Subsequent papers will describe other species studied and comparative aspects of the study. Methods This study used standard mark-recapture tech- niques with snakes trapped alive in drift fence traps (Fitch 1951, 1960; Platt 1969). Thirty to fifty trapping stations were used on a study area 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 each end. These traps set without bait intercepted movement of the animals. Most measurements were made in the labo- 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 measured on a triple beam balance. The snakes 41 42 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY were released within three to four days at the site of capture. Marking was accomplished by clipping or branding subcaudals or ventrals so that each snake was individually recognizable. Individual vari- ations in color pattern and scutellation were also recorded so that almost all recaptured snakes were individually identified with certainty. Sex- ing was accomplished by probing through the vent for the hemipenial sacs and was checked later by body proportions. Food records were obtained by palping fecal matter from the intes- tine and forcing stomach contents back up the gullet into the mouth for identification and then repalping into the stomach. Previous studies of growth have used snout- vent lengths, total lengths and/or weights as a measure of size. I used snout-vent length (SVL) because it is one of the least variable measures of size. Weight is more affected by the stage of the feeding cycle or the reproductive cycle while total length is affected by partial loss or differ- ential growth of the tail. Only measurements of live snakes were used in the analysis. Growth rates from recapture records were cal- culated by averaging growth increments during the period between captures for samples of re- captured snakes. Recapture records were used to calculate growth rates only ifthree 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 until September. First-year snakes were defined in this study as those caught between September of their hatching year and the end of the next August. Records of recaptured first-year and old- er bullsnakes were readily distinguished by plot- ting the SVLs of snakes with respect to capture date. Growth rates were also calculated from the changes in mean SVL of population samples of identified age. Frequency distributions were cal- culated for the lengths of all bullsnakes caught in each two-week interval throughout the trap- 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 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 May to the end of October or early November over a nine year period, 1 966-1 974. Growth was not continuous throughout the year. On my study area it was usually most rapid in early summer but occurred throughout the period of activity and trapping; probably little or no growth oc- 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 October as the growth season. Absolute growth rates were calculated as growth increment in SVL per month (30 days) excluding the period from 1 November to 30 April. Rela- tive growth rates were calculated as the growth increment per month per 100 mm median SVL. The median SVL was defined as the midpoint between the lengths at two successive captures or between the mean lengths of two successive population samples. Some studies (Carpenter 1952; Fukada 1959, 1960, 1978) have used the initial length at first capture to calculate relative growth rates. The median length is more similar to the length of the snake during the growing period. Relative growth rates calculated from median lengths are less affected by the duration of the period between captures. Mean values in this paper are usually accom- panied by one standard error. Homogeneity of variances was tested by an F test. Differences in the central tendency of different samples were tested by Student's / test for samples having sim- ilar variances and by the Mann-Whitney U test when the variances were heterogeneous (Cox 1980). Regression equations of weight on length were calculated by Bartlett's method (Simpson et al. 1960). Rodent populations, principal prey of bull- snakes, were sampled by the same drift fence traps used to capture snakes and by 100 baited small 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 con- tinuously from May through October while bait- ed live traps were operated for a few nights per month through the summer (May-August). Ro- dents 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 number caught per 100 trap nights (TN). Study Area The Sand Prairie Natural History Reservation is 80 acres (32.4 hectares) of prairie on sand dunes VERTEBRATE ECOLOGY AND SYSTEMATIC S 43 Tabii 1. Rodents trapped on the Sand Prairie Nat- ural History Reservation in Kansas. Dnl'i fence traps Hailed traps Year No cil nap station days i rsD) No. rodents 100 TSD Medium Small sized sized speeies species No rodents 100 TN Medium Small si/ed sized species species No. of trap niehls (TN) 1967 1968 1969 1970 1971 1972 1973 1974 7317 9989 19.775 17.076 19,962 15.272 17.014 17,908 0.2 3.2 0.3 0.2 0.1 0.3 0.1 0.2 6.4 9.8 1.1 1.9 0.4 4.5 0.8 1.1 2369 1 161 1430 1088 1061 1518 2134 1747 1.3 3.4 2.0 1.3 1.5 1.8 0.6 managed as a natural area. Prior to its acquisition by Bethel College in 1 965. it was used as a pasture but was never cultivated. All snakes used in my analyses were captured on this study area or on 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 reserva- tion are dominated by little bluestem (Andropo- gon scoparius). Forbs and other genera of grasses ( Triplasis, Aristida and Panicum) also occur. The unliooded lowlands have dense tall grass com- munities 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 augustifolia) are com- mon on the uplands and buttonbush (Cephalan- thus occidentalis) and black willow (Sa/ix nigra) in the lowlands. The area is poorly drained and its low depressions between sand dunes are rel- atively wet, having ponds, shallow marshes or dry ground depending upon the amount ofrecent rainfall. A more complete description of the study area can be found in Piatt (1973. 1975). Results Prey Populations. — Prey of bullsnakes on the study area were predominantly rodents. Trap- ping success (Table 1 ) provides a rough measure of the size and activity of rodent populations. Medium-sized rodents, prairie voles (Microtus T\nn 2. Proportions of bullsnakes (Pituophis mel- anoleucus) 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. Year Summer (Ma>-Aug.) Per- centage contain ing loud x" \utumn (Sept. -i K i i 1966 1967 1968 1969 1970 1971 1972 1973 1974 46 43 31 55 65 15 42 44 61% 72% 52% 58% 26% 33% 52% 68% 2.43 7 54** 0.66 27.94** 1.72 6.85** 4.46* 44 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 3. Absolute and relative growth rates of bull- snakes {Pituophis melanoleucus) in Kansas ( 1 966-1 974) determined from recapture records. N = number of useable recapture records. Mean followed by ± 1 stan- dard error. Table 4. Absolute and relative growth rates of bull- snakes (Pituophis melanoleucus) older than one year in Kansas determined from recapture records. The prob- able ages are based on size (see text). N = number of useable recapture records. Means followed by ± 1 stan- VERTEBRATE ECOLOGY AND SYSTEMATICS 45 Table 5. Snout-vent lengths of bullsnakes (Pituophis melanoleucus) in successive months during their first 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. 46 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Tabi i 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 small samples (<5). VERTEBRATE ECOLOGY AND SYSTEMATICS 47 Table 7. Snout-vent lengths and success of feeding for juvenile bullsnakes caught in half-month intervals in their first 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 caught. Mean SVL Extreme SVL Half-moruh inters als Half-month intervals Year S 1-15 S 16-30 O 1-15 O 16-30 S 1-15 S 16-30 O 1-15 O 16-30 A. Lower lh of 48 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY TABLE 8. Absolute growth rates ofbullsnak.es (Pitu- ophis melanoleucus) trapped more than once during a year in Kansas. N = number of useable recapture rec- ords. 1969 and 1972 were omitted because of the small number of recaptures. Means followed by ± 1 standard error. Age: first year summer only Age: >one year Year VERTEBRATE ECOLOGY AND SYSTEMATICS 49 Table 10. Absolute growth rates (GR) of recaptured bullsnakes (Pituophis melanoleucus) in Kansas after different lengths of intervals between captures. N = useable recapture records. Means followed by ± 1 standard error. 50 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY II00-- I000-- ]> 900-- — 800+ to UJ _J > CO 700 600 500-- 400-- I f I A S FIRST i—I—I—I—I—t- -I—I—I—I—h S -H 1 1 1 1 H- M J J A S FOURTH MJJASO MJJA SECOND 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 mean lengths of snakes in monthly samples of first-year snakes (see text for age designations). Dashed lines represent mean growth rates in a poor year (1970-1971) and good years ( 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), feed- ing 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 ofjuveniles in the autumn were variable. Low production, low survival and/or low growth ofjuveniles 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 51 yond the first year was estimated from the growth rates of recaptured snakes. The lower dashed line represents the average growth in 1970-1971. a poor growth year, and the upper dashed line rep- resents growth determined from the combined samples for 1967-1968 and 1968-1969. good years. The data were inadequate to estimate the monthly variation in growth for snakes older than one year so the growth rate is applied uniformly through the active season. This growth curve in- dicates that bullsnakes were approximately 790 mm SVL at one year of age in August. 950 mm SVL at two years and 1030 mm SVL at three years. These estimates may be slightly low for the second and third years, since growth was probably more rapid before August than after August. If older snakes grew at the rate of 35- 40 mm per year (Table 4). a bullsnake 1 100 mm SVL would be five years old and one 1200 mm SVL would be seven to eight years old. This is similar to the estimate by Imler ( 1 945) that bull- snakes in Nebraska reach a total length of 49 inches (1245 mm; ca. 1100 mm SVL) in five to six years. Parker and Brown (1980) found that gopher snakes (P. m. desertkola) in Utah re- quired 1 8-20 years to reach 1 200 mm SVL. Fitch ( 1 949) estimated that P. m. catenifer in central California reached a SVL of more than 800 mm at two years of age. Growth of bullsnakes in cen- tral Kansas was comparable to that of bullsnakes in Nebraska but more rapid than populations of the same species in Utah and California. Different studies ofgrowth rates in snakes can- not be precisely compared because of differences in methods used. But comparisons can be made on the basis of the general magnitude of growth in the first year of life. Information for some colubnd species on growth during the first year are presented in Table 12. The species are ar- ranged in order of increasing hatchling size. Growth rate of snakes in their first year is pos- itively related to both size of hatchlings and to normal adult size (Table 12). Hatchling size was either listed in the reports cited or was deter- mined from the growth curves reported and was rounded to the nearest five mm. Normal adult size was usually determined from length distri- butions as the mode of the snakes forming the largest size group in a population sample. The SVL was rounded to the closest 10 mm in small snakes, closest 50 mm in moderate-sized snakes and closest 100 mm in large snakes. If a fre- 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 rela- tion between first-year growth rate and size: a) Geographic variation in growth rate probably was mediated through environmental limi- tation. The studies of growth of Thamnophis and Nerodia in Michigan with a short growing season reported lower first-year growth than studies of related forms in Kansas. The growth rates of Coluber constrictor mormon and Pi- tuophis melanoleucus deserticola in Utah, where ecosystem primary production is low- er, were much lower than those ofrelated sub- species in Kansas. b) Taxonomic differences in growth are evident. Species of the genus Thamnophis have high growth rates in proportion to hatchling and adult size while the species of Elaphe and Lampropeltis studied appear to have low rel- ative growth rates. c) There is a relationship between first-year growth rate and age at sexual maturity. Fe- males of most of the species of moderate- sized 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 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 low growth rates, Lampropeltis triangulum in 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 longer to mature than the faster growing sub- species from Kansas. The smallest species of Elaphe studied. E. quadrivirgata, matures in one year but the larger species, E. climaco- phora and E. obsoleta, mature in three years and have relatively slower growth rates. First-year growth in viperid snakes ranged from 70-370 mm (10-45 mm/month) (Fitch 1949. 1 960; Gibbons 1 972: Klauber 1956: Prestt 1971; Volsoe 1944; Wharton 1966). Growth rates in elapid snakes in Australia up to 12 months of age ranged from 70 to 4 1 mm in a growth season 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 53 Table 12. 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 54 SPECIAL PUBLICATION- MUSEUM OF NATURAL HISTORY of eight months (Shine 1978. 1980). It appears that bullsnakes on the Sand Prairie Reservation in Kansas have one ofthe highest first-year growth rates that has been reported for snakes in free- living populations. Summary The growth of bullsnakes (Pituophis melano- leucus sayi) was studied by mark-recapture tech- niques on the Sand Prairie Natural History Res- ervation in south central Kansas. During the nine years of the study, 1966 through 1974, 709 cap- tures of47 1 bullsnakes were made. Growth rates were calculated from the records of recaptured snakes and from the mean snout-vent lengths of first-year bullsnakes in monthly samples. No sig- nificant sexual differences in size or growth rates of bullsnakes were found. Bullsnakes grew rapidly for the first year of life (absolute growth rate of 70.4 mm/month) and reached 790 mm SVL at one year of age. Growth rates declined to 40% of the first-year rate in the second year, to 20% in the third year and to 1 0% and less in older snakes. Growth rates were significantly lower in 1971 when prey populations and feeding rates of bull- snakes were low. Prey populations and feeding rates were very high in 1968 and growth rates were highest then. Growth rates calculated from recapture rec- ords 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 differential mortalitv of smaller snakes. Acknowledgments 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 per- sons assisted with the field study and were in- dispensable to the completion of this investiga- tion: Victor Claassen, Steven G. Hetzke, Marilyn Johnson. Mark Matthies, Scott Matthies. Ka- mala Piatt. Stanley Senner, Patricia Senner. Rob- ert C. Waltner and James Wedel. Douglas Harms, Richard Piatt, and William Schmidt assisted with the data analysis. Literature Cited Blanc hard. F. N. and Finster, E. B. 1933. A method of marking living snakes for future recognition with a discussion of some prob- lems and results. Ecology. 14:334-349. Brown. W. S. and Parker, W. S. 1984. 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 Trib- ute to Henry S. Fitch. Univ. Kansas Mus. Nat. Hist. Spec. Pub., 10. Carpenter, C. C. 1952. Growth and maturity of three species of Thamnophis in Michigan. Copeia. 1952(4): 237-243. Clark. D. R.. Jr. 1 970. Ecological study of the worm snake Carpho- phis v£7vw.s(Kennicott). Univ. Kansas Publ. Mus. Nat. Hist.. 19(2):85-194. 1974. 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 ofGeneral Ecology. Wm. C. Brown Co., Dubuque. Iowa, viii + 237 pp. Feaver, P. E. 1977. The demography of a Michigan population of Natrix sipedon with discussions of ophid- ian growth and reproduction. Ph.D. Thesis, Univ. of Michigan. Fitch, H. S. 1949. Study of snake populations in central Cali- fornia. Am. Mid. Nat.. 4 1(3):5 13-579. 1950. A new style live-trap for small mammals. Jour. Mammal.. 31(3):364-365. 1951. A simplified type of funnel trap for reptiles. Herpetologica. 7:77-80. 1 960. Autecology of the copperhead. Univ. Kansas Publ. Mus. Nat. Hist.. 1 3(4):85-288. 1 963a. Natural history of the black rat snake (Elaphe o. obsoleta) in Kansas. Copeia. 1963(4):649- 658. 1963b. Natural history of the racer Coluber con- strictor. Univ. Kansas Publ. Mus. Nat. Hist.. 15(8):35 1-468. 1965. An ecological study of the garter snake, Thamnophis sirtalis. Univ. Kansas Publ. Mus. Nat. Hist., 1 5(10):493-564. 1975. A demographic study of the ringneck snake ( Diadophis punctatus) in Kansas. Univ. Kansas Mus. Nat. Hist. Misc. Publ.. 62:1- 53. VERTEBRATE ECOLOGY AND SYSTEMATICS 55 Fitc h, H. S. and Fleet, R. R. 1970. Natural history of the milk snake (Lampro- peltis triangulum) in northeastern Kansas. Herpetologica. 26(4):387-396. Frye, J. C. and Leonard, A. B. 1952. Pleistocene geology of Kansas. Univ. Kan- sas, State Geol. Surv. Kansas Bull.. 99:1- 230. Fl IKADA, H. 1959. Biological studies on the snakes. VI. Growth and maturity ofNatrix tigrina tigrina (Boie). Bull. Kyoto Gakugei Univ., Ser. B, 15:25- 41. 1960. Biological studies on the snakes. VII. Growth and maturity of Elaphe quadrivirgata (Boie). Bull. Kyoto Gakugei Univ.. Ser. B, 16:6-21. 1972. Growth and maturity of some Japanese snakes (Review). The Snake, 4:75-83. 1978. Growth and maturity of the Japanese rat snake, Elaphe climacophora (Reptilia, Ser- pentes, Colubridae). J. Herpetol.. 12(3):269- 274. Gibbons, J. W. 1972. Reproduction, growth and sexual dimor- phism in the canebrake rattlesnake (Crotalus horridus atricaudatus). Copeia, 1972(2):222- 226. Heyrend. F. L. and Call. A. 1951. Symposium: a snake den in Tooele County, Utah. Growth and age in western striped rac- er and Great Basin rattlesnake. Herpetolog- ica, 7:28-40. Imler, R. H. 1945. Bullsnakes and their control on a Nebraska wildlife refuge. Jour. Wildl. Mgt., 9(4):265- 273. Klal'ber, L. M. 1937. Statistical study of the rattlesnakes. IV. The growth of the rattlesnake. Occ. Papers San Diego Soc. Nat. Hist.. 3:1-56. 1956. Rattlesnakes, their habits, life histories, and influence on mankind, vol. 1. Univ. Cali- fornia Press, Berkeley, xxx + 708 pp. Parker, W. S. and Brown, W. S. 1 980. Comparative ecology oftwo colubrid snakes. Masticophis t. taeniatus and Pituophis mcl- anoleucus descrticola, in northern Utah. Milwaukee Public Mus. Publ. Biol. Geol., 7: vii + 104 pp. Platt, D. R. 1969. Natural history of the hognose snakes Het- erodon platyrhinos and Heterodon nasicus. Univ. Kansas Publ. Mus. Nat. Hist.. 15(8): 351-468. 1973. Vascular plants of the Sand Prairie Natural History Reservation, Harvey County. Kan- sas. Trans. Kansas Acad. Sci.. 76( 1 ):5 1—73. 1975. Breeding birds of Sand Prairie Natural His- tory Reservation, Harvey County, Kansas. American Birds, 29(6): 1146-1151. Prestt, 1. 1971. An ecological study of the viper Vipera berus in 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): 261-333. 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. Herpe- tologica, 34:73-79. 1980. Reproduction, feeding and growth in the Australian burrowing snake Vermice/la an- nulata. J. Herpetol.. 14(l):71-77. Simpson, G. G., Roe, A., and Lewontin, R. C. 1960. Quantitative Zoology. Harcourt. Brace. & Co., New York, viii + 440 pp. Volsoe, H. 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 cotton- mouth, Agkistrodon piscivorus Lacepede, of Cedar Keys, Florida. Copeia, 1966(2): 149- 161. 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 in Snakes Patrick T. Gregory Introduction The temperate zone, especially at higher lati- tudes, is characterized by a pronounced season- ality. Relatively few species of reptiles occur in such environments and those that do must be adapted to contend with this climatic variability. The most critical season for reptiles in the tem- perate zone is winter, when conditions may be- come much 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 ad- aptations which enhance their chances for sur- vival over winter, mainly by depressing meta- bolic rate in hibernation below that predicted on a simple Q ln basis (Moberly 1963; Mayhew 1965; Aleksiuk 1976; Gatten 1978; Patterson and Da- vies 1978; Johansen and Lykkeboe 1979). This depression is interpreted as an important mech- anism for conserving energy at a time when loss- es cannot be replaced (Gregory 1982). In addi- tion to these physiological adaptations, however, the need to hibernate has had other important ecological and evolutionary effects on reptiles. In this paper, I discuss ecological aspects of one particular phenomenon in reptilian overwinter- ing: communal denning of snakes. Communal denning occurs mainly in snakes among reptiles, although it has been described for some lizards and turtles (Woodbury and Hardy 1949; Weintraub 1968; Ataev 1974; Vitt 1 974). Communal hibernation has been reported for a few snakes in the southern hemisphere (e.g., 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 northern snakes hibernate communally. In communal denning, large numbers of oth- erwise solitary animals aggregate at localized sites to pass the winter. Both single and mixed species aggregations occur (Gregory 1982). This type of behavior is often conspicuous. Snakes which or- dinarily are not locally abundant suddenly be- come extremely abundant in both fall and spring, the beginning and end of hibernation, respec- tively. This conspicuousness sometimes works to the disadvantage ofsnakes by advertising their presence to predators, including man. Rattle- snakes in particular have suffered major declines due to raids on denning populations (Klauber 1 972; Galligan and Dunson 1974) and it is likely that preservation ofsome species depends in part on protection of communal dens (e.g., Crotalus horridus. Brown et al. 1982). On the other hand, the conspicuousness and abundance of snakes at some dens have provided us with an opportunity to learn much about the structure and dynamics ofsnake populations that we might not otherwise have learned (Brown 1973; Gregory 1977a; Par- ker and 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 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 em- phasize 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 hope to provide at least a partial answer to the question: "Why do snakes den communally?" Major Features of Communal Denning Although considerable variation exists from one case to another, several important features emerge from an examination of studies of com- munal dens. These are: type of site used for hi- bernation, spatial relationship of den to summer habitat, size and structure of denning popula- tions, and fall and spring activity of snakes at denning sites. 1. Type of site.—Communal dens are usually relatively permanent 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- snake {Crotalus viridis) usually hibernates com- 57 58 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 1 . Dispersal distances of snakes between hibernacula and summer ranges (modified from Gregory 1 982). VERTEBRATE ECOLOGY AND SYSTEMATICA 59 2. Relationship of den to summer habitat.— Snakes which hibernate singly or in very small groups may use sites within the summer range (Fitch and Glading 1 947; Naulleau 1 966). Com- munal dens of snakes, however, are frequently separated by fairly long distances from the sum- mer range, necessitating an annual migration back and forth between the two. Distance travelled ranges from a few hundred m to several km (Ta- ble 1): perhaps length of migration is inversely related to the availability of sites suitable for housing snakes in winter, but this idea has not been tested. In Coluber constrictor, mean dis- persal distance may be correlated with popula- tion density; presumably, in years with high numbers, individuals which disperse farther es- cape intraspecific competition for resources (Brown and Parker 1976). Some snakes which move between discrete denning and summer areas show a highly direc- tional form of dispersal (Gregory and Stewart 1975). while others do not (Parker 1976). Indi- vidual snakes often return to the same den or denning area year after year; measures of den fidelity of snakes in successive years are often in 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 defi- nition of what constitutes a den or denning area varies from study to study so that results are not necessarily comparable. In addition, distance be- tween 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 species: Homing to specific den complexes less than 1000 m apart is almost 100% in Coluber constrictor \n Utah (Brown 1973; Brown and Par- ker 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 ( 1 967) and Lang (1971). 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- bernating sites are abundant. This is important since high rates of mortality during hibernation have been reported for many snakes, especially the young (Bailey 1948; Carpenter 1953; Hirth 1 966a; Viitanen 1 967: Lang 1 969, 1971; Gregory 1977a, 1982; Parker and Brown 1980). Parker and Brown (1974, 1980) suggest that high mor- tality figures in such studies may be an artifact ofhandling 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 mat- ing usually occurs at the den site (see below), communal denning produces a large departure from panmixia. Over large areas, isolation of dif- ferent denning sites might ultimately be an im- portant contributing factor to differentiation within species (Gannon 1978). At the local level, 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- ing occurs away from the den when individuals from different hibernacula may come into con- tact (Gregory 1977a). Inter-den mating has been observed in some cases (Brown 1973; Brown and Parker 1976). Finally, there is no particular rea- son 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 popula- tions.— Sizes of overwintering aggregations of snakes have been reviewed by Klauber (1972). Parker and Brown (1973). and Gregory (1982). Most aggregations probably consist ofmuch few- er than 100 individuals of all species combined. Some denning populations, however, may in- clude a few to several hundred individuals of a given species (Criddle 1 937; Viitanen 1 967; Lang 1969; Klauber 1972; Parker 1976). The largest denning populations known are those of Tham- nophis sirtalis in Manitoba, where numbers at one den fluctuated between about 4000 and 8000 in a four-year period (Gregory 1977a). Sampling the different size age groups in a snake population in proportion to their relative abundance is difficult because young snakes are smaller and often more secretive than adults. Nevertheless, it seems clear that young-of-year and/or juveniles are frequently absent (Viitanen 1967; Gregory 1977a: Sexton and Hunt 1980) or greatly underrepresented (Hirth et al. 1969; 60 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Klauber 1972; Parker and Brown 1973, 1980; Parker 1976; Brown and Parker 1976) at com- munal hibernacula. Prestt (1971), however, in contrast to Viitanen's ( 1 967) observations of the same species ( Vipera berus), found young hiber- nating with the adults. In some small species, the 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- itat. If the young are born in the summer range a long way from the den, it may simply be too expensive energetically to make the journey if 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 their size. This is probably the case for Tham- nophis sirtalis in Manitoba, where the adults, but no young-of-year, hibernate in limestone sinks several kilometres from the summer habitat (Gregory and Stewart 1975; Gregory 1977a). Young of this species are known to hibernate communally with two species of small snakes in ant mounds in nearby Minnesota (Lang 1971); this is presumably what occurs in the Manitoba summer habitat, where ant mounds are abun- dant. Young born or hatched closer to the den. on the other hand, may be more likely to hiber- nate with the adults (but see Viitanen 1967). Gravid females of many snake species show a 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 1971; Gregory, unpubl. obs.; see section on "Communal Denning and Mating Behavior in Thamnophis"). Other examples of gravid fe- males occasionally being found at dens are given by Preston (1964) and Galligan and Dunson (1979). Parker and Brown (1980) argue that young snakes hibernate elsewhere as a defense against intra- and interspecific predation by adults (in this case, Masticophis taeniatus and Coluber constrictor). The hypothesis that there is some disadvantage to young snakes in hibernating with the adults is supported by the observation that most young Masticophis using communal dens do not survive to age one and most one-year- olds at dens are not known to have used the dens the previous year (Parker and Brown 1980). However, young Thamnophis sirtalis hibernate with adults at a communal hibernaculum in Brit- ish Columbia despite occasional predation on them by adult T. elegans at the same den (Greg- ory, unpubl. obs.). Quantitative assessment of these ideas awaits further study, but it is clear that there is an important ontogenetic change in hibernation behavior in many species of snakes. 4. Fall 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; 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., Lang 1971). In other cases, however, individual snakes may remain active in the vicinity of the den for a large part ofthe fall and/or spring period (e.g., Viitanen 1967; Prestt 1971; Gregory 1974; Parker and Brown 1980), usually without feed- ing. The advantages of remaining active above ground at the den, rather than seeking cooler conditions below ground, are not clear, since such activity is energetically very expensive (Parker and Brown 1980). The significance of activity at dens is obvious in cases where snakes mate at the den site or nearby (Viitanen 1 967; Prestt 1971; Gregory 1974, 1977a; Bennion and Parker 1976; Parker and Brown 1980). Some species, how- ever, apparently mate away from the den (Brown 1973; Brown and Parker 1976; Parker and Brown 1980). Fall mating occurs in some snakes (Trap- ido 1940; Rahn 1942; Saint Girons 1957; Greg- ory 1977a), but in most cases is only an occa- sional phenomenon and less intense than in spring, which is the major breeding season for most temperate zone snakes. Prolonged activity at dens in fall is therefore generally not explained by mating behavior. Perhaps spring and fall activity is related in other ways to the reproductive cycle. For ex- ample, male Vipera berus bask at dens in fall to promote spermatogenesis, which will be com- pleted during basking the following spring (Vol- soe 1944). Vipera species, however, are different from all other temperate zone snakes, in which spermatogenesis is completed well before hiber- VERTEBRATE ECOLOGY AND SYSTEMATICS 61 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 vi- tellogenesis in fall (Aldridge 1 979b), but presence or absence of this pattern has not been correlated with fall activity or lack of it. In females of all 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 den site for the advantage of readily available shelter at times when cold weather could arise suddenly. We do not yet know enough about details of reproductive cycles (and factors af- fecting them) or fall and spring activity periods of most snakes to be able to correlate these fea- tures. Why Do Snakes Den Communally? Certain disadvantages appear to be inherent in the habit of communal hibernation. First, an- 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 toll of Thamnophis sirtalis at dens in Manitoba in early spring when the ground vegetation cover is sparse (Gregory 1977a). In- dividuals hibernating singly at isolated locations would be much less 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- tively persecuted (Klauber 1972; Galligan and Dunson 1979). However, human collection is a relatively recent phenomenon and cannot be considered a long-term selective force.] Another possible disadvantage of communal denning is related to the fact that the den and summer hab- itat may be quite far apart. In such cases, snakes have to migrate, often through unfavorable hab- itat, expending energy and possibly exposing themselves to a higher risk of predation. The question therefore arises as to why snakes den communally. Very small aggregations of snakes (see examples in Parker and Brown 1 973) may simply be fortuitous and irregular in oc- currence, but large aggregations probably have a different basis. There are at least three possible reasons, not mutually exclusive, for communal denning: 1. low availability of suitable hibernat- ing 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 communal denning is that it may lead to more efficient utilization 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 com- munal denning is necessary for this system to operate and even if so, it is more likely to be a consequence of communal hibernation rather than a reason for its occurrence in the first place. Shortage of suitable hibernacula is undoubt- edly the main cause of communal overwintering in many cases. This argument has been used to explain winter aggregations of some lizards (Weintraub 1 968; Vitt 1974) and the rattlesnakes Crotalus viridis (Gannon 1978) and C. horridus (Brazaitis 1 980) and the occurrence ofmore than one species in large communal dens (e.g.. Car- penter 1953; Hirth el al 1969; Lang 1969). Smaller species of snakes may be less influenced 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- ited availability of hibernacula is expected to be particularly serious in cold climates, where hi- bernation at considerable depth is critical for sur- vival. This correlates well with the observation that communal denning is an especially well de- veloped phenomenon at higher latitudes. Gan- non (1978) feels that availability of hibernacula is an important factor limiting the distribution of Crotalus viridis in southern Saskatchewan and Alberta. On the other hand, several authors have noted that there may be apparently usable hi- bernacula which go unused in any winter, even at high latitudes (Viitanen 1967; Lang 1971; Klauber 1972; Gregory 1 977a). Lang (1971) con- cluded that availability of ant mound hibernac- ula was therefore not a limiting factor on num- bers ofthree species ofsmall 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 62 SPECIAL PUBLICATION- MUSEUM OF NATURAL HISTORY (Parker and Brown 1973). In addition, what ap- pears to be a suitable den to the observer may not be seen as such by snakes; we need to know more about what qualities make a good hiber- nating site and to assess these qualities at poten- tial sites before we can reach a conclusion re- garding availability of hibernacula. The argument that snakes hibernate commun- ally so that they can aggregate below ground and reduce heat loss is difficult to support. It was put forward by White and Lasiewski (1971). with particular reference to rattlesnakes. In favor of this idea is the observation that rattlesnake dens blasted open in winter sometimes reveal large masses of animals (KJauber 1972); however, such behavior could be due to disturbance. Aleksiuk (1977) has also shown that Thamnophis sirtalis tend to huddle under cold conditions, but there is no evidence that this actually happens in the den during hibernation. Snakes hibernating in communal dens are 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 iso- lated individuals do not differ in body temper- ature from grouped individuals (Brown et al. 1 974). Use ofenergy reserves during hibernation per se is probably very low in most cases (Parker and Brown 1980), consistent with the observa- tion that hibernation usually takes place at a low temperature (e.g.. Brown et al. 1974; Sexton and Hunt 1980; Brown 1982; Gregory 1982), not a high one; use ofenergy reserves may be very high during activity at dens in fall and spring, but this 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 ad- aptation of many hibernating reptiles seems to be that metabolism is 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 hi- bernation, then it is not surprising that snakes hibernate at fairly low temperatures, contrary to the predictions of White and Lasiewski (1971). Snakes hibernating with conspecifics at com- munal 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 that mating opportunities at communal sites should be frequent simply because of the large 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 en- hance mating chances since individuals dispers- ing from a small area should come into contact more often than when widely scattered (Parker and Brown 1 980). As in the case ofthe "refuging" hypothesis above, it may be argued that high probability of reproductive success is not a pri- mary reason for the occurrence of communal denning, but simply a secondary advantage of it. This problem is somewhat circular, however, since it is also possible that the prior evolution of early spring mating has resulted in selection for individuals that seek hibernacula used by conspecifics, or that the two have evolved jointly. We need studies which aim to unravel this ques- tion. 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 mat- ing behaviors to evolve which take advantage of this situation. An example is provided by the garter snake, Thamnophis sirtalis. In this species, the different mating strategies of the two sexes seem 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 commun- ally denning species which show different be- haviors. Communal Denning and Mating Behavior in Thamnophis The common garter snake, Thamnophis sir- talis, is the most widespread species of snake in North America. While this species does not den communally throughout its range, such behavior is well developed in the northern parts of 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 1 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 VERTEBRATE ECOLOGY AND SYSTEMATICS 63 are mainly limestone sinks, formed by the col- lapse of the ground surface into subterranean caves. The major den examined in this studs (Den 1 ) is a large, bowl-like depression about 20 m long x 12 m wide x 3 m deep; the bottom of the bowl is riddled with cavities leading under- ground. These dens occur on ridges between large marsh belts. Dens are abundant in such areas and are frequently less than 1 km apart. Popu- lations using dens may be very large. Den 1 is estimated to have housed as many as 8000 snakes during one winter, but population size fluctuates drastically from one year to another, apparently in response to variations in weather (Gregory 1977a). Den populations are exclusively adult (Fig. 1); it is not known where the young hiber- nate in this area. The summer habitat of these snakes is in the marshes between the ridges. In- dividuals may move as much as 18 km between den and summer range; migrations are unidirec- tional, with all animals moving south in summer despite the fact that suitable marshes are also found in other directions (Gregory and Stewart 1975). Despite these long migrations and the rel- ative closeness of dens to one another, homing success of individual snakes to the same den in successive years is about 96% (Gregory 1977a). The den is a central feature in the annual cycle of Interlake T. sirtalis. The hibernation period may be as long as six months; in addition, the fall and spring activity periods in the vicinity of the den may occupy up to 1 yh months each (Greg- ory 1977a). In extreme cases, therefore, individ- ual snakes may spend only three months away from the den during the year, and this is the only time in which feeding takes place (Gregory and Stewart 1975). Although occasional fall mating occurs, virtually all mating occurs at the den in spring after emergence (Gregory 1974, 1977a). Mating activity of males is apparently stimu- lated by the change from the cool conditions ex- perienced in hibernation to the warmer condi- tions above ground in spring (Aleksiuk and Gregory 1974; Hawley and Aleksiuk 1975; Crews and Gartska 1982; Gartska et al. 1982). Male courtship activity is directly related to temper- ature (Hawley and Aleksiuk 1975), but declines as the mating season progresses (Aleksiuk and Gregory 1974; Camazine et al. 1980). The tem- perature change associated with emergence from hibernation also stimulates sexual receptivity of females, but does not affect their attractivity to males (Licht and Bona-Gallo 1982). However, 300 380 460 540 620 700 780 860 SVL(mm) Fig. 1 . Size frequency distribution of T. sirtalis at Den 1 in fall 1972. Animals are grouped into intervals of 20 mm snout-vent length (SVL). n = sample size. Data from Gregory (1977a). Gartska 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; females which have warmed up may not be sex- 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 cold, active season. Time to reproduce and per- form other essential functions is therefore at a premium. Under such conditions, we should ex- pect the evolution of a mating system which maximizes the efficiency and success of mating for both sexes early in the season following hi- bernation. The two sexes have, in effect, different reproductive strategies: Males should mate as often as possible since this is their only means of increasing fitness; females, on the other hand, need mate only once per season (but see Gibson and Falls 1975 for evidence of multiple insem- ination 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 re- production. 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 situation. A reasonable substitute for a true con- 64 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. Numbers of individuals of each sex of T. sirtalis caught at Den 1 in fall and spring of four over- wintering seasons (data from Gregory 1977a). VERTEBRATE ECOLOGY AND SYSTEMATICS 65 7 SEPT. 13 19 25 OCT. 13 19 25 1 .8 6 Q- 4 .2 oo 4 V J ki .. II Witt iln I * i i 25 1 APR. 13 19 MAY 25 31 6 JUNE 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. daily samples does not vary greatly over the spring period (Fig. 2). In contrast to males, however, females apparently emerge throughout the spring period and spend little time at the den, dispersing very soon after emergence; road counts of dis- persing snakes indicate mostly females leaving early in the spring and increasing proportions of 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 weath- er is sometimes cool, females are courted as soon as they emerge, or even while emerging (Aleksiuk and Gregory 1974). Typically, many males si- multaneously court a single female, creating a writhing mating "ball'"' (Aleksiuk and Gregory 1974). Not surprisingly, the head-to-head ori- entation of male and female shown by many col- ubrid snakes is not required for successful court- ship in this species (Gillingham and Dickson 1980). Courtship and mating take several min- utes, but it is not usually possible to see which male manages to copulate with the female. This contrasts with the observations of others (e.g., Devine 1977), in which unsuccessful males leave before the successful male has finished copulat- ing. However, the numbers involved in mating activity in the Interlake are much larger than those reported elsewhere, obscuring actual cop- ulation. Mating almost always occurs on the ground, but males may follow females into low bushes and mate there (Gregory 1975b). Follow- ing copulation, the mating group breaks up rap- idly; the males seem to have no further interest in the female, which becomes unattractive for a day or more and even intolerant of further court- ship (Crews and Gartska 1982; Gartska et al. 1982), but turn to other emerging females in- stead. Devine ( 1 977) and Ross and Crews (1977) have shown that male garter snakes can distinguish between mature, non-mated and recently mated females and only court the former. The cues used are apparently pheromonal. The female attrac- tiveness pheromone is a non-volatile lipid, re- lated 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- 66 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 1.0 .8 .6 Q a u £ 1.0 .8 .6 .4 .2 SEPT. 13 19 25 OCT. 7 13 19 i 25 UL I flu 11 6 6 ii ii 25 1 APR. 7 13 19 MAY 25 31 6 JUNE Fig. 3. Proportion of recaptures from same season (P-recaps) in daily samples of T. sirtalis at Den 1 for fall 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 3- 30 (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. tected by the male via the vomeronasal system, and may or may not be the same as the trailing pheromone, which allows species-specific trail- ing of females by males and has its most pro- nounced effect during the spring mating period (Ford 1978, 1981, 1982; Ford and Low 1982). In any case, males are not sensitive to the female attractiveness pheromone early in the season when mating opportunities are very low; how- ever, as the season progresses, females become slightly more abundant relative to males, and males become sensitive to the pheromone and more discriminating about potential mates (Gartska et a/. 1982). Mated females are unattractive to males be- cause of a male-inhibiting pheromone. Follow- ing copulation, a plug forms in the cloaca of the female (Devine 1975); this copulatory plug is apparently manufactured in the renal sex seg- ment of the male (Crews and Gartska 1 982). The 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 1.0r .6 Q a u . 16 cm PL often lacked one or more annuli. they were excluded from this analysis. Statistical tests follow Ott (1977). Means are followed by ± one standard deviation. Results and Discussion Growth and Sexual Maturity. — One hundred thirteen Malaclemys were examined from the 77 78 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 80° 40' Fig. 1. Merritt Island National Wildlife Refuge. Shaded areas represent lagoonal waters. Indian River study site is shown by cross-hatching. Banana River study site by cross. VERTEBRATE ECOLOGY AND SYSTEMATICA 79 E o X 12- 8 - H 2 3 —r- 4 O cr h- 20 i < _J CL 16 12- 8 H T 3 ~r- 4 5 nr 6 AGE Fig. 2. Relationship between age and plastron length in 53 female and 13 male Malaclemys from the Indian River. Vertical bars represent sample range. Indian River and 44 from the Banana River. Fifty-three of the Indian River turtles bore dis- tinct 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 ofthe abdominal scute, such as that noted by Moll and Legler ( 1 97 1 ) for tropical 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 a particular age class observed in Malaclemys frequently occurs in other turtles (Gibbons 1 968: Ernst 1971. 1975. 1977: Plummer 1977b). Growth of the sexes is relatively constant and similar for the first 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 decline in growth as sexual maturity is reached (see below). Fig. 3 shows the relationship be- tween percent growth/year and plastron size. Most 80 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 100 -| £ 80 > LL! h- < DC o DC CD 60 - \- 40 - 20 - <> 4) II I) II T -L .L _L <> I 3.5 - 1 1 1 1 1 1 1 5.5 7.5 9.5 11.5 PLASTRON LENGTH (CM) 13.5 15.5 Fig. 3. Relationship between growth rate (%/year) and plastron length for Indian River Malaclemys. Vertical bars represent sample range. rapid growth occurs at PL 3-3.9 cm, followed by a sharp decrease, and then a more gradual decline in growth to <5%/year in mature individuals. This pattern is similar to that ofmost other fresh- water turtles, especially the genera Chrysemys and Pseudemys (see Bury 1979 for review). Lim- ited 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 Riv- er population (PL = 17.7 cm) would be approx- imately 15 years old. Longevity in this popula- tion 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 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 in ini- tial size is insufficient to account for the differ- ences in Fig. 4. Gibbons ( 1 967) showed that even local populations of Chrysemys picta varied widely in growth rates because of differences in local feeding habits and food quality. Most data suggest that the feeding habits ofMalaclemys are relatively similar throughout its range (Cagle 1952; Wood 1976; Hurd et al. 1979; R. Seigel, pers. obs.; but see Cochran 1976), with no com- parable dramatic differences such as Gibbons ( 1 967) 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 VERTEBRATE ECOLOGY AND SYSTEMATICS 81 14 12-| ? o E10 o z LLt _l Z 8- o cc l- (/) < c 14 3 4 AGE 5 J J 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 some- what affected. The differences in Fig. 4 may re- flect the longer activity and growing season of M. I. 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 (Hilde- brand 1932). No data on the activity season of Louisiana terrapins are available, but from a cli- matic viewpoint, it is probably more similar to Florida than North Carolina. The smallest female showing evidence of sex- ual 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- rity in Malaclemys was related to size rather than age, and my results support this idea. Table 1 shows the size and age at maturity for Malacle- mys from different parts of the range. Size at 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 north- temperate turtles, and made the following con- clusions: 1) males often mature earlier and at a smaller size than females; 2) growth is most rapid before maturity is reached; 3) in temperate re- gions, individuals in southern populations ma- ture earlier than northern conspecifics: 4) sexual 82 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 1. Size and age at sexual maturity for male and female Malaclemys terrapin from different parts of the range. Subspecies (locality ) VERTEBRATE ECOLOGY AND SYSTEMATICS 83 20-i 10 - \° o\ >O 2 LU D O LU DC BANANA RIVER N =44 r^ io - 84 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY >o 40 -s VERTEBRATE ECOLOGY AND SYSTEMATICS 85 barriers restraining movement). Plummer ( 1 977a) found that temporary movements of Trionyx muticus out of his Kansas study site greatly in- creased the variability of his population size es- timates. Although Malaclemys at Merritt Island showed relatively long-term (ca. 18 months) fi- delity to a particular area (Seigel 1979), it is prob- able that short-term movements took place at both study sites, so the above estimates may be somewhat 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- pling area was 1.62 acres, yielding a density of 131.1 individuals/acre. These figures are some- what higher than most reports ofdensity in fresh- water turtle populations (Bury 1 979), but are not 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 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 unreal- istically low estimate of the population bound- aries, and/or b) the tendency of Malaclemys to form large aggregations during the breeding sea- son (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. Summary The growth rates, age at maturity, population size and population structure of the Florida east coast terrapin, Malaclemys terrapin tequesta were studied from 1977 to 1979 at the Merritt Island National Wildlife Refuge, Brevard County, Flor- ida. Data from two areas (Indian and Banana rivers) are presented. Growth was most rapid immediately after hatching, declining to <5%/ year in mature turtles. Females matured at a plas- tron length of 13.5-14.0 cm, at an age of 4-5 years. Male terrapins reached maturity at a plas- tron 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 re- productive strategies between the sexes; females benefit from large body size via increased repro- ductive potential, whereas males attain only a small body size, but reach maturity earlier than females. The two study populations differed sig- nificantly in size structure, with the Banana Riv- er population having relatively more individuals in the larger size classes. This may reflect higher mortality among Indian River females. The sex ratios of both populations were significantly dif- ferent from 1:1, with females outnumbering males by at least 5:1. Schnabel population size esti- mates for the Indian and Banana rivers were 404.7 and 212.5, respectively, and it appears that Malaclemys may attain a considerable density and biomass in local areas. Acknowledgments Assistance in the field was provided by E. Scott Clark, Timothy R. Claybaugh, John D. Galluzzo, Mary T. Mendonca, Boyd Thompson, and Sher- ry Williams. Thanks go to the U.S. Fish and Wildlife Service for making field sites accessible, and for logistical support. I also thank Gunther Schlager for statistical advice. The critical re- views of William E. Duellman, Carl H. Ernst, Henry S. Fitch, and an anonymous reviewer im- proved this manuscript. Particular appreciation goes to my wife, Nadia, for constant help both in the field and during preparation of the manu- script, and to the late James D. Anderson for advice and encouragement. This research was supported by NASA contract NASI 0-8986, to L. M. Ehrhart. Literature Cited Allen, J. F. and Littleford, R. A. 1955. Observations on the feeding habits and growth ofimmature diamondback terrapins. Herpetologica, 11:77-80. Auger, P. J. and Giovannone. P. 1 979. On the fringe ofexistence. Diamondback ter- rapins at Sandv Neck. Cape Natur.. 8:44- 58. Berry. J. F. and Shine, R. 1980. Sexual size dimorphism and sexual selection in turtles (Order Testudines). Oecologia, 44: 185-191. Bull, J. J. and Vogt, R. C 1979. Temperature-dependent sex determination in turtles. Science, 206:1 186-1 188. Burger. J. 1976a. Behavior of hatchling diamondback terra- pins (Malaclemys terrapin) in the field. Co- peia. 1976:742-748. 86 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 1976b. Temperature relationships in nests of the northern diamondbaek terrapin. Malacle- mys terrapin terrapin. Herpetologica, 32:4 1 2- 418. 1977. Determinants of hatehing success in dia- mondbaek terrapin, Malaclemys terrapin. Amer. Midi. Natur.. 97:444-464. Burger, J. and Montevecchi, W. A. 1975. Nest site selection in the terrapin Malacle- mys terrapin. Copeia. 1975:1 13-1 19. Burnley, J. M. 1969. Diamondbaek terrapin. Int. Turtle and Tor- toise Soc. J.. 3:32-34. Bury. R. B. 1979. Population ecology of freshwater turtles. Pp. 571-602. In Harless, M. and Morlock, H. (eds.), Turtles, Perspectives and Research. Wiley-Interscience, New York. Cagle. F. R. 1952. A Louisiana terrapin population (Malacle- mys). Copeia, 1952:74-76. Carr, A. F. 1952. Handbook of turtles. Comstock Publ. As- soc, Ithaca. 542 p. Cochran, J. D. 1978. A note on the behavior of the diamondbaek terrapin, Malaclemys t. terrapin (SchoepfF) in Maryland. Bull. Md. Herpetol. Soc, 14: 100. CONANT, R. 1975. A field guide to reptiles and amphibians of eastern and central North America. Hough- ton Mifflin Co., Boston. 429 p. Ernst, C. H. 1971a. Growth of the painted turtle, Chrysemys pic- ta. in southeastern Pennsylvania. Herpe- tologica, 27:135-141. 1971b. Population dynamics and activity cycles of Chrysemys picta in southeastern Pennsyl- vania. J. Herpetol., 5:151-160. 1975. Growth of the spotted turtle, Clemmvs gut- tata. 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. 1 974. A new coding system for hardshelled turtles. Trans. Kentucky Acad. Sci., 35:27-28. Finneran, L. C. 1948. Diamond-back terrapin in Connecticut. Co- peia, 1948:138. Fitch, H. S. 1981. Sexual size differences in reptiles. Misc. Publ., Mus. Natur. Hist., Univ. of Kansas 70: 1-72. Gibbons, J. W. 1967. Variation in growth rates in three popula- tions of the painted turtle, Chrysemys picta. Herpetologica. 23:296-303. 1 968. Population structure and survivorship in the painted turtle, Chrvsemvs picta. Copeia, 1968:260-268. 1970. Sex ratios in turtles. Res. Popul. Ecol., 12: 252-254. Graham, T. E. 1979. Life history techniques. Pp. 73-95. In Har- less, M. and Morlock. H. (eds.). Turtles, Per- spectives and Research. Wiley-Interscience, New York. HlLDEBRAND, S. F. 1929. Review of experiments on artificial culture of diamond-back terrapin. Bull. U.S. Bur. Fish., 45:25-70. 1932. Growth of diamondbaek terrapins size at- tained, sex ratio and longevity. Zoologica, 9: 551-563. Hurd, L. E., Smedes, G. W. and Dean, T. A. 1979. A ecological study of a natural population of diamondbaek terrapins (Malaclemys t. terrapin) in a Delaware salt marsh. Estuaries, 2:28-33. Jackson, C. G., Jr. and Ross, A. 1971. Molluscan fouling of the ornate diamond- back terrapin, Malaclemys terrapin macro- spilota. Herpetologica, 27:341-344. Jackson, C. G., Jr., Ross, A. and Kennedy, G. 1973. Epifaunal invertebrates of the ornate dia- mondbaek terrapin, Malaclemys terrapin macrospilota. Amer. Midi. Natur., 89:495- 497. Lawler, A. R. and Musick, J. A. 1972. Sand beach hibernation by a northern dia- mondbaek terrapin, Malaclemys terrapin terrapin (SchoepfF). Copeia, 1972:389-390. Moll, E. O. and Legler, J. M. 1971. The life history of a neotropical slider turtle, Pseudemys scripta (SchoepfF). in Panama. Bull. Los Angeles Co. Mus. Natur. Hist. 1 1: 1-102. Montevecchi, W. A. and Burger. J. 1975. Aspects of the reproductive biology of the northern diamondbaek terrapin. Malacle- mys terrapin terrapin. Amer. Mildl. Natur., 94:166-175. Nichols, J. D. and Chabreck, R. H. 1980. On the variability of alligator sex ratios. Amer. Natur.. 116:125-137. Ott, L. 1977. An introduction to statistical methods and data analysis. Duxbury Press. Belmont. 730 p. Overton, W. S. 1969. Estimating the numbers of animals in wild- life populations. Pp. 403-455. //; Giles, R. (ed.). Wildlife Management Techniques. Wildlife Soc Washington. Plummer, M. V. 1 977a. Activity, habitat and population structure in the turtle, Trionyx muticus. Copeia. 1977: 431-440. 1977b. Reproduction and growth in the turtle, Tri- onyx muticus. Copeia, 1977:441-447. Reid, G. K. 1955. Reproduction and development in the northern diamondbaek terrapin, Malacle- mys terrapin terrapin. Copeia, 1955:310-31 1. VERTEBRATE ECOLOGY AND SYSTEMATICA 87 Ross, A. and Jackson, C. G.. Jr. 1972. Barnacle fouling of the ornate diamondback terrapin, Malaclemys terrapin macrospilota. Crustaceana, 22: 203-205. Seigel, R. A. 1 978. Simultaneous mortality in the diamondback terrapin. Malaclemys terrapin tequesta Schwartz. Bull. N.Y. Herpetol. Soc, 14:31- 32. 1979. 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 ter- rapins, Malaclemys terrapin tequesta. J. Herpetol., 14:87-89. 1980b. Courtship and mating behavior of the dia- mondback terrapin, Malaclemys terrapin te- questa. J. Herpetol., 14:420-421. 1980c. Nesting habits of diamondback terrapins (Malaclemys terrapin) on the Atlantic Coast of Florida. Trans. Kansas Acad. Sci., 83:239- 246. 1983. Occurrence and effects of barnacle infesta- tions on diamondback terrapins (Malacle- mys terrapin). Amer. Midi. Natur.. 109:34- 39. Sergeev, A. 1937. Some materials to the problem of reptilian post-embryonic growth. Zool. J. Moscow, 1 6: 723-735. Sexton, O. J. 1959. A method for estimating the age of painted turtles for use in demographic studies. Ecol- ogy. 40:716-718. Wood, R. C. 1976. $25 per egg. N.J. Outdoors. 3:14-15. 26. Yearicks, E. F., Wood, R. C. and Johnson, W. S. 1981. Hibernation of the northern diamondback terrapin, Malaclemys terrapin terrapin. Es- tuaries, 4:78-80. Yntema, C. L. 1979. Temperature levels and periods of sex de- termination during incubation of eggs of Chelydra serpentina. J. Morphol., 159:17- 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 Introduction The cricket frog, Acris crepitans, is a useful subject for population studies since it is generally abundant throughout most of the year and tends to form separate and distinct populations. It oc- curs in a variety of habitats, such as along lakes, pond, rivers, streams and occasionally tempo- rary ponds or rain pools and even relatively dry stretches of intermittent streams. Most Acris ag- gregate 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 ofalgae 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 Py- burn (1958, 1961a, 1961b), Blair (1961), Fer- guson et al. (1967), Bayless (1969b), Labanick ( 1 976), and Johnson and Christiansen ( 1 976), on A. crepitans. Nevo (1973a, 1973b) has studied both species and Bayless (1969a) studied sym- patric populations of both species. Some comparisons were made with popula- tions of Acris from other locations, but the main objective of this study was to determine if there were any differences in the ecology of popula- tions of cricket frogs in different habitats sepa- rated by only a few kilometers. 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 Kansas Natural History Reservation (KUNHR). about 1 1 km northeast of Lawrence, and the oth- er in an open reservoir and 1 1 rectangular ponds at the University Fish Laboratory (FL) on the southwestern part of the campus. The Kansas River lies between the two populations as a pos- sible barrier to gene flow. The pond at the Reservation was created by the construction of an earthen dam impounding water on its northeastern side. Water overflowing the pond drains down a stream to the southwest and into a small creek that empties into the Kan- sas River about four km east of Lawrence. The pond has a maximum circumference of about 435 m. The northeastern end of the pond is shal- low and swampy with numerous willows (Salix) along its edge. Honey locust (Gleditsia triacan- 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- es. 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, 1965) and Fitch and McGregor (1956). At the Fish Lab the reservoir is on a south- facing slope and the 1 1 ponds are located about 90 m south ofthe reservoir. Each pond is drained through pipes that empty into a small stream south of the ponds. The stream continues south until it reaches the Wakarusa River, which enters the Kansas River about 1 1 km east of Lawrence. 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 saplings of Populus and Salix, which occur in about equal numbers. Methods A total of 2492 frogs were captured at the Nat- ural History Reservation, and 1077 were cap- tured at the Fish Laboratory. Owing to the large numbers of individuals that were sometimes present, frogs were marked serially rather than individually. Areas (not exceeding 100 m long and 4.5 m wide) were marked off at each locality. All individuals captured in each area were given a unique mark for that area and date. Frogs cap- tured during the initial sampling period and sub- 89 90 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 1. Estimated reproductive output for.4cris cre- pitans in two populations in northeastern Kansas. See text for explanation. Location VERTEBRATE ECOLOGY AND SYSTEMATICS 91 35 361 288 54 148 JO 30h U) c -J 25 890 142 220 C (D ^ 20 +-» =J o " 151 47 ||. II ii 10 L 274 65 J I69 n II ' J J 1 32 33ri i \ 91 u ir J- 127 J L J LJASON MAM J J KUNHR 35 141 T 97 56 JZ 30 -4-> U) c I- >-OO CO -8 _ < Fig. 1. Opheodrys aestivus: Number of various sized follicles and amount of body fat in mature different times of the year. The horizontal lines of the dicegrams are means; vertical lines are ranges: delimit 95% confidence limits. OE = oviducal eggs. females at rectangles were either partially or totally unshelled. For totally shelled eggs the modal embryonic stage (Zehr 1 962) at oviposition was 25 (N = 37, range 21-27). The range of stages in a single clutch of 7 eggs was 25-26. For partially shelled eggs or those that were inviable at or soon after ovipo- sition (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 given in Table 5. Incubation ranged 36-43 days and averaged Table 3. Opheodrys aestivus: Number of snakes (>35.0 cm SVL) determined to be gravid by palpation in the field during various times in 1978. Period 108 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY i.o- X Ll) >QO CD X UJ X o I- _J ° 35 40 45 50 SNOUT-VENT LENGTH (cm) Fig. 2. Opheodrys aestivus: The relationship of clutch weight/post-reproductive body weight and snout-vent length for snakes collected in 1979. The regression equation is Y = .00 IX + .588 (r = .024, P > .75). 1.36 ± 0.23 g; / = .01, P > .90) and have about the same SVL (13.9 ± .10, 14.0 ± .10 cm; / = .02, P > .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 non- reproductive female weight (C/B) (Clark 1970; Fitch 1975; Pianka and Parker 1975; Shine 1977). Because in O. aestivus there is no parental care (egg brooding or oviducal retention) most of the reproductive investment should be contained in the egg itself and therefore the ratio C/B should be representative of reproductive effort (but see Vitt 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 variation in C/B is explained by body size. Larger snakes produce both larger eggs and larger clutch- es (Fig. 3). There is a possible trend toward small- er eggs with increasing clutch size (Fig. 4) al- though there is great variation (r 2 = 1.7%) and the regression is 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 ovipa- rous snake. From the limited data available for a comparison of geographic variation in repro- ductive attributes, other reports appear to con- form with this population. In southern Louisiana Tinkle ( 1 960) and in Illinois Morris (1982) found similar results in O. aestivus with regard to size at sexual maturity, the ovarian cycle, and repro- ductive potential. Apparently, mating is limited to spring in this population although fall mating may occur in other populations (Richmond 1956). Table 5. Opheodrys aestivus: Egg and hatchling statistics. All data are expressed as x ± 1 SE (N); range. Wgt. (g) Max. width (cm) Max. length (cm) SVL (cm) Shelled eggs Hatchlings 1.62 ± .015(190); 1.17-2.26 1.37 ± .016(144); .82-1.76 9.9 ± .04(190); 8.4-11.9 24.8 ± .23(190); 16.2-34.2 13.9 ± .07(144); 10.7-16.1 VERTEBRATE ECOLOGY AND SYSTEMATICS 109 10- LiJ 8M CO 5 6 h- _l o 4 r-X C5 UJ 5J < LxJ 2 - 1.8 - .6- w 1.4 1.2- _ 110 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY VERTEBRATE ECOLOGY AND SYSTEMATICS 11 1.4 16 1.8 EGG WEIGHT (g) 2.0 Fig. 5. Opheodrvs aestivus: The relationship of hatchling weight and egg weight. The regression equation is Y = .79X + .115 (r= .85. P < .001). frequency of female reproduction in several snakes (summarized in Wharton 1966; Gibbons 1972). Production of equal numbers of male and fe- male 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 1977). Because metabolism decreases with body weight in snakes (Galvao et al. 1965) propor- tionally more energy may be available for repro- duction. The risks involved when time and en- ergy are allocated to reproduction may decrease survivorship and therefore the expectation of fu- 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 Carpho- phis vermis (Clark 1 970) C/B increases with body size(=age). In Diadophis punctatus (Fitch 1975), Masticophis taeniatus (Pianka and Parker 1975), and O. aestivus (present study) C/B remains con- stant with body size. In Notechis scutatus and Pseudechis porphyriacus (Shine 1977) C/B de- creases with body size. Pianka and Parker ( 1 975) and Pianka ( 1976) suggested that correlations be- tween reproductive effort and reproductive value might be greater in multiple-brooded species than in single-brooded species where proximal factors such as resource availability might have a greater effect. In all of the above studies the snakes were single-brooded. However, in a study of annual reproductive variation in O. aestivus (Plummer 1983) it was shown that C/B and other repro- ductive attributes did not vary between years in which snakes stored greatly different quantities of lipids. Even if reproductive effort remains con- stant with age (as in O. aestivus), the absolute energy allocated to reproduction actually in- creases. The increased energy available in O. aes- tivus is reflected in the production of larger eggs and larger clutch sizes (Fig. 3). Fecundity in snakes is often related to body size (Fitch 1970; Shine 1977; Aldridge 1 979; present study). Shine ( 1 978) found that in about 66% of species (including O. 112 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY aestivus) females attain a larger body size than males. Shine suggested that one reason for this disparity was that selection has favored large body sizes in the females because of greater fecundity. Another reason for increased body size might be that larger snakes produce larger eggs which pro- duce larger hatchlings (Fig. 5). In general, larger hatchlings should enjoy higher survivorship and 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 for snakes, if female fitness was increased by pro- ducing larger eggs then selection should favor either larger parental body size or decreased clutch sizes (Pianka 1976). Smith and Fretwell ( 1974), Pianka (1976), and Stewart ( 1979) discuss models which predict that with a constant reproductive effort, an increased female size may result in either larger clutches or larger sized eggs. These models assume a neg- ative correlation between clutch size and egg weight. In O. aestivus there is no statistical re- lationship between clutch size and egg weight (Fig. 4). Although the correlation between SVL and egg size is not strong (/' = .46), it appears that in this population correlates of female body size are selection for increased clutch size as well as for increased egg size. Summary 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- ture at 36-40 cm SVL and breed annually there- after. Ovarian follicles measuring 1-5 mm in di- ameter are present in mature snakes throughout the year. Rapid yolking of follicles occurs in the spring and ovulation begins in late May. Extra- uterine transfer of ova is common. One clutch is produced per year (.v = 6. 1 eggs). Coelomic fat bodies cycle annually and presumably provide energy for vitellogenesis and ovulation. Ovipo- sition occurs in late June and July. Ninety per- cent ofthe eggs laid were fertile and 90% of fertile eggs hatched in the laboratory. At oviposition the modal embryonic stage (Zehr 1962) was 25 (range 21-27). Incubation averaged 39 days. The sex ratio of hatchlings was not significantly dif- ferent from 1:1. Male and female hatchlings are similar in length and weight. Reproductive effort (jc = .64) did not change with body size. Larger females produce both larger clutches and larger 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. Pat- terson, 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 made helpful suggestions regarding the manu- script. M. Groves and J. Huckeba willingly typed the numerous versions of the manuscript. I owe much to H. S. Fitch who stimulated and refined my interests in living organisms under natural conditions. To him this paper, and this entire volume, is rightfully dedicated. This study was supported in part by grants from Sigma Xi and Harding University. Literature Cited Aldridge, R. D. 1 979. Female reproductive cycles of the snakes Ar- izona elegans and Crotalus viridus. Herpe- tologica, 35:256-261. Anderson, P. 1965. The Reptiles of Missouri. Univ. Missouri Press. 330 p. Blanchard, F. N. 1933. 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 ofOkla- homa snakes. Herpetologica, 14:1 13-1 15. Clark, D. R. 1970. Age-specific "reproductive effort"' in the worm snake Carphophis vermis (Kennicott). Trans. Kansas Acad. Sci., 73:20-24. Conant, R. 1938. The Reptiles of Ohio. Amer. Mid. Nat., 20: 1-200. Conant, R. 1975. A field guide to reptiles and amphibians of eastern and central North America. 2nd ed. Houghton Mifflin Co., 429 p. Conant, R. and Downs, A., Jr. 1940. Miscellaneous notes on the eggs and young of reptiles. Zoologica, 25:33-48. VERTEBRATE ECOLOGY AND SYSTEMATICS 113 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. DHEW Publ. No. 77-491. 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., 62:1- 53. Galvao, P. E., Tarasantchi, J. and Guertzen- STEIN, P. 1965. Heat production of tropical snakes in rela- tionship to body weight and body surface. Am. J. Physiol., 209:501-506. Gibbons, J. W. 1972. Reproduction, growth and sexual dimor- phism in the canebrake rattlesnake (Cwtalus horridus atricaudatus). Copeia, 1972:222- 227. Guidrv, E. V. 1953. Herpetological notes from Southeastern Texas. Herpetologica, 9:49-56. McCauley, R. H., Jr. 1945. The Reptiles of Maryland and the District ofColumbia. Hagerstown. Maryland, 194 p. Morris, Michael A. 1982. Activity, reproduction, and growth of Opheodrys aestivus in Illinois (Serpentes: Colubridae). Nat. Hist. Misc., 214:1-1 1. Mount. R. H. 1975. 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 and embryos, and the evolution of viviparity within the class Reptilia. Biol. Rev.. 52:71- 105. PlANKA, E. R. 1976. Natural selection of optimal reproductive tactics. Am. Zool.. 16:775-784. Pianka, E. R. and Parker, W. S. 1975. Age-specific reproductive tactics. Amer. Nat., 109:453-464. Plummer, M. V. 1981. Habitat utilization, diet and movements of a temperate arboreal snake (Opheodrys aes- tivus). J. Herpetol., 15:425-432. 1983. Annual variation in stored lipids and repro- duction in green snakes (Opheodrys aesti- vus). Copeia. 1983:741-745. Richmond, N. D. 1956. Autumn mating 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., 25:655-666. 1978. 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: 228-234. Smith, C. C. and Fretwell, S. D. 1 974. The optimal balance between size and num- ber of offspring. Am. Nat., 108:499-506. Smith. P. W. 1961. Amphibians and Reptiles of Illinois. Bull. Illinois Nat. Hist. Survey, 28:1-298. Stewart, J. R. 1979. The balance between number and size of young in the live bearing lizard Gerrhonotus coeruleus. Herpetologica, 35:342-350. Tinkle, D. W. 1960. A population of Opheodrys aestivus (Reptil- ia: Squamata). Copeia, 1960:29-34. Turner, F. B. 1977. The dynamics of populations of squamates, crocodilians. and rhynchocephalians. Pp. 157-264. In Gans, C. and Tinkle, D. W. (eds.). Biology of the Reptilia. Academic Press. Vitt, L. J. and Congdon, J. D. 1978. Body shape, reproductive effort, and relative clutch mass in lizards: Resolution of a par- adox. Amer. Nat. 112:595-608. Webb, R. G. 1970. Reptiles of Oklahoma. Univ. Oklahoma Press, 370 p. Wharton, C. H. 1966. Reproduction and growth in the cotton- mouth Agkistrodon piscivorus Lacepede. of Cedar Keys, Florida. Copeia, 1 966: 149-1 6 1 . Wright, A. H. and Wright, A. A. 1957. Handbook of Snakes. Vol. I. Comstock Publ.. 564 p. Zehr. D. R. 1962. Stages in the normal development of the common garter snake, Thamnophis sirtalis 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 in Central Panama A. Stanley Rand Introduction Iguana iguana, laying up to 6 dozen eggs in a single clutch, is one of the most prolific lizards in the new world. Prized as food and heavily hunted in many parts of its range, its conserva- tion and the possibilities for sustained yield har- vesting have been discussed (Fitch et al. 1983). Though reproductive potential is important in any understanding of population dynamics, only one detailed study of the number of eggs which female iguanas produce has been published. Fitch and Henderson (1977) for Nicaragua. The pres- ent paper describes the size and weight of clutch- es produced by female iguanas and their relation to female size in Panama. The clutch size and reproductive investment in Iguana iguana is compared with that described for other lizards. Materials and Methods Clutch size data were collected from females caught during the nesting season, late January to early March, between 1968 and 1980 in the vi- cinity of Panama City and Gamboa, Republic of Panama. Some females were caught and allowed to nest in a large 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 be- tween female size and clutch size, the few females deliberately shot were selected because they were very large or very small. The following mea- surements were taken on 30 females: snout-vent length (SVL), female wet weight without eggs, clutch wet weight before significant hydration and clutch size (number of eggs). In some cases clutch volume, clutch dry weight (oven dried at 105°C) were also measured. Regressions were compared using covariance (Snedecor 1956). Results The results from examination of females with oviductal eggs are given in Table 1 . Sizes of these animals were not distributed normally nor do they suggest that the population can be divided into age classes. Even though animals were col- lected to emphasize the extremes, only one fe- male was below 300 mm SVL. Because sampling was not random, ranges are probably more ac- curate representations of the population than are means. Female weight is closely correlated with SVL (N = 30, /• = 0.86. P < .001) (Fig. 1) par- ticularly if logs of both are plotted (N = 30, r = 0.91, P < .001). Number of eggs per clutch ranged from 9 to 71 (N = 30, mean = 40.6) and was closely pos- itively correlated with female size. The correla- tion of egg number with SVL (N = 30, r = 0.78, P < .001) (Fig. 2) is about equal to that with female weight (N = 30, r = 0.79. P < .001). A better predictor of the number of eggs that a female will lay, and one that can be used in the field, is her weight before she has laid her eggs (i.e., her own body weight plus the weight of her clutch) (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). Mean egg weight per clutch (clutch weight/num- ber of eggs) ranged from 9.3 to 16.0 g (N = 28, mean = 13.1). Eggs within a clutch appear quite uniform in size. Larger females tended to lay larger eggs but the correlation ofmean egg weight to SVL, though significant (N = 28, r= 0.55, P < .01), is not as high as the correlations already cited. The single very small female with her very- small eggs contributes greatly to this correlation; if she is excluded the correlation is lower (N = 27. /• = 0.35. .01 < P < .05). The water content of the eggs varied little, 57- 67% (N = 11, mean = 62.3%). There was no sig- nificant correlation of water content either with egg weight or female size. The percentage that the clutch contributed to the combined weight of female and clutch (rel- ative clutch mass of Vitt and Congdon 1978) ranged from 19.7 to 39.9% (N = 28. mean = 30.3%). It shows a weak positive correlation with female SVL (N = 28. r = 0.30. .01 < P < .05) which depends heavily on the single small fe- 115 116 SPECIAL PUBLICATION- MUSEUM OF NATURAL HISTORY Table 1. Clutch VERTEBRATE ECOLOGY AND SYSTEMATICS 117 2000 1000 — 400 - J_ ±. ± X X v x X x x x x x X x x L 250 300 350 SNOUT VENT LENGTH, IN MM WO 450 Fig. 1. The relationship between weight and snout-vent length in Panamanian iguanas. so that females of the same weight in two pop- ulations lay about the same number of eggs but have a lower total clutch weight in Nicaragua. The relative clutch mass and the investment per egg is not significantly different between the two populations. No other equally extensive data on Iguana iguana clutch sizes has been published. Hirth ( 1 963) reported SVL and clutch size for 7 females from Tortuguero, Costa Rica (Fig. 2). These clutch sizes are intermediate between those of Pana- manian 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. Hoogmoed ( 1973) reports clutch sizes from 24 to 57 in Surinam. Muller (1972) reports clutches from 14 to 70 at Santa Marta, Colombia, and egg weights aver- aging 13.0 g (12.4-14.0). Detailed comparison with South American populations awaits more data. Wiewandt (1983) has compared reproductive patterns among iguanine lizards. He distin- guished three groups of genera on ecological grounds: 1) those in mainland deserts of tem- perate North America (Dipsosaurus and Sau- romalus); 2) those on dry subtropical islands (Cy- clura); 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. In Iguana iguana, though its eggs are small relative to female size, the weight of its total clutch, relative to that of the female, is about the same as it is for the three other iguanine species for 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 within the iguanines in having very few, very- large eggs with a high investment per offspring but a low investment per clutch (Carpenter 1 966). Wiewandt attributes the reproductive pattern in Iguana iguana and Ctenosaura similis to the relatively high predation pressure on young liz- ards 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 118 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 80 60 CO U3 5/|0 LLl 00 20" + X X VERTEBRATE ECOLOGY AND SYSTEMATICS 119 HO.. 30-- oo CD £20-- 10- ++ *t + +4 + + + lol — ' ' ' — 260 — ' "" SNOUT VENT LENGTH AT MATURITY, IN 300 Fig. 4. The relationship between clutch size and minimum snout-vent length at maturity. + = single brooded iguanas (Tinkle et al. 1970), 1 = J 'vblyrhynchus cristatus, (Carpenter 1966), 2 = Sauromalus obesus, 3 = Cten- osaura similis, 4 = Cyclura cahnata, 5 = Cyclura coronuta stejnegeri (2-5, Wiewandt 1983), 6 = Iguana iguana from Panama, 7 = Iguana iguana from Nicaragua (Fitch and Henderson 1977). the second group. In contrast to the largely trop- ical 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 single brooded group as well as the data for Igua- na iguana from Panama and Nicaragua and 4 other iguanines. As Wiewandt noted, Ambly- rhynchus cristatus, Cyclura cahnata and C. co- ronuta stejnegeri and Sauromalus obesus have few eggs for their size. Fig. 4 shows that Cteno- saura 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 we do have some data (Brachylophus faciatus, Sauromalus varius and S. hispidus, Conolophus subcristatus, and Iguana delicatissima) all seem to 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 parameter in a reproductive strategy. It is not, however, a very good index of reproductive effort (Tinkle and Hadley 1975; Vitt and Congdon 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 Ballin- ger and Clark (1973) and Vitt (1978) have shown that calorific content per unit weight is quite con- stant for the eggs of a variety of numbers of liz- ards. Vitt (1978) has shown that the ratio of ca- lorific content of clutch to calorific content of female is similar to the ratio of wet weight of clutch to wet weight of female. That our ratio of dry to wet weight of iguana eggs lies within the range that Vitt reported for other lizards is sup- port for our assumption that iguana eggs are probably not too different from other lizards in calorific 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 Cong- don (1978, Table 2) give for 17 much smaller North American iguanid lizards. Not surprising- 120 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. A comparison between clutch sizes of Iguana iguana from Panama and Nicaragua. VERTEBRATE ECOLOGY AND SYSTEMATICS 121 Table 2. Continued. 122 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Conservation. Garland STPM Press. New York. Hikih. H. F. 1963. Some aspects of the natural history of Igua- na iguana on a tropical strand. Ecology. 44(3): 613-615. HOOGMOED, M. S. 1973. Notes on the herpetofauna of Surinam. IV. The lizards and amphisbaenians ofSurinam. W. Junk. The Hague. I-IX: 14-19 pp. MlLLER, H. 1972. Ukologische und ethologische studien an Iguana iguana L. (reptilia: Iguanidae) in Ko- lumbien. Zoologische Bertrage N. F. 1 8: 109— 131. Pianka, E. R. and Parker. W. S. 1 975. Age-specific reproductive tactics. Amer. Nat.. 109:453-464. Rand. W. M. and Rand, A. S. 1 976. Agonistic behavior in nesting iguanas: a sto- chastic analysis of dispute settlement dom- inated by the minimization of energy cost. Z. Tierpsychol. 40:279-299. Snedecor, G. W. 1956. Statistical methods. Iowa State Univ. Press, Ames. Iowa. pp. 1-534. Tinkle. D. W., Wilbur, H. M. and Tillev, S. G. 1970. Evolutionary strategies in lizard reproduc- tion. Evolution, 24(l):55-74. Tinkle, D. W. and Hadlev, N. F. 1975. Lizard reproductive effort: caloric estimates and comments on its evolution. Ecology, 56: 427-434. Vim, L. J. 1978. Caloric content of lizard and snake (Reptilia) eggs and bodies and the conversion ofweight to caloric data. J. Herpetol., 12:65-72. Vitt, L. J. and Congdon, J. D. 1978. Body shape, reproductive effort, and relative clutch mass in lizards: resolution of a para- dox. Amer. Natur.. 1 12(985):595-608. WlEWANDT, T. A. 1 983. Evolution of nesting patterns in iguanine liz- ards. In Burghardt, G. M. and Rand, A. S. (eds.). Iguanas of the World: Behavior, Ecol- ogy and Conservation. Garland STPM Press, New York. Williams, G. C. 1966. Adaptation and natural selection. Princeton University Press, Princeton, New Jersey, pp. 1-307. Vertebrate Ecology ami Systematics—A Tribute to Henry S Fitch Edited by R. A. Seigel, L. E. Hum. I I .. Knight. I Malaret and N. L. Zuschlag 1984 Museum of" Natural Histoiv I he I 'inversus of Kansas. Lawrence Are Anuran Amphibians Heavy Metal Accumulators? Russell J. Hall and Bernard M. Mulhern Introduction Concern about heavy metals in the environ- ment has increased recently, partly as a result of increased awareness of their potential effects, and also because of the prospect of expanded use of fossil fuels in 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 1956) indicated extremely high levels (up to 1 640 ppm) in Bufo marinus; average copper concen- trations in livers were generally much lower in other species, although one species of marine fish had higher average concentrations. Surprisingly high concentrations of lead in the livers of some frogs from a remote and apparently uncontam- inated area were reported by Schroeder and Tip- ton (1968). Gale el al. (1973) found up to 1590 ppm (dry weight) of lead in tadpoles from a con- taminated area and these results suggest that they have a much greater ability to concentrate en- vironmental lead than do the other species sam- pled. The iron content of one sample of Rana catesbeiana tadpoles analyzed in our own labo- ratory reached the startling level of 19.000 ppm (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 information from the literature in order to ascertain whether the con- centrations of metals in amphibians fall outside the normal range of variation of other animals. This paper will assemble data which may bear on the questions of whether amphibians are par- ticularly susceptible to 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- tamination. Methods Adult amphibians collected at the Patuxent Wildlife Research Center were generally ob- tained from the Island Marshes, artificial habi- tats built for waterfowl management, or they were picked up on service roads on wet nights. Larval amphibians and fish were captured by seine or dip net from Harding Spring or Mabbott ponds: both are shallow, moderate-sized artificial ponds surrounded by wooded areas. The research cen- ter is 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 mon- itoring organochlorine and heavy metal levels in certain animals found in different areas of the center. A sample of 10 leopard frogs (Rana pip- iens) obtained from the National Fish and Wild- life Health Laboratory, Madison. Wisconsin was also analyzed for heavy metal levels. Tissue samples were homogenized in a blender and a 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 deter- mine mercury levels. Digestion for mercury anal- ysis used the method described by Monk (1961). Mercury was determined by cold vapor atomic absorption spectrophotometry using the method of Hatch and Ott (1968) with a Coleman model 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 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 of concentrated nitric acid and 1 ml of concen- trated hydrochloric acid, transferred to a 50 ml polypropylene centrifuge tube, and diluted with dionized water. Analysis was by flame atomic absorption spectrophotometry with a Perkin-El- mer model 703 equipped with a deuterium arc background corrector, an AS-50 autosampler. and a PRS-10 printer. The lower limit of reportable 123 24 SPECIAL PUBLICATION -MUSEUM OF NATURAL HISTORY Table 1. Cadmium in amphibians. VERTEBRATE ECOLOGY AND SYSTEMATICA 125 Table 2. Copper in amphibians. PPM Sample Area Tissue Wet weight Dp. ueight Reference Tadpoles (sp.) 126 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Tabi f 3. Lead in amphibians. VERTEBRATE ECOLOGY AND SYSTEMATICA 127 Table 4. Mercury in amphibians. 128 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 5. Zinc in amphibians. VERTEBRATE ECOLOGY AND SYSTEMATICS 129 Table 6. Other elements in amphibia. 130 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY and Pickering 1958). Pendleton and Hanson (1958) looked at cesium- 137 uptake in a variety of organisms following addition of 6 pCi/ml of the radionuclide to the water of a concrete-lined pond. They analyzed concentration factors after approximately 90 days when levels of Cs-137 in the water had apparently stabilized. They found high concentrations in bullfrog tadpoles, with the bulk of the material stored in the gut fraction. Still higher levels were found in spadefoot toad (Scaphiopus hammondi) tadpoles and adult bull- frogs. Comparison ofAnuran 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 Sr- 85, body remainders still had greater accumu- lations 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 main- tained 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 experi- mental 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 settle to the bottom. Tadpoles consume them there and accumulate high levels before various processes had distrib- uted the contaminants more generally through- out the the system. Support for the assertion that feeding habits rather than physiological factors produced the high levels observed in tadpoles is seen in Brungs' data on adult bullfrogs; they ac- cumulated much lower levels of all the metals than did tadpoles, and less than detritivores such as crayfish. Relatively high concentrations of Cs- 1 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 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 be- tween dosing and observation; both the relative amounts of Cs- 1 37 in tadpoles compared to oth- er animals and its absolute concentrations de- clined 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 of the body tended to be higher than those in fish from the same waters. These results support the idea that the uptake of the metals is through the diet. Comparison of tadpole gut and contents with those ofbass and bluegills indicates an approximate 10-fold greater concentration of the three metals in the amphibian samples, also supporting Brungs' (1963) contention that feed- ing habits produce the higher levels in amphib- ians. 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 ofsediment were high- est in lead. Getz et al. (1977) concluded that physical contact with silt and the direct ingestion of lead in silt and detritus were important in uptake. They believed that food chain concen- tration did not occur. Pooled samples of tadpoles of two species, and fish collected from two nearby ponds, are com- pared in Fig. 1 . These results do not closely cor- respond to those metal levels reported by Gale et al. (1973) and Jennett et al. (1977) nor the radionuclides documented by Brungs ( 1963) be- VERTEBRATE ECOLOGY AND SYSTEMATICA 131 3i 2 H Pb M B G 2- 1- Cu 300i 200 100- Mg M P B G M P B G o- 132 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 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- ences, but they indicate that conditions favoring the uptake of specific metals do not always result in the greatest 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 Steinwascher 1978, 1979) and the distribution of the metals within the environment. Distri- bution of metals in the ponds seems to differ from that in the systems examined by other au- thors, perhaps because our areas were essentially uncontaminated and had stable levels of most of the metals rather than a single treatment (Brungs 1 963) or a continuous (Jennett et al. 1977) influx 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 in- dicators of environmental contamination. them 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- gested in the case of lead (Getz et al. 1977; Jen- nett et al. 1977). Residual metals in uncontam- inated areas, or those which have been in the ecosystem for some time, should tend to become more widely dispersed (Brungs 1 963) and to pro- duce 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. Acknowledgments H. M. Ohlendorfand 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 W. N. Beyer, E. H. Dustman, and J. C. Lewis. Literature Cited Conclusions 1 ) Adult amphibians ofcertain species can ac- cumulate extremely high levels of copper in the liver. It seems likely that dietary imbalances or metabolic factors, rather than high environmen- tal levels, result in this accumulation. It has been shown that some anurans are protected from these high copper levels, but individuals with such ac- cumulations may be toxic to their predators. There is little evidence that adult amphibians can concentrate other metals to a greater extent than other vertebrates. 2) 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 extensive literature, not reviewed here, on the toxic effects of metals on amphibians and other aquatic vertebrates which indicates that these organisms are suscep- tible to poisoning by metals. Doubtless their un- usual powers ofaccumulation can sometimes re- sult in metals in tissues reaching toxic levels. 3) The apparent tendency for tadpoles to pick up metals from surface sediments might make 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. 1963. The relative distribution of multiple radio- nuclides in a freshwater pond. Ph.D. Thesis. Ohio State Univ., Columbus. Ohio: 97 p. Byrne. A. R., Kosta, L. and Stegnar, P. 1975. The occurrence of mercury in amphibia. En- viron. Lett., 8:147-155. Dmowski, K. and Karolewski, M. A. 1979. Cumulation of zinc, cadmium and lead in invertebrates and in some vertebrates ac- cording to the degree of an area contami- nation. Ekologia Polska, 27:333-349. Domby, A. H., Paine, D. and McFarlane, R. W. 1 977. Radiocesium dynamics in herons inhabiting a contaminated reservoir system. Health Physics, 33:523-532. Dustman, E. H., Stickel, L. F. and Elder, J. B. 1 970. Mercury in wild animals: Lake St. Clair 1 970. Pp. 46-52. In Hartung. R. and Dinman, B. D. (eds.). Environmental Mercury Contam- ination. Ann Arbor Sci. Publ., Ann Arbor, Michigan. Fleischer, M., Sarofim, A. F., Fassett, D. W., Hammond, P., Shacklette, H. T., Nisbet, I. C. T. and Epstein, S. 1974. Environmental impact of cadmium: A re- view by the panel on hazardous trace sub- stances. Environ. Hlth. Perspectives. Exp. Issue No. 7:253-323. VERTEBRATE ECOLOGY AND SYSTEMATICS 133 Gale, N. L., Wixson, B. G.. Hardie, M. G. and Jennett, J. C. 1973. Aquatic organisms and heavy metals in Mis- souri's New Lead Belt. Water Resour. Bull.. 9:673-688. Getz, L. L., Haney, A. Q., Larimore, R. W.. Mc- Nurney, J. W., Leland. H. V.. Price, P. W.. Rolfe, G. L., Wortman, R. L., Hudson, J. L., Solomon, R. L. and Reinbold, K. A. 1977. Transport and distribution in a watershed ecosystem. Ch. 6 In Boggess, W. R. (ed.). Lead in the environment. National Science Foundation RANN Program NSF/RA 770214. Goldfisc her, S., Schiller, B. and Sternwieb, I. 1970. Copper in hepatocyte lysosomes of the toad, Bufo marinus L. Nature, 228:172-173. Hatch, W. R. and Ott. W. L. 1968. Determination of sub-microgram quantities of mercury by atomic absorption spectro- photometry-. Anal. Chem., 40:2085-2087. Ireland, M. P. 1977. Lead retention in toads Xenopus laevis fed increasing levels of lead-contaminated earthworms. Environ. Pollut., 2:85-92. Jennett. J. C, Wixson, B. G.. Lowsley, I. H.. PURUSHOTHAMAN, K... BOLTER, E., HEM- PHILL, D. D., Gale, N. L. and Tranter, W. H. 1977. 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 Foundation RANN Program NSF/RA 770214. LOVETT. R. J., GUTENMANN, W. H., PAKKALA, I. S.. Youngs, W. D.. Lisk. D. J.. Burdick. G. E. and Harris, E. J. 1972. A survey of the total cadmium content of 406 fish from 49 New York State freshwa- ters. J. Fish. Res. Brd. Canada, 29:1283- 1290. Lucas, J. W. and Pickering, D. C. 1958. Direct absorption of dissolved strontium-90 and yttrium-90 by tadpoles of Rana tem- porary. Nature, 182:1242-1243. Monk, H. E. 1961. Recommended methods of analysis of pes- ticide residues in food stuffs. Report by the Joint Mercury Residues Panel Anal., 86:608- 614. National Resear( h Coun< ii . 1977. Copper. National Academy of Sciences, Washington, DC. 115 p. 1979. Iron. University Park Press. Baltimore. MD. 248 p. Pasanen, S. and Koskela, P. 1974. Seasonal changes in calcium, magnesium, copper and zinc content of the liver of the common frog, Rana temporaria L. Comp. Biochem. Physiol., 48A:27-36. Pendleton, R. C. and Hanson, W. C. 1958. Absorption of Cesium-137 by components of an aquatic community. Proc. 2 nd United Nations Conf. on Peaceful Uses of Atomic Energy, 18:419-422. Rolfe, G. L., Haney, A. and Reinbold, K. A. 1977. Environmental contamination by lead and other heavy metals, Vol. II. Ecosystem anal- ysis. Final Rept. National Science Founda- tion RANN Program, Inst, for Envtl. Stud. Univ. Illinois Urbana-Champaign. Sarata, U. 1938. Studies in the biochemistry ofcopper. XXX. Seasonal changes in the amount and distri- bution of copper in tissues of the cultivated bullfrog. Japan. J. Med. Soc, 4:65-69. Schroeder, H. A. and Balassa, J.J. 1966. Abnormal trace elements in man: Arsenic. J. 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: chromi- um. J. Chron. Dis.. 15:941-964. Schroeder, H. A. and Tipton, I. H. 1968. The human body burden of lead. Arch. En- 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. 1978. The effect of coprophagy on the growth of Rana caiesbeiana tadpoles. Copeia. 1978: 130-134. 1979. Competitive interactions among tadpoles: Responses to resource level. Ecology. 60: 1172-1183. Wagemann, R., Snow, N. B.. Rosenberg, D. M. and Lutz, A. 1 978. Arsenic in sediments, water and aquatic bio- ta in lakes from the vicinity of Yellowknife. Northwest Territories, Canada. Arch. En- viron. Contam. Toxicol., 7:169-191. Part III Feeding and Behavior Vertebrate Ecolog> and Systematics—A Tribute to Henrj S Fitch Edited by R. A. Seigel, L E Hunt. .1 I Knight. I. Malarct and N. L. Zuschlag • IV84 Museum of Natural Historv The University of Kansas. Lawrence Energetics of Sit-and-Wait and Widely-Searching Lizard Predators Robin M. Andrews Introduction The foraging tactics of insectivorous lizards, like those of most other predators, appear to be dichotomous (Pianka 1978). In North America, iguanid lizards exemplify the "sit-and-wait" tac- tic in which prey are sought passively from a fixed perch site. Sight of a moving prey item elicits ambush 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- tics may represent a fundamental means of par- titioning the food niche (Pianka et al. 1979). Each tactic apparently gives maximal foraging effi- ciency (in time or energy units) under conditions of varying prey abundance (Norberg 1977) or structural configuration of the habitat (Stamps 1977). The sit-and-wait and the widely-searching tac- tics are each associated with an "adaptive syn- drome" of predator characteristics (Eckhardt 1979). In addition to characteristics strictly re- lated to foraging, the adaptive syndromes of ig- uanid and teiid lizards differ markedly in several ways. Iguanids have more stereotyped responses to novel items in their environment than do teiids (Regal 1978). Iguanids are strongly territorial, teiids lack home range defense (Stamps 1977). Iguanids escape predators by cryptic behavior and, once discovered, by the use ofknown routes to hiding places. In contrast, teiids rely on rapid flight to escape their predators (Vitt and Congdon 1978; Schall and Pianka 1980). Clutch size per unit body weight is higher for iguanids than for teiids (Vitt and Congdon 1978). Although many aspects of the adaptive syn- dromes of sit-and-wait and widely-searching predators have been described, the energetic costs and benefits of each tactic are unknown. For ex- ample, the low searching costs of the sit-and-wait tactic are often associated with relatively low pre- ferred body temperatures. Sceloporine and an- oline iguanids have preferred body temperatures of 35°C or less even in well insolated environ- ments (Blair 1960; McGinnis 1966; Andrews, unpublished data; Huey and Webster 1 976; Ben- nett 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 body temperatures ofabout 40°C (Asplund 1 970; Schall 1977; Bennett and Gorman 1979). Thus, for many iguanids (notable exceptions are desert lizards such as Holbrookia and Callisaums), the metabolic cost of foraging is low compared to that of teiids not only because of the low levels of activity associated with the sit-and-wait tactic but because of low activity temperatures. The major objectives of this study were to an- swer two questions: 1 ) What are the relative en- ergy intakes of lizards using sit-and-wait and widely-searching tactics when both forage in the same habitat? 2) Does the proportional alloca- tion of assimilated energy to production and me- tabolism differ for lizards using the sit-and-wait tactic and the widely-searching tactic? Lizard Subjects and Field Sites Field studies were conducted in the Chiricahua Mountains of Arizona. The lizard subjects were Sceloporus jarrovi (viviparous, Goldberg 1971) and Cnemidophorus exsanguis (parthenogenetic. Cole and Townsend 1977). These species are an ideal pair for comparative studies of feeding be- havior and energetics. First, they are broadly sympatric in oak-pine-juniper woodland. Sec- ond, 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 1977; Ballinger 1979). Observations were made from 1 1 July to 8 August 1979. At this time, the fat reserves of both species are increasing rapidly (Goldberg 1972; Schall 1978). Since energy stored by female lizards prior to winter inactivity contributes di- rectly to the development ofoffspring or eggs that will be produced the following spring (Hahn and Tinkle 1965; Gaffney and Fitzpatrick 1973), en- ergy available for fat storage is directly related to the reproductive effort of both S. jarrovi and C. exsanguis. Moreover, adult S. jarrovi females. 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. Prey taxa Drv wt. (%) Ash (%) S jarrovi C. exsanguii Coleoptera Tenebrio molitor, adult Dermestes caninus, adult Dermestes caninus, larvae Phyllophaga sp., adult Chaulognathus pennsylvanicus, adult Lepidoptera (various moths) Orthoptera Blattella germanica. adult male 59 61 34.0 VERTEBRATE ECOLOGY AND SYSTEMATICS 139 their lengths were estimated roughly at 5 mm intervals. A major assumption of this method of esti- mating food intake is that lizards are active every day and that they defecate regularly. From ob- servations made near my site B, Simon and Mid- dendorf ( 1976) found that the percent of adult S. jarrovi active every day was 75% in July and 100% in August. Thus, the assumption of daily activity is probably valid for S. jarrovi but has not been tested for C. exsanguis. Observations on 5. jarrovi and C. exsanguis maintained in large cages under simulated field conditions sug- gest that defecation occurs at least every morning following the attainment of preferred body tem- peratures (see also Cowles and Bogert 1944). Ash contents of faeces, urinary wastes, and the various prey types used in the laboratory exper- iments were measured by heating samples for 1 h at 550°C in an ashing oven. The mass of all materials is presented as ash-free dry weight. Activity Periods and Body Temperatures Any lizard seen was considered active. Because 5. jarrovi individuals were readily found during all daylight hours. I assumed that their activity period potentially spanned 10-12 h. In contrast. C. exsanguis individuals were encountered most 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 about 2 km through site A were counted. Body (cloacal) temperatures (Tb ) were mea- sured immediately after capture with a Schult- heis quick-reading thermometer. Temperatures of lizards which avoided capture for more than a minute were not taken to avoid bias. Shaded air temperatures were taken at 1 m and at 1 cm above the place where the lizard was first seen. Results Food Intake by Free- ranging S. jarrovi and C. exsanguis Various species of beetles made up about 60% of the insects eaten by lizards in the feeding ex- periments and moths and cockroaches made up the other 40% (Table 1 ). This particular feeding regime was similar to the natural diets of the two Table 2. Prey items of Sceloporusjarrovi and of Cne- midophorus exsanguis in July-August 1979. Propor- tion of total prey is given for each species followed by modal length category in parentheses. Pre\ laxa S \UTTOVi unguis Coleoptera (adults) Formicidae Lepidoptera (adults) Hymenoptera (adults) Araneida Orthoptera Miscellaneous* .306(5-10) .518 (<5) 0.0 .082(5-10) .023 (<5) .023(10-15) .047 (<5) .189(5-10) .500(5-10) .122(10-15) 0.0 .067 (<5) .078(10-15) .044(5-10) 90 * S. jarrovi: 4 Homoptera-Hemiptera; C. exsanguis: 2 Homoptera-Hemiptera. 1 mantid. 1 Chilopoda. lizard species (Table 2). Judging by both fre- quency and size, beetles were probably the most important component of the diets of both S. jar- rovi and C. exsanguis. Orthoptera were probably the second most important component of the diet of S. jarrovi and Lepidoptera were probably the second most important component of the diet of C. exsanguis. Ants were not used in the feeding experiments although they comprised about half of the items eaten by both species in the field. Because of their small size (bulk) their contribution to total energy intake was probably low. Using stomach contents to evaluate the diet 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 I, -= F*CFF-'*W- 0.83 and I„ = U*CFU-'*W-°-83 where I, and I u are the respective estimates of food intake based on faecal and urinary produc- tion. F and U are faecal and urinary production (mg dry wt) during the 48 h of confinement, re- spectively. CFF and CFU are the factors which convert F and U to food intake for faecal and urinary production, respectively, and W 3 is live body weight in g raised to a power of 0.83 to adjust for weight specific metabolic rates (Ben- 140 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 3. Daily food intake (I, and I u ) of S. jarrovi and C. exsanguis based on production of faecal and urinary material (sec text for details). Spec ies and site In I, ± SE (mg'g~°-"-d" In I„ ± SE (mg-g-°" d ') S. jarrovi—A S. jarrovi— B C. exsangi 0.05), mean values were used to estimate food intake. Respective mean values ( ± SE) for S. jarrovi and C. exsanguis were CFF = 0.56 ± 0.053 and 0.24 ± 0.024 and CFU = 0.2 1 ± 0.027 and 0. 1 9 ± 0.027. Natural log transformations were used to normalize the I, and the I u data for statistical analyses. Some female-sized male S. jarrovi were included in the analyses (5 of 9 and 9 of 29 individuals on Sites A and B, respectively) since the faecal production of these males did not differ from females on either site (P > 0.05, two-tailed /-tests). Al- though individuals were captured at various times during the day (see below), regression analyses indicated that time of capture was not related to food intake (P > 0.05). Differences in I, and I u among the C. exsanguis females and the two populations of S. jarrovi (Table 3, Fig. 1) were statistically significant (P < 0.05, analysis of variance). A posteriori tests showed that C. exsanguis females had a signifi- cantly (P < 0.05) greater I, and I u than both the S. jarrovi populations, and that the S. jarrovi populations did not differ from one another for either I, or I u (P > 0.05, Duncan's multiple range tests). Therefore, in subsequent analyses the data for S. jarrovi females have been combined. The two estimates of food intake for C. exsanguis differed by only 4% on a In scale (Table 3) and by only 1 1% on an arithmetic scale. Since the correlation between U and I 3 for S. jarrovi was not statistically significant, I u was not deter- mined. Activity Periods and Thermoregulation Scleoporus jarrovi and C. exsanguis differed considerably in the apparent length of their ac- tivity periods. I made observations from about 0800 to 1730 h with comparable times spent in the field in the morning and in the afternoon. The number of S. jarrovi individuals observed in the morning and the afternoon was very sim- ilar. 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 activity was in the morning (Table 4). Body temperatures of S. jarrovi were depen- dent on weather conditions (Figs. 2 and 3). On site A where temperatures were measured under sunny conditions, S. jarrovi individuals main- tained relatively constant T hs (Mean ± SE = 34.2 ± 0.36°C). In contrast, on site B about one- half of temperature measurements were taken under overcast or intermittently cloudy condi- tions. At these times, Tbs averaged 31.1 ± 0.6 PC. During sunny conditions Tbs averaged 35.8 ± 0.34°C. Body temperatures of C. exsanguis were in- dependent of ambient temperatures (Fig. 1), av- eraging 40.0 ± 0.3 PC. The one individual with a T b of 34°C had probably just emerged from a burrow. Discussion During the July-August study period, 5. jar- rovi females, using sit-and-wait tactics, had a sig- nificantly lower intake of food than did C. exsan- guis females which were using widely-searching VERTEBRATE ECOLOGY AND SYSTEMATICA 141 CO 00 o E 100i 60 20 A A A • A A a* A A 4 a A*A A A. * 10 15 20 W eg) Fig. 1. Food intake based on faecal production (I r) of field-collected Sceloporus jarrovi (Site A. filled circles, and Site B. filled triangles) and Cnemidophorus e.xsanguis (Site A, open triangles) as a function of body weight. tactics. The result is particularly interesting in that S. jarrovi 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 abun- dances of prey available to them. Thus, the widely-searching tactic appears to be more effi- cient both in terms of time spent and energy acquired. In order to compare the energy that S. jarrovi and C. e.xsanguis females have available for pro- duction, I, and I u were partitioned into their ma- jor components as I = R + P + FU where I is food intake. R is metabolism. P is production, and FU is the combined faecal and urinary wastes. The energy value of food intake was determined as I times 5800. the mean caloric 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 C. e.xsanguis (Johnson and Lillywhite 1979; Andrews and Asato 1977). The parameters used to estimate field metabolism of a 12 g S. jarrovi female are from Table 3 and Appendix A. Table 1. of Congdon (1977). Since his Ash Spring site and my sites were located within a few km of one another, I have used his July-August deter- minations directly. The field metabolism of C. e.xsanguis was estimated from metabolic data collected on Cnemidophorus murinus, a West In- dian species. Metabolic rates of C. murinus were determined under standard conditions for both resting individuals and for individuals moving Table 4. Numbers ofCnemidophorus individuals seen during 30 min censuses conducted on 3 and 4 August 1979. Shaded air temperatures 1 m above ground are shown for the time the census was began. Both days were sunny. Census period 142 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY O o LU DC Z) h- < DC LU Q_ LU 45 35 25 A A A A ZA A AA A A A A & v> • • 10 11 T ~ 4 6. 25 - 010,

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 Introduction Snakes are prominent predators in many ter- restrial, aquatic, and tropical marine commu- nities, 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 un- dulation) is energetically more efficient than te- trapody (Chodrow and Taylor 1973). Perhaps most importantly, these "limbless tetrapods" possess an extremely flexible jaw apparatus that permits the ingestion of large prey items without the assistance of limbs or mastication (Gans 1961). Although many species swallow prey alive and struggling, others immobilize it by constric- tion, venom injection, or a combination of these methods (Gans 1978; Greene and Burghardt 1978; Kardong 1980). It is now clear that venom delivery systems comprise at least three grades of structural com- plexity and that these have evolved indepen- dently in several lineages ofsnakes (see Gans and Gans 1978; Savitzky 1978; 1980; Kardong 1980; Cadle. in press, for extensive discussion and re- views). Opisthoglyphs (many species of colu- brids) possess enlarged, grooved teeth on the pos- terior ends ofotherwide normal, elongate, toothed maxillae. Proteroglyphs (elapids and hydro- phiids) have one or two enlarged, canaliculate, anterior teeth on each short, nonmobile or slight- ly 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- peridae show that these snakes often strike and release prey, then relocate it before swallowing (e.g., Klauber 1956; Duellemeijer 1962; Nalleau 1966; Minton 1969; Kardong 1975; Chiszar and Scudder 1 980). Although there are isolated notes on the feeding behavior of opisthoglyphs and proteroglyphs in the literature (e.g.. Armitage 1965, Lambins 1967, for African elapids). the only extensive accounts are for certain sea snakes (Voris et al. 1978; Radcliffe and Chiszar 1980). With few exceptions (e.g.. Shine 1977, for ela- pids; Voris et al. 1978. for hydrophiids), we also know very little about the dietary ecology of pro- teroglyphs. This general lack of descriptive stud- ies on proteroglyphs, particularly terrestrial forms, hampers broader considerations of functional morphology, adaptive radiation, and commu- nity structure in snakes (cf. Arnold 1972; Rabb and Marx 1973; Kardong 1980; Savitzky 1980; Greene. MS). In the present paper I provide a description of feeding behavior, an ecological characterization of the food habits, and a discussion of factors affecting diet composition in a venomous coral snake. Micrurus fulvius. This species occurs in the southeastern United States and northeastern Mexico, in habitats ranging from subtropical swamps and lowland forests to semiarid scrub (Wright and Wright 1957). It is a northern rep- resentative of an essentially Neotropical radia- tion of the cosmopolitan front-fanged family Elapidae(Roze 1967; see Savitzky 1978. and Ca- dle and Sarich 198 1 for contrasting views on the relationships ofcoral 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 ( 1 973a. 1 973b). Quinn (1979). and Jackson and Franz (1981) summa- rized some aspects of the biology of this species. Methods Behavioral Observations. — Sixty-five com- plete feeding sequences on live and dead prey by- four captive coral snakes were observed (one fe- male, three males; total lengths 52.5-85.0 cm; from Dallas. Hidalgo, and Nacogdoches Coun- ties, Texas). The snakes were individually housed in glass terraria that measured 32 x 32 x 62 cm 147 148 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY or 27 x 32 x 52 cm. Each cage had a gravel sub- strate covered with leaf litter, a water bowl, and at least 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 occasionally rose to 30°C. A 100 W bulb on top of the perforated metal cover of each tank raised the temperature at one end to ca. 24-26°C for 10 hr each day. Observations were timed with a stop watch and recorded on audio tape or with a 35 mm camera and electronic flash. Captive coral snakes were offered live or dead prey, as available, ofthe following species: Anolis carolinensis, Eumeces tetragrammus, E. fascia- tits. Scincella lateralis, Carphophis amoenus, Coluber constrictor, Diadophis punctatus, Elaphe obsoleta, Heterodon platyrhinos, Nerodia ery- 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 jig- gled 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), con- sisting of an 80 x 80 cm piece of white duck cloth (28 strands/cm 2 ) in a plastic swimming pool. An octagonal trail lane with segments 20 cm on an outer side and 1 cm wide was marked on the cloth with small, faint broken lines of indelible ink. Prior to 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 1- gal plastic jar. The snake was released by lifting the jar, and its behavior observed under a 60 W red light positioned so that the arena was very dimly lit. The cloth arenas were machine washed, rinsed, and dried after each test. Diet Studies. — Museum specimens were opened with a ventral incision and the orienta- tion of each prey item in the gut was recorded. The identity and approximate total length (TL) of each item was determined if possible, often on the basis ofa tail or a tail and posterior portion of a body, by comparisons with published in- formation and intact reference specimens. Ad- ditional records were obtained from conversa- tions or correspondence with collectors and from the literature (Matthes I860; Hay 1893; Mitchell 1903; Strecker 1908; Schmidt 1932; Loveridge 1938, 1944; Klauber 1946; Ruick 1948; Minton 1949; Clark 1949; Telford 1952; Curtis 1952; Highton 1956; Martin 1958; Kennedy 1964; Myers 1965; Neill 1968; Chance 1970; Malloy 1971; Fisher 1973; Jackson and Franz 1981). Snout-vent (SV), tail, and head lengths of pre- served coral snakes were measured when possi- ble. Because many museum specimens had dam- aged heads, SV was used for comparisons with prey TL. I estimated the weights ofcommon prey 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 I 'irginia striatula. The average total lengths of all snakes in east Texas were taken as the midpoints of the ranges for adults given in Con- ant (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 subsam- ples: "east Texas" (mixed deciduous and pine forests), "north central Texas" (tall grass-prairie- forest ecotone), "central Texas" (forested hill country of the Edwards Plateau and the extreme eastern edge of the Chihuahuan Desert), and "south Texas" (semiarid thorn scrub and sub- tropical forest, see Gould 1969, for vegetation regions). Records from elsewhere in the species range are grouped as "Florida" and "other" (Ar- kansas, Louisiana, South Carolina, and Mexico). Feeding Behavior The description that follows incorporates pub- lished accounts (Grijs 1898;Ditmars 1907, 1912; Clark 1949) and my observations. Feeding be- havior is discussed in six groups of sequentially and functionally related motor patterns to facil- itate future comparisons with other snakes. Encountering Prey. — Methods of encounter- ing prey should be included in discussions of feeding behavior, because snakes use species- typical postures and strategies for obtaining food. VERTEBRATE ECOLOGY AND SYSTEMATICA 149 Prey might be located by some type of searching, trail following, "sitting and waiting," (Pianka 1966), or a mixed strategy (Tollestrup 1980; Chiszar and Scuddcr 1980); each of these tech- niques might be enhanced by behavioral or mor- phological specializations. For example, search- ing and trail following utilize stereotyped poking behavior (in coral snakes, see below) and highly specialized receptor systems (e.g., facial pits in boids and some viperids). "Sitting and waiting" is probably more efficient when accompanied by camouflage (Fitch 1 960) or caudal luring (Greene and Campbell 1972). When a coral snake had not fed for several days, it crawled slowly over the substrate and poked its head in and out of the leaf litter. This involved repeated forward and lateral head movements, and was accompanied by frequent tongue flick clusters. At times a snake crawled slowly beneath a large leaf or a small piece of bark and soon emerged from the opposite side, still moving its head from side to side and flicking its tongue. When a coral snake was searching, any movement of an object in the terrarium elicited pointing and, if it was not a large object, ap- proach behavior. When an acceptable prey item caused the approach, it was seized and eaten. Unsuccessful attempts to capture prey were fol- lowed by more searching behavior. F. R. Gehlbach (pers. comm.) observed similar crawling and poking movements by two free- living coral snakes on the Santa Ana Wildlife Refuge. Hidalgo County. Texas, one of which I 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, moved its head from side to side, and poked its head into the surface litter. Neill also stated that the snake's tail made "con- stant rapid, probing motions" in the leaves, and that at times "the hind part of the creature was thrown nearly as far forward as the head." He observed similar behavior in a captive snake, and suggested that the head and tail movements served to flush small reptiles and amphibians from cover. These observations suggest that crawling and head-poking in ground litter are motor patterns normally used by coral snakes to locate potential prey items. However, neither Gehlbach nor I observed use of the tail in for- aging, and I doubt that it is a normal behavior, at least for coral snakes in Texas. The threshold for tail waving, an important component ofcoral snake antipredator behavior (Gehlbach 1972; Greene 1973b), is often very low for this species (pers. obs.), and perhaps the snake observed by Neill was responding defensively to tactile or vi- brational stimuli. Several species of small snakes deposit chem- ical trails that serve as attractant pheromones (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 kept in ajar of wet moss for several days before the snakes and moss were put in a coral snake's cage. The coral snake was crawling on the leaves and encountered the moss. It moved its head back and forth over the moss for approximately five minutes and frequently flicked its tongue. Then it crawled across the cage, generally fol- lowing the route taken by one ofthe earth snakes. The coral snake soon found the prey in a corner and ate it. During staged encounters with ground skinks {Scincella lateralis), a coral snake fre- quently paused for several seconds in the exact spot where a skink had recently rested and point- 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 snakes, a small colubrid snake (adult Storeria dekayi or I 'irginia striatula) was allowed to crawl around the alley one time. In each case the coral snakes crawled away from the central release point, paused briefly and pointed at the trail, and moved offthe cloth. A second block of trials used trails laid by a small snake or a skink (adult female Eumeces fasciatus) making four circuits of the octagon in five minutes. One coral snake responded to two snake trails with pointing and then escape behavior, but followed a skink trail for one complete circuit and two additional turns on the octagon. The other coral snake followed trails laid by S. dekayi (two trials) and V. stria- tula (one trial) for one complete circuit, seven lane segments, and three lane segments, respec- tively. It followed two lane segments of a skink trail before crawling off of the cloth. These ob- servations suggest that known prey species can leave trails which are perceived and followed by coral snakes. Additional experiments using more coral snakes, more prey species, and more trials are required before comparisons with the exten- 150 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY sive study by Gchlbach et al. (1971) are war- ranted. Trail following behavior was stereotyped and similar to that described by Gehlbach et al. ( 1 97 1 ) for blind snakes. Leptotyphlops dulcis. A coral snake crawled slowly from the release site, point- ed and flicked its tongue at the trail, then turned 90° and began following it. The snake's head re- mained elevated while it crawled, and there were frequent tongue-flick clusters. At each corner it overshot 2-4 cm, paused, pointed and tongue- flicked at the cloth, moved its head from side to side, turned back onto the trail, and resumed crawling. If a wire was jiggled on the cloth in front of a coral snake it pointed and approached rapidly. The available captive and field observations imply that coral snakes actively search for prey, but the frequency and extent of foraging move- ments are unknown. There is no evidence that free-living coral snakes use a "sit and wait" strat- egy to ambush prey, but the behavior of captives suggests that they might. My snakes were fre- quently seen coiled with head raised and pro- truding from beneath the edge of a piece of bark or pile of leaves. Such snakes responded to near- by movements by pointing, tongue-flicking, and approaching. Recognition and Approach. — Recognition of prey probably begins as soon as a coral snake points toward a stimulus, and incorporates vi- sual and chemical cues. Captives approached any small movement, such as a wire jiggled in the leaves or a finger moved against the glass from outside of the terrarium. Larger moving objects, such as a hand or a piece of bark, usually elicited pointing and then rapid head withdrawal and crawling. This seemed especially likely if the ob- ject was moved suddenly. Approach was accompanied by tongue-flick clusters, which evidently convey the necessary stimuli for seizing or avoiding a potential prey item. Coral snakes quickly approached to within 2 cm of large coleopteran larvae, cricket frogs (Acris crepitans), and newborn mice, but then withdrew without seizing them. Small live water snakes (Nerodia sp.) were also approached and rejected, and in most cases they had discharged the cloacal sac contents. However, rapid prey movements seemed to result in a quicker attack and to override aversive chemical cues. Dead Nerodia were usually refused when stationary or pulled slowly but were attacked when pulled more rapidly. In 10 incomplete feeding sequences, a prey item was grasped and immediately released, or maneuvered for a short time and then re- leased. This suggests that a coral snake continues to receive input from the prey after it is seized, perhaps via either oral sensory papillae (Burns 1969; Greene, unpublished) or the Jacobson's 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, 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 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 an approaching predator at a distance of several centimeters and often slipped away unseen. Dur- ing 1 1 attempts on these lizards by a coral snake, I observed eight misses, two tail autotomies (skink escaped unharmed), and one capture. These were during staged confrontations on a 32 x 62 cm substrate of gravel and scattered leaves, and the only capture occurred when the snake trapped a skink in a corner. Small live prey snakes pre- sented a slower and more elongate target, and were captured without difficulty; each of 23 at- tempts was successful. Ditmars (1907) and Clark (1949) stated that Micrurus fulvius immobilizes its prey with ven- om before swallowing, but Ditmars (1912) re- marked that the venom is of little value in sub- duing "cold blooded" animals. My observations indicate that this species typically holds prey at the point of seizure until paralysis and then be- gins pre-ingestion maneuvers (see below). Slight movements ofthe 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 151 Table 1. Pre-ingcstion latencies (in seconds) for coral snakes. Micrurus fulvius, dealing with live and dead prey. Ranges, means, standard deviations, and sample sizes are given. Latency Snake no. 4 Snake no. 5 Time between seizure and onset of pre-ingestion maneuvers (live prey) Time between seizure and onset of pre-ingestion maneuvers (dead prey) Time between last prey body movement and onset of pre-ingestion maneuvers 290-595 (jc = 434.2 ± 132.5) N = 6 0-85 (a- =28.6 ± 32.3) N= 10 63-190 (v = 99.2 ± 52.6) N = 5 70-940 (a- = 400.9 ± 334.9) N = 8 0-290 (a= 73.7 ± 86.5) N= 10 0-152 (a =71.8 ± 73.3) N = 4 This tended to untangle a small, writhing snake, and it 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- ruleus (Rosenberg 1967; see also Savitzky 1978). In two instances a coral snake bit and quickly released an adult female Eumeces fascialus that struggled violently. One of the skinks was im- mediately recaptured. The other lizard crawled slowly for several centimeters and went under a piece of bark. It was soon followed by the coral snake and regrasped. Both skinks subsequently- made only feeble movements and were eventu- ally eaten. Pre-ingestion Maneuvers. — Coral snakes nor- mally do not release prey prior to swallowing it. Pre-ingestion maneuvers are probably evoked by tactile and/or chemical cues (cf. Nalleau 1966) and inhibited by prey movements. If prey move- ments inhibit the coral snake, the time between seizure and the onset of pre-ingestion maneuvers should be longer with live prey than with dead prey. The mean pre-ingestion handling times with live and dead prey (Table 1) differed significantly for each of two coral snakes (P < .01, Mann- Whitney c'test). If prey movements inhibit the 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 < .01, Mann-Whitney c'test) but not for the other snake (P > .90). I interpret the large variances and the equivocal results of the last comparison as resulting from individual differences and from the use of different sizes and species of prey in the feeding trials. Captive and free living coral snakes almost always swallowed prey head first, and scale over- lap on the prey item was used as a cue in locating its anterior end (Greene 1976). Alternating jaw movements, typical of snakes (Gans 1961). were used to shift along the prey's body prior to swal- lowing. In one instance a small stick in the mouth of a coral snake prevented it from shifting over 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 sub- stance, regrasped the prey by the snout, and swal- lowed it. In all other feeding sequences, prey snakes were not released before they were swal- lowed. Swallowing.— After the prey's head had been shifted down the throat, it was swallowed by re- peated series ofalternatingjaw movements. These were separated by brief pauses and accompanied by lateral movements of the entire head. Ac- cording to McDowell (1970), Micrurus belongs to a group of elapids in which "the palatine is erected along with the maxilla during maximum protraction of the palate." This presumably oc- curs 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 swal- lowing. During swallowing a coral snake sometimes rolled about its long axis, perhaps using the prey's inertia to achieve better contact between its teeth 52 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. Frequency of prey items by taxon in eastern coral snakes, Micrurus fulvius. Abbreviations refer to east Texas (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 all localities (T). Number of coral snakes containing prey for each sample is in parentheses. Prey species E VERTEBRATE ECOLOGY AND SYSTEMATICS 153 Table 2. Continued. 154 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 3. Seasonal incidence of three prey types in coral snakes, Micrurus fulvius, from Texas. Number of coral 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. Seasons Skinks Juvenile large snakes Other prey Total Spring. March-May (33) Summer. June-August (16) Fall, September-December (33) Total (82) 12 (.29) VERTEBRATE ECOLOGY AND SYSTEMATICA 155 500- 156 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 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 size, absence of effective anti- 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 (Ta- ble 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 as- pects 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, ofMacArthur and Pianka 1966). Food value includes the energy and other nutritional factors present in a prey UJ 4 O HI a o 3 LL O 2 cc LU CO ^ 1 VERTEBRATE ECOLOGY AND SYSTEMATICS 157 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 to- tal energy budget of a large snake than it would to that of a smaller individual. Whether this is actually true for large and small coral snakes is not known, (ii) Occurrence of young individuals of large prey snakes is seasonally restricted, and they are thus not a predictable resource for coral snakes at all times of the year, (iii) The size con- figurations of terrestrial snake communities in temperate forests can be discontinuous; in east Texas they contain several species of very small, moderate, and large snakes, but very few me- dium-small species relative to adult M. fulvius (Fig. 3: Appendix, Note 3). In other words, be- cause of (ii) and (iii), large coral snakes in the southeastern United States probably rarely en- 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). The disparity is increased by the lower vulner- ability of skinks (see above and Vitt et al. 1977) and perhaps by their capacity for inflicting a pow- erful bite on the predator. In other words, skinks are probably more costly to handle than small snakes ofequivalent weight and provide less food value than small snakes of equivalent handling cost. Ideally, coral snakes should add skinks to their diets only when location costs are reduced proportionate to the increased handling costs these lizards impose. This suggests an explana- tion 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 observa- tions 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 tails were for very small M. fulvius, for which they might have been proportionately large items). Either skinks (or skink tails) are propor- tionately more valuable than small snakes in per gram food value (cf. Clark 1 97 1 ; B. E. Dial, pers. comm.) or the overall expectation of finding "better" items is low enough to make them worth chasing in spite of the very low average payoff per attempt. These considerations suggest that eastern coral snakes attack and sometimes eat substantial numbers of intuitively non-ideal prey (skinks, relatively small snakes). That they do so is per- haps surprising, because many snakes apparently feed infrequently on relatively heavy items (Greene 1983, MS) and such predators might be especially able to defer feeding until a highly prof- itable 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 amount of food, rather than to maximize the intake of energy in a given time period or prey- encounter (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. (ii) According to MacArthur (1972:62), "an animal should elect to pursue an item ifand 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 un- derlies much subsequent literature on optimal foraging (e.g., Pyke et al. 1977; Krebs and Davies 1978; Morse 1980), although MacArthur (1972: 59) noted that it assumed "a fairly clear statistical expectation ofthe resources [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 predict- ability of finding a "better" item nearby is ex- tremely low (see also Godley 1 980). If this is true, MacArthur's formulation is trivial for such pred- ators, 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- logical factors (Appendix, Note 2), and by rela- tive prey abundances (through their effects on encounter rates), rather than by more complex strategic "decisions" on an item by item basis 158 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY (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. Summary Tais paper reports the first extensive survey of feeding biology in a New World proteroglyph- ous snake, Micrurus fulvius. Foraging behavior was described on the basis of anecdotal field re- ports 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. Pre- ingestion movements were apparently inhibited by the prey's struggles and directed by scale over- lap. 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, relocat- ed, 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 en- tirely 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 sea- sonal availability of particular prey species, rath- er than shifts in the general types of prey taken. Large coral snakes sometimes eat larger prey than do smaller individuals, but they also continue to 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 con- siderations suggest that diet breadth in Micrurus fulvius might be constrained by naive prey pref- erences, morphological constraints, and relative prey abundance rather than by strategic "deci- siGns" on an item by item basis. Acknowledgments This paper is extensively revised from part of a thesis submitted to The University of Texas at Arlington in partial fulfillment of the require- ments for a Master of Arts degree. I am partic- ularly indebted to the chairman of my commit- tee, W. F. Pyburn, for his encouragement and counsel over the years, and to F. R. Gehlbach for serving as visiting committee member, loan- ing me his trailing arenas, and the gift of the live coral snake that began the study. The following curators and institutions allowed me to examine specimens: C. J. Cole, C. W. Myers, and R. G. Zweifel, American Museum of Natural History; R. L. Drewes and A. E. Leviton, California Acad- emy of Sciences; C. J. McCoy, Carnegie Museum of Natural History; W. J. Voss, Fort Worth Mu- seum of Science and History; M. A. Nickerson, Milwaukee Public Museum; B. Hinderstein, Sam Houston State University; F. L. Rainwater, Ste- phen F. Austin State University; B. C. Brown, Strecker Museum, Baylor University; A. H. Cha- ney, Texas A&M 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 Nat- ural 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 Me- morial Museum, University of Texas; A. C. Echternacht, University ofTennessee; and W. F. Pyburn, University of Texas at Arlington. I also thank R. L. Anderson, K. F. Barnes, J. R. Dixon, F. S. 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 Dallas Zoo; S. Brums, G. M. Burghardt, R. Franz, H. S. Harris, D. R. Jackson, J. E. Joy, J. P. Ken- nedy, R. L. Lardie, H. D. Lehmann, 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 His- tory; Theodore Roosevelt Memorial Fund, American Museum of Natural History; the Na- VERTEBRATE ECOLOGY AND SYSTEMATICS 159 tional Science Foundation: and the Committee on Research, University of California, Berkeley. Literature Cited Ardrey, R. 1970. The social contract. Atheneum Publishers. New York. Armitage, W. W. 1965. Observations on differences in morphology and behavior of Dendroaspis angiisticeps and D. polvlepis. J. Herp. Assoc. Africa. (1):12- 16. Arnold, S. J. 1972. Species densities of predators and their prey. Amer. Nat., 106:220-236. 1 980. The microevolution of feeding behavior. Pp. 409-453. In Kamil, A. C, and Sargent. T. D. (eds.). Foraging Behavior: Ecological, Ethological. and Psychological Approaches. Garland STPM Press. New York. Beavers, R. A. 1 976. Food habits ofthe western diamondback rat- tlesnake, Crotalus atrox, in Texas (Viperi- dae). Southwest. Nat.. 20:503-515. Bennett. A. F. and Dawson, W. R. 1 976. Metabolism. Pp. 1 27-233. In Gans, C. and Dawson, W. R. (eds.). Biology of the Rep- tilia. Vol. 5 (Physiology A). Academic Press, New York. Bl RGHARDT, G. M. 1970. Chemical perception in reptiles. Pp. 241- 308. In Johnson, J. W., Moulton, D. G. and Turk, A. (eds.). Communication by Chem- ical Signals. Appleton-Century-Crofts, New York. Burns, B. 1969. Oral sensory papillae in sea snakes. Copeia. 1969:617-619. Cadle, J. E. In Press. Problems and approaches in the interpre- tation of the evolutionary history of ven- omous snakes. Mem. Inst. Butantan. Cadle, J. E. and Sarich, V. M. 1 98 1 The phylogenetic position and historical bio- geography of the New World coral snakes. J. Zool., 195:157-167. Campbell, J. A. 1973. A captive hatching of MicrurusJulvius (Ser- pentes. Elapidae). J. Herp.. 7:312-315. Chance, B. 1970. A note on the feeding habits of Micrurus fulvius fuhius. Bull. Maryland Herpetol. Soc, 6:56. Chiszar, D. and Scudder, K. M. 1980. 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, R. E. and Taylor, C. R. 1973. Energetic cost of limbless locomotion in snakes. Fed. Proc. 32:422. Clark, D. R. 1964. Reproduction and sexual dimorphism in a population ofthe rough earth snake. I 'irginia striatula (Linnaeus). Texas J. Sci.. 16:265- 295. 1971. The strategy of tail-autotomy in the ground skink, Lvgosoma laterale. J. Exp. Zool.. 1 76: 295-302. Clark, R. R. 1949. Snakes of the hill parishes of Louisiana. J. Tennessee Acad. Sci.. 24:244-261. Cirtis, L. 1952. Cannibalism in the Texas coral snake. Her- petologica. 8:27. CONANT. R. 1975. A field guide to reptiles and amphibians of eastern and central North America. Hough- ton Mifflin Co., Boston. Ditmars, R. L. 1907. The reptile book. Doubleday. Page and Co.. Garden City. 1912. The feeding habits of serpents. Zoologica. 1: 197-238. Dullemeijer, P. 1962. Some remarks on the feeding behaviour of rattlesnakes. Proc. Kon. Ned. Akad. We- tensch.. 64:383-396. Fisher, C. B. 1973. Status of the flat-headed snake. Tantilla gracilis Baird and Girard, in Louisiana. J. 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 sexlineatus). Univ. Kansas Publ. Mus. Nat. Hist.. 11:11-62. 1 960. Autecology of the copperhead. Univ. Kansas Publ. Mus. Nat. Hist., 13:85-288. 1970. 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 pata- gonica Woodward Part II. Jaw machinery of the earliest snakes. Forma et Functio. 3:205- 221. Gans, C. 1961. The feeding mechanism of snakes and its possible evolution. Amer. Zool.. 1:21 7-227. 1978. Reptilian venoms: some evolutionary con- siderations. Pp. 1-42. In Gans, C. and Gans. K. A. (eds.). Biology of the Reptilia. Vol. 8 (Physiology B). Academic Press, New York. Gans, C, and Gans, K. A. (eds.) 1978. Biology of the Reptilia. Vol. 8 (Physiology B). Academic Press, New York. Gehlbach. F. R. 1 972. Coral snake mimicry reconsidered: the strat- egy of self-mimicry. Forma et Functio. 5: 311-320. Gehlbach, F. R., Watkins, J. F. and Kroll, J. C. 1971. Pheromone trail-following studies of typhlo- 160 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY pid, leptotyphlopid, and colubrid snakes. Behaviour, 40:282-294. Gentry, J. B. and Smith, M. H. 1968. Food habits and burrow associates of Pero- myscus polionotus. J. Mamm., 49:562-565. Godley, J. S. 1980. Foraging ecology of the striped swamp snake, Regina alleni, in southern Florida. Ecol. Monogr., 50:411-436. Gould, F. W. 1 969. Texas plants: a checklist and ecological sum- mary. Texas Agr. Exper. Stat. MP-585 (re- vised ed.), 121 pp. Greene, H. W. 1 973a. The food habits and feeding behavior ofNew World coral snakes. Master of Arts Thesis, The University of Texas at Arlington. 1973b. Defensive tail display in snakes and am- phisbaenians. J. Herp., 7:43-61. 1976. Scale overlap, a directional sign stimulus for prey ingestion by ophiophagous snakes. Z. Tierpsych., 41:113-120. 1980. Review of behavioral ecology: an evolution- ary approach, by Krebs, J. and Davies, N. B.J. Mamm., 61:168-169. 1983. Dietary correlates of the origin and radiation of snakes. American Zool., 23:431-441. MS. A diet-based model of adaptive radiation in snakes. Manuscript in preparation. Greene, H. W. and Burghardt, G. M. 1 978. Behavior and phylogeny: constriction in an- cient and modern snakes. Science, 200:74- 77. Greene, H. W. and Campbell, J. A. 1 972. Notes on the use of caudal lures by arboreal green pit vipers. Herpetologica, 28:32-34. Grijs, P. 1898. Beobachtungen an Reptilien in der Gefan- genschaft. Elaps fulvius. Zool. Gart., 39:276- 279. Hay, O. P. 1892. Description of a supposed new species of Storeria from Florida, Storeria victa. Sci- ence, 19:199. HlGHTON, R. 1956. Systematics and variation of the endemic Florida snake genus Stilosoma. Bull. Florida St. Mus., 1:73-96. HOLLING, C. S. 1959. Some characteristics of simple types of pre- dation and parasitism. Can. Entomol., 91: 385-398. Hornocker, M. 1 970. An analysis ofmountain lion predation upon mule deer and elk in the 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 Flor- ida. Herpetologica, 37:213-228. Kardong, K. V. 1975. Prey capture in the cottonmouth snake (Ag- kistrodon piscivorus). J. Herp., 9:169-175. 1980. Evolutionary patterns in advanced snakes. Amer. Zool., 20:269-282. Kassing, E. F. 1961. A life history study ofthe Great Plains ground snake, Sonora episcopa episcopa Kennicott. Texas J. Sci., 13:185-203. Kennedy, J. P. 1 964. Natural history notes on some snakes ofeast- ern Texas. Texas J. Sci., 16:210-215. Klauber, L. M. 1 946. The glossy snake, Arizona, with descriptions of new subspecies. Trans. San Diego Soc. Nat. Hist., 10:311-398. 1956. Rattlesnakes, their habits, life histories, and influence on mankind. Univ. California Press, Berkeley. Krebs, J. R. and Davies, N. B. 1978. Behavioural ecology: an evolutionary ap- proach. Sinauer Assoc, Sunderland, Mas- sachusetts. Lambiris, A. J. 1967. Observations on the feeding behavior of Naja haje and Naja nicricollis. J. Herp. Assoc Af- rica, (3):36. Lank.es, J. 1928. Micrurus (Elaps) frontalis D. B. im Terra- rium. 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 inges- tion behavior of rodent-eating snakes. Psy- chon. 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. MacArthur, R. H. 1972. 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. Maiorana, V. C. 1978. Behavior of an unobservable species: diet selection by a salamander. Copeia, 1978:664- 672. Malloy, J. F. 1971. Food habits of snakes in an east Texas State Fish Hatchery. Master of Science Thesis, S. F. Austin State University, Nacogdoches. Martin, P. S. 1958. A biogeography of reptiles and amphibians in the Gomez Farias region, Tamaulipas. Mexico. Misc. Publ. Mus. Zool. Univ. Mich- igan, 101:1-102. Marx, H. and Rabb, G. B. 1972. Phyletic analysis of fifty characters of ad- vanced snakes. Fieldiana: Zool., 63:1-321. Matthes, B. 1860. 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 161 and G. aus den Vereinigten Staaten von Nord-Amerika und Elaps corallinus Pr. Max. aus Brasilien. Pp. 52-59. //? Drechsler, A. (ed.), Denkschriften der Naturwissenschaft- lichen Gesellschaft Isis ze Dresden. Dresden. McDowell, S. B. 1970. On the status and relationships of the Sol- omon Island elapid snakes. J. Zool., 161: 145-190. Mertens, R. 1956. Beobachtungen an Korallenschlangen im Terrarium. Aquar. Terrar. Z., 9:74-77. Minton, J. E. 1949. Coral snake preyed upon by bullfrog. Co- peia, 1949:288. Minton, S. A. 1969. The feeding strike of the timber rattlesnake. J. Herp., 3:121-124. Mitchell, J. D. 1903. The poisonous snakes of Texas with notes on their habits. Trans. Texas Acad. Sci., 5: 19-48. Mole, R. R. 1924. The Trinidad snakes. Proc. Zool. Soc. Lon- don, 1924:235-278. Morse, D. H. 1 980. Behavioral mechanisms in ecology. Harvard Univ. Press, Cambridge. Myers, C. W. 1965. Biology of the ringneck snake, Diadophis punctatus, in Florida. Bull. Florida St. Mus., 10:43-90. Nalleau, G. 1 966. La biologie et le comportement predateur de 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 Al- iens' Reptile Inst., 1:47-60. 1957. Some misconceptions regarding the eastern coral snake, Micrurus fulvius. Herpetologica. 13:111-118. 1968. 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 Jer- sey. Pianka, E. R. 1966. Convexity, desert lizards and spatial heter- ogeneity. Ecology, 47:1055-1059. PlCKWELL, G. V. 1972. The venomous sea snakes. Fauna, 4:17-32. Pvke, G. H., Pulliam, H. R. and Charnov, E. L. 1977. Optimal foraging: a selective review of the- ory' and tests. Q. Rev. Biol., 52:137-154. Quinn, H. R. 1979. Reproduction and growth of the Texas coral snake (Micrurus fulvius tenere). Copeia, 1979: 453-463. Rabb, G. B. and Marx, H. 1973. Major ecological and geographical patterns in the evolution of colubrid snakes. Evolu- tion, 27:69-83. Raix i iffe, C. W. and Chiszar, D. A. 1 980. A descriptive analysis of predatory behavior in the yellow lipped sea krait (Lalicauda col- ubrina).}. Herp.. 14:422-424. Rosenberg, H. 1967. Histology, histochemistry, and emptying mechanism of the venom glands of some elapid snakes. J. Morph., 123:133-156. Roze, J. A. 1967. A checklist of the New World venomous cor- al snakes (Elapidae), with descriptions of new forms. Amer. Mus. Novitates, 2287:1-60. Rliick, J. D. 1948. Collecting coral snakes, Micrurus fulvius te- nere, in Texas. Herpetologica, 4:215-216. Savitzkv, A. H. 1978. The origin of the New World proteroglyph- ous snakes and its bearing on the study of venom delivery systems in snakes. Doctoral dissertation. University of Kansas. Law- rence. 1 980. The role ofvenom delivery strategies in snake evolution. Evolution, 34:1194-1204. Schmidt, K. P. 1932. Stomach contents of some American coral snakes, with the description of a new species of Geophis. Copeia, 1932:6-9. Schoener, T. W. 1969. Optimal size and specialization in constant and fluctuating environments; an energy-time approach. Brookhaven Symp. Biol., 22: 103- 114. 1971. Theory of foraging strategies. Ann. Rev. Ecol. Syst.. 2:369-404. Seib, R. L. 1981. Size and shape in a neotropical burrowing colubrid snake, Geophis nasalis, and its prey. American Zool., 21:933. Shaw, C. E. 1971. The coral snakes, genera Micrurus and Mi- cruroides, of the United States and northern Mexico. Pp. 157-172. In Biicherl, W., and Buckley, E. (eds.). Venomous Animals and Their Venoms. Volume 2, venomous ver- tebrates. 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. SOCHLIREK, E. 1955. Einige Freilandbeobachtungen an Koral- lenschlangen der USA. Aquar. Terrar. (Ber- lin), 2:26-28. Strecker, J. K. 1908. The reptiles and batrachians of McLennan County. Texas. Proc. Biol. Soc. Washington. 11:69-84. Telford, S. R. 1952. A herpetological survey in the vicinity of Lake Shipp, Polk County. Florida. Quart. J. Florida Acad. Sci., 15:175-185. Tollestrup, K. 1980. Sit-and-wait predators versus active forag- ers: do they exist? Amer. Zool., 20:809. 162 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Vitt, L. J., Congdon, J. D. and Dickson, N. A. 1977. Adaptive strategies and energetics of tail au- totomy in lizards. Ecology, 58:326-337. 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. 1981. Size and proportion relationships between the beaked sea snake and its prey. Biotro- pica, 13:15-19. Wright, A. H. and Wright, A. A. 1957. Handbook ofsnakes of the United States and Canada. Comstock Publishing Associates. Ithaca, New York. Zegel, J. C. 1975. Notes on collecting and breeding the eastern coral snake, Micrurus fulviusfulvius (L.). Bull. S. E. Herp. Soc, 1(6):9— 10." Appendix Note 7. — Much of the recent literature on foraging theory is 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- 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 Wil- son 1978; Greene 1980). Maiorana (1978) referred to such hypothetical diet parameters as "ideal," a more appropriate label than optimal in view of the proce- dures 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 1 96 1 ; Greene, MS). Cir- cumstantial evidence suggests that coral snakes have relatively restricted gapes: they have fewer scale rows (15) than most snakes (implying less interscalar skin and consequently less capacity for stretching), rela- tively shorter quadrate bones, and relatively shorter mandibular elements than many other snakes (Marx and Rabb 1972; Greene, MS). As in most other elapids, the maxillae are greatly foreshortened and. at most, slightly mobile. Apparently the palatine and pterygoid bones are mostly restricted to anterior-posterior and vertical movements (McDowell 1970), and this is also suggested bv lateral movements of the entire head dur- ing swallowing (see above, and Frazzetta 1970). My casual observations of captive elapids (Bungarus, Mi- crurus, Naja, Ophiophagus, Walternnesia) indicate that they have more difficulty swallowing prey of a partic- ular relative diameter than do some snakes in other families (e.g., boids, viperids, Atractaspis, some colu- brids). Note 3. — The taxa in Fig. 3 and their average total lengths (to nearest cm) are: Tantilla gracilis (19), Vir- ginia striatula (22), V. valeriae (22), Storeria occipi- tomaculata (23), 5. dekayi (28), Tropidoclonion linea- tum (30). Diadophis punctatus (31), Cemophora coccinea (44), Opheodrys vernalis (44), Sistrurus mil- iaris (45), Lampropeltis triangulum (49), Heterodon platyrhinos (68), O. aestivus (69), Agkistrodon contor- trix (76), L. calligaster (92), Elaphe guttata (99), L. getulus ( 1 07), Coluber constrictor (112), Farancia aba- cura (120), Masticophis flagellum (130), Crotalus hor- ridus (130), Pituophis melanoleucus (132), and E. ob- soleta (145). 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 inter- mediate length are either rare in east Texas (C. cocci- nea, O. vernalis) or proportionately stout for their lengths (A. contortrix, H. platyrhinos, S. miliaris). Note 4—1 attempted to test the responses of two 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 pref- erences for snakes and lizards, because these comprise almost all known natural prey for all sizes of coral 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; see Burghardt 1970, and Arnold 1980, for reviews of the roles of naive preferences and experiential factors in the recognition of food by snakes). 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 i 1 4X4 Museum of Natural History. The University of Kansas. Lawrence The Role of Chemoreception in the Prey Selection of Neonate Reptiles Pennie H. von Achen and James L. Rakestraw Introduction The role of chemoreception in the prey selec- tion of neonate reptiles was the focus of our re- search on ten species of Kansas snakes and two species of Kansas lizards. Recent investigators, most notably Burghardt ( 1 970b. 1971,1 973) have demonstrated innate chemical preferences in cer- tain snakes and lizards for the kinds of prey nor- mally eaten in the wild. However, such innate feeding preferences are subject to some degree of variation, including geographic variation paral- leling those in the animals' natural diets (Burg- hardt 1970a; Arnold 1977). Furthermore, dis- tinct polymorphism within local populations, even within broods, has been demonstrated, ap- parently serving to prevent overspecialization, thereby permitting better utilization of available food resources (Arnold 1977; Burghardt 1975; Gove and Burghardt 1975). Elimination of visual and olfactory senses re- sults in unaltered prey attacks in at least some kinds ofsnakes (Wilde 1938; Burghardt and Hess 1968; Burghardt 1970b). Snakes with vomero- nasal nerve lesions fail to respond differentially to chemical cues (Halpern and Frumin 1979). Therefore, the primary receptor of this chemical information appears to be the vomeronasal sys- tem: the tongue, Jacobson's organ, and associ- ated nerves. This system likely evolved as prim- itive lizards accidently picked up chemicals with the tongue while drinking, eating, and mating. With increased sensitivity of Jacobson's organ and concomitant facilitating manipulations of the tongue, the system became proficient at respond- ing to airborne chemicals (Gove 1979). Since 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 the object (Burghardt 1967). Past works have dealt predominantly with nat- ricines (Burghardt 1967, 1969, 1975; Burghardt and Hess 1 968; Sheffield et a/. 1 968). While these snakes displayed a strong response when tested (many tongue flicks and attacks) others, such as Lampropeltis getulus, appear to be less respon- sive to this methodology (Brock and Myers 1979). Exploration of saurian chemosensory mechanisms, although scant, has shown innate predisposition to chemical stimuli in some species of Eunieces (Loop and Scoville 1972; Burghardt 1973) and Gerrhonotus (Burghardt 1977). The innateness of the behavior implies an evo- lutionary origin; thus one might expect a phy- logenetic basis for the differential reliance upon 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- mates and whether these responses differ among species and among higher taxa regardless of life histories. Methods Subjects. — Twenty-three broods of snakes, representing ten species (one hundred and thirty- four individuals), and four broods of lizards rep- resenting two species (thirty-one individuals) were born in captivity to gravid females captured in central or eastern Kansas (Table 1). The females were kept in individual containers until partu- 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 kept in gallon jars in moist wood frass with 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- ity. Preparation. —A variety of prey animals were collected (Table 2). Although most investigators have followed Burghardt's extract preparation technique (1968). Carr and Gregory (1976) sug- gest that since reptiles presumably respond to odors emanating from the surface of the prey, rubbing a moistened cotton swab over the prey 163 164 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 1. Data on reptiles at time of testing. Species Capture site of gravid female (Kansas county) Brood si/e Age (days) letup. Time (CDT) Eumeces fasciatus Ophisaurus attenuatus Coluber constrictor Lampropeltis calligaster Diadophis punctatus Storena dekavi Thamnophis sirtalis Thamnophis radix Nerodia sipedon Agkistrodon contortrix Sistrurus catenatus Crotalus viridis Johnson VERTEBRATE ECOLOGY AND SYSTEMATICS 165 Scoring. — Burghardl's tongue flick-attack score (1967), an arbitrary value system based on the number of tongue flicks and length of attack la- tencies, was used to calculate a "response profile" for each species (Table 2). 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: Score = base unit + (60-attack latency) The base unit is the 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 species plus the trial length minus the attack la- tency. A reptile which did not attack was given a score identical with the number oftongue flicks emitted towards the swab. Czaplicki ( 1 975) found a high test-retest reliability when this scoring method was used (r = .86). 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. Results In all species a preference was shown for one or more types of prey items over the control swab. Using the Wilcoxon Signed Ranks Test to investigate differences between the control and the most preferred swab (those with the highest tongue flick-attack score for each species), we determined that for four species the difference was significant (P < 0.05) (Fig. 1). In addition. Diadophis punctatus showed a significant re- sponse (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 sirtalis. a paired /-test showed highly significant results (P < 0.005). Preferred stimuli generally corresponded to the prey species in the natural diet as indicated by food samples from the local population. Lizards. — Eumeces fasciatus responded strongly to grasshoppers and spiders (Table 2). Although grasshoppers received the most atten- tion, they are outnumbered two : one by spiders in stomach contents (Fitch 1954). Perhaps this is because as adults, most grasshoppers are too large for even an adult skink to subdue and in- gest. Spiders and orthopterans combined com- prise the bulk of the diet in the wild. Only 6% of the skinks attacked harvestman swabs, com- pared to 47% that attacked the grasshopper swab. Harvestmcn constitute a minor food source in the local population. Ophisaurus attenuatus likewise showed a strong response to spider (P < 0.015) and orthopteran 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 pop- ulation (Fitch, pers. comm.). Although the crick- et swab elicited as many attacks as the spider swab, the spider trials were characterized by shorter attack latencies and more tongue flicks. Snakes.— Coluber constrictor showed m i n imal responses to all prey swabs. Of the three prey offered, worms are not included in their natural diets, mice are often eaten, but crickets are the most frequently consumed prey (Fitch 1 963). The mouse swab elicited the strongest response, al- though that was negligible. No attacks were made on any swab. Lampropeltis calligaster eats mainly small mammals and reptiles (Fitch 1978). While the mouse swab surprisingly aroused no more inter- est than the distilled water swab, the swabs of red sided garter snake and eastern yellow bellied racer did elicit some response as expected. Since ringneck snakes have been found in stomach and scat contents (Fitch 1978). it is somewhat sur- prising that ringneck swabs received so little at- tention (Table 2). No attacks were made. The diet of local populations of Diadophis punctatus is composed almost exclusively of 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 pref- erence for the earthworm swab (P < 0.0001), supporting Collins' (1974) assessment that lo- cally they eat primarily earthworms. There was a weaker, but still significant (P < 0.01) response to the pond snail (Table 2). Thamnophis sirtalis preys chiefly on mam- mals, frogs and worms (Fitch 1965). Fish are not well represented in the diet of the local popula- 166 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. Tongue-flick attack (TFA) scores for 12 species of Kansas reptiles. Pre\ swab Number of attacks Mean TFA score ± SE Eumeces fasciatus (P < 0.023, N = 17)a Distilled water Harvestman (Leiobunum vittatum) Differential grasshopper {Melanoplus differentialis) House spider (Achaeranea tepidariorum) Ophisaurus attenuatus (P < 0.015. N = 14) a Distilled water Differential grasshopper (Melanoplus differentialis) Cricket (Acheta assimilis) Prairie wolf spider (Lycosa rabida) Coluber constrictor (t = 1.39, N = 13. P < 0. 10) b Distilled water Cricket (Acheta assimilis) House mouse (Mus musculus) Earthworm (Allolobophora caliginosa) Lampropeltis calligaster (P < 0.18, N = 5)a Distilled water House mouse (Mus musculus) Red-sided garter snake (Thamnophis sirtalis) Eastern yellow-bellied racer (Coluber constrictor) Prairie ringneck snake (Diadophis punctatus) Diadophis punctatus (P < 0.14, N = 17)a Distilled water Pond snail (Physa hawni) Mealworm (Tenebrio molitor) Earthworm (Allolobophora caliginosa) Storena dekayi (P < 0.0001, N = 56 )a Distilled water Pond snail (Physa hawni) Earthworm (Allolobophora caliginosa) Mealworm (Tenebrio molitor) Thamnophis 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 contortnx (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) 1 VERTEBRATE ECOLOGY AND SYSTEMATICS 167 Table 2. Continued. Prev swab Number of attacks Mean TFA score t SE Sistrurus catenatus (P < 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 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) 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 169 In their test on six families of lizards. Bissinger and Simon (1979) noted that the difference in frequency of tongue extrusions in different fam- ilies might indicate the relative importance of the vomeronasal system. Accordingly, mean re- sponses to preferred prey swabs provide the basis for comparing each species' differential sensitiv- ity to chemical cues. Fig. 1 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 dis- tilled 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 interesting pattern. Ranked by their levels of se- lective chemical discrimination, the reptiles are generally arranged into subfamilies (based on Dowling's [1975] classification). The three species with the lowest response scores. Coluber constrictor, Lampropeltis calli- gaster, 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 small animal stimulates the racer to pursue and attack. Olfaction appears to play a minor role in the finding and capture of prey. 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- cant response. Lampropeltis calligaster, unlike the natricines usually used in these works, is a constrictor. Oth- er investigators using colubrid constrictors have found their responses towards test swabs differ somewhat from that of natricines. Brock and Myers (1979) were unable to find any significant difference between the control and prey swabs for ingestively naive L. getulus. However, Wil- liams and Brisbin ( 1978) found that adult L. ge- tulus had an innate preference for certain prey extracts despite restricted diets. Burghardt and Abeshaheen (1971), working with another col- ubrid constrictor, Elaphe vu/pina, found that, in contrast to a garter snake's immediate attack and ingestion of prey, E. vulpina moved more delib- erately and hesitantly, often taking several hours to constrict and ingest the prey. Perhaps a sixty second trial is insufficient for such snakes. The importance of the vomeronasal system to the Crotalinae seems not to be strongly associ- ated with prey detection. Previous work on rat- tlesnakes (Chiszar and Radclifle 1976; Chiszar et al. 1 978) has demonstrated that visual or ther- mal signals are needed to elicit attacks and rel- atively few tongue flicks are emitted prior to striking. Once the strike has occurred, however, tongue flicking is initiated and continues while the snake trails the stricken prey and investigates the carcass, presumably to locate the head. Tongue flicking again commences after the prey is 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 accom- plished by odor alone. The fourth snake in an intermediate position is a natricine, Nerodia sipedon. Unlike the other natricine tested, N. sipedon exhibited a relatively low degree of tongue flicking and made no prey attacks. Burghardt (1968) has reported similar findings for this species. Because this snake gen- erally feeds in the water, volatile chemical cues may not be as useful as visual or tactile cues. 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 mechan- ical stimuli are adequate. Attack frequency in- creases when diffuse chemical cues accompany visual ones. However, attack can be induced by chemical cues alone. The three snakes with the highest response scores, Storeria dekayi, Thaninophis sirtalis, and Thamnophis radix, are terrestrial natricine non- constrictors that overpower, hold and swallow their prey. Their high response scores corrobo- rate previous findings based on essentially the same testing technique (Burghardt 1967, 1969, 1970a, 1971; Burghardt and Hess 1968). Prey attacks were common and tongue flick activity was among the highest of all the snake species tested, suggesting a strong dependence on che- mosensory methods of prey selection. Sheffield ct al. (1968) note that prey attacks were always 170 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY preceded by at least one tongue flick that actually touched the swab, and we noted only two excep- tions among the one hundred sixty-five individ- uals tested in this study. One E. fasciatus at- tacked a harvestman swab immediately upon introduction, and an O. attenuatus likewise at- tacked a cricket swab. These attacks were not surprising since many lizards are known to re- spond to visual cues, including movement. How- ever, Chis/ar et al. (1976) noted that juvenile Thamnophis flick their tongue more often than do adults. It is possible that other senses grad- ually replace some of the dependence on che- 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 prelim- inary tongue flicks by the predator. Lizards are generally thought to be less depen- dent on Jacobson's organ than snakes. Terrestrial lizards, however, usually have better developed olfactory/vomeronasal organs and concomitant decreased vision as compared with arboreal liz- ards (Gravelle 1980). E. fasciatus exhibited a relatively strong chemical preference when tested, placing it among the natricines in response scores. While Loop and Scoville (1972) found no dif- ferential tongue flicking or prey attack behavior in a congener, E. inexpectatus, our findings sup- port the conclusion of Burghardt ( 1 973) that there is innate chemical recognition of prey. Burghardt also noted, as we did, the very low rate of tongue flicking in Eumeces compared with snakes in similar tests. Ophisaurus attenuatus displayed the highest level of response of all the species tested, both in terms of tongue flicks and attacks. This is con- 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) dem- onstrates that the tongue flick pattern of these 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 rel- atively long time in different species and indi- viduals. Moreover, tests to date are based on adult food preferences. It is known that the young of some species prefer different prey than do the adults (Mushinsky and Lotz 1980). If more were known about juvenile diets, utilization of prey items which would generate maximum responses could be assured. Furthermore, we tentatively conclude that various obscure inhibiting factors altered results to varying degrees, depending on the species and perhaps on the individual. The gentle momen- tary 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 hyper- sensitive 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 re- sponses to light intensities and to temperatures may 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 ofconfusing 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 that nonvolatile macromolecules were the attack stimulus for Thamnophis. Cowles and Phelan (1958) theorized that the external nares, receiv- ing olfactory stimuli, were highly sensitive but oflow discrimination, alerting the snake to visual stimuli of movements and initiating lingual air sampling and subsequent specific analysis by Ja- cobson's organ. In other words, olfaction con- veys 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 sig- nificant 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 Ja- cobson's organ. However, Elaphe (Burghardt and Abeshaheen 1971) and Nerodia (Dunbar 1979) have demonstrated discrimination of prey on the VERTEBRATE ECOLOGY AND SYSTEMATICA 171 basis of volatile chemical cues alone. In our tests, the constrictor L. calligaster displayed substan- tial discriminatory behavior when both touching and non-touching tongue flicks are considered. By contrast, only a weak response was made di- rectly towards the swab. On the other hand, N. sipedon, similarly appraised on the basis of all tongue tlicks made, responded more to the con- trol than to the scented swabs. Clearly more in- vestigation is needed to determine the most meaningful scoring procedure to use. With some reservations due to the foregoing constraints, we conclude that purposeful tongue extrusions and attacks do consistently vary among 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. Summary Inexperienced young oftwo 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 flicks and attacks, var- ious levels of discriminatory behavior were re- corded, suggesting the relative importance of chemoreception in prey selection for each species. The reptiles tested can be ranked according to their levels of responsivity to chemical stimuli. Tending to conform with subfamily groupings, the colubrinae show the lowest level of discrim- inatory behavior, natricines, the highest level for snakes, and crotalines, intermediate between the two. The saurian Eumcces ranks among the nat- ricines, 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. encouragement, information on feeding habits, and critical comments regarding this manuscript. Li I I KATURE C III I) Arnoi i). S. J. 1977. Polymorphism and geographic variation in the feeding behavior of the garter snake Thamnophis elegans. Science, 197:676-678. 1978. Some effects of early experience on feeding responses in the common garter snake. Thamnophis sirtalis. Animal Behav., 26:455- 462. Bissinger. B. E. and Simon. C. A. 1979. Comparison of tongue extrusions in repre- sentatives of six families of lizards (Reptilia. Lacertilia). J. Herp.. 1 3(2): 1 33-1 39. 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. Chemical-cue preferences of inexperienced snakes: comparative aspects. Science, 157: 717-721. 1968. Chemical preference studies on newborn snakes of three sympatric species of Natrix. Copeia. 4:732-737. 1 969. Comparative prey-attack studies in newborn snakes of the genus Thamnophis. Behavior. 33:77-114. 1 970a. Intraspecific geographical variation in chem- ical food cue preferences of newborn garter snakes ( Thamnophis sirtalis). Behavior. 36: 246-257. 1970b. Chemical perception in reptiles. Pp. 241- 308. In Johnston, J. W., Moulton. D. G. and Turk. A. (eds.).. Communication by Chem- ical Signals. Appleton-Century-Crofts, New York. 1971. Chemical-cue preferences of newborn snakes: influence of prenatal maternal experience. Science, 171:921-923. 1973. Chemical release of prey attack: extension to naive newly hatched lizards, Eumcces fas- cial us. Copeia. 1:178-181. 1975. Chemical prey preference polymorphism in newborn garter snakes. Thamnophis sirtalis. Behaviour. 52:202-225. 1 977. The ontogeny, evolution, and stimulus con- trol of feeding in humans and reptiles. Pp. 253-275. //; Rare. M. and Mailer. O. (eds.). The Chemical Senses and Nutrition. Aca- demic 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. 1975. Role of tongue and senses in feeding of naive 172 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY and experienced 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 palatable prey induced by delayed illness. Animal Learn, and Behav.. 1(4):3 17-320. Carr. C. M. and Gregory, P. T. 1976. Can tongue flicks be used to measure niche sizes? Canadian J. Zool., 54:1389-1394. Chiszar, D.. Carter, T., Knight, L., Simonsen. L. and Taylor, S. 1976. Investigatory behavior in the plains garter snake (Thamnophis radix) and several ad- ditional species. Animal Learn, and Behav.. 4(3):273-278. Chiszar, D. and Radcliffe, C. W. 1976. Rate oftongue 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. 1978. Chemosensory searching for wounded prey by rattlesnakes is released by striking: a rep- lication report. Herp. Review, 9(2):54-56. Collins, J. T. 1974. Amphibians and reptiles in Kansas. Univ. Kansas Mus. Nat. Hist. Pub. Educ. Ser. 1: 1-283. Cowles, R. B. and Phelan. R. L. 1958. Olfaction in rattlesnakes. Copeia, 2:77-83. Czaplicki, J. A. 1975. Habituation of the chemically elicited prey- attack response in the diamond-backed water snake, Natrix rhombifera rhombifera. Her- petologica. 31:403-409. Dowling, H. G. 1975. A provisional classification of snakes. In 1974 Yearbook of Herpetology. HISS Publica- tions. New York. Drummond, H. M. 1979. Stimulus control ofamphibious predation in the northern water snake (Nerodia sipedon sipedon). Zeitsch. fur Tierpsychol.. 50:18- 44. Dunbar. G. L. 1979. Effects of early feeding experience on chem- ical preference of the northern water snake. Natrix sipedon sipedon (Reptilia, Serpentes, Colubridae). J. Herp.. 13(2): 165-169. Fitch. H. S. 1954. Life history and ecology ofthe five-lined skink (Eumeces fasciatus). Univ. Kansas Publ., Mus. Nat. Hist., 8:1-156. 1 960. Autecology of the copperhead. Univ. Kansas Publ., Mus. Nat. Hist., 13:85-288. 1963. Natural history of the racer Coluber con- strictor. Univ. Kansas Publ.. Mus. Nat. Hist., 15:351-468. 1965. An ecological study of the garter snake, Thamnophis sirtalis. Univ. Kansas Publ., Mus. Nat. Hist., 15:493-564. 1975. A demographic study of the ringneck snake {Diadoplus punctatus) in Kansas. Univ. Kansas Mus. Nat. Hist. Misc. Publ., 62:1- 53. 1 978. A field study of the prairie kingsnake. Trans. Kansas Acad. Sci. 81(4):353-363. Gove, D. 1979. A comparative study of snake and lizard tongue-flicking, with an evolutionary hy- pothesis. Zeitsch. Fur Tierpsychol., 51:58- 76. Gove, D. and Burghardt, G. M. 1975. Responses of ecologically dissimilar popu- lations of the water snake Natrix sipedon si- pedon to chemical cues from prey. J. Chem. Ecol., 1:25-40. Gravelle, K. and Simon. C. A. 1980. Field observations on the use of the tongue- Jacobson's organ system in two Iguanid liz- ards, Sceloporusjarrovi and Anolis trinitatis. Copeia, 2:356-359. Halpern, M. and Frumin, N. 1 979. Roles of the vomeronasal and olfactory sys- tems in prey attack and feeding in adult gar- ter snakes. Physiol, and Behav., 22:1183- 1189. Loop, M.S. and Scoville, S. A. 1972. Response of newborn Eumeces inexpectatus to prey-object extracts. Herpetologica, 28: 254-256. Mushinsky, H. R. and Lotz, K. H. 1 980. Chemoreceptive responses of two sympatric water snakes to extracts of commonly in- gested prey species. J. Chem. Ecol., 6(3):523- 535. 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., A(2):7-12. Wilde, W. S. 1938. 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 king- snakes (Lampropeltis getulus) to several prey odor stimuli. Herpetologica, 34:79-83. Vertebrate Ecology and Systematics—A Tribute to Henr> S. Fitch Edited by R. A. Seigcl. L. E. Hunt. .1 L Knight. L Malaret and N. L. Zuschlag j 14X4 Museum of" Natural History, The University of Kansas. Lawrence Ecology of Small Fossorial Australian Snakes of the Genera Neelaps and Simoselaps (Serpentes, Elapidae) Richard Shine Introduction Most of continental Australia is arid, but the herpetofauna ofthese enormous deserts is poorly known. Recent studies have clarified the tax- onomy of Australian desert reptiles (e.g., Storr 1967, 1979; Greer 1979; Cogger 1979), but the ecology of these animals remains virtually un- studied. The present paper is based on dissec- tions 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 gen- era are closely related to each other. All of the Neelaps and Simoselaps species are characterized by small body size (< 50 cm snout- vent length), bright colouration, and fossorial habits. Five Simoselaps species show a pro- nounced upturned edge on the rostral scale, pre- sumably as an adaptation to burrowing. In this regard, as well as in general appearance and hab- its, Neelaps and Simoselaps are strikingly con- vergent with small sand-dwelling snakes from other continents (e.g., Chilomeniscus, Chionac- tis, Ficimia, Gyalopion in North America; Pro- symna and Elapsoidea in Africa). Although Neelaps and Simoselaps species may be among the most abundant snakes over most of Australia, they have attracted little study. Storr (1967, 1979) recently has revised the group, de- scribing several new species in the process. Ac- cording to Storr ( 1 967), Neelaps and Simoselaps are not sufficiently distinct to warrant generic separation; Storr considers that they belong to the single large genus Vermicella (together with the Bandy-Bandy, V. annulata). Cogger (1979) prefers to recognize the three genera separately. I follow Cogger's (1979) nomenclature in the present study, because of my subjective impres- sion that Vermicella {sensu Storr) is too heter- ogeneous a group. Geographic distributions of the Neelaps and Simoselaps species are given by Storr (1967, 1 979) and Cogger (1979) and are briefly summarized in Table 1. Within these two genera, at least five distinct "species-groups" are evident: (i) Neelaps bimaculatus and N. calonotus are slender unbanded species of south-western Australia. (ii) The Simoselaps "bertholdi group" (ber- tholdi, anomala, littoralis. minima) are all short heavy-bodied snakes with distinct yellow-and-black bands and lacking an up- turned rostral (Fig. 1 ). (iii) The Simoselaps "semifasciatus group" {semifasciatus, approximans, incinctus, ro- peri) have less distinct bands along the body {incinctus lacks bands). The rostral has a sharply upturned, angular leading edge (Fig. 1). (iv) Simoselaps australis and S. fasciolatus may not be closely related to each other. Both species resemble S. semifasciatus in general shape and colour; 5". australis has a sharp- edged rostral whereas S. fasciolatus does not. (v) Simoselaps warro is a rainforest species of northeastern Queensland, and is so aber- rant that it is only doubtfully included in this genus (Storr 1979). Methods I examined all specimens (N = 953) ofNeelaps and Simoselaps in the collections of the Western Australian Museum, the South Australian Mu- seum, the National Museum, the Queensland Museum and the Australian Museum. I took the following data from each specimen: (i) snout- vent length (SVL), measured by running a tape measure along the body; (ii) gut contents; (iii) reproductive maturity or immaturity (criteria were: males— large testes or opaque efferent ducts: females— gravid, large oviducts, or ovarian fol- licles >5 mm); and (iv) diameters of ovarian follicles or oviducal eggs, in mature females. Growth rates were estimated from seasonal dis- tributions ofbody sizes; this method is explained in more detail below. 173 174 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 1. Sample sizes, body sizes and sexual size dimorphism in Neelaps and Simoselaps. VERTEBRATE ECOLOGY AND SYSTEMATICA 175 41 * • ^ ^^^^ +<» 176 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. Prey items found in stomachs of Neelaps and Simoselaps species. Pre) items VERTEBRATE ECOLOGY AND SYSTEMATICS 177 IU u 3- OO O, 20 34 Fig. 2. 22 24 26 28 30 32 MEAN SNOUT -VENT LENGTH OF ADULT 00 (cm) Fecundity and inferred body-size at hatching in Neelaps and Simoselaps species. 3.99, P > .20) or S. semifasciatus (N = 88, 3d.f., X 2 = 7.73, P > .05). Neither are the seasonal dis- tributions of these two species different from each other (N = 205, 3 d.f, x 2 = 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. semi- fasciatus, feeding is most common in summer, and ceases during winter. Discussion Body Sizes.— The tendency for females to grow larger than 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 Caco- phis, Furina, Drysdalia and Vermicella; males larger in 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). Food Habits. — Published literature generally is in error on this subject. Glauert suggested that most Simoselaps species eat "insects and other small forms of life, including frogs and small lizards" (1957, p. 40). Kinghorn ( 1 964) suggested that the diet of Neelaps bimaculatus consisted mainly of small insects. Mackay (1949) believed that S. australis fed on slugs, beetle larvae, and worms. McPhee (1979) noted that S. semifas- ciatus probably ate only insects. Worrell (1963) correctly asserted that S. bertholdi feeds on skinks. Gow (1976) suggested that N. bimaculatus feeds on insects, and that 5. australis probably does also (as well as feeding on skinks). Gow (1976) credited 5". bertholdi with feeding on insects, frogs and lizards, 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 178 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY CO LU < z IS) 61 2- S. bertholdi co Z 6-1 4- 2- 8 1 1 1 1 1 1 1 10 — i — 12 14 S. semifasciatus .iiiiiii - 1 — 16 SNOUT-VENT LENGTH (cm) —r— 18 H Fig. 3. Body-size distributions of juvenile Simoselaps bertholdi and S. semifasciatus collected during the 'non-growing season" (April through October). prey type of Neelaps and the Simoselaps "ber- tholdi group," and squamate eggs are the only prey taken by the S. "semifasciatus group. " Stud- ies on several other small Australian elapids have revealed an analogous situation: published lit- erature is almost unanimous in suggesting that invertebrates are the main food items, but dis- sections show that lizards comprise the major portion of the diet. This is true in Cacophis, Fu- rina, 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- sia. 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 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 ofthe snakes. Waite (1929), Worrell (1963) and Gow (1976) note that S. bertholdi may be active diurnally, unlike all the other (nocturnal) Simoselaps for which records are available. Simoselaps ber- tholdi subdues its scincid prey by constriction, in the same manner as do pythons (Bush 1981). The specialization of the Simoselaps "semi- fasciatus group" on squamate eggs was an un- expected finding. No other Australian snakes are known to feed predominantly upon eggs, al- though 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 con- tained lizards, mice, frogs and orthopterans). Oc- casional oophagy has been recorded in several other Australian elapids. These include small species such as Cacophis harrietae (N = 4 eggs), C. squamulosus (N = 9) (Shine 1980a), Deman- sia olivacea (N = 1), D. psammophis (N = 2) (Shine 1980c), and Drysda/ia coronoides (N = 6) (Shine 1981a). Eggs have also been found in the stomachs of large species: Austrelaps superbus (N = 2), Pseudechis porphyriacus (N = 3) and Pseudonaja textilis (N = 8) (Shine 1977c). How- ever, eggs form only a small part of the diet in all of these species. In contrast, squamate eggs were the only food recorded in stomachs of Simoselaps roperi (N = VERTEBRATE ECOLOGY AND SYSTEMATICS 179 Nee/ops O x x » OO " O * 1 1 ! 1 1 1 1 1 1 1 1 S JFMAMJJASOND < —i O 20 15- 10- S. bertholdi O O O Ox ox OX OOO O J O x oo M 25 n 20 <> 10 20 15- 10 5- o 10, MONTH S auitralu o° ° o I I I 1 1 1 1 1 1 1 1JFMAMJJASOND S semifosciohjs OO O OO o ^ O o OO o OCXDOOCO OO O o M M J T" -T- O N Fig. 4. Monthly variation in diameter of the largest ovarian follicle in mature female Neelaps and Simoselaps: in Neelaps graph, circles show A', bimaculatus, crosses show N. calonotus; in S. bertholdi graph, circles show S. bertholdi, crosses show 5. littoralis. In all graphs, dots show oviducal eggs. 3) and S. semifasciatus (N = 27). were common in 5. australis (3 of 4 food items) and present in S. fasciolatus (1 of 4 food items). The lack of recognizable embryos in all these eggs suggests that the oophagous Simoselaps raid nests soon after oviposition. 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. However, some oophagous snakes take eggs at all stages ofembryonic development (e.g., Pro- stvwja-Broadley 1979; Oligodon— Wall 1921). Snakes that feed primarily on squamate eggs may show morphological adaptations to this diet. A clear correlate of oophagy within Neelaps and Simoselaps is the shape of the rostral scale. The "shovel-nosed" species (i.e., those with an up- turned angular edge to the rostral) are egg-eaters (S. "semifasciatus group," S. australis), whereas the snakes lacking this feature (Neelaps and other Simoselaps) are saurophagous. A survey of pub- lished literature reveals the same general corre- lation among snakes in general, but there are many exceptions. Reptile eggs are an important dietary component of several "shovel-nosed" species (e.g., Prosymna— Broadley 1979: Phyl- lorhynchus— Klauber 1940; Salvadora— Blair 1960; Oligodon— Wall 1921), and an occasional component in others (e.g., Rhinocheilus lecontei tessellatus—Sha-w and Campbell 1974; Aspide- laps— Branch 1979; Heterodon— Shaw and Campbell 1974). However, some "shovel-nosed" species do not feed on reptile eggs: Lytorhynchus is saurophagous (Minton 1 966) and the extensive array of North American desert colubrids with upturned rostrals (e.g.. Chionactis, Chilomenis- cus, ticimia, Gyalopion) feed only on inverte- brates (e.g., Shaw and Campbell 1974). Similar- ly, the upturned rostral is lacking from at least two snake species that feed mainly on squamate eggs (Cemophora coccinea— Palmer and Tre- gembo 1970; Elapsoidea sundevalli— Branch 180 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 3. Fecundity of Neelaps and Simoselaps species: Table gives values to solve the equation y where y = clutch size and x = 9 SVL (cm). Regression fit by least squares. ax + b VERTEBRATE ECOLOGY AND SYSTEMATICS 181 i < CO < - O : -2o O z o O a. O 182 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY parent: scincid and pygopodid lizards are the only prey taken by Neelaps species and by the 57- moselaps bertholdi species-group, whereas the Simoselaps semifasciatus species-group feeds ex- clusively on squamate eggs. Oophagy is common also in 5. australis, and recorded in 5. fascio- 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, in saurophagous as well as oophagous species. At least five of the thirteen species studied are oviparous. In mature females, vitellogenesis commences in spring, with ovulation in summer. Fecundity is low (mean clutch sizes 2.5 to 5.3), and is correlated with mean adult body size in an interspecific comparison. Body size at hatch- ing also increases with mean adult body size. Females attain larger body sizes than males, and mature at larger sizes. Analysis of body-size dis- tributions suggests that sexual maturity is at- tained at 20 to 32 months of age. Acknowledgments This study would not have been possible with- out the full co-operation of the following cura- tors: G. M. Storr (Western Australian Museum), A. Edwards (South Australian Museum), J. Cov- acevich (Queensland Museum), J. Coventry (Na- tional Museum of Victoria), A. E. Greer and H. G. Cogger (Australian Museum). Too all ofthem, I am grateful. I thank especially Allen Greer for his 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 squa- mate ecology have laid the foundation for all subsequent work in this field. Literature Cited Blair, W. F. 1960. The Rusty Lizard. A population study. Univ. Texas Press, Austin. 185 pp. 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. Bush, B. 1981. Reptiles of the Kalgoorlie-Esperance region. Vanguard Press, Perth. 46 pp. Cogger, H. G. 1 979. Reptiles and Amphibians of Australia. A. H. & A. W. Reed, Sydney. 608 pp. Glauert, L. 1957. A Handbook of the Snakes of Western Aus- tralia. W. A. Naturalists Club, Perth. 62 pp. Gow, G. F. 1976. Snakes of Australia. Angus & Robertson, Sydney. 88 pp. Greer. A. E. 1 979. Eremiascincus, a new generic name for some Australian sand swimming skinks. Rec. Aust. Mus., 32(7):321-338. Kinghorn, J. R. 1964. The Snakes of Australia. Angus & Robert- son, Sydney. 197 pp. Klauber, L. M. 1940. 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. 1 949. The Australian coral snake. Proc. Roy. Zool. Soc. N.S.W., 1949:36-37. McDowell, S. R. 1972. The species of Stegonotus (Serpentes, Col- ubridae) in Papua New Guinea. Zool. Med- ed., 47:6-26. McPhee, D. R. 1979. The Observer's Book of Snakes and Lizards of Australia. Methuen, 157 pp. Minton, S. A., Jr. 1966. A contribution to the herpetology of West Pakistan. Bull. Amer. Mus. Nat. Hist., 134: 27-184. Moll, E. O. and Legler, J. M. 1971. The life history of a neotropical slider turtle, Pseudemvs scripta (SchoepfT), in Panama. Bull. L.A. County Mus. (Sci), 1 1:1-102. Palmer, W. M. and Tregembo, G. 1970. Notes on the natural history of the scarlet snake Cemophora coccinea copei Jan in North Carolina. Herpetologica, 26:300-302. Shaw, C. E. and Campbell, S. 1974. Snakes of the American west. A. A. Knopf, N.Y. 329 pp. Shine, R. 1977a. Reproduction in Australian elapid snakes. I. Testicular cycles and mating seasons. Aust. J. 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. Herpe- tologica, 34:73-79. 1978b. Sexual size dimorphism and male combat in snakes. Oecologia. 33:269-278. VERTEBRATE ECOLOGY AND SYSTEMATICS 183 1980a. Comparative ecology of three Australian snake species of the genus Cacophis (Ser- pentes: Elapidae). Copeia 1980:831-838. 1980b. Reproduction, feeding and growth in the Australian burrowing snake 1 'ermicella an- nulate. J. Herpetol.. 14:71-77. 1 980c. Ecology of eastern Australian whipsnakes of the genus Demansia. J. Herpetol., 14:381- 389. 1981a. Venomous snakes in cold climates: ecology ofthe 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. The genus Vermicella (Serpentes, Elapidae) in Western Australia and the Northern Ter- ritory. J. Roy. Soc. Western Aust. 50:80-92. 1979. Revisionary notes on the genus Vermicella (Serpentes, Elapidae). Rec. West. Aust. Mus., 8:75-79. Waite, E. R. 1 929. The Reptiles and Amphibians of South Aus- tralia. Government Printer, Adelaide. 270 pp. Wall, W. F. 1921. Ophidia Taprobanica. or the Snakes of Cey- lon. H. R. Cottle, Government Printer, Co- lombo. 581 pp. Worrell, E. 1963. Reptiles of Australia. Angus & Robertson, Sydney. 207 pp. 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 rela- tively rare in collections and are something of a curiosity since they possess two enigmatic char- acteristics: all taxa have extremely long tails and exhibit a high incidence of tail injuries; and one oftwo basic color patterns exhibited by members of the genus has coral snake-like banding re- stricted to the anterior portion of the body. In addition, they have peculiar hinged, shovel-like teeth (Savitzky 1981). The purposes of this paper are first to sum- marize what is known about the natural history of Scaphiodontophis and, second, to test a hy- pothesis related to coral snake mimicry. Note.— Scaphiodontophis taxonomy has long been in a chaotic state. However, an unpublished dissertation by Morgan (1973) dealing with the entire colubrid subfamily Sibynophiinae has clarified the situation. Morgan recognized two monotypic species of Scaphiodontophis: S. an- nulatus (including S. carpicinctus and S. zeteki listed by Peters and Orejas-Miranda [1970]) and S. venustissimus. I follow Morgan's (1973) clas- sification in this paper. Natural History Habitat. — Scaphiodontophis ranges from southern Tamaulipas, Mexico to northern Co- lombia. S. annalatus is primarily a rainforest inhabitant (Alvarez del Toro 1960; Duellman 1965; Martin 1955; Neill and Allen 1959; Stuart 1935,1 958; Wilson and Meyer 1 982). It has also been taken in dense scrub forest (Duellman 1 965). pine savanna and parkland (Henderson and Hoevers 1975), coffee groves (Slevin 1939; Tay- lor and Smith 1943), banana plantations (Roze 1969) and citrus groves (McCoy 1970). It is a leaf litter species (Alvarez del Toro 1960; Hen- derson 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). Food and Feeding Behavior. — Both species of Scaphiodontophis are stenophagous. apparently feeding almost exclusively on scincid lizards of the genus Sphenomorphus in nature (Alvarez del Toro 1960; Landy etal. 1966; Scott 1969; Stuart 1 948; Taylor and Smith 1 943; pers. observ.), but also taking Gymnophthalmus (Teiidae) and Eu- meces (Scincidae) (Alvarez del Toro, in litt.). I observed a captive S. venustissimus prowling on the leaf litter-covered floor of its cage during the day. 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. venustis- simus 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- ably uncommon in Scaphiodontophis and prey is usually swallowed alive and with incredible speed (Alvarez del Toro 1960; R. W. Van De- vender, in litt.: pers. observ.). I recorded amount of time necessary for prey (Scincella lateralis) ingestion by a 41.0 cm SVL 5. annulatus with a stopwatch on eight occasions. The watch was started as soon as the snake grasped the lizard and it was stopped when the lizard's body (tails were removed immediately prior to feeding) was no longer visible. Mean time of ingestion for six lizards between 40-47 mm 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 mm SVL took 5.2 sec and 20.5 sec for inges- tion. One 45 mm SVL skink which did not have its tail 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. Defensive Behavior. — The only display ob- served in Scaphiodontophis has been tail and body thrashing. J. W. Wright (in litt.) observed it in S. annulatus in the field in northern Guatemala. An S. annulatus "was found in a fallen bush that was overgrown with herbs and grasses along with a considerable amount of leaf litter. Mv attention 185 186 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY y*~-**>^^s ' "*3«v3k Fig. 1. Scaphiodontophis venustissimus from Limon, Costa Rica. (Photo by R. W. Van Devender.) was called to the snake because of the noise it was making. I heard the thrashing right up to the point ofclose inspection. . . . The banded portion 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 prat- tling) ofsome snakes, as much ofthe body moved as well." Likewise, I have observed that S. an- nulatus 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. venustissi- mus. This may be comparable to the tail thrash- ing 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.) col- lected three S. annulatus eggs from beneath a rotting log in Veracruz, Mexico on 3 August 1 964. One egg was opened on 14 September and it contained a living male snake of 1 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 2 bands, 1 with 4 bands). A specimen of S. ven- ustissimus from Limon Prov., Costa Rica had a clutch of three eggs; the gravid snake was 43.5 cm SVL (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 187 Fig. 2. Scaphiodontophis annulatus from Honduras. (Photo by L. W. Porras.) painted in rings but not with coral snake colors, did not cause alarm and were attacked by the birds with little or no hesitation. Likewise, when 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 un- painted end. Other end-third models, in color and pattern combinations as for the solid models, had pecks directed mostly or entirely at the paint- ed ends. The S. venustissimus (Fig. 1) pattern has black bands bordered by yellow bands and red inter- spaces: this pattern typically covers the entire length of the body and the tail. The venter is yellowish and marked with small dark spots. S. venustissimus almost invariably exhibits this pattern. In 5. annulatus (Fig. 2) banding is fre- quently restricted to the anterior part of the body or to the entire 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 ground color with 3 rows of dark spots, giving the impression of stripes. Again, the venter is 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 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 damage is predator inflicted (see below) and since S. annulatus is a living example of Smith's ( 1 975, 1977) "end-third" model and .S\ venustissimus a living example of her "solid ring" model. I hy- 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 IKS SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Fig. 3. 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. annu- latus were banded the entire length of their body and tail. These specimens were regarded as hav- ing the 5. venustissimus-type pattern. Two spec- imens I examined had a "dugandi" pattern (Roze 1969; Henderson 1983) in which the anterior portion of the body is banded, the posterior por- tion is not banded, but the tail is banded (Fig. 3). These specimens were discounted from the statistical treatment. An additional number of snakes was examined but disregarded because their tail injuries appeared recent (i.e., the pos- terior end 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 mi- croscope was used to examine tails in which damage was not obvious, but with which the possibility existed that a minor injury may have occurred. Results. — Twenty-seven of 48 (56.3%) snakes with the 5. annulatus pattern had sustained tail injuries while 15 of 29 (51.7%) with the S. ven- ustissimus pattern had suffered tail injuries. Us- ing a chi-square contingency table, the incidence of injured vs. uninjured tails in S. annulatus- patterned and S. venustissimus-patlerned snakes was compared. Differences were found to be not significant (P > .05, x 2 = 0.1494, 1 d.f.). Discussion Scaphiodontophis venustissimus resembles a coral snake, primarily Micrurus nigrocinctus (Dunn 1954; Hecht and Marien 1956; Savage and Vial 1974). S. annulatus is, however, enig- matic in that individuals frequently have band- ing restricted to the anterior portion of the body. Perhaps, as suggested by Echternacht (1973) mimicry may just be in a developmental stage "in some 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 189 Tabu 1. Ratios of tail length to SVL and total length in snakes various adaptive zones as compared to male Scaphiodontophis annulatus. ( ienera species" Tail length SVL" ( ienera species" Tail length loial length*** S. annulatus* Terrestrial Arboreal terrestrial Arboreal Micrurus 1/1 190 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Do these data suggest that long-tailed snakes are more susceptible to predation? Possibly, but alternatively it suggests that such species are more successful at escaping conflicts with predators 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 pre- hensile tails (such as Dipsas catesbyi and Im- antodes cenchoa) may show a low incidence of tail breaks because the tail is usually wrapped around a branch making it more 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 se- rious and escape less likely. Being nocturnal may also decrease the incidence of predation. The frequency of broken or regenerated tails in lizards has been used as an indicator of pre- dation intensity (e.g., Brooks 1967; Parker 1972; Parker and Pianka 1975; Schall and Pianka 1980; Vitt et al. 1977), although Schoener (1979) and Jaksic 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- ples. Some size class samples show tail breakage incidence at 100% (e.g., Clark [1971] for Scin- cella lateralis). Taylor and Smith (1943) attrib- uted the high incidence of stub tails in Scaphio- dontophis to disease, but Taylor later (1954) reported that 5. venustissimus apparently breaks its tail deliberately when restrained by it. On three occasions he grabbed a snake by the 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 in- dication of tail disease in preserved specimens of Scaphiodontophis and therefore attribute the 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- 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 tail lost is gone forever. To compensate for this, Scaphiodonto- phis, and possibly other snake species (e.g., Den- drophidion dendrophis and Rhadinaea deci- piens), have evolved a tail long enough to sustain several predator-inflicted breaks. If a predator were to grab a Scaphiodontophis near the base of the tail, that may be the only attack the tail (and the snake) could sustain. Liner ( 1 960) found that the tail of a Pliocerus elapoides "was given off like that of a lizard" when he tried to pick it up caudally. Wilson ( 1 968) examined caudal ver- tebrae in P. elapoides and described a fracture plane on the expanded transverse processes. He also found very shallow grooving on the trans- verse processes of "a few caudal vertebrae" from a single specimen of S. zeteki nothus (=S. an- nulatus). (Morgan [1973], however, found no evidence of a fracture plane in the caudal ver- tebrae 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 verte- brae of Pliocercus and perhaps Scaphiodontophis is a point of sufficient weakness that allows the vertebrae to break when the snake is seized by the tail. The selective advantage of this adapta- tion seems obvious. As in lizards, the essential portion to the animal survives while the tail re- mains behind to occupy a predator." Although the results of the chi-square test do not support my hypothesis (i.e., Scaphiodonto- phis 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- alikes are preyed upon by birds (Howell 1957; Pough 1964; Skutch 1971; Smith 1969) and mammals (Jackson 1979) despite having apo- sematic patterns. Since these snakes are going to be preyed upon despite their aposematic colors and pattern, a tail which can be easily autoto- mized can only be advantageous. Alternately, Scaphiodontophis annulatus is es- sentially bi-patterned, whereas S. venustissimus is uni-patterned. Since 5. annulatus is bi-pat- terned, a predator has the option of choosing which pattern to attack: a dull-colored, striped VERTEBRATE ECOLOGY AND SYSTEMATICS 191 Fig. 4. Scaphiodontophis anmdatus (41.0 cm SVL) from Honduras in a typical diurnal posture. (Photo by R. W. Henderson.) pattern, or a brightly colored banded pattern. I suggest that the predator would more often attack the dull, striped pattern and Smith's (1975, 1977) experiments support this; naive birds would at- tack a dowel if the coral snake pattern was re- stricted 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 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 preda- tors. In addition, the coral snake pattern, even without benefit of a tail display (Gehlbach 1972) may inhibit and confuse a predator and cause it to go for the 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 au- totomizeable. Other leaf litter snakes may be ex- 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 tail damage and color pattern in Scaphiodontophis: 1) Scaphiodontophis {anmdatus and venustis- simus) 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 poten- tially confusing and/or inhibitive to potential predators. 2) An alternative interpretation of pattern function in Scaphiodontophis is based on the an- tipredator strategies of flight and defense, and on the phenomenon of flicker fusion (Jackson et al. 1 976). Brattstrom (1955) suggested that the coral snake pattern conceals the bearer by a disruptive effect when it is immobile and, when fleeing, the banded pattern prohibits the predator from shift- ing focus rapidly forward to maintain the snake in its field of vision and the snake may therefore escape. Jackson et al. (1976) concurred with 192 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Brattstrom (1955) that a regularly banded pat- tern "may represent a compromise between the strategies of disruptive concealment and gener- ation of a deflective illusion during flight." Sca- phiodontophis annulatus bears a pattern that can be regularly or irregularly banded anteriorly and is essentially striped posteriorly. According to Jackson et al. (1976), a striped pattern is asso- ciated more with flight than defense, and they came to the conclusion that "The aposematic- mimetic functions might be most useful with predators, like birds, that can perceive color while the disruptive-deflective effect might be of greatest value against color-insensitive mam- malian predators. . . ." Observations of captive S. annulatus, even in relatively small enclosures where there is not op- portunity for prolonged movement, have illus- trated to me the confusing effect of the pattern. Even with a priori knowledge of the appearance of 5. annulatus, it takes several seconds to de- termine where the head is 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 the rest of it under a cover object (Fig. 4). I be- lieve, like Jackson et al. (1976) and as stated by Pough (1976), that "... a pattern of brightly col- ored crossbands can be both cryptic and con- spicuous depending on the light conditions, the visual capacity of the predator involved, and the behavior of the snake." 3) Finally, Wright's field observations of S. annulatus, along with its unusual color pattern, suggests a function of tail thrashing apart from defense. Sphenomorphus cherriei is apparently the primary prey species of Scaphiodontophis, and it would not be surprising that such extreme stenophagy would give rise to anatomical and possibly behavioral characteristics adapted to Sphenomorphus predation. Savitzky (1981) has noted a number of anatomical peculiarities in Scaphiodontophis which apparently are adapta- tions for swallowing hard-bodied prey (i.e., scin- cid lizards). Sphenomorphus cherriei is, like Sca- phiodontophis, a rainforest, leaf litter inhabitant (Fitch 1973; Stuart 1958; pers. observ.) 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 by thrashing its tail through the leaf litter? Al- though the moving tail could attract the attention of potential predators (as it did Wright), the long autotomizeable tail, which would be the part of the snake most likely attacked, is adapted to sus- tain such attacks. The motionless, coral snake- patterned head would attract less attention and is mimetically colored. Summary 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- tophis venustissimus has coral snake-like colors and pattern the entire length of the body and tail, but S. annulatus usually has the coral snake colors and pattern restricted to the anterior one-half (or less) of the body. Preserved S. venustissimus ex- hibit fewer tail injuries, but not significantly so, than S. annulatus. Assuming that the tail injuries are predator inflicted, the incidence of injuries in the two species suggests that the coral snake pattern and colors confer no selective advantage to S. venustissimus over S. annulatus. I conclude that the patterns of both snakes are confusing and inhibitory to potential predators: in venus- tissimus because it is a coral snake mimic, and in annulatus because it is bi-patterned and a predator must choose which pattern to attack (most likely the non-coral snake-like posterior). The tails of both species are autotomizeable and potentially able to sustain several breaks. Another potential function of the color and pattern of 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. Acknowledgments 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. F. Pyburn, A. H. Savitzky, H. K. Voris, J. W. Wright, and R. G. Zweifel. For the use of photographs I thank Louis Por- ras and R. W. Van Devender. Miguel Alvarez del Toro, R. W. Van Devender and J. W. Wright provided useful field obser- vations of Scaphiodontophis. VERTEBRATE ECOLOGY AND SYSTEMATICA 193 H. W. Greene, M. A. Nickerson, J. 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. Literature Cited Alvarez del Toro, M. 1960. Reptiles de Chiapas. Instituto Zoologico de Estado (Chiapas), Tuxtla Gutierrez. Alvarez del Toro, M. and Smith, H. M. 1958. Notulae herpetologicae Chiapsiae II. Her- petologica. 14:15-17. Brattstrom, B. H. 1955. The coral snake "mimic" problem and pro- tective coloration. Evolution, 9:217-219. Brooks, G. R.. Jr. 1967. Population ecology of the ground skink, Ly- gosoma laterale (Say). Ecol. Monogr., 37:71- 87. Clark, D. R. 1971. The strategy of tail autotomy in the ground skink, Lvgosoma laterale. J. Exp. Zool., 1 76: 295-302. Dixon, J. R. and Soini, P. 1977. The reptiles of the Upper Amazon Basin, Iquitos Region, Peru II. Crocodilians, turtles and snakes. Contrib. Biol. Geol. Milwaukee Public Mus., (12): 1-91. DUELLMAN, W. E. 1965. Amphibians and Reptiles from the Yucatan Peninsula, Mexico. Univ. Kansas Mus. Nat. Hist. Publ., 15(2):577-614. 1979. The biology of an equatorial herpetofauna in Amazonian Ecuador. Univ. Kansas Mus. Nat. Hist. Misc. Publ., 65:1-352. Dunn, E. R. 1954. The coral snake "mimic" problem in Pan- ama. Evolution, 8:97-102. Echternacht, A. C. 1973. The color pattern of Sonora michoacanensis (Duges) (Serpentes. Colubridae) and its bear- ing on the origin of the species. Breviora, (410):1-18. Fitch, H. S. 1973. A field study of Costa Rican lizards. Univ. Kansas Sci. Bull., 50:39-126. Gehlbach, F. R. 1 972. Coral snake mimicry reconsidered: the strat- egy of self-mimicry. Forma et Functio. 5: 311-320. Greene, H. W. 1973. Defensive tail displays by snakes and am- phisbaemans. J. Herpetol., 7:143-161. Hecht, M. K. and Marien, D. 1956. The coral snake mimic problem: a reinter- pretation. J. Morphol., 98:335-356. Henderson, R. W. 1983. An unusually patterned Scaphiodontophis (Serpentes: Colubridae) from Costa Rica. Brenesia, 19/20:359-362. Henderson. R. W., Binder, M. H. and Sajdak, R. A. 1981. Ecological relationships of the tree snakes Uromacer catesbyi and U. oxyrhynchus (Colubridae) on Isla Saona, Republica Do- minicana. Amphibia-Reptilia, 2:153-163. Henderson. R. W. and Hoevers, L. G. 1975. A checklist and key to the amphibians and reptiles of Belize. Central America. Milwau- kee Public Mus. Contrib. Biol. Geol., (5):1- 63. Howell, T. R. 1957. Birds of a second-growth rain forest area of Nicaragua. Condor 59:74-1 1 1. Jackson, J. F. 1979. Effects of some ophidian tail displays on the predatory behavior of grison (Galictis sp.). Copeia, 1979:169-172. Jackson, J. F., Ingram, W. Ill, and Campbell, H. W. 1976. The dorsal pigmentation pattern of snakes as an antipredator strategy: a multivariate approach. Amer. Natur.. 110:1029-1053. Jaksic, F. M. and Fuentes, E. R. 1980. Correlates of tail losses in twelve species of Liolaemus lizards. J. Herpetol., 14:137-141. Landy, M. J., Langebartel, D. A., Moll, E. O. and Smith, H. M. 1966. A collection of snakes from Volcan Tacana. Chiapas, Mexico. J. Ohio Herpetol. Soc. 5: 93-101. Liner, E. A. 1960. A new subspecies of false coral snake (Pli- ocercus elapoides) from San Luis Potosi, Mexico. Southwest Nat., 5:217-220. Martin, P. S. 1955. Herpetological records from the Gomez Far- ias region of southwestern Tamaulipas. Mexico. Copeia, 1955:173-180. McCoy, C. J. 1970. The snake fauna of Middlesex. British Hon- duras. J. Herpetol., 4:135-140. Morgan, E. C. 1973. Snakes of the Subfamily Sibynophiinae. Ph.D. Diss., University Southwestern Lou- isiana. 205 pp. Myers, C. W. 1967. The pine woods snake, Rhadinaea Jlavilata (Cope). Bull. Florida State Mus., 1 1:47-97. 1 974. 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. 1959. Studies on the amphibians and reptiles of British Honduras. Publ. Res. Div. Ross Al- len's Reptile Inst., 2(1): 1-76. Parker, W. S. 1972. Ecological study of the western whiptail liz- ard, Cnemidophorus tigris gracilis, in Ari- zona. Herpetologica. 28:360-369. Parker, W. S. and Pianka, E. R. 1975. Comparative ecology of populations of the 194 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY lizard Via stansburiana. Copeia 1 975:6 1 5— 632. Peters, J. A. and Orejas-Miranda, B. 1970. Catalogue of the neotropical Squamata: Part I. Snakes. Bull. U.S. Nat. Mus., (297): 1-347. PlANKA, E. R. 1970. Comparative autecology of the lizard Cne- midophorus tigris in different parts of its geo- graphic range. Ecology, 51:703-720. Pough, F. H. 1964. A coral snake "mimic" eaten by a bird. Co- peia, 1964:223. 1976. Multiple cryptic effects of crossbanded and ringed patterns of snakes. Copeia, 1 976:834- 836. Roze, J. A. 1969. Una nueva coral falsa del genero Scaphio- dontophis (Serpentes: Colubridae) de Co- lombia. Caldasia, 10(48):355-363. Savage, J. M. and Vial, J. L. 1974. The venomous coral snakes (genus Micru- rus) of Costa Rica. Rev. Biol. Trop., 21(2): 295-349. Savitzky, A. H. 1981. Hinged teeth in snakes: An adaptation for swallowing hard-bodied prey. Science, 212: 346-349. Schall, J. J. and Pianka, E. R. 1980. Evolution of escape behavior diversity. Amer. Natur., 115:551-566. SCHOENER, T. W. 1979. Inferring the properties ofpredation and oth- er injury-producing agents from injury fre- quencies. Ecology, 60: 1 1 10-1 1 1 5. Scott, N. J. 1969. A zoogeographic analysis of the snakes of Costa Rica. Ph.D. Diss., Univ. So. Califor- nia. Sharma, B. D. 1980. A rare case of autotomy seen in Amphiesma stolata (Linn., Serpentes: Colubridae). The Snake, 12:60. Skutch, A. F. 1971. A naturalist in Costa Rica. Univ. Florida Press, Gainesville. Slevin, J. R. 1939. Notes on a collection of reptiles and am- phibians from Guatemala. Proc. California Acad. Sci., 23(26):393-414. Smith, N. G. 1969. Avian predation of coral snakes. Copeia, 1969:402-404. Smith, S. M. 1975. Innate recognition of coral snake pattern by a possible avian predator. Science, 187:759- 760. 1977. Coral-snake pattern recognition and stimu- lus generalization by naive great Kiskadees (Aves: Tyrannidae). Nature, 265(5594):535 - 536. Stuart, 1935. 1948. 1958. Taylor, 1954. Taylor. 1943. Vitt, L. 1977. Vitt, L. 1973. Wilson. 1968. Wilson. 1982. Zug, G. 1979. L. C. A contribution to a knowledge of the savan- na region of central Peten, Guatemala. Misc. Publ. Mus. Zool. Univ. Michigan, (29): 1-56. The amphibians and reptiles of Alta Vera- paz, Guatemala. Misc. Publ. Mus. Zool. Univ. Michigan, (69): 1-109. A study of the Uaxactun-Tikal area of north- ern 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. E. H. and Smith, H. M. A review of the American sibynophiine snakes, with a proposal ofa new genus. Univ. Kansas Sci. Bull., 29:301-337. J., Congdon, J. D. and Dickson, N. A. Adaptive strategies and energetics of tail au- totomy in lizards. Ecology, 58:326-337. J. and Hulse, A. C. Observations on feeding habits and tail dis- play of the Sonora coral snake, Micruroides euryxanthus. Herpetologica, 29:302-304. L. D. A fracture plane in the caudal vertebrae of Pliocercus elapoides (Serpentes: Colubridae). J. Herpetol., 1:93. L. D. and Meyer, J. R. The snakes of Honduras. Milwaukee Public Mus. Publ. in Biol, and Geol., 6:1-159. 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 Edited by R. A. Seigel. L. E. Hunt. .1 L Knight. L. Malaret and N. L. Zusehlag l'»N4 Museum of Natural History. The University of Kansas. Laurence Dominance in Snakes Charles C. Carpenter Dominance behavior has been suggested by a number ofobservers as an explanation for a func- tion of combat rituals in snakes. Other reported 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 or avoidance, strength and weakness and physically overcome. It is not my aim to review all of the literature using these terms, but significant ref- erences will be cited below. Confusion or uncertainty arises from the lack of criteria for measuring or determining domi- nance in snakes, and what the functions ofdom- inance might be. Recent definitions of domi- nance 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 wanting. The criteria or behaviors used by observers to judge dominance vary with the group being ob- served; what may be used for lizards, birds or mammals, may not apply for snakes. The op- erational definition I shall use for dominance behavior in snakes is as follows: Dominance is an interaction between two snakes in which one individual performs certain actions (physical or otherwise) which ultimately causes the other in- dividual to avoid these actions (subordinate). Interactions between snakes are difficult to ob- serve in the field, although the potential is there for certain species patiently observed. In captive snakes the evidence for social dominance is more direct and easily observed. Social behaviors are known to occur for snakes during the interactions of courtship and mating, combat rituals, and possibly, parental care and could be associated with different types of ag- gregations 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 activity or home ranges (Seibert and Hagen 1 947; Stickel and Cope 1947; Fitch 1949; Carpenter 1952; Barbour et al. 1969) suggests that snakes may interact in the field. But while these studies show limited movement of certain species of snakes, few indicate interactions in the field be- tween individuals. Kennedy ( 1 965) and Bennion and Parker ( 1976) recorded interactions between male Masticophis and Andren (1975) and Volsoe (1944) have observed males showing aggression related to breeding sites in I 'ipera berus. Combat rituals are the most easily recognized agonistic interactions between male snakes (Carpenter and Ferguson 1977). What must be involved in snake interactions which may result in the dominance of one in- dividual? There must be species recognition be- tween conspecifics. I believe the evidence is good that an individual snake recognizes its own species, especially after tongue contact— impli- cating the vomeronasal modality or channel for this recognition (Noble and Clausen 1 936; Kubie et al. 1978; Devine 1976). Sex recognition must also occur and this appears to be rapid between conspecifics and accomplished by chemical cues (Noble 1937; Froese 1980). perhaps sometimes visually. Individual recognition or identification involves the ability of one member of a species to identify an individual conspecific. There is no satisfactory evidence that a snake distinguishes one conspecific individual from another, though I see no reason they could not accomplish this by detecting individual differences in chemical or visual cues. Individual recognition is consid- ered important in establishing social herarchies in groups of birds and mammals. The observa- 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 and one female Python molurus. Combat Rituals.—Over the past few years I have observed many interactions in captivity be- tween conspecific male snakes. I believe I rec- ognize dominance as a result of most of these interactions whether or not actual combat rituals occurred. Numerous descriptions of combat rit- uals are available from the literature (Carpenter and Ferguson 1977) and I will use aspects of those descriptions which relate to the establish- es 196 SPECIAL PUBLICATION- MUSEUM OF NATURAL HISTORY Fig. 1. Dominant male crawling over subordinate male in submissive posture. Crotalus viridis. ing of physical superiority and those behaviors which indicate a dominant-subordinate relation- ship between two snakes. During a combat ritual between two conspe- cific male snakes, one, or both individuals at- tempt to obtain a higher or superior position by assuming a posture with the head and anterior trunk higher in display than his adversary (Cro- talus— Carpenters al. 1976; Sistrurus— Carpen- ter 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 down by quickly looping around him and push- ing or throwing (topping) him down (Crotalus, op. cit.) or by lowering the hovering head and anterior trunk down on the anterior region of his adversary {Lampropeltis, op. cit.) forcing the lower snake down and pinning his anterior region to the substrate. These actions are obvious phys- ical attempts by one individual to force another into a lower posture. When one individual has been forced to a low- er or prostrate posture (head and anterior trunk) the superior snake may then return to the solic- itation display (Carpenter et al. 1976) of the combat ritual with the other again rising in re- sponse and repeating attempts at topping. Such actions may continue for an hour or more with numerous topping bouts. During these bouts it appears that the more aggressive snake keeps his posterior trunk region over some posterior part of the other snake. Often after a combat ritual has been proceed- ing for sometime one individual attempts to crawl away from the contact with the other, sometimes 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- ations the area for flight is small so that the dom- inant can easily reestablish contact. The domi- nant may again perform the solicitation display before the subordinate with the latter sometimes rising in response. If the subordinate 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 subor- dinate (Fig. 2). If the subordinate tries to flee or move away, the dominant will continue to crawl over or lie on top ofhim. Ifthe subordinate raises his head, the dominant may respond by moving over his head region, as if forcing him to main- tain his submissive (negative) posture (Fig. 3). Each time the subordinate shows activity, the dominant may crawl away, and when the sub- ordinate 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 at- tempt to flee or again assume the submissive posture. During many ofmy attempts to induce combat rituals in a variety of male snakes, no tendency to perform a ritual was exhibited, or only one individual might rise to a solicitation posture. In such instances, the non-responder, or one of the individuals, appears to act subordinate, assum- ing a submissive posture without overt physical contact, or may actively try to avoid the other which appears to be dominant. From these ac- tions it appears that dominance may occur with- out contact between two males. Observations of Dominance The following brief summaries are from un- published notes on complete videotapes of the VERTEBRATE ECOLOGY AND SYSTEMATICS 197 Fie. 2. Dominant male Crotalus viridis lying on subordinate male Crotalus molossus. interactions of staged encounters between paired male snakes. 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 fiat, vibrating his tail when pinned by the larger male. The subordinate male burrowed complete- ly beneath the sand. Lampropeltis getulus holbwoki. — The larger of the two males on first contact immediately hovered over the smaller male which tried to avoid contact. The larger male persisted in crawl- ing 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 vibrated his tail rapidly. The smaller male finally burrowed under the sand and escaped. 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 separat- ing, the smaller male fled from approach of the larger male. Lampropeltis calligaster.—A male L. c. calli- gaster 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. A slightly smaller 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 male. The larger male hovered and pinned the smaller male within 2Vi min and then persisted during the 35 min they were together, intermit- tantly, in pursuing, crawling over, hovering and pinning the smaller male, until he was removed. Over this time the larger male bit the smaller male 1 1 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 fe- male added at this time. The smaller male (sub- ordinate) remained active during the entire ep- isode moving about the chamber, climbing the sides and pushing its rostrum against the glass front of the chamber. Crotalus atrox. — M\er approximately 10 min of combat ritual, the more aggressive male per- forming solicitation displays appeared dominant when the other male ceased to display, tried to escape, was pursued, and then assumed a sub- missive coil with the dominant crawling over him. Sistrurus miliarias.— The initiator made con- tact with the other male which jerked in re- sponse, then the initiator crawled over the other which assumed a loose coil with head down. They separated and as the initiator again made contact the other violently thrashed away pursued by the initiator. The initiator crawled over the fleeing male. The submissive male lay coiled and when contacted waved his tail vertically. Sistrurus catenatus. —On contact one male im- mediately assumed a submissive coil, head down. As the superior 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 times. 198 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY in Fig. 3. Submissive (negative) posture of a male Lampropeltis getulus holbrooki. 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 subor- dinate crawled off and was pursued by the dom- inant 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 pursued and crawled upon by the dominant. Vipera lebentina.—One male, after performing solicitation display persistently, crawled over and lay upon the other which assumed a coil (sub- missive posture) with its head down. 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 for 3 to 4 sec. After this reciprocal biting episode, all displaying ceased and both males lay quiet with no further interaction. Crotalus v. viridis. — One male initiated the combat ritual with solicitation displays, the other responded, but soon moved off, pursued by the initiating male. They again displayed intermit- tantly and the initiator started to crawl over the other male, which tried to retreat (violently) and then assumed a submissive coil with the now dominant initiating male crawling back and forth over the subordinate. Crotalus viridis.— The lighter male persistently followed and crawled over the darker male. When the darker male tried to move from contact he was pursued by the light male. The darker male 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 tail (no rattling). The C viridis stayed on top of the C molossus, crawling back and forth and at times rising to display. The C. molossus tried to crawl away but the C 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 actions of the C. viridis. A male Agkistrodon piscivorus was then placed in the chamber with the C. viridis. The C. viridis immediately performed a solicitation display and persisted in displaying for some time finally il- liciting a solicitation display response from the A. piscivorus. In two topping attempts, the C viridis appeared dominant, forcing the A. pisciv- orus down. At one time the A. piscivorus waved his tail as the C viridis crawled over him. It appeared that the A. piscivorus was trying to avoid the persistent actions of the C. viridis. Observations by Others Published records of species for which domi- nance, dominance-like behavior, submissive be- havior and territoriality have been stated, sug- VERTEBRATE ECOLOGY AND SYSTEMATICS 199 gested or inferred follow. A briefreference is made to my interpretation of each account. Boidae.— Python molurus (Barker et al. 1979). Dominance with linear hierarchy. Combat ritual with spur gouging and escape. Individual rec- ognition. Sanzinia madagascariensis (Carpenter et al. 1978). Dominance-subordination resulting from combat ritual with spur gouging. Colubridae.— Elaphe obsoleta bairdi (Brecke el al. 1976). Combat ritual with strong rivalry. Elaphe o. obsoleta (Rigley 1971). Combat ritual suggesting dominance as a result. Coronella aus- triaca (Andren and Nilson 1 976). Males bite while fighting. Coluber viridiflavus (Guibe and Saint Girons 1955). Combat ritual with the victor first to mate. Lampropeltis triangulum (Shaw 1951). Flight by subordinate after combat ritual. Lam- propeltis getulus holbrooki (Carpenter and Gil- lingham 1977). As a result of a combat ritual one male exhibited dominance actions, the other male subordinate behaviors. Lampropeltis mexicana alterna (Murphy et al. 1978). Dominance as a result of combat ritual. Lampropeltis pyrome- lana (Martin 1976). Aggression with biting and chasing. Masticophisf.JJagellum (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). Territoriality suggested in contest for supremacy. Elapidae. — Demansia textilis (Fleay 1937. 1951). Dominance was indicated with the "weaker" subordinate retreating, the "stronger" male intimidating rivals. Pseudechis porphyria- cus (Fleay 1937, 1951). Territorial "right" sug- gested from combat ritual. Crotalidae.—Agkistrodon piscivorus leucosto- ma (Perry 1978). Suggests territoriality and dom- inance from combat ritual. Crotalus adamanteus (Wagner 1962). During combat ritual the dom- inant has more stamina. Crotalus atrox (Jenni 1966). Dominance apparent for victory and de- feat clearly defined as result of combat ritual. Crotalus atrox (Foree 1949). Territoriality sug- gested. Crotalus cerastes (Lowe and Norris 1950). Territoriality suggested and discussed as a result of combat ritual. Crotalus horridus atricaudatus (Sutherland 1958). Dominance a possible result of combat ritual. Crotalus lepidus klauberi (Car- penter et al. 1976). Dominance apparent result of combat ritual. Crotalus ruber (Shaw 1948). Dominance apparent from the result of rivalry in combat ritual. Crotalus v. viridis (Thorne 1977). Social domination and territoriality suggested from combat ritual. Sistrurus miliarius (Carpen- ter 1979). Dominance as a result of combat rit- ual. Viperidae.— Vipera sp. (Prior 1933). Territo- riality suggested. Vipera aspis (Naulleau 1970). Territoriality suggested. Vipera berus (Andren 1975). Indicates winner and loser, with chasing, in combat rituals. Vipera berus (Volsoe 1944). Dominance suggested, with winner of combat ritual pursuing female. Vipera berus (Guibe and 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. Discussion In a recent paper (Carpenter 1977) I discussed the role of different signal channels in commu- nication 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 other agonistic interactions involving contact (tactile) signals are tongue flicking, crawling over, dorsal crawl, lying on. pinning, topping, pushing down, entwining, spur use and biting, and are apparently used in determining dominance. Vi- sual signals are those of vertical or oblique dis- plays, hovering, pursuit, and perhaps approach. The subordinate snake signals its submission by avoiding the dominant, thrashing on contact, fleeing (retreat), tail waving, submissive (nega- tive) posture, and sometimes burrowing, which are visual signals. If the subordinate raises his head (visual) or begins to move (tactile or visual) these are signals to which the dominant will re- spond. The submissive posture fits the criterion for submissive postures in other animals, that is, offering the lowest or smallest profile. I believe the evidence is strong that in many instances the interplay of these agonistic signals results in individual male snakes becoming dom- inant and their adversaries becoming subordi- nate. What is the function of individual dominance in male snakes? The resources over which male 200 SPECIAL PUBLICATION- MUSEUM OF NATURAL HISTORY snakes might compete are a mate, food, and space. The strongest evidence of possible competition for a mate is in Vipera berus. A male having localized and identified a reproductive female will fight with, and repeatedly chase off, other males (Andren 1975;Volsoe 1944), with possible temporary territoriality. The evidence for com- petition for food and space is less evident, though combat rituals are noted in the presence of food (Shaw 1951; Sutherland 1958). Since many of my observations occurred in the absence of a female or food, dominance does occur in the absence of these resources. We need to know more about mating strategies in snakes and how these may relate to domi- nance. Shine (1978) provided data that "reveals a high correlation between the occurrence of male 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 is a natural consequence of intra- sexual male competition. We need evidence of a resource reward for dominance. What are the taxonomic relationships ofdom- inance? Since combat rituals have been observed in the Boidae, Colubridae, Elapidae, Crotalidae and Viperidae, this spread suggests that domi- nance is likely to be a phenomenon occurring in all groups of snakes. The fact that it is recorded mostly for the larger species of snakes may be due to the difficulty, or lack, of observing the smaller species. For those species where aggregations are com- mon and multiple courtships occur (two or more males courting a female at the same time, i.e., some natricine colubrids), it appears that combat rituals do not occur and that dominance is not likely to occur. Is there individual recognition between male snakes and is this a necessary attribute of dom- inance (this is suggested in Python molurus by Barker et al. 1979, where a linear hierarchy oc- curred)? When does the onset of dominance interac- tions occur, i.e., at a certain size, age, sexual ma- turity? Most combat rituals have been observed in large (presumably sexually mature) males. The establishment of dominance, whether by combat rituals or other dominance actions, ap- pears to fulfill the ritualistic function of gaining superiority without significant physical harm. The general lack of biting during combat rituals and dominance-subordinate encounters supports this. Certain of the actions seen performed by the dominant, 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- sume a submissive posture, or retreat and be chased by a male. The similarity of these sub- ordinate actions by a male to those of a courted female may provide communication signals that lead the dominant male to homosexual action, i.e., aligning next to the subordinate male and attempts to tail search and effect intromission, though chemical signals should direct otherwise. Is dominance related to larger size? This ap- pears to be true for the Lampropeltis getulus hol- brooki observed, but more detailed measure- ments of size and weight are needed to verify this assumption. The observations of Crotalus viridis indicate that this male dominated larger males of different species (perhaps an artifact of captivity). The determination of the existence and sig- nificance of dominance as a social factor in nat- urally occurring populations of snakes will be difficult. I hope that my observations will create an awareness of this phenomenon ofdominance by other herpetologists and stimulate them to watch for this behavior in the field. Summary The existence of dominance-subordinate re- lationships between individual conspecific snakes has been suggested by observers recording com- bat rituals, mostly from captive encounters. Us- ing an operational definition of dominance in snakes based on one snake exhibiting superiority through various actions over a subordinate which 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 very infrequent. 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 pos- ture, repeating its dominant actions. VERTEBRATE ECOLOGY AND SYSTEMATICS 201 A series of observations of male encounters for nine species ofsnakes all indicate dominance- subordinate relationships using the above be- haviors. Literature records for 25 species ofsnakes suggest the occurrence ofdominance-like behav- ior. The evidence that dominance functions to give priority for a resource has not been adequately demonstrated, and is often observed in the ab- sence 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. Vi- peridae). More information will be needed to determine if different mechanisms are used in determining dominance in these and other groups. The dominance actions of snakes are likely to be related to the behaviors used in combat rituals, while subordinate behaviors are likely to be more similar between families. Acknowledgments I wish to thank James B. Murphy and his staff in the Department of Herpetology at the Dallas Zoo, Dallas, TX and Frank Bryce of the Ven- Am Laboratory in Cache, Oklahoma for the courtesies extended in the use of their facilities. Literature Cited Andren, C 1975. Social behavior of Vipera berus during the reproductive period. Norwegian J. Zool.. 24: 234-235. Andren. C and Nilson. G. 1976. Hasselsnoken (Cornonella austriaca)— a 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 eastern worm snake. Carphophis amoenus amoenus. Ecology. 50:470-476. Barker, D. G., Murphy, J. B. and Smith. K. W. 1 979. Social behavior in a captive group of Indian pythons. Python molurus (Serpentes, Boi- dae) with formation of a linear social hier- archy. Copeia, 1979:466-471. Bennion, R. S. and Parker. W. S. 1976. Field observations on courtship and aggres- sive behavior in desert striped whipsnakes, Masticophis t. taeniatus. Herpetologica, 32: 30-35. Brecke. B. J., Murphy. J. B. and Seifert, W. 1976. An inventory of reproduction and social be- havior in captive Baird's ratsnakes, Elaphe obsoleta bairdi (Yarrow). Herpetologica, 32: 389-395. Brown, J. L. 1975. The Evolution of Behavior. W. W. Norton & Company. Inc.. New York. 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. 1979. A combat ritual between two male pygmy rattlesnakes (Sistrurus rnilianus). Copeia, 1979:638-642. Carpenter, C. C. and Ferguson, G. W. 1977. Variation and evolution of stereotyped be- havior in reptiles. Pp. 335-554. In Gans. C. and Tinkle, D. W. (eds.). Biology of the Rep- tilia Vol. 7. Academic Press, London. Carpenter, C. C, Gillingham, J. C. and Murphy, J. B. 1976. The combat ritual of the rock rattlesnake (Crotalus lepidus). Copeia. 1976:764-789. Carpenter, C. C. and Gillingham, J. C. 1977. A combat ritual between two male speckled kingsnakes (Tampropeltis getulus holbrooki: Colubridae, Serpentes) with indications of dominance. Southwestern Naturalist, 22: 517-524. Carpenter, C. C, Murphy. J. B. and Mitchell, L. A. 1978. Combat bouts with spur use in the Mada- gascan boa (Sanzmia madagascariensis). Herpetologica, 34:207-212. Devine, M. C. 1976. Species discrimination in mate selection by free living male garter snakes and experi- mental evidence for the role of pheromones. Herpetological Reviews, 1976:(abstract). Fitch, H. S. 1949. Study of snake populations in central Cali- fornia. 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. Fleay, D. 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. 1980. Reptiles. Pp. 39-68. In Roy. M. A. (ed.). Species Identity and Attachment. Garland STPM Press, New York. 202 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Guibe, J. and Saint Girons, H. 1955. Espace vital et territorire chez les reptiles. La Nature, 245:358-362. Jenni, B. 1966. Combat dance of the male rattlesnake rarely seen by man. Outdoor Oklahoma, 22:6-7. Kennedy. J. P. 1965. Territorial behavior in the eastern coach- whip. Masticophis flagellum. Anat. Rec. 151: 499 (abstract). KiBiE, J. L., Vagvolgyi, A. and Halpern, M. 1978. Roles of the vomeronasal and olfactory sys- tems in courtship behavior of male garter snakes. J. Comp. Physiol. Psychol., 92:627- 641. Lowe, C. H., Jr. and Norris, K. W. 1950. Aggressive behavior in male sidewinders, (Crotalus cerastes), with a discussion of ag- gressive behavior and territoriality in snakes. Natur. Hist. Misc., (66): 1-1 3. Martin, B. E. 1 976. Notes on breeding behavior in a captive pair of Sonoran mountain kingsnakes, (Lampro- peltis pvromelana). Bull. Maryland Herpe- tol. Soc, 12:23-24. Murphy, J. B., Tyron, B. W. and Brecre, B. J. 1978. An inventory of reproduction and social be- havior in captive gray-banded kingsnakes, Lampropeltis mexicana alterna (Brown). Herpetologica, 34:84-93. McCann, C. 1935. Male rat snake {Zamenis mucosus) fighting. J. Bombay Natur. Hist. Soc, 38:409. Naulleau, G. 1970. Espace vital et territorire chez Vipera aspis. In Richard, G. (ed.), Territoire et Domi- nance Vital. Serie Ecologie et Ethologie. No. 1. Masson et Cie, Paris. Noble, G. K. 1937. The sense organs involved in the courtship of Storeria, Thamnophis and other snakes. Bull. Amer. Mus. Natur., 73:763-725. Noble, G. K. and Clausen, H. J. 1 936. The aggregation behavior of Storeria dekayi and other snakes with special reference to the sense organs involved. Ecol. Monog., 6: 269-316. Perry, J. 1978. An observation of "dance behavior" in the western cottonmouth, Agkistrodon pisciv- orus leucostoma (Reptilia, Serpentes, Viper- idae). J. Herpetology, 12:429-430. Prior, H. T. J. 1933. The dance of the adders. A remarkable ex- ample of reptilian rivalry. Countryside, 9: 492-493. RlGLEY, L. 1971. "Combat dance" of the black rat snake, Elapheo. obsoleta. J. Herpetology, 5:65-66. Seibert, H. C. and Hagen, C. W., Jr. 1 947. Studies on a population of snakes in Illinois. Copeia, 1946:6-22. Shaw, C. E. 1 948. The combat "dance" ofsome crotalid snakes. Herpetologica, 4:137-145. 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, W. H. and Cope, J. B. 1 947. The home ranges and wanderings of snakes. Copeia, 1947:127-136. Sutherland, I. D. W. 1958. The "combat dance" of the timber rattle- snakes. Herpetologica, 14:23-24. Thorne, E. T. 1977. Sybille Creek snake dance. Wyoming Wild- life, 41:14. VOLS0E, H. 1 944. Structure and seasonal variation of the male reproductive organs of Vipera bents. Spolia Mus. Zool. Hauniensis (Copenhagen), 5:1- 157. Wagner, R. T. 1962. Notes on the combat dance in Crotalus ada- rnanteus. Bull. Phildephia Herpetol. Soc, 10: 7-8. Wilson, E. O. 1975. 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 < l l*S4 Museum of Natural History. The UniversitJ of Kansas. 1 awrence An Experimental Study of Variation in Habitat Selection and Occurrence of the Deermouse, Peromyscus maniculatus gracilis John H. Fitch Introduction Habitat selection has important consequences at the levels of both the individual and the pop- ulation. Many organisms must actively select the type of habitat in which to live from a variety of accessible choices. The choice of a particular habitat exposes the organism to a specific set of selective pressures that can profoundly affect its survival and breeding success (Partridge 1978). Variations in habitat selection may lead to mod- ifications of gene frequencies at the population level (Doyle 1975). A positive correlation between the spatial dis- tribution of a population and a specific habitat 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 (Kettlewell 1959. 1965), interspecific competi- tion (Sheppe 1961: Turner 1961). intraspecific competition (Kluyver and Tinbergen 1953). or by internal morphological or physiological con- straints (Bursell 1960; Klopfer 1969). Habitat selection and its relationship to habitat occur- rence must therefore be verified experimentally (Meadows and Campbell 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 un- der controlled conditions. The relationship be- tween habitat selection and occurrence has been experimentally verified and reviewed in verte- brate species offish (Sale 1969; Casterlin and Reynolds 1978). amphibians (Wiens 1970, 1972), reptiles (Kiester et al. 1 975). birds (Klopfer 1 965; Hilden 1965), and mammals (Harris 1952; Wecker 1963; Fitch 1979). Patterns of habitat selection may vary within and among populations of a species or even sub- species. Intrapopulation variation in habitat se- lection has been documented experimentally in plankton (Doyle 1975, 1976) and in mice (Fitch 1979). Interpopulation variation in habitat se- lection has not been extensively studied. Miller (1973) reported that prairie deer mice (Pero- myscus maniculatus bairdi) from North Dakota selected a simulated forest habitat corresponding to their natural habitat; in Michigan, where P. m. bairdi occurs only in grassland habitats, in- dividuals selected that simulated habitat in favor of forest. Little information is available, how- ever, on variations in habitat selection among local or regional populations occupying the same biome. The purpose of this study was to investigate the local variations in habitat selection among two populations of the woodland deer mouse. Peromyscus maniculatus gracilis, in relation to observed differences in habitat occurrence. Hab- itat occurrences of populations were first verified in a forest habitat and an open lichen-grass hab- itat 36 kilometers apart by trapping studies. An- imals 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 variations in habitat occurrence reflected real dif- ferences in habitat selection or whether they in- dicated acceptance of suboptimal habitats be- cause of competition. The woodland deer mouse occurs throughout the northeastern United States and southeastern Canada from Pennsylvania north to southern Quebec and from western Manitoba east to Maryland (Hall and Kelson 1959). In Michigan. P. m. gracilis is widely distributed throughout the Upper Peninsula, on many islands in Lake Michigan, and as far south as Missaukee County in the lower peninsula of the state (Burt 1946). This subspecies is generally restricted to wood- land habitats and occurs most commonly in up- land deciduous associations of maple, birch, and beech in northern Michigan (Dice 1925; Blair 1941). In New York. P. m. gracilis is restricted to woodland habitats even in forest-grassland ecotonal areas (Klein 1960). Harris (1952) re- ported that P. m. gracilis from the Upper Pen- 203 204 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY miles Fig. 1. General map of Michigan showing the location of the Kingston Plains (see 1) and the Cusino Wildlife Refuge (see 2). insula of Michigan consistently selected simu- lated forest models in preference to grassland models under controlled laboratory conditions. Fitch (1979) discovered a population of P. m. gracilis inhabiting an open lichen-grass habitat in the Kingston Plains, Alger County, Michigan, approximately 36 kilometers from the Cusino Wildlife Refuge, where Harris (1952) obtained his animals. In the present study, trapping was done in both study areas to confirm patterns of habitat occurrences, and animals from both lo- calities were then tested for habitat selection. Study Areas Cusino The Cusino Wildlife Refuge is located in the southern part of Alger County, Michigan, ap- proximately 47 kilometers south of Lake Supe- rior (Fig. 1). The region is characterized by low- land swamp conifer habitats, upland associations of hardwoods, and infrequent man-made clear- ings. Most of the forests in this part of the Upper Peninsula were logged over in the late 1 800's and are now secondary growth. Lowland swamp conifer habitats are charac- terized by associations of northern white cedar {Thuja occidentalis), black spruce (Picea mar- iana), and balsam (Abies balsamea). Alder (Al- nus rugosa), Labrador tea (Ledum groenlandi- curri), and wintergreen (Gaultheria procumbens) are found commonly in this habitat. The upland hardwood forests have associa- tions of striped maple (Acer pensylvanicum), sugar maple (Acer saccharum), and American beech (Fagus grandifolia). Elderberry (Sambu- cus pubens), bedstraw (Galium sp.), and bracken fern (Pteridium aquilinum) are commonly found in upland forests. Man-made clearings are dominated by blue- grass (Poa nemoralis) and orange hawkweed (Hieracium aurantiacum). VERTEBRATE ECOLOGY AND SYSTEMATICA 205 4 \ \ ?$*& ' , '- - XT" ' - < ' v#" - :» MP Fig. 4. Pine-Oak, 10 road km SW El Salto, Durango (both photographs 1 1 July 1970 by author). 305 m (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). Mixed Boreal-Tropical (Fig. 5) This habitat is unique, is relatively sharply de- limited, and is somewhat transitional between the Pine-Oak and, at lower elevations, the Trop- ical-Deciduous. The Mixed Boreal-Tropical cov- ers rugged, mountainous terrain at the highest elevations in large barrancas and canyons, and is best developed on south-facing slopes. Steep boulder-strewn hillsides with rock outcrops, in- terrupted by small, relatively level areas, are cov- ered in most places with a tall pine-oak woodland and often a dense understory of herbs, shrubs, and thick tangles of vines. Common pines are Pinus oocarpa, P. teocote, and P. lumholtzi (the VERTEBRATE ECOLOGY AND SYSTEMATICS 223 "pino triste," an indicator species for this re- gion). Oaks include Quercus macrophylla (broadleaf). Q.fulva, and Q. viminea; other large trees are madrono {Arbutus glandulosa), and in some places magnolia {Magnolia shiedeana) and hop-hornbeam {Ostrya virginiana). The under- story, dense in places, includes the large shrubs Tithonia calva. Rhus terebinthifolia, and Cer- cocarpus macrophyllus, tall herbs of the genus Stevia (spp.). the tree-like Bocconia arborea, thick brambles of Rubus, and some poison ivy {Tox- icodendron). 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 with bromeliads (most common are Tillandsia exserta and T. benthamiana) and mistletoe {Psit- tacanthus, usually on oaks). Along with mos- quitoes, biting black flies {Simulium) are a nuisance in the rainy season. The Mixed Boreal- Tropical, recognized elsewhere in Durango (Webb and Baker 1 962), has some resemblance to a cloud forest. This region occurs for approximately 5 1 km (32 mi) along Highway 40 between elevations of 2408 (7900) and 1798 m (5900 ft). Some 8 or 9 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. About 1 5 km (9 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 m (2900 ft). The oak woodland consists of both deciduous and evergreen species of Quercus. The probable climax vegetation, modified by clearing on many hillsides, is most lush in the shaded and narrow mountainous ravines and arroyos and in the larg- er canyon bottoms. Trees and shrubs include morning-glory tree {Ipomoea arborescens), gua- vas {Psidium), sugar apple {Annona squamosa) pricklenut (Guazuma ulmifolia). coleto {Oreo- panax peltatum), sandboxtree {Hura polyandra). the large eardrop tree {Enterolobium cyclocar- pum), the Acacia-hke Lysiloma divaricata, large figs {Ficus), as well as the genera Brosunum, Cei- ba, 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 m (5900 and 400 ft). Thorny acacias become increasingly abundant at lower elevations and with the advent of organ- pipe cactus indicate the transition to the coastal lowland Thorn-Scrub; this rather broad transi- tional zone seems to extend from near Chupa- deros to the vicinity of Concordia. Thorn-Scrub (Fig. 7) The vegetation ofthe Thorn-Scrub forms dense thickets, averaging about 7 to 9 m (25-30 ft) in height, 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 with extensive rock outcrops in some places. The plant cover consists principally of species ofAca- cia (mostly A. cymbispina), Mimosa, Cassia, Caesalpinia, and Bursera, and the guamuchil {Pithecollobium sonorae). The organ-pipe cactus {Pachycereus pecten-arboriginum) is scattered and characteristic. Terrestrial bromeliads and prickly pear, as well as some ofthe plants ofthe Tropical- Deciduous, are of occasional occurrence. Along the coast, water hyacinth mats are common in the rainy season in roadside sloughs. Large co- conut palms {Cocos nucifera) occur near the beach 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 pur- poses. 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 the text. Each locality is followed by descriptive remarks, including approximate road distance from other localities, elevation, and faunal re- gion. Most localities are indicated by signposts. All place-names along the transect route are not mentioned below. Elevations are approximate; they may vary many meters (depending on air pressure) when recorded at the same place at different times. Each entry terminates with a number in parentheses, which indicates its geo- graphic position on the maps (Figs. 1 and 2). Durango 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). Coyotes, Hacienda: Ranch 7 km (4 mi) east El Salto, 2454 m (8050 ft), Pine-Oak (19). Durango (Ciudad): Capital of state and eastern terminus of transect, 320 km ( 198 mi) from Ma- zatlan and 92.5 km (57.5 mi) east El Salto; mile- age from highway at Parque Guadiana on west side of city, 1905 m (6250 ft), Mesquite-Grass- land (26). El Espinazo (Espinazo del Diablo): Ridge (Devil's Backbone) connecting two ranges with drops of several hundred meters on either side; roadside stop and scenic view where monument commemorates dedication of completion of highway on 30 November 1960; 13 km (8 mi) west Puerto Buenos Aires and 20 km (12.3 mi) east Revolcaderos, 2377 m (7800 ft), Mixed Bo- real-Tropical (13). El Mil Diez: Small village, 2 km north highway at 1.2 km (0.8 mi) west El Salto, 2515 m (8250 ft), Pine-Oak (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 26.3 km (16.4 mi) west Las Adjuntas and 4.8 km (3 mi) east Buenos Aires, 2484 m (8150 ft), Pine-Oak (16). Las Adjuntas: Small village 17.7 km (1 1 mi) west El Salto, 2515 m (8250 ft). Pine-Oak (17). Llano Grande: Large ejido 22.2 km (13.8 mi) east El Salto and 70.3 km (43.7 mi) west Duran- go, 2408 m (7900 ft), Pine-Oak (20). Los Bancos: Small village observed about 1 km south of highway, 7 km (4.2 mi) west Puerto Buenos Aires and about 6 km (3.3 mi) east El Espinazo, 2286 m (7500 ft). Mixed Boreal-Trop- 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 ft), Mesquite-Grassland/Pine-Oak transition (25). Mimbres: Small village in Barranca de los Mimbres, 14.5 km (9 mi) west Rio Chico and 30.9 km (19.2 mi) east Llano Grande. 2225 m (7300 ft), Mesquite-Grassland/Pine-Oak transi- tion (23). Navios: Small village about 11.3 km (7 mi) west Rancho Santa Barbara and 1 7.7 km ( 1 1 mi) east Llano Grande, 2438 m (8000 ft), Pine-Oak (21). Puerto Buenos Aires: Roadside stop for scenic view, 3.2 km (2.2 mi) west Buenos Aires and 13 km (8 mi) east El Espinazo, 2560 m (8400 ft). Pine-Oak (15). Rancho Santa Barbara (formerly Weicher Ranch): Cattle ranch 1.7 km (1.1 mi) west Mimbres and 29 km (18 mi) east Llano Grande. 2256 m (7400 ft), Mesquite-Grassland/Pine-Oak transition (22). Revolcaderos: Small village 40.5 km (25.2 mi) west La Ciudad and 10.9 km (6.7 mi) east El Palmito, Sinaloa, 2042 m (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-Grass- land/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 Chupaderos: Small village-truck stop at bridge across Rio Chupaderos, 5.3 km (3.3 mi) west Fig. 5. Mixed Boreal-Tropical. 13 road km SW El Palmito. Sinaloa (both photographs 13 July 1973 by Richard C. Lovelace, Jr.). 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.). 228 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY turnofF to Copala and 17.5 km (10.9 mi) east Concordia, 244 m (800 ft). 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 ar- royo 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- paderos, 579 m (1900 ft), Tropical-Deciduous (6). El Batel: Small village 3.5 km (2.2 mi) east Potrerillos and 3.4 km (2.1 mi) west Loberas, 1646 m (5400 ft), 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 Boreal- Tropical (11). Loberas: Roadside stop for scenic view west- ward (microondas station, and small group of casitas 0.8 mi to the east, erected in 1970's), 3.4 km (2.1 mi) east El Batel, 1922 m (6300 ft). Mixed Boreal-Tropical/Tropical-Deciduous transition (10). Mazatlan: Seaport-tourist resort on small pen- insula and western terminus of transect, about 22.5 km (14 mi) northwest Villa Union, 15 m (50 ft), Thorn-Scrub ( 1 ). Panuco: Mining settlement 10 km (6 mi) by dirt road off highway at Km signpost 70, 1 . 1 km (0.7 mi) east Copala and 16.9 km (10.5 mi) west Santa Lucia, 640 m (2100 ft), Tropical-Decid- uous (5). Potrerillos: Highway construction village 8.5 km (5.3 mi) east Santa Lucia and 2.4 km (1.5 mi) west Santa Rita, 1615m (5300 ft), Tropical- Deciduous (8). Presidio (Presidio de Mazatlan): Same as Villa Union (see Conant 1969:89, footnote); name re- 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, 1 100 m (3600 ft), Tropical-De- ciduous (7). Santa Rita: Rancho and restaurant-bus stop, 1.1 km (0.7 mi) west El Batel and 2.4 km (1.5 mi) east Potrerillos, 1676 m (5500 ft), Tropical- Deciduous (9). 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 highways 40 and 15 about 1 km south of town, 30 m (100 ft), Thorn-Scrub (2). Composition of Herpetofauna This section documents the occurrence of the kinds of amphibians and reptiles in the five her- petofaunal regions along the transect. Introduced species (Gehyra mutilata, Ramphotyphlops bra- minus), the estuarine crocodile (Crocodylus acu- tus), sea turtles, and the sea snake {Pelamis pla- turus) are not included. Species will doubtless be added, especially in the Sinaloan tropical habi- tats. Known ranges probably will be extended northward (e.g., Eumeces parvulus) or southward (e.g., Syrrhophus interorbitalis, Phyllorhynchus decurtatus). The hiatus in geographic range of some excluded species will perhaps be rectified where records of occurrence to the north and south are not now available for the transect area (e.g., Pseudoewycea 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 distribution patterns-barranca corridors). Assignment of some species to faunal regions probably will be altered pending further data of collection (e.g., Ctenosaura pectinata, and sev- eral snakes, especially Boa constrictor and Ox- ybelis aeneus, which are here restricted to the Thorn-Scrub, probably occur in the adjacent Tropical-Deciduous). Taxa are assigned to a par- ticular faunal region based on their overall dis- tribution; several species (e.g., Bufo kelloggi, B. marinus, B. marmoreus, Pachymedusa dacni- color, Smilisca baudini, Coniophanes lateritius) represented by many localities in, and assigned only to, the Thorn-Scrub penetrate eastward to the Chupaderos-Copala region, which is barely into but near the transition to the Tropical-De- ciduous. No attempt is made to indicate relative abundance ofspecies in particular faunal regions. Some species, however, are represented by only one locality in a region (e.g., Hylactophryne tara- humaraensis, Sceloporus clarki, S. nelsoni, Eu- meces brevirostris, Pituophis deppei, and Crota- lus molossus in Mixed Boreal-Tropical; VERTEBRATE ECOLOGY AND SYSTEMATICS 229 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- Scrub. Some species, excluded for various rea- sons, are discussed below. Phrynohyas venulosa (Laurenti).— This large arboreal tree frog is reported from Presidio, Si- naloa by 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 sta- tus and distribution of ranid frogs of the R. pip- iens 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 re- gion in Durango seem to represent the recently described R. chiricahuensis (Platz and Mecham 1979). Another taxon, presumably a subspecies of R. berlandien, occurs in the Mesquite-Grass- land of Durango. Lepidochelys olivacea (Eschscholtz). — The rid- ley may be the most abundant species of sea turtle in the vicinity of Mazatlan. Carapaces and/ or skulls were found on the beach north of Ma- zatlan on 12 August, 6 and 8 June, and 23 July (different years). Local fishermen say sea turtles (probably Lepidochelys) are caught in the vicinity of Mazatlan in July, August, and September; in- dividuals float at the surface and are relatively easy to catch. Flesh and eggs are used locally for food. Some turtles nest near Mazatlan (North Beach and Isla de la Piedras) in May and early June, but adults are not captured then. The larg- est rookery in the general area is said to be south near Acaponeta, Nayarit. Gehyra mutilata (Wiegmann). — This intro- duced lizard is abundant at night on the walls of beachfront establishments in Mazatlan. In- creased urbanization along the beach north of Mazatlan as observed in the years since 1955 has favored the dispersal of this species. A hatchling (ca. 20 mm SVL, later destroyed) was active (night of 22 August) among window fixtures on the sec- ond 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 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 ornatus lateralis (Boulenger).— Hardy and McDiarmid (1969:141-142) discuss a questionable record for Presidio and mention the nearest locality as 36 miles north Mazatlan. These two localities are the southernmost for the species on the west coast of Mexico. Occurrence of the species in the Mazatlan-Villa Union area needs verification. Coluber constrictor oaxaca Jan. — Wilson (1966) records one specimen of this snake from Coyotes, Durango. No other specimens are known from the state. The status of this species in Du- rango requires further study. Geagras redimitus Cope. — Hardy and Mc- Diarmid (1969:162) discuss the occurrence of this species of snake in Sinaloa based on the locality of "Mazatlan." The only other few known records of occurrence in Mexico are from the states of Michoacan and Oaxaca. Mazatlan may refer to another place of that name in Guerrero or Oaxaca (see comments concerning type-lo- cality of Sphaerodactylus torquatus by Taylor 1947:304-305). Accounts of Species The amphibians and 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. Distribution of Herpetofauna The herpetofauna considered for distribution- al purposes consists of 145 taxa— 2 salamanders (1.4%), 35 frogs (24.1%), 5 turtles (3.4%). 33 liz- ards (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 230 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 1 . Frequency (number and percentage) of higher taxa ofamphibians and reptiles in the five herpetofaunal regions (MG, Mesquite-Grassland; PO, Pine-Oak; MBT, Mixed Boreal-Tropical; TD, Tropical-Deciduous; TS, Thorn-Scrub). Taxa VERTEBRATE ECOLOGY AND SYSTEMATICS 231 Boreal-Tropical, Tropical-Deciduous, and Thorn-Scrub; one of these (Masticophis mento- varius) that seems established in the Mesquite- Grassland is discussed below under Barranca Corridors. Sixteen taxa (2 frogs. 2 turtles, 3 liz- ards. 9 snakes) occur only in the Tropical-De- ciduous and Thorn-Scrub. These variable dis- tributional patterns are depicted in Fig. 2. An unusual distributional pattern, not influ- enced by the east-west course of the transect, is the occurrence of four species in only the Mes- quite-Grassland and Thorn-Scrub. This pattern, represented by Scaphiopus couchi, Bufo punc- tatus, Hypsiglena torquata, and Arizona elegans (two subspecies), is explained as southern atten- uations of geographic ranges of Nearctic species on either side of the Sierra Madre. Barranca Corridors.— The Sierra Madre Oc- cidental is cut by many large barrancas and ar- 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 head- waters draining the vicinity of Ciudad Durango. Documentation of tropical species far to the east in these barrancas has been previously noted by Webb and Baker (1962) and by Crossin et al. (1973). Tropical species may extend into or very near non-tropical areas. Dispersal of tropical species eastward in bar- ranca corridors is exemplified by: 1) Anolis neb- 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 Durango for Trimorphodon tau (Univ. New Mexico 22790) in Mesquite-Grassland, 3) the locality of "ca. 10 mi SW Durango" for Elaphe 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- less continuity of populations of Hylactophryne august i (disjunct along transect route, see species account) provided by rocky barranca habitats (an intervening locality for the species from such a habitat is 6 miles southeast Llano Grande. UTEP). Faunal Assemblages In an effort to determine the degree to which each of the four transition zones acts as a barrier to east-west dispersal, 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- ards, 6 snakes), Pine-Oak/Mixed Boreal-Tropi- cal; 21 (5 frogs, 8 lizards, 8 snakes). Mixed Bo- real-Tropical/Tropical-Deciduous; and 24 (2 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- sition 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 to the eastern dispersal of tropical species. Ek- man's (in Udvardy 1969:274) formula A + B (total taxa of two compared regions)/C (shared taxa), in which the highest numerical value in- dicates the greatest faunal change, also marks the most abrupt transition between the Pine-Oak and Mixed Boreal-Tropical with a value of 6.73 (Mesquite-Grassland/Pine-Oak. 4.67; Mixed Boreal-Tropical/Tropical- Deciduous, 3.67; Tropical-Deciduous/Thorn-Scrub, 4.87). A slightly different manipulation of the num- bers of total taxa in. and shared taxa between, each region marks the Mixed Boreal-Tropical as having the greatest discrepancy between per- centages of shared taxa with adjacent regions (i.e., 65.6% of the taxa in this region is shared with the Tropical-Deciduous and only 34.4% is shared with the Pine-Oak, a difference of 3 1 .2%). In the Tropical-Deciduous the discrepancy percentage of shared taxa with adjacent faunal regions is 6.7% (53.3% shared with Thorn-Scrub, 46.6% with Mixed Boreal-Tropical), and in the Pine- Oak is 10.2% (31.4% shared with Mixed Boreal- Tropical and 41.6% with Mesquite-Grassland). These data indicate the greatest east-west faunal break in the Mixed Boreal-Tropical, with the herpetofauna mostly aligned to the western trop- ical regions. Although various formulas (primarily to ad- just for the differences in magnitude of the two compared regions) have been proposed to indi- cate degree of faunal resemblance, all show the same general trend (as does Ekman's formula, see above). For example, the values based on the formulas ofJaccard,C/N, + N 2 - C x 100, and Simpson, C/N, x 100 (in Udvardy 1969:273). 232 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY and of Duellman (1965:677), 2C/N, + N 2 x 100 (here modified to avoid decimal fractions) when applied, respectively, to the four faunal transi- tions along the transect are: Mesquite-Grassland/ Pine-Oak, 25.5, 44. 1 , and 40.6; Pine-Oak/Mixed Boreal-Tropical, 19.6, 34.4, and 32.8; Mixed Bo- real-Tropical/Tropical-Deciduous, 37.5, 65.6, and 54.6; 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 Trop- ical-Deciduous and Mixed Boreal-Tropical, about the same degree of resemblance between Mes- quite-Grassland and Pine-Oak as between Thorn- Scrub and Tropical-Deciduous, and the least re- semblance between Pine-Oak and Mixed Boreal- Tropical. The distributional data, as well as the abrupt climatic change personally experienced in winter (see section on Physiography and Climate) em- phasizes the distinction between the Pine-Oak and Mixed Boreal-Tropical. This transition cor- responds to that between two major herpeto- faunal assemblages, the classic Neotropical and Nearctic zoogeographical realms. Northern Nearctic and southern Neotropical species over- lap on either side of the Sierra Madre along the east-west trending transect. Nearctic Herpetofauna.— Aside from that of the Pine-Oak and most of the herpetofauna of the Mesquite-Grassland, some species of Nearc- tic affinities also have extended their ranges far to the south on the Pacific side of the Sierra Madre into tropical habitats. Of the 93 taxa in the three tropical regions along the transect, 18 (19.4%) have Nearctic affinities, and all but one of them (Gyalopion recorded only once from the adjacent Tropical- Deciduous) are restricted to the coastal Thorn- Scrub. These 18 taxa, some of which seem to be near their southernmost extent of range (marked with asterisk), consist of four frogs (Scaphiopus couchi, Bufo kelloggi, *Bufo punctatus, *Gastro- 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, Phyllo- rhynchus browni, Pituophis tnelanoleucus, Sal- vadora deserticola, Sonora aemula, Tantilla ya- quia, *Micruroides euryxanthus). Some of these species are represented by tropically adapted subspecies. Neotropical Herpetofauna.—Taxa of tropical affinities occur not only in the Pacific coastal Thorn-Scrub, Tropical-Deciduous, and Mixed Boreal-Tropical, but also in the Mesquite-Grass- land that is composed mostly of Nearctic species. The tropical species in these two regions of the transect represent two different tropical assem- blages. 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 (Eleutherodac- tylus hobartsmithi, Tomodactylus nitidus, T. saxatilus, Syrrhophus teretistes, Hyla bistincta, H. smaragdina, Gastrophryne usta), four lizards (Sceloporus bulleri, S. heterolepis, S. utiformis, Eumeces colimensis), and four snakes (Dryado- phis c/iftoni, D. melanolomus, Rhadinaea hes- peria, Leptodeira maculata). Of the 34 taxa in the Mesquite-Grassland, 4 ( 1 1.8%) are judged to have tropical affinities with the Mesa Central, the southern tropical highland of the Mexican Plateau. The four taxa include one frog (Bufo occidentalis), one turtle (Kinoster- non integrum subsp.), one lizard (Sceloporus spi- nosus), and one snake (Pituophis deppei). Acknowledgments Field work was financed by grants from the Bache Fund ofthe National Academy ofSciences (1961, Grant 463), the National Science Foun- dation (1962, part of Grant G-23042 to William W. Milstead), and the Penrose Fund ofthe Amer- ican 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 helpful field companions, I am most grateful to Rollin H. Baker and his wife Mary, J. Keever Greer, Leslie C. Drew, and Rudolph A. Scheib- ner. Field work through the years has been en- hanced by the cooperation and hospitality ex- tended by many local residents, especially Rodolfo Corrales and Fidel Gutierrez of Ciudad Durango. Summary In studying the distribution of 145 kinds of amphibians and reptiles across the Sierra Madre Occidental from Cd. Durango, Durango to Ma- VERTEBRATE ECOLOGY AND SYSTEMATICS 233 zatlan, Sinaloa, five herpetofaunal regions are recognized (from east to west)— Mesquite-Grass- land. Pine-Oak, Mixed Boreal-Tropical, Tropi- cal-Deciduous, and Thorn-Scrub. An enumera- tion of localities places each of the 145 taxa in one or more region and provides for a distri- butional analysis of the herpetofauna along the transect route. The total number of taxa is high- est in the coastal Sinaloan Thorn-Scrub, and snakes are the most abundant component in each region. Each region contains endemic taxa with the highest percentage in the Thorn-Scrub. The most abrupt faunal break is between the Pine- Oak and Mixed Boreal-Tropical, which also marks the transition between the Nearctic and Neotropical zoogeographical realms. New state records include Diadophis punctatus and Pituo- phis deppei for Sinaloa. and Dryadophis cliftoni and Sahadora bairdi for Durango. Literature Cited Anderson, J. D. 1 960. Storeria storerioides in western Mexico. Her- petologica, 16:63-66. 1961. The life history and systematics of Ambys- toma rosaceum. Copeia, 1961:371-377. Armstrong, B. L. and Murphy, J. B. 1979. The natural history of Mexican rattlesnakes. Univ. Kansas Publ. Mus. Nat. Hist., Spec. Publ., (5): 1-88. Baker. R. H. and Greer, J. K. 1 962. Mammals 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. Hist., (5)10:326-328' 1885. Catalogue of the lizards in the British Mu- seum (Natural History). Vol. II. Taylor & Francis, London. 1887. Catalogue of the lizards in the British Mu- seum (Natural History). Vol. III. Taylor & Francis. London. Chrapliwy. P. S. and Fugler, C. M. 1955. Amphibians and reptiles collected in Mexico in the summer of 1953. Herpetologica. 1 1: 121-128. Chrapliwy. P. S., Williams, K. and Smith, H. M. 1961. Noteworthy records of amphibians from Mexico. Herpetologica. 17:85-90. Conant, R. 1963. Semiaquatic snakes of the genus Thamno- phis from the isolated drainage system of the Rio Nazas and adjacent areas in Mexico. Copeia, 1963:473-499. 1969. A review of the water snakes of the genus Natrix in Mexico. Bull. Amer. Mus. Nat. Hist.. 142:1-140. Crossin, R. S. 1967. The breeding biology of the tufted jay. Proc. Western Foundation Vert. ZooL 1 :265-299. Crossin, R. S.. Sot it. O. H., Webb, R. G. and Baker, R. H. 1973. Biotic relationships in the Canon del Rio Mezquital. Durango, Mexico. Southwestern Nat., 18:187-200. Dixon, J. R. 1969. Taxonomic review of the Mexican skinks of the Eumeces brevirostris group. Contnb. Sci. Los Angeles County Mus.. ( 1 68): 1—30. Dowling, H. G. 1960. A taxonomic study of the ratsnakes. genus Elaphe Fitzinger. VII. The triaspis section. Zoologica. 45:53-80. Duellman, W. E. 1960. A taxonomic study of the middle American snake, Pituophis deppei. Univ. Kansas Publ. Mus. Nat. Hist., 10:599-610. 1965. A biogeographic account of the herpetofauna of Michoacan. Mexico. Univ. Kansas Publ. Mus. Nat. Hist., 15:627-709. 1970. Hylid frogs ofmiddle America. Monogr. Mus. Nat. Hist. Univ. Kansas. (1): 1-753. Dunn, E. R. 1936. The amphibians and reptiles of the Mexican expedition of 1934. Proc. Acad. Nat. Sci. Philadelphia. 88:471-477. Frost, J. S. and Bagnara, J. T. 1 976. A new species of leopard frog (Rana pipiens complex) from northwestern Mexico. Co- peia. 1976:332-338. Gadow, H. 1 905. The distribution of Mexican amphibians and reptiles. Proc. Zool. Soc. London. 2:191-244. Garstka, W. R. 1982. Systematics of the mexicana species group of the colubrid genus Lampropeltis, with an hypothesis [of] mimicry. Breviora, (466): 1- 36. Gehlbach, F. R. 1965. Herpetology of the Zuni Mountains region, northwestern New Mexico. Proc. U.S. Nat. Mus., 116:243-332. Gloyd. H. K. 1940. The rattlesnakes, genera Sistrurus and Cro- talus. Chicago Acad. Sci. Spec. Publ.. (4):i- vii, 1-270, pis. 1-31. Gunther, A. C. L. G. 1885-1902. Reptilia and Batrachia. Biologia Cen- trali-Americana. Dulau and Co.. London. Hardy, L. M. 1975. A systematic revision of the colubrid snake genus Gyalopion. J. Herpetol.. 9:107-132. Hardy, L. M. and McDiarmid. R. W. 1969. The amphibians and reptiles of Sinaloa, Mexico. Univ. Kansas Publ. Mus. Nat. Hist.. 18:39-252. Johnson, J. D. 1977. The taxonomy and distribution of the neo- tropical whipsnake Masticophis mentovarius (Reptilia, Serpentes. Colubridac). J. Herpe- tol.. 11:287-309. 234 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Klauber, L. M. 1949. The subspecies of the ridge-nosed rattle- snake, Crotalus willardi. Trans. San Diego Soc. Nat. Hist., 11:121-140. Langebartel, D. A. 1959. A new lizard (Sceloporus) from the Sierra Madre Occidental of Mexico. Herpetologica, 15:25-27. Lowe, C. H., Jones, C. J. and Wright, J. W. 1968. A new plethodontid salamander from So- nora, Mexico. Contrib. Sci. Los Angeles County Mus., (140): 1-11. McCoy, C. J., Jr. 1964. Notes on snakes from northern Mexico. Southwestern Nat., 9:46-48. McDiARMID, R. W. 1968. Variation, distribution and systematic status of the black-headed snake Tantilla yaquia 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 Mex- ico: new records from Sinaloa and the Tres Marias Islands. Contrib. Sci. Nat. Hist. Mus. Los Angeles County, (275): 1-1 7. McDiarmid, R. W. and Scott, N. J., Jr. 1970. Geographic variation and systematic status of Mexican lyre snakes of the Trimorphodon tan group (Colubridae). Contrib. Sci. Nat. Hist. Mus. Los Angeles County, ( 1 79): 1 — 43. Myers, C. W. 1 974. 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 microhylid genus Gastrophrvne. J. Herpe- toL, 6:111-137. Platz, J. E. and Mecham, J. S. 1979. Rana chiricahuensis, a new species of leop- ard frog {Rana pipiens complex) from Ari- zona. Copeia, 1979:383-390. Reeve, W. L. 1952. Taxonomy and distribution of the horned lizards genus Phrynosoma. Univ. Kansas Sci. Bull., 34:817-960. RlEMER, W. J. 1955. Comments on the distribution of certain Mexican toads. Herpetologica, 11:17-23. Robinson, M. D. 1979. Systematics of skinks of the Eumeces bre- virostris species group in western Mexico. Contrib. Sci. Nat. Hist. Mus. Los Angeles County, (3 19): 1-1 3. Rossman, D. A. and Blanev, R. M. 1 968. A new natricine snake of the genus Adelophis from western Mexico. Occas. Pap. Mus. Zool. Louisiana State Univ., (35): 1-1 2. Schmidt. R. H., Jr. 1976. A geographical survey of Sinaloa. South- western Stud., Texas Western Press, Univ. Texas El Paso, (50): 1-77. Smith, H. M. 1938. The lizards of the torquatus group of the ge- nus Sceloporus Wiegmann, 1828. Univ. Kansas Sci. Bull. (1936), 24:539-693. 1939a. An annotated list of the Mexican amphibi- ans and reptiles in the Carnegie Museum. Ann. Carnegie Mus., 27:31 1-320. 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 VI. Guide to Mexican turtles. Bib- liographic Addendum III. John Johnson, Vermont (1979), pp. xviii + 1044. Smith, R. B. 1971. Seasonal activity and other aspects of the ecology of terrestrial vertebrates in a neo- tropical 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. Mamm., 55:45-65). Tanner, W. W. 1961. A new subspecies of Conopsis nasus from Chihuahua, Mexico. Herpetologica, 17:13- 18. Tanner, W. W., Dixon, J. R. and Harris, H. S., Jr. 1972. A new subspecies of Crotalus lepidus from western Mexico. Great Basin Nat., 32:16- 24. Taylor, E. H. 1947. A review of the Mexican forms of the lizard genus Sphaerodactvlus. Univ. Kansas Sci. Bull., 31:299-309. Thomas, R. A. and Dixon, J. R. 1976. A re-evaluation of the Sceloporus scalaris group (Sauria: Iguanidae). Southwestern Nat., 20:523-536. Thompson, F. G. 1957. A new Mexican gartersnake (genus Tham- nophis) with notes on related forms. Occas. Pap. Mus. Zool. Univ. Michigan, (584): 1- 10. Tihen, J. A. 1949. A review of the lizard genus Barisia. Univ. Kansas Sci. Bull., 33:217-256. 1954. 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. F. 1969. Dynamic zoogeography. Van Nostrand Reinhold Co., New York. Webb, R. G. 1 960. Notes on some amphibians and reptiles from northern Mexico. Trans. Kansas Acad. Sci., 63:289-298. VERTEBRATE ECOLOGY AND SYSTEMATICS 235 1 966. Resurrected names for Mexican garter snakes. Thamnophis cyrtopsis (Kennicott). Tulane Stud. Zool.. 13:55-70. 1 967. Variation and distribution of the iguanid liz- ard Sceloporus bulleri, and the description of a related new species. Copeia. 1967:202- 213. 1968. The Mexican skink Eumeces lynxe (Squa- mata. Scincidae). Publ. Mus. Michigan State Univ., Biol. Sen. 4:1-28. 1969. Variation, status, and relationship of the iguanid lizard Sceloporus shannonorum. Herpetologica. 25:300-307. 1972. Resurrection of Bufo mexicanus Brocchi for a highland toad in western Mexico. Herpe- tologica, 28:1-6. 1976. A review of the garter snake Thamnophis elegans in Mexico. Contrib. Sci. Nat. Hist. Mus. Los Angeles County. (284): 1-1 3. 1977. Comments on snakes of the genus Geophis (Colubridae) from the Mexican states of Du- rango and Sinaloa. Southwestern Nat., 21: 543-559. 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. 1 970. Three new species ofHannemania (Acarina, Trombiculidae) from amphibians ofwestern Mexico. Bull. So. California Acad. Sci.. 69: 65-73. Wilson. L. D. 1 966. The range of the Rio Grande racer in Mexico and the status of Coluber oaxaca (Jan). Her- petologica, 22:42-47. 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. 1 954b. Notes on the distribution of some reptiles in western Mexico. Herpetologica. 10:145-149. 1956. A survey of the frogs of the augusti group, genus Eleutherodactylus. Amer. Mus. Nov- it.. (1813): 1-35. 1967. Eleutherodactylus augusti. Cat. Amer. Am- phib. Rept:4 1.1-4 1.4. Appendix This appendix provides accounts of the 1 45 taxa that formed the data base for the analysis of distributional patterns. Accounts are brief, the primary intent being only to provide locality records that indicate occur- rence in one or more ofthe faunal regions. Place-names are explained in the gazetteer and geographically ori- ented 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: AMNH— American Museum of Natural History CSULB— California State University Long Beach (specimens now presumably in Los An- geles County Museum Natural History) KU — Museum of Natural History. University of Kansas MCZ— Museum of Comparative Zoology. Har- vard University MSU — The Museum, Michigan State Universiu UIMNH — University of Illinois Museum ofNatural History UMMZ— Museum ofZoology. University of Mich- igan USNM — National Museum of Natural History UTEP— Laboratory for Environmental Biology. University of Texas at El Paso The reference Hardy and McDiarmid (1969), often cited for Sinaloan localities, is abbreviated to "H-M: [page number]"; some localities cited by them are mod- ified 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) owing to contemplated publication of more detailed information elsewhere. Class Amphibia Order Caudata Family Ambystomatidae Ambystoma rosaceum Taylor. Durango: Vicinity El Salto to near La Ciudad (Anderson 1961): 1.6 km W Buenos Aires (Welbourn and Loomis 1970:69. 71). Pine-Oak. Ambvstoma tigrinum subsp. Durango: Navios. 1 mi S Navios (AMNH); 4 mi NE Navios (UTEP): 13 mi N Durango (MCZ). The taxonomic status of these sal- amanders in Durango is uncertain. Mesquite-Grass- land and Pine-Oak. Order Anura Family Pelobatidae Scaphiopus couchi Baird. Durango: Vicinity Duran- go (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 metamorphosed toadlets were active in daytime on 1 1 July about temporary rain pools 1 mi N Mazatlan: 26 (UTEP) ranged in length from 9 to 19. averaging 13.8 mm. Mesquite- Grassland and Thorn-Scrub. Scaphiopus multiplicatus Cope. Durango: Several specimens within seven-mile radius Durango. 10 miW Metates. 15 mi E Coyotes (MSU): 6 mi SE Llano Grande (UTEP). Mesquite-Grassland and Pine-Oak. Family Leptodactylidae Eleutherodactylus hobartsmithi (Taylor). Sinaloa: Vicinity Chupaderos [24.8 mi E jet hwys 40-15 = ca. 1 mi E Chupaderos]. Santa Lucia (H-M:73). Tropical- Deciduous. Eleutherodactylus occidentals Taylor. Sinaloa: 7.1 mi E Concordia. 9.8 mi E Concordia. 15.7 mi E Con- cordia (H-M:74): 5 mi SW Copala. 2 mi SW Santa Lucia (MSU). The two MSU specimens were active at 236 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY night on dirt roads in the rainy season (30 July and 1 August 1960). Food in the stomach of the Copala spec- imen (identified by Dr. George W. Byers, Department of Entomology, University of Kansas) consisted of: Neuroptera, Myrmeliontidae (1 larva); Diptera, Ti- pulidae, Limonia; Lepidoptera, Noctuidae (1 larva); Phalangida (2 specimens); and many earthworm frag- ments. Tropical-Deciduous. Eleutherodactylus vocalis Taylor. Sinaloa: Vicinity Chupaderos [MCZ Concordia locality close to Chu- paderos; 24.8 mi E jet hwys 40-15 = ca. 1 mi E Chu- paderos] east to Santa Lucia and Potrerillos (H-M:75). 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 individ- uals (KU) of the more frequently observed Eleuthero- dactylus vocalis (Webb 1960:289), were obtained at night as they perched on boulders wet from splashing water ofa cascading stream near Santa Lucia. Tropical- Deciduous. Hylactophryne augusti latrans (Cope). Durango: 2.5 mi W Tapias (AMNH). Durangan specimens, tenta- tively assigned to H. a. latrans, may represent inter- grades with H. a. cactorum (Zweifel 1967:41.3). Mes- quite-Grassland. Hylactophryne tarahumaraensis Taylor. Durango: 10 mi NW Las Adjuntas, 6 mi WSW Las Adjuntas (Zweifel 1956:29); 5.5 mi SW El Salto (KU); 6 mi SW El Salto (UTEP); 2 mi E El Espinazo (CSULB). Pine- Oak and Mixed Boreal-Tropical. Leptodactylus melanonotus (Hallowell). Sinaloa: Mazatlan-Villa Union area, 11 and 12 mi NE Con- cordia (H-M:78, as L. occidentalis). Thorn-Scrub. Syrrhophus teretistes Duellman. Sinaloa: Several lo- calities extending from 3.4 mi NE Concordia to vicin- ity El Batel [47.2 mi NE Villa Union = 1.4 mi NE El Batel] (H-M:78, as S. modestus). Tropical-Deciduous. Tomodactylus nitidus petersi Duellman. Durango: 0.5 mi W Revolcaderos (MSU); 49 mi NE Concordia, Si- naloa [=ca. 1.5 mi SW Revolcaderos] (H-M:78). Si- naloa: Vicinity Santa Lucia east to Santa Rita and El Batel [47.2 mi NE Villa Union = 1.4 mi NE El Batel] (H-M:78); 8 (road) mi W El Palmito (UTEP). Mixed Boreal-Tropical and Tropical-Deciduous. Tomodactylus saxatilus Webb. Durango: 0.5 mi W Revolcaderos (MSU); 23.5 km SW Buenos Aires [=7.5 mi NE Revolcaderos] (Welbourn and Loomis 1970: 7 1 ). Sinaloa: 8 (road) mi W El Palmito (H-M:79). Mixed 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 Du- rango, NE city" (UMMZ); 2 mi NE Coyotes, 9.7 mi NE El Salto, 10 mi W Metates (Webb 1972: 1-2). Mes- quite-Grassland and Pine-Oak. Bufo debilis insidior Girard. Durango: 5 mi S Du- rango (MSU). Mesquite-Grassland. Bufo kelloggi Taylor. Sinaloa: Mazatlan- Villa Union area east to 4.7 mi NE Concordia and Copala (H-M: 80). Thorn-Scrub. Bufo marinus (Linnaeus). Sinaloa: Mazatlan-Villa Union area east to 2 mi ENE Copala and 1 mi W 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). Thorn- Scrub. Bufo mazatlanensis Taylor. Sinaloa: Mazatlan-Villa Union area east to 0.6 mi W Santa Lucia and 27.2 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. Bufo occidentalis Camerano. Durango: 3 mi W Du- rango (UTEP), 10 mi SW El Salto (KU), 1 1 mi W La Ciudad (AMNH), 0.5 mi W Revolcaderos (MSU), 4 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 Batel] (H-M:85). Mesquite-Grassland, Pine-Oak, Mixed Boreal-Tropical, and Tropical-Deciduous. Bufo punctatus Baird and Girard. Durango: Durango (AMNH); 2.5 mi W Tapias, Rio Chico (UTEP). Si- naloa: 2 mi E Mazatlan, about 3 mi SE Mazatlan (H-M: 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. Mes- quite-Grassland and Thorn-Scrub. Family Hylidae Hyla arenicolorCope. Durango: Localities extending from Cerro de Mercado [=ca. 3 km N Durango] to 5 km W El Espinazo (Duellman 1970:698). Sinaloa: 44 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 El Palmito (UTEP). Mesquite-Grassland, Pine-Oak, Mixed Boreal-Tropical, and Tropical-Deciduous. Hyla bistincta Cope. Durango: 5 km W El Espinazo (Duellman 1970:698). Sinaloa: 1.6 km E Santa Lucia (Duellman 1970:699). Mixed Boreal-Tropical and Tropical-Deciduous. Hyla eximia Baird. Durango: Localities vicinity Du- rango 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 Pine- Oak. Hyla smaragdina Taylor. Sinaloa: Localities extend- ing from Copala east to Potrerillos [27.2 mi E Con- cordia = 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: Mazatlan- Villa Union area east to 3.2 km SW Copala and 12 mi VERTEBRATE ECOLOGY AND SYSTEMATICA 237 NE Concordia [=ca. 2 mi SW Copala] (H-M:92-93). Thorn-Scrub. Pternohyla fodiens Boulenger. Sinaloa: Mazatlan- Villa Union area east to 4.7 mi NE Concordia (H-M: 94). Thorn-Scrub. Smilisca baudini (Dumeril and Bibron). Sinaloa: Mazatlan-Villa Union area east to 1 6.5 mi E Concordia [=2.3 mi NE Copala] (H-M:95-96). Thorn-Scrub. Triprion spatulatus spatulatus (Giinther). Sinaloa: Localities in Mazatlan-Villa Union area east to vicinity Concordia (H-M:88). Thorn-Scrub. Family Microhylidae Gastrophryne olivacea (Hallowell). Sinaloa: Maza- tlan-Villa Union area east to 5 km SW Concordia (H-M: 98). Thorn-Scrub. Gastrophryne usta (Cope). Sinaloa: Presidio and Venadillo [near Mazatlan], east to 4 km NE Concordia and 4.5 mi NE Concordia (H-M:99); 9 mi N Mazatlan. 1 1 mi SE Villa Union (Nelson 1 972: 131). Thorn-Scrub. Hypopachus variolosus (Cope). Sinaloa: Mazatlan- Villa Union area east to 4.5 mi NE Concordia and 9 km NE Concordia (H-M: 100, as H. o. oxyrrhinus). Thorn-Scrub. Family Ranidae Rana pustulosa Boulenger. Sinaloa: Several localities vicinity Santa Lucia (2 1 km in error for 2. 1 km E Santa Lucia for KU 4463 1), 7.2 mi W Santa Rita (H-M: 103); 14 mi SW El Batel. 10 mi NE El Batel [=ca. 1 mi W El Palmito] (Zweifel 1954a: 13 1-1 32. H-M: 103): cave below (west) Copalita [see Copala] (LACM). The above records of occurrence combine those listed by Hardy and McDiarmid (1969:103) for R. pustulosa and R. sinaloae. 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- species of A', integrum are recognized— Sinaloan spec- imens having coarse yellow-blotched head patterns in Thorn-Scrub and Tropical-Deciduous, and Durangan specimens that lack contrasting blotched head patterns in Mesquite-Grassland and Pine-Oak. Fragmentary data on eggs deposited by a captive female (Rio del Presidio. Sinaloa) are: 9 eggs (total), deposited 6 October (5). 8 October (2), and 1 1 October (2 eggs); weights (taken 6-13 October) ranged from 3.74 to 5.30, averaging 4.2 gm; measurements (taken 1 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. Family Emydidae Pseudemvs scripta ornata (Gray). Sinaloa: Mazatlan- Villa Union area (H-M: 106. Smith and Smith 1980: 518). Thorn-Scrub. Rhinoclemmys pulcherrima rogerbarboun (Ernst). Sinaloa: Mazatlan, Presidio de Mazatlan. 9 mi W Con- cordia. 7.5 mi E Concordia. Santa Lucia (H-M: 107. Smith and Smith 1980:397). Thorn-Scrub and Trop- ical-Deciduous. Order Squamata— Suborder Sauria Family Gekkonidae Coleonyx variegatus fasciatus (Boulenger). Sinaloa: 10 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 (H-M:l 14). Over the relatively flat terrain in the im- mediate 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- Scrub. Family Iguanidae Anolis nebulosus (Wiegmann). Durango: 0.5 mi W Revolcaderos (MSU, UTEP). Sinaloa: Mazatlan-Villa Union area east to 5 km SW El Palmito (H-M:l 15- 1 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 N Mazatlan (H-M:l 16). Thorn-Scrub. Callisaurus draconoides bogerti Martin del Campo. Sinaloa: Several localities in immediate vicinity Ma- zatlan (H-M:l 19). Individuals seem mostly restricted to the leeward side of low sand dunes. Two gravid females were obtained on June 27; hatchlings and adult males and females were captured on 22 August (dif- ferent years). Progressive urbanization along the beach- front north of Mazatlan, as witnessed in years since 1955, has eradicated suitable habitat for Callisaurus. Thorn-Scrub. Ctenosaura pectinata (Wiegmann). Sinaloa: Maza- tlan-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 talons, alighted, and when startled, released the pre- sumed prey unharmed (ca. 4 mi W Concordia. 30 June 1961). The record of Ctenosaura from La Ciudad. Du- rango is in error (see Conant 1969:86). Thorn-Scrub. 238 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Holbrookia approximates subsp. Durango: Durango, Rio Chico (AMNH); 4 mi E Durango (KU). The no- menclature is in accord with the as yet unpublished data of Ralph W. Axtell. Mesquite-Grassland. Holbrookia elegans elegans Bocourt. Sinaloa: Ma- zatlan-Villa 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 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 igua- nas 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. Duran- go: 5 mi N Durango (Reeve 1 952:9 1 8, KU). Mesquite- Grassland. Phrvnosoma orbiculare bradti Horowitz. Durango: Coyotes (Smith 1 939a:3 1 5), El Salto (Smith 1 942:36 1 ), 10 mi E El Salto (Reeve 1952:940), La Ciudad (Bou- lenger 1885:242). Pine-Oak. Sceloporus bulleri Boulenger. Durango: 0.5 mi W Revolcaderos (Webb 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- cordia^ ca. 2 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- Deciduous. Sceloporus clarki boulengeri Stejneger. Sinaloa: Ma- zatlan-Villa Union area east to 5 km SW El Palmito (H-M: 1 32-1 34). A gravid female (seven eggs) was ob- tained on 2 August 1960 (5 mi SW Copala). Mixed Boreal-Tropical, Tropical-Deciduous, and Thorn- Scrub. Sceloporus grammicus microlepidotus Wiegmann. Durango: 7 mi SW Las Adjuntas, 2 mi E El Salto (Chrapliwy and Fugler 1955:124); La Ciudad (Boulen- ger 1885:233, Gunther 1 890[1 885-1 902]:72); Rancho Santa Barbara (MSU); Buenos Aires (AMNH). Pine- Oak. Sceloporus heterolepis shannonorum Langebartel. Durango: 0.5 mi W Revolcaderos (Webb 1969:302, 307). Sinaloa: 5 km SW El Palmito, 8 mi (13 km) W El Palmito, 19.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas] (H-M: 1 39, Webb 1 969:307); 37 miles by road 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 W Durango (AMNH); 10 mi W Metates (UTEP); El Salto (Dunn 1936:473); La Ciudad (Boulenger 1885: 224, Gunther 1890(1885-1 902]:69); 1 6 mi SW La Ciu- dad, 17 mi NE El Batel, Sinaloa [=ca. 0.5 mi W Re- volcaderos] (Zweifel 1954b: 145). Sinaloa: 4.7 mi W El Palmito (UTEP), 10 mi NE El Batel (Zweifel 1954b: 145), 1 km NE Santa Lucia (H-M: 135). Mesquite- Grassland, Pine-Oak, 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 Chrapli- wy. Durango: Durango, Coyotes, La Ciudad (Smith 1938:614); El Salto, 10 mi W Durango (Smith and Chrapliwy 1958:268); 4 mi SW Coyotes (Chrapliwy and Fugler 1955:124); Buenos Aires (AMNH). Mes- quite-Grassland and Pine-Oak. Sceloporus scalaris Wiegmann. Durango: 9.9 mi W Durango (Ernest A. Liner, pers. comm.); localities ex- tending from vicinity Neveria [=4 mi W La Ciudad] to 10 mi W Metates (Thomas and Dixon 1976:535). Mesquite-Grassland and Pine-Oak. Sceloporus spinosus spinosus Wiegmann. Durango: Durango (Smith 1939b:93), 4 mi E Durango (KU), 2.5 mi W Tapias (UTEP). Mesquite-Grassland. Sceloporus utiformis Cope. Sinaloa: Vicinity Villa Union east to Santa Lucia area (H-M: 140). Tropical- Deciduous and Thorn-Scrub. Urosaurus bicarinatus tuberculatus (Schmidt). Si- naloa: Mazatlan-Villa 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). Tropical- Deciduous and Thorn-Scrub. Family Scincidae Eumeces brevirostris bilineatus Tanner. Durango: Localities extending from 33 mi W El Salto [=near Los Bancos] east to Coyotes (Dixon 1969:14); localities ex- tending from 24 km SW El Salto to 16 km E Llano Grande (Robinson 1979:11). Pine-Oak and Mixed Bo- real-Tropical. Eumeces callicephalus Bocourt. Sinaloa: 5 mi (8 km) N Mazatlan, Presidio (H-M: 143, Robinson 1979:12); 1 mi N Mazatlan (MSU, UTEP). Four of these skinks (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 three adults, 67. 72, and 73 mm SVL, have blue-brown tails and two have indistinct pale stripes on head and neck. The hatchling of 27 mm (colors in 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 after- noon 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 herb- covered area, 5 mi SW Copala (just below Chupade- ros). The small specimen was about 45 mm SVL, had a blue tail, whitish venter, broad white lateral stripes on anterior half of body, and a blackish head and back; VERTEBRATE ECOLOGY AND SYSTEMATICS 239 the pale head stripes were reddish. Tropical-Decidu- ous. Eumeces lynxe belli (Gray). Durango: 30 mi E El Salto. Rancho Santa Barbara (Webb 1968:22). Pine- Oak. Family Teiidae Cnemidophorus costatus subspp. Sinaloa: Mazatlan- Villa Union area east to 5 km SW El Palmito (H-M: 147, 148). Two subspecies, C. c. huico and C. c. ma- zatlanensis, intergrade in the transect area (not differ- entially influenced by faunal regions) and are not recognized in this study. Mixed Boreal-Tropical. Trop- ical-Deciduous, and Thorn-Scrub. Cnemidophorus scalaris scalaris Cope. Durango: 5 mi S Durango (MSU); 2.5 mi W Tapias, Rio Chico (UTEP). Mesquite-Grassland. 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. Si- naloa: 7.2 mi E Santa Lucia. 19.2 km NE Santa Lucia. 5 mi SW El Palmito (H-M: 151). Mixed Boreal-Trop- ical and Tropical-Deciduous. Family Helodermatidae Heloderma horridum horridum (Wiegmann). Sina- loa: Mazatlan area east to 2 mi ENE Copala (H-M: 153). Thorn-Scrub. Order Squamata— Suborder Serpentes Family Leptotyphlopidae Leptotyphlops humilis dugesi (Bocourt). Sinaloa: Mazatlan, Presidio (H-M: 156). Thorn-Scrub. Family Boidae Boa constrictor imperator Daudin. Sinaloa: Maza- tlan- Villa Union area east to 10 mi NE Concordia (H-M: 156). Thorn-Scrub. Family Colubridae Adelophisfoxi Rossman and Blaney. Durango: Va mi E El Mil Diez (Rossman and Blaney 1968). Pine-Oak. Arizona elegans expolita KJauber. Durango: 4.4 mi ESE Durango (UMMZ). Mesquite-Grassland. Arizona elegans noctivaga Klauber. Sinaloa: 1.1 mi N Mazatlan (H-M: 156). Thorn-Scrub. Coniophanes lateritius lateritius Cope. Sinaloa: 8 km N Villa Union and about 30 mi NE Villa Union [=ca. 3 mi NE Copala] (H-M: 157). Thorn-Scrub. Conopsis nasus nasus Gunther. Durango: 32.5 mi W Durango City [=ca. 7 mi W Rancho Santa Barbara]. 5 mi E El Salto (Tanner 1961:17); 0.5 mi W Revol- caderos (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. Diadophis punctatus subsp. Durango: 32 mi W Du- rango [=ca. 6.5 mi W Rancho Santa Barbara] (McCoy 1964:47); 20 mi NW Los Coyotes (Gehlbach 1965: 307); 1.6 mi E El Palmito, Sinaloa (UTEP). Sinaloa: 1 mi W 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- tergrades (D. p. dugesi x D. p. regalis). Pine-Oak and Mixed Boreal-Tropical. Dryadophis cliftoni Hardy. Durango: 0.5 mi W Re- volcaderos (MSU), ca. 1 1 km W 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 W Santa Rita (H-M: 158). The Durango specimens are the first recorded from that state. Mixed Boreal-Tropical and Tropical-Deciduous. Dryadophis melanolomus stuarti Smith. Sinaloa: 8 km N Villa Union (H-M: 159). Thorn-Scrub. Drymarchon corais rubidus Smith. Sinaloa: Maza- tlan-Villa Union area east to 2.2 km NE Santa Lucia (H-M: 160). Tropical-Deciduous and Thorn-Scrub. Drymobius margaritiferusfistulosus Smith. Sinaloa: Mazatlan, 5 mi N Mazatlan, Presidio (H-M: 161). Thorn-Scrub. Elaphe triaspis intermedia (Boettger). Sinaloa: Ma- zatlan-Villa Union 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: 1 .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 Km 1 1 77 (AMNH); 5 km SW El Palmito. 1 9.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas] (H-M: 163. Webb 1977:551). Mixed Boreal-Tropical. Gyalopion quadrangulans (Gunther). Sinaloa: Ma- zatlan-Villa Union area, and 3.2 km SW Santa Lucia (H-M:168-169. Hardy 1975:116). Tropical-Decidu- ous 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 W Tapias (UTEP). Sinaloa: Mazatlan-Villa Union area east to 2.7 km NE Chupaderos (H-M: 170-1 71). Nomenclature follows Hardy and McDiarmid (1969: 170). Mesquite-Grassland and Thorn-Scrub. Imantodes gemmistratus latistratus (Cope). Sinaloa: Several localities vicinity Mazatlan. 2.2 km NE Santa Lucia (H-M: 1 72-173). Tropical-Deciduous and Thorn- Scrub. Lampropeltis getulus splendida (Baird and Girard). Durango: 5.1 mi ESE Durango (UMMZ). Mesquite- Grassland. Lampropeltis mexicana (Garman). Durango: Mimbres (MCZ); Rio Chico, Rancho Santa Barbara (Garstka 1982:31). Pine-Oak. 240 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Lampropeltis triangulum sinaloae Williams. Sina- loa: Mazatlan-Villa Union area east to 6 km SW Con- cordia (H-M: 175, as L. t. nelsoni). Thorn-Scrub. Leptodeira maculata Hallowell. Sinaloa: Mazatlan- Villa Union area east to Santa Lucia (H-M: 1 76). Trop- ical-Deciduous and Thorn-Scrub. Leptodeira punctata (Peters). Sinaloa: Several rec- ords in Mazatlan-Villa Union area (H-M: 177-1 79). Thorn-Scrub. Leptodeira septentrionalis polysticta Giinther. Sina- loa: Three specimens from north of Mazatlan (the near- est, 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 Batel, Presidio (H-M: 180). Tropical-Deciduous and Thorn-Scrub. Leptophis diplotropis (Giinther). Sinaloa: Mazatlan- Villa Union area east to 1 9.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas] and 10.6 mi E Santa Lucia [=ca. 1 mi E Loberas] (H-M: 182). Mixed Boreal-Tropical, Tropical-Deciduous, and Thorn-Scrub. Masticophis bilineatus Jan. Sinaloa: Mazatlan-Villa Union area east to 34 mi E Villa Union [=ca. 4 mi SW Santa Lucia] (H-M: 183). Tropical-Deciduous and Thorn-Scrub. Masticophis flagellum linearulus Smith. Durango: about 10 km SSE Durango (UTEP). Mesquite-Grass- land. Masticophis mentovarius striolatus (Mertens). Sina- loa: 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, Tropi- cal-Deciduous, and Thorn-Scrub. Masticophis taeniatus girardi Stejneger and Barbour. Durango: 8 mi W Durango (AMNH). Mesquite-Grass- land. Nerodia valida valida (Kennicott). Sinaloa: Maza- tlan-Villa Union area east to 24.8 mi E Villa Union [=ca. 1 mi E Chupaderos] (H-M: 187-188); Chupade- ros on Rio Chupaderos, 5 mi SW Copala (Conant 1 969: 88). Thorn-Scrub. Oxybelis aeneus auratus (Bell). Sinaloa: Mazatlan- Villa Union area, 4.4 mi SW Concordia, 20 mi E Villa Union [=ca. 4 mi W Chupaderos] (H-M: 189). Thorn- Scrub. Phyllorhynchus browni Stejneger. Sinaloa: 10 km N Mazatlan (H-M: 191). Thorn-Scrub. Pituophis deppei deppei (Dumeril). Durango: Coy- otes, Llano Grande (Duellman 1960:605); 3 mi W Du- rango (UTEP). Sinaloa: 4 km E Loberas (12 km W El 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. 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 Santa Lucia, 19.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas] (H-MT94-195; Myers 1974:243); 2 km E Loberas, Km 1 175 (Myers 1974:243). Mixed Boreal-Tropical and Tropical-Deciduous. Rhadinaea laureata (Giinther). Durango: Coyotes, lOmiEElSalto, 10 mi SW El Salto (Myers 1974:244); 1 mi S Navios (AMNH); 6 mi SE Llano Grande (UTEP). Pine-Oak. Rhinocheilus lecontei antonii Duges. Sinaloa: Ma- zatlan-Villa Union area east to 1.4 mi E Concordia (H-M: 195-1 96). Thorn-Scrub. Salvadora bairdi Jan. Durango: 24 road mi W La Ciudad [=ca. 1 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 W El Palmito (MSU). The MSU specimen was found DOR, as well as another badly mashed specimen (from 4 mi 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- ous. Salvadora deserticola Schmidt. Sinaloa: 9 mi N Ma- zatlan, 10 mi S Villa Union (H-M: 199). Thorn-Scrub. Salvadora grahamiae grahamiae lineata Schmidt. 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). Thorn- Scrub. Storeria storerioides (Cope). Durango: La Ciudad, El Salto (Anderson 1960:63). Sinaloa: 19.2 km NE Santa Lucia [=ca. 2.3 mi E Loberas], 9.6 mi SW 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 WSW Durango (MCZ), 2.5 mi W Tapias (MSU), Rio Chico (UTEP). Mesquite-Grassland. Tantilla vaquia Smith. Sinaloa: 5.8 mi N Mazatlan. 16 mi 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). Si- 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). Duran- go: 12 mi N Durango, 10 mi W Metates, Rio Chico (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 localities extending from Hacienda Coyotes to 1 mi W 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: 487). Mesquite-Grassland. Thamnophis eques virgatenuis Conant. Durango: Localities extending from near Coyotes west to 3 mi E Las Adjuntas (Conant 1963:490). Pine-Oak. Thamnophis melanogaster canescens Smith. Duran- go: Durango, 6 mi E Durango, 10 mi N Durango, Rio Chico, Mimbres, Coyotes, Hacienda Coyotes, 6 mi SW VERTEBRATE ECOLOGY AND SYSTEMATICS 241 El Salto, S side El Mil Diez (Conant 1963:481-482). Mesquite-Grassland and Pine-Oak. Thamnophis nigronuchalis Thompson. Durango: 5.6 mi W El Salto (Thompson 1957:1). 6 mi SW El Salto (UTEP). Two snakes from Coyotes and from 33 mi ENE El Salto [=Mimbres], although referred to Tham- nophis 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, Bi- bron, 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). Trop- ical-Deciduous and Thorn-Scrub. Tropidodipsas philippi (Jan). Sinaloa: 31.6 mi N Ma- zatlan (H-M:210). Thorn-Scrub. Family Elapidae Micruroides euryxanthus neglectus Roze. Sinaloa: 16.3 mi NNW 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 Con- cordia, 6.5 km SW Concordia (H-M:2 11-21 2). Thorn- Scrub. Family Viperidae Agkistrodon bilineatus bilineatus (Giinther). Sinaloa: 7.5 mi N Mazatlan. Mazatlan, Presidio (H-M:213); 4 mi SE Villa Union (UTEP). Thorn-Scrub. Crotalus basiliscus basiliscus (Cope). Sinaloa: Ma- zatlan-Villa Union area east to 19.2 km NE Santa Luc- cia (H-M:2 14-2 1 5); 4 km S Santa Lucia. 5 mi W Con- cordia (Armstrong and Murphy 1979:6): 5 mi W El Palmito, 8 road mi W El Palmito (UTEP). Mixed Bo- 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- ris. Durango: 15 mi W La Ciudad and 16 mi SW La Ciudad [both ca. 2-3 mi W El Espinazo], 1 km W Los Bancos, 1 1 mi W La Ciudad [=ca. 2 mi E El Espinazo] (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 1 6); 1 mi W Durango-Sinaloa state line, 4.8 mi E Santa Rita, 12.5 mi W El Palmito [=ca. 0.5 mi E Potrerillos] (Tanner, Dixon, and Harris 1 972: 1 6-17). Mixed Boreal-Tropical and Tropical-De- ciduous. Crotalus molossus nigrescens Gloyd. Durango: Coy- otes, El Salto (Gloyd 1940:164): 16 km W Durango. 8.3 km E Coyotes, Los Bancos (Armstrong and Murphy 1979:33). Mesquite-Grassland. Pine-Oak. and Mixed Boreal-Tropical. Crotalus pricei pricei Van Denburgh. Durango: Las Adjuntas, near Coyotes, 14 mi ENE El Salto. Llano Grande, Los Bancos (Armstrong and Murphy 1979: 38). Pine-Oak and Mixed Boreal-Tropical. Crotalus scutulatus scutulatus (Kennicott). Durango: 5 mi S Durango (MSU), 10 mi W Durango (AMNH). 2.5 mi W Tapias (UTEP). Mesquite-Grassland. Crotalus stejnegeri Dunn. Sinaloa: 2.2 km NE Santa Lucia (H-M:217); between 10 and 15 mi NE Concor- dia. 1 mi E Concordia (McDiarmid, Copp. and Breed- 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). Pine- Oak. 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 i 1984 Museum of Natural History. The University of Kansas. Lawrence Systematic Review of the Percid Fish, Etheostoma lepidum* Alice F. Echelle, Anthony A. Echelle, and Clark Hubbs Until recently, the greenthroat darter, Etheo- stoma lepidum (Baird and Girard). was known only from south and central Texas in east-flowing drainages of the Edwards Plateau (Strawn 1 955a, 1957). However. Hubbs and Echelle (1972) re- ported that Roster's (1957) "Etheostoma sp." in the Pecos River drainage of New Mexico is E. 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 previously recognized distribution. Hubbs and Echelle (1972) noted that the New Mexico pop- ulation had declined since the time ofW. J. Ros- ter's collections in the 1940's and 1950's. The disjunct occurrence and declining status of the New Mexico population prompted the present review ofthe species. Other studies ofgeographic variation in morphological characters of E. lep- idum are Strawn's (1955a, 1961) descriptions of variation in dorsal fin color and five meristic characters of Texas populations. Hubbs (1967) described variation in survival of offspring from intra- and interspecific crosses involving several Texas populations, and Hubbs and Delco (1960) described some aspects of geographic variation in egg complements of Texas populations. Materials and Methods Collections examined. — Museum abbrevia- tions in the following list are: UNM = Univer- sity of New Mexico Collection of Vertebrates; TNHC = 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- versity. Collections used for counts and mea- surements are as follows (letters in parentheses refer to localities as given in Tables 1-5; com- plete locality data available from authors): E. lepidum: Pecos River drainage: (A) OSU 1 1 342. 1 1 343. 1 1 344; (B) UNM 50. 55; (C) UNM 52. 63; (D) UNM 51, 57. 59; (E) UNM 65, 66, * Dedicated to Henry S. Fitch and Virginia R. Fitch, esteemed parents and friends. 67. 3064; (F) UNM 53; (G) UNM 49. Colorado River drainage: (H) TNHC 2435. 3096; (I) TNHC 207 1 . 3 1 2 1 . 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 1 5; (O) TNHC 3475. 3536. 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 made at minimum depth of caudal peduncle- scales above lateral line begins with the scale row above the 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- lows: = no scales: 1 = one to several imbedded or exposed scales covering less than one-half the area; 2 = partially unsealed, but scales covering more than one-half the area; 3 = completely scaled. 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 of pelvics. Etheostoma (Oligocephalus) lepidum (Baird and Girard) Figs. 1-2. Tables 1-5 Types and Nomenclature. — Baird's and Gi- rard'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 in- clude one syntype at the University of Michigan Museum of Zoology (UMMZ 86335) and six syntypes at the U.S. National Museum of Nat- ural History (USNM 744). Following article 74 243 244 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY m * >• it*. ' <* Fig. 1. Mature adults of Etheostoma lepidum from Sago Spring, upper end of Unit 4, Bitter Lake National Wildlife Refuge, Roswell, New Mexico. OSU 11342, 2 June 1971. Top, male, 45.1 mm SL. Bottom, female, 37.7 mm SL. of the International Code of Zoological Nomen- clature we designate as paralectotypes of E. lep- idum, UMMZ 86335, and under a new catalog number (USNM 223024), five specimens of USNM 744. The sixth specimen of USNM 744 retains the original number as the lectotype. Gi- rard ( 1 859a) referred to E. lepidum from the up- per Nueces (Leona River) as Poecilichthys lepi- dus and, in a paper published in the same volume (Girard 1859b), described Oligocephalus leonen- sis as a new species, also from Leona River. Two syntypes bearing the latter name are at the Mu- seum ofComparative Zoology (MCZ 24580) with the information that they were collected in the Nueces River by J. H. Clark and sent to MCZ from USNM in 1853. Evermann and Kendall (1894) regarded O. leonensis a junior synonym of E. lepidum. However, Collette and Knapp (1966) noted that the syntypes have "well-de- veloped 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 lepidogenys, was described by Evermann and Kendall (1894), from Comal Springs of the Gua- dalupe River drainage as a result of a mistaken comparison (Hubbs, Kuehne, and Ball 1 953) with E. spectabile rather than E. lepidum. The two syntypes for the latter description are deposited at the National Museum of Natural History (USNM 44840). Hubbs, Kuehne, and Ball (1953) and Hubbs and Echelle (1972) used the trino- mial, E. lepidum lepidogenys in reference to, re- spectively, Guadalupe River populations and "the Guadalupe-Colorado river stocks." VERTEBRATE ECOLOGY AND SYSTEMATICS 245 33- 30 27 w- 100 246 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY breeding tubercles (Collette 1965), a character which sets these three species apart from E. spectabile. Diagnosis. —A member of the subgenus Oli- gocephalus characterized by the following com- bination 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 so, rarely completely scaled; lateral line moderately arched, incomplete, total lateral line scales 44-60, pored scales, 19-42. Description. — Counts are presented in Tables 1-4. Measurements are shown in Table 5. Gen- eral 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 ac- count, ranges for Strawn's (1961) data are pre- sented in brackets. Fin rays of spinous dorsal 7-12 [4-1 1], mod- ally 9 in all populations except two tributaries of Colorado River where mode = 10. Soft dorsal fin rays 9-13 [8-14], most frequent mode =11. Anal soft rays 5-8 [4-9], rarely 5. Anal spines 1 or 2; in New Mexico populations, 2 more com- mon than 1; in 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 12- 19. 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 lateral line. Belly usually completely scaled, but occasional specimens with small naked area anteriorly or on the midventral line and rarely lh to more than Vi naked. Variation in codes for anterior squa- mation is given in Table 4. Opercles completely naked. (The comment by Moore 1 968, that spec- imens from San Saba River have scaled opercles is incorrect.) Other areas of anterior squamation generally naked or lightly covered, but with spec- imens from Blue Spring. New Mexico, and San Saba and Guadalupe rivers, Texas, more heavily scaled. Cheeks completely naked in all specimens 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 gen- erally as described for Blue Spring, but specimens from San Saba and Guadalupe rivers more fully scaled, occasional specimens with cheeks com- pletely covered with large overlapping scales. Moore's ( 1 968) comment that the cheeks are ful- ly 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-cov- ered with scales, rarely (two specimens from Blue Spring) completely covered. Breast completely naked 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 samples, but specimens often have scales similar to those described for Blue Spring. Preopercular pores 5-7. modally 6. Mandib- ular pores 2-7. modally 4. Infraorbital pores 2- 8, modally 6 with canal interrupted and 4 pores anteriorly, 2 posteriorly. Supratemporal pores generally 1-3, modally 2. Supratemporal canal usually interrupted middorsally. but occasion- ally uninterrupted; when uninterrupted, an extra 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 ex- tends dorsally between lateral bars and well above lateral line. Suborbital bar pronounced, extends downward from orbit to well below angle ofjaw. Ventrolateral sides ofhead, branchiostegal mem- branes and breast blue green. Pectoral fins with 6-7 variably distinct vertical rows ofdusky spots on rays, usually a reddish to rust-colored spot in axil. Pelvic fins dark blue green to blackish me- sially with, in more brightly colored individuals, scattered red, orange, or rust-colored spots lat- erally. Anal fin blue green, darker at base, often with 2-3 horizontal rows composed of 2-3 small. red, orange or rust-colored interradial spots. Caudal fin with 5-7 vertical rows composed of dusky to reddish brown spots on rays. First dor- VERTEBRATE ECOLOGY AND SYSTEMATICA 247 Table 1 . Caudal peduncle scale counts in E. lepidum and E. grahami. Locality letters refer to museum collection numbers as listed in Materials and Methods. Species 248 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. Lateral line scale counts in E. lepidum and E. grahami. Locality letters as in Table 1. VERTEBRATE ECOLOGY AND SYSTEMATICS 249 Table 3. Fin ray counts in E. lepidum and E. grahami. Locality letters as in Table 1. Species and 250 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 4. Codes for anterior squamation in E. lepidum and E. grahami. See Methods for description of codes. Locality letters as in Table 1. Species and locality Cheek Opercle Nape Breast E. lepidum New Mexico A 15 B 38 C 11 D 26 E 43 12 F 23 G 20 Texas H I J K L E. grahami M 10 3 N 16 6 O 2 8 31 4 4 16 18 13 27 29 13 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.1 2 1.2 6 1.6 0.7 0.3 0.2 0.3 1.2 15 37 11 26 55 23 20 35 27 29 40 42 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11 2.8 16 2.6 15 3.0 11 20 8 26 7 22 12 32 22 3 24 6 12 21 10 34 1 3 5 17 16 36 1 1 4 12 0.3 VERTEBRATE ECOLOGY AND SYSTEMATICA 251 National Wildlife Refuge— see Fig. 2) are closely associated with dense vegetation. In Texas the species occurs most abundantly in vegetated rif- fles when in sympatry with E. spectabile, but a shift 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 ofTexas populations has not been substantially altered by human activities. How- ever, pumping of water from underground aqui- fers will 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 W. J. Roster's early collections and a compila- tion of other records assembled by J. E. Sublette of Eastern New Mexico University. Although never abundant in the Pecos River proper (Ros- ter, pers. comm.). several large collections (32- 330 specimens) were made in the 1940's and early 1 950"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 (Ar- izona State University #0936) was taken from Cottonwood Creek. N of Artesia, Eddy County. Large populations are now known only from three localities: 1 ) Blue Spring, a 4 km spring and spring run ofthe Black River drainage. 8 km E ofWhites City. Eddy County. 2) small springs and water- fowl management ponds on the Bitter Lake Na- tional 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 (D. Jester, pers. 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 Pen- asco. 1.7 km S. 29.2 km EMayhill. Chaves Coun- ty, an area sufficiently isolated from presently known major populations that it is difficult to explain the single specimen as a stray individual from known areas of dense concentration. Thus, a permanent population may occur somewhere in the Rio Penasco drainage. The virtual elimi- nation of E. lepidum from the Pecos River and the mouths of its major tributaries has resulted from extensive habitat alterations which include channel straightening, brush removal, irrigation diversions, and pollution from oil fields, munic- ipalities, and cattle feedlots. The populations at Blue Spring and Bitter Lake National Wildlife Refuge are well protected by the present own- erships and seem in no immediate danger of elimination. Strawn (1955b) noted that greenthroat darters "will spawn repeatedly when kept at tempera- tures ranging from ... [16] (and probably lower) to [23°C] " Hubbs (1961. 1967) described developmental temperature tolerances (7-29°C) of the South Concho River and Nueces River populations. Hubbs and Strawn ( 1 957) noted that, in the Guadalupe River, at a springfed locality with relatively uniform temperatures, the breed- ing season is 10 to 1 2 months long with depressed breeding in mid-summer, while at a locality with more variable water temperatures, breeding oc- curred from November through April. The New Mexico populations seem to respond similarly. Based on dissection of females larger than 35 mm 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 collec- tions from the Pecos River and the mouths of larger tributaries include two February collec- tions in which 7 of 8 females were ripe, one July collection in which only 1 of 22 females was ripe, and six August collections in which, excluding a collection from the mouth of Black River, only 1 of 24 was ripe: in the excluded collection. 6 of 14 females were ripe. Thus. New Mexico pop- ulations apparently experience depressed breed- ing 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 Table 5. Means and, in parentheses, standard deviations of standard length (SL) and proportional measure- ments, as thousandths of SL, for E. lepidum and E. grahami.* VERTEBRATE ECOLOGY AND SYSTEMATICA 253 Table 5. Continued. 254 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 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 Lake National Wildlife Refuge, near Roswell. The New Mexico population conforms with published reports on breeding season in Texas populations; ripe fe- males occur during most months of the year, but breeding condition declines in summer. It is hypothesized that, in Pliocene times, E. lepidum and E. grahami diverged in allopatry, the latter in the ancestral Rio Grande and the former in more northern drainages of the Gulf Coast. In late Pliocene or early Pleistocene the lower Pecos River (presently occupied by E. gra- hami) eroded headward and presumably cap- tured the middle to upper Pecos from the Col- orado River or another drainage ofcentral Texas. The latter event brought E. lepidum into the Pe- cos River and created the present disjunct dis- tribution of the species. At present E. lepidum and E. grahami are separated by a saline, 300 km segment of the Pecos River. Acknowledgments We are indebted to William J. Koster for the loan of valuable early collections of E. lepidum from New Mexico, to James E. Sublette for his efforts at ferreting out obscure museum records, to Sublette and Mike Hatch for locality records in New Mexico, to the New Mexico Fish and Game Department, the Texas Parks and Wildlife Department, and the U.S. Fish and Wildlife Ser- vice for collecting permits, to Reeve M. Bailey (UMMZ), Bruce B. Collette and Susan Karnella (USNM) and William L. Fink (MCZ) for infor- mation regarding type specimens in their care, to Royal D. Suttkus (TU) and Robert F. Martin (TNHC) for loans of specimens, and to Warren Bounds, Larry Kline, and Delbert Boggs for as- sistance and permission to collect on property under their control. Reeve M. Bailey and an anonymous reviewer provided valuable com- ments on the manuscript. Part of this work was done at The University of Oklahoma Biological Station. Literature Cited Bailey, R. M. and Gosline, W. A. 1955. Variation and systematic significance of ver- tebral counts in the American fishes of the family Percidae. Misc. Publ. Mus. Zool., Univ. Michigan, 93:1-44. Baird, S. F. and Girard, C 1 853. Description ofnew species of fishes collected by Mr. John H. Clark, on the U.S. and Mex- ican Boundary Survey, under Lt. Col. Jas. D. Graham. Proc. Acad. Nat. Sci. Philadel- phia, 6:387-390. Blair, W. F. 1950. The biotic provinces of Texas. Texas J. Sci., 2:93-117. Collette, B. B. 1965. Systematic significance of breeding tubercles in fishes of the family Percidae. Proc. U.S. Nat. Mus., 117:567-614. Collette, B. B. and Knapp, L. W. 1966. Catalog of type specimens of the darters (Pisces, Percidae, Etheostomatini). Proc. U.S. Nat. Mus., 119:1-88. Distler, D. A. 1968. Distribution and variation of Etheostoma spectabile (Agassiz) (Percidae. Teleostei). Univ. Kansas Sci. Bull., 48:143-208. Echelle, A. A. and Echelle, A. F. 1978. The Pecos River pupfish. Cyprinodon pe- cosensisn. sp. (Cyprinodontidae), with com- ments on its evolutionary origin. Copeia, 1978:569-584. Eddy, S. and Underhill, J. C. 1978. How to Know 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- sin, 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 Re- gion. Cranbrook Inst. Sci. Bull., 26:1-213. Hubbs, C. 1961. Developmental temperature tolerances of four etheostomatine fishes occurring in Tex- as. Copeia, 1961:195-198. 1 967. Geographic variations in survival of hybrids between etheostomatine fishes. Bull. Texas Mem. Mus., 13:1-72. Hubbs, C. and Delco, E. A. 1 960. 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 VERTEBRATE ECOLOGY AND SYSTEMATICS 255 Jordan 1929. Jordan 1930. KOSTER 1957. Leonar 1975 MOORE, 1968. fecundity of the greenthroat darter, Etheo- stoma lepidum (Girard). Ecology, 38:596- 602. D. S. Manual of the Vertebrate Animals of the northeastern United States, Inclusive of Ma- rine Species. World Book, N.Y. D. S., Evermann, B. W. and Clark. H. W. Check-list of the Fishes and Fish-like Ver- tebrates of North America north of the Northern Boundary of Venezuela and Co- lombia. Rep. U.S. Fish Comm., 1928. , W. J. Guide to the Fishes of New Mexico. Univ. New Mexico Press, Albuquerque. d, A. B. and Frve, J. C. Pliocene and Pleistocene deposits and mol- luscan faunas, east-central New Mexico. New Mexico Bur. Mines and Min. Res. Memoir 30. G. A. Fishes. Pt. II. Pp. 22-165. In Blair, W. F., el ah, Vertebrates of the United States. McGraw Hill Book Co., N.Y. Strawn, K. 1955a. A method of breeding and raising three Tex- as darters, 1. Aquar. J., 26:408-412. 1 955b. A method of breeding and raising three Tex- as darters. 2. Aquar. J.. 27:12-14, 17. 31. 1957. The influence of environment on the meris- tic counts of the fishes, Etheostoma grahami and E. lepidum. Unpubl. Ph.D. dissertation, Univ. Texas, Austin. 1961. A comparison of meristic means and vari- ances of wild and laboratory-raised samples of the fishes, Etheostoma grahami and E. lepidum (Percidae). Texas J. Sci., 13:127- 159. Thomas, R. G. 1972. The geomorphic evolution of the Pecos Riv- er 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 Anolis fitchi, A New Species of the Anolis aequatorialis Group from Ecuador and Colombia Ernest E. Williams and William E. Duellman The aequatorialis group of the iguanid lizard genus Anolis is characteristic of cloud forests on the slopes of the Andes in northwestern South America. Vast areas of these forests still remain inaccessible, and the Anolis aequatorialis group, like many other elements of the cloud forest fau- na, has remained poorly known. The only pub- lished field observations on any of these lizards are those of Henry S. Fitch. Therefore, it is par- ticularly appropriate to associate his name with a handsome new species in the eulaemus 1 sub- section of the group. Anolis fitchi new species Frontispiece Holotype. — KXJ 142865, an adult male, from 16.5 km (by road) north-northeast ofSanta Rosa, Provincia Napo, Ecuador, 1 700 m elevation, ob- tained on 1 9 October 1971 by William E. Duell- man. Paratypes.—ECUADOR: Provincia Napo: KU 142864, 142866, same data as holotype; KU 142867-69. Rio Azuela at Quito-Lago Agrio road, 1740 m, William E. Duellman and Joseph T. Collins, 20-21 October 1971: MCZ 158324, same locality, Kenneth Miyata, 24 February 1979; KU 164162, Rio Salado, 1 km upstream from Rio Coca, 1410 m, William E. Duellman, 7 October 1974; Ku 164163-65, same locality, William E. Duellman and Alan H. Savitzky, 18 March 1975; KU 178960, same locality David C. Cannatella, 18 July 1977; KU 178961, same locality, Martha C. Lynch, 17 July 1977; MCZ 124350-51, "Loreto region." collector and date 1 By a lapsus, Williams (1976) used the name ae- quatorialis group in a table and eulaemus group in a key on the following page. The intention was to use the oldest name in each group as the nominate form. Anolis aequatorialis Werner 1894 antedates A. eulae- mus Boulenger 1 908, 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 distinct subgroups, which therefore are called the A. aequatorialis subgroup and the A. eulaemus subgroup. unknown; AMNH 28900, "Volcan Sumaco," Carlos Olalla, January, 1924; USNM 214869, "upper Rio Napo," Jorge Olalla, date unknown; USNM 214870, La Alegria on Rio Chingual, ±3 km N Sibundoy, ±20 km N La Bonita, 1930 m. James A. Peters, 24 June 1962. COLOMBIA: KU 169823-26, Departamento de Putumayo: 10.3 km W of El Pepino, 1440 m, William E. Duellman, 27-29 September 1974. 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 phalanx I, rather than continuous with the latter ("Norops condition" as understood by Boulenger 1885). Anolisfitchi is similar to A. eulaemus Bou- lenger 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 scales in multiple lines) and in the presence of a moderate-sized mottled or spotted dewlap in fe- males (female dewlap rudimentary with dark skin in eulaemus). Anolis fitchi is similar also to A. ventrimaculatus Boulenger but differs by having 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- erate dewlap in females (no trace of a dewlap in female A. ventrimaculatus). Description (Counts for Holotype in Parenthe- ses).— Anterior head scales small, multicarinate. tuberculate. or wrinkled; 1 1-18 (16) scales across snout between second canthals; some scales within shallow frontal depression larger than those immediately anterior to depression; 6-9 (9) scales bordering rostral posteriorly; 10 or 11 scales between supranasals dorsally; anterior na- sal above, or just behind, suture between rostral and first supralabial (Fig. 1). Supraorbital semi- circles separated by 1-3 (3) scales; no differen- tiated supraocular disc, but some scales slightly enlarged, keeled; one moderately elongate su- praciliary followed by one or two shorter scales. 257 258 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Fig. 1. Dorsal view of head of holotype of Anolis fitchi. first and second largest; 7-10 (8) rows of loreals, uppermost longest. Temporal and supratempor- al scales granular; no differentiated double in- tertemporal line of enlarged scales; scales in depression surrounding interparietal distinctly but variably enlarged, posterior and posterolater- al ones grading abruptly into dorsals and supra- temporals; interparietal smaller than ear, sepa- rated from supraorbital semicircles by 3-6 (4) scales. Suboculars separated from supralabials by one row of scales or narrowly in contact; 8- 11 (11) supralabials from rostral to middle ofeye (Fig. 2). Mental semidivided, wider than long, posteriorly in contact in an approximately trans- verse line with 5-8 (6) scales between infrala- bials; sublabials not clearly differentiated; me- dian throat scales small, swollen, grading into 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 females extending just pos- terior to level of axilla. Two to four middorsal rows of body scales slightly enlarged, keeled, swollen, subimbricate; lateral granules swollen, pointed, juxtaposed; ventrals larger than dorsals, imbricate or sub- imbricate, smooth, tending to be in transverse rows. Some larger scales on limbs multicarinate; scales on dorsum of hand large, multicarinate; supradigital scales multicarinate; digital expan- sions narrow; 21-24 (22) lamellae under pha- langes II and III of fourth toe. Tail compressed but without dorsal crest; ver- ticils not distinct; two enlarged middorsal rows of scales; postanals weakly enlarged in males, sometimes not evident. Color in Preservative.— Dorsum pale brown with broad dark brown middorsal blotches con- fluent 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 pale brown with broad brown blotches becoming indistinct posteriorly. Venter dull tan, flecked or not with dark brown; throat dark with lighter spots or transverse streaks. Male dewlap dark brown with tan scales. Female dewlap blotched black on brown. In both sexes a more or less conspicuous complex light, often white, blotch containing black spots or oblique streaking just above dewlap. Color in Life (See Frontispiece).— Dorsum ol- ive-green to tan with dark brown markings, with or without yellowish tan flecks and/or round spots laterally; often a tan vertebral stripe in females; venter brown to yellowish green; male dewlap dark brown with yellowish tan to yellowish green scales; female dewlap mottled or spotted; iris dull 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. cality: KU 1 78960, <5: "Green with reddish-brown markings. Hint of yellow along lateral surfaces. 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 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 green with dark brown transverse markings. Venter pale brown with dark brown flecks. Dew- lap 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 flecks proximally and orange bars distally. Iris pale blue." (Field notes, W. E. Duellman). Measurements ofHolotype (mm). — Snout-vent 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 elevations of 1410-1930 m on the eastern slope of the Andes (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- azon Basin in northern Provincia Napo (Cordi- llera del Due, 1 150 m; Bermejo, 720 m; Puerto 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, Ec- uador and adjacent Departamento de Putumayo. Colombia, are included in the type series. Spec- imens from farther south (Provincias Pastaza. Tungurahua, and Morona-Santiago. Ecuador, and Departamento Amazonas, Peru) obviously are close to A. fitchi. However, these specimens ap- pear to differ in coloration and may represent more than one taxon. The present samples are 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 261 Table 1. Scale characters in the Anolis eulaemus subgroup. 262 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Table 2. Dewlap and body pattern differences in the eulaemus subgroup (pattern differences in preservative emphasized because color in life of eulaemus is unknown). Dewlap 95 A. eulaemus A. vent ri'macular us A. gemmosus A.fitchi A. eulaemus A. ventrimaculatus A. gemmosus Large, skin dull, scales lighter, minute, in multiple lines (5-6) separated by naked skin. Large, skin dark or light, scales light, large, in single or double lines sepa- rated by naked skin. Large, skin dark at base or all light, scales light, large, in multiple lines (3-4) separated by naked skin. Large, skin dark, scales light, large, in single or double rows separated by naked skin. BODY PATTERN Side of Neck Pattern dull, vague; a poorly bounded black blotch just above the dewlap. A narrow light line from labials arching over the upper margin of the ear and continuing to the shoulder; above this a black blotch. A light line from labials not arching above ear, no black blotch above it. Rudimentary, represented by folds of skin that are emphasized by inter- vening dark pigment. None. None. Moderate, skin mottled or spotted, scales light, large, in single or double rows separated by naked skin. 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 Uniform dark, rarely some vague light spots. Faint bluish tinge on side of neck, no well-defined pattern. A.fitchi A. eulaemus A. ventrimaculatus A. gemmosus A. tiic/11 A. eulaemus A. ventrimaculatus A light blotch containing black spots or oblique streaks contiguous with the base of the dewlap. Throat Gray, lighter lateralh. Llniform dark or very weakly vermicu- late (juveniles may show bold ver- miculation). Nearly uniform, at most shades of pur- ple (juveniles may have dark spots on a light ground). Some light streaks or spots on a light ground. Dorsum "Purplish brown above with rather in- distinct transverse bars on back and large round lighter spots on sides." Boulenger 1908. Middorsum uniform dark or crossed by narrow dark bars containing light spots. Flanks boldly spotted with lighter. A black blotch in front of shoulder just behind a light blotch containing black spots. Uniformly dark. Boldly vermiculate, dark on light. Weakly to strongly vermiculate, dark on light. Light marks or spots on dark back- ground. Oblique narrow white bars with some- what irregular margins meeting in forward pointing angles middorsally and separated by wide areas of uni- form brown. A longitudinal dorsal light zone bound- ed by dorsolateral lightbands or a narrow middorsal light line or a se- ries of middorsal multiply-shaped fig- ures. Flanks with strong to weak or absent spotting or vermiculation. VERTEBRATE ECOLOGY AND SYSTEMATICS 263 Table 2. Continued. Dewlap <3<5 99 A. gemmosus A. fitchi A. eulaemus A. ventrimaculatus A. gemmosus A. fitchi Dark transverse bands widest dorsally separated by areas with bold and ir- regular spotting. Transverse bands ta- pering on flanks which become en- tirely spotted or pale dorsum and flanks nearly uniform. Broad dark transverse bands dorsally. Flanks more or less boldly spotted or vermiculate. Venter Venter obscurely but densely vermicu- late. dark on light. Densely and finely spotted (juveniles also vermiculate). Belly with spotting encroaching from flanks or nearlv uniform bluish. Belly darkish, edge invaded by dark spots. A middorsal light zone with dark mar- gins or a narrow middorsal light line or short transverse bands not extend- ing onto flanks. Flanks patternless or with some dark spotting. A middorsal light zone with dark mar- gins or dark transverse bands nar- rowed in center (butterfly pattern). Flanks may be obscurely vermiculate or spotted with darker. Sides of venter pepper and salt becom- ing more uniform brown in center. Light with weak dark vermiculations or spotting. Venter immaculate or more or less densely but obscurely vermiculate and spotted. Belly with dark spots or markings, most prominent laterally. 1 All comments on the female of eulaemus are based only on AMNH 1 18980. Anolis eulaemus was described from a unique male type (BMNH 1946.8.13.31) from Pavas (near San Antonio). Departamento Valle, Co- lombia, and A. ventrimaculatus from two syn- types— an adult female and a juvenile from the Rio San Juan, Intendencia Choco, Colombia. We here designate the adult female syntype (BMNH 1 946.8. 13.5) as the lectotype; it is uncertain that the faded juvenile is the same species. Recently collected material from cloud forests in the Departamento Valle, Colombia, including material from the vicinity of San Antonio, pro- vides an excellent match for the female syntype of A. ventrimaculatus, which lacks any trace of a dewlap. Males from the vicinity of Lago Cal- ima, Departamento Valle, recently collected in numbers along with females of A. ventrimacu- latus, agree with these females and not with the type ofA. eulaemus in the small size of the scales surrounding the interparietal and differ sharply from A. eulaemus in the squamation of the dew- lap. The type of A. eulaemus has minute lateral 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. ventri- maculatus, the lateral scales are large, as in A. fitchi, larger than the ventrals, and in single series 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 at Pueblo Rico, Risaralda (formerly Caldas). This speci- men makes more plausible Boulenger's inexact locality "Rio San Juan" for the syntypes of A. ventrimaculatus. Pueblo Rico is near the source of the Rio San Juan, and M. G. Palmer, who collected both syntypes ofA. ventrimaculatus and the type ofA. eulaemus, is reported by Boulenger (191 1), in the same paper in which he described A. ventrimaculatus, to have obtained Lepto- gnathus (=Dipsas) sancti-joannis at "Pueblo Rico, slopes ofSan Juan River, Colombian Choco. 5200 feet." It is possible that Niceforo Maria's spec- imen is topotypic or near-topotypic, but Boulen- ger'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 collected recently near the type locality. A single male (AMNH 1 1 0495) was collected at Lago Cal- 264 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY 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 Pen- as Blancas (where A. ventrimaculatus also oc- curs), one from the "Farallones de Cali, Pi- chinde," and one from "region alta cerca al Lago Calima." The probable female of the species is represented by AMNH 118980 from "moun- tains above the north side of Lago Calima ( 1 700 m)."- All of these localities are on the Pacific versant of the Cordillera Occidental in Depar- tamento 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 A. 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 with a dark blotch at its base. Anolis gemmosus from the western slope of the Andes in Ecuador, the third previously de- scribed member of the eulaemus subgroup of the aequatoriahs species group has not previously been associated with this group. Williams ( 1976) placed it in the punctatus group in error. The digital dilations of gemmosus are narrow as in the aequatoriahs group, not wide as in the punc- tatus series (cf. the key in Williams 1976). How- ever, A. gemmosus (maximum ru balsamea U4 religiosa f?0 Acacia. 220. 223 cvmbispina 223 farnesiana '"" Acer ->nd pensylvanicum -U4 rubrum 205.206 saccharum 204 Achaearanea tepidariorum 1 66 Acheta assimilis 166 Acris 89, 90. 92. 93. 98. 100. 101. 126 crepitans 89. 90. 92. 93. 95-98. 101. 150. 166 gryllus 89 acntus, Crocodylus — ° adamanteus, Crotalus 199 Adelophisfoxi 239 aemula, Sonora -32, 240 aeneus, Oxybelis 228 aeneus auratus, Oxybelis 240 aequatonalis. Anolis 257. 260, 261, 264 aestivus, Opheodrys 105-1 12. 148, 152. 154, 155 affinis. Pituophis melanoleucus 240 Agave 220 Agkistrodon l )4 bilineatus bilineatus 24 1 contortrix 36, 58, 153. 164. 166-168. 198 198 piscivorus iyo piscivorus leucostoma 1 99 albiventris, Sceloporus horridus 238 Alligator mississipiensis 84 Allolobophora caliginosa 1 66 Alnus 220 204 rugosa -w alpestris. Cladina 205. 206 alterna, Lampwpeltis mexicana 199 Amblyrhynchus 1 2 1 119 cnstatus Ambystoma rosaceum. 235 235 tigrinum.... ^JJ americana. Ulmus 89 americanus Bufo 126 Fraxinus °v amoenus, Carphophis 148 _ andianus, Anolis 265 Andropogon hallii 43 scoparius 4 - angustifolia. Pnmus 4 3 Annona 223 squamosa ^- J annulata, I ermicella ' annulatus, Scaphiodontophis 185-192 annulifera, Tropidodipsas 241 Anolis. 156.185.264 aequatonalis 257.260.261.264 andianus 265 carolinensis ' 48 eulaemus 257. 261-264 fasciatus 261 fitchi fronlispiece, 257-263, 265 gemmosus • 260-265 mints 26 1 nebulosus 231. 237 parilis 260, 261 punctatus utowanae 229, 237 ventrimaculatus 257, 261-265 anomala. Simoselaps 173. 174. 176. 180 antonii, Rhinocheilus lecontei 240 Aparallactus capensis approximans Holbrookia 238 Simoselaps 173. 174. 176. 180 Aprasia ' 76 - ' 78 repens Aquilegia 220 aquilinum. Ptendium 204. 205 arborea, Bocconia 223 arborescens, Ipomoea 223 Arbutus 220 glandulosa 223 Arctostaphvlos 220 arenicolor, Hyla 230. 236 Aristida Arizona elegans 1 52. 230-232 elegans expolita 239 elegans noctivaga 239 arnyi. Diadophis punctatus 52 Aspidelaps r- 199 aspis. I ipera assimilis. Acheta atricaudatus, Crotalus horridus 199 atrox, Crotalus 58 , 197, 199 attenuatus, Ophisaurus... 164-166. 168. 170. 171 augusti, Hylactophryne 230.231 august i cactorum, Hylactophryne.... augusti latrans. Hylactophryne 236 aurantiacum, Hieracium 204, 205 auratus, Oxybelis aeneus 240 Austrelaps superbus 1 7 8 australis, Simoselaps 173. 174. 176 177. 179. 180. 182 austnaca, Coronella 1" avenaceum, Sorghastrum B Elaphe obsoleta '" Sa/vadora 233.240 bairdii, Pcromyscus maniculatus .... 203. -1- balsamea. Abies 204 Bansia imbricata alians -39 basiliscus, Crotalus basiliscus .... 241 basi/iscus basiliscus, Crotalus.... - 24^ baudim. Smilisca 228. 23 267 268 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Begonia 223 belli Eumeces lynxe 239 Pseudoeurycea 228 benthamiana, Tillandsia 223 berlandieri, Rana 229 bertholdi, Simoselaps 173-182 berus, Vipera 58, 60, 68, 195, 199, 200 Betula cerulea 205, 206 bicarinatus tuberculatus, Urosaurus 238 bilineatus Agkistrodon bilineatus 24 1 Eumeces brevirostris 238 Masticophis 240 bilineatus bilineatus, Agkistrodon 24 1 bimaculatus, Neelaps.... 173, 174, 176, 177, 179, 180 biscustatus, Trimorphodon biscutatus 24 1 biscutatus biscutatus, Trimorphodon 24 1 bistincta, Hyla 232, 236 Blattella germanica „ 138 Boa constrictor 228 constrictor imperator 239 Bocconia arborea 223 bocourti, Tantilla 228 bogerti, Callisaurus draconoides 237 Boleosoma lepida 243 Bombina 124, 125, 127, 128 Bothrops godmani 198 boulengeri, Sceloporus clarki 238 Bouteloua 220 brachycercum, Phrynosoma douglassi 238 Brachylophus fasciatus 1 1 9 bradti, Phrynosoma orbiculare 238 braminus, Ramphotyphlops 228 brevirostris, Eumeces 228, 230 brevirostris bilineatus, Eumeces 238 Brosimum 223 browni, Phyllorhynchus 232, 240 Bubo virginianus 2 1 2 Bufo 1 24- 1 28 americanus 126 cognatus 236 compactHis 236 debilis insidior 236 kelloggi 228, 232, 236 marinus 123, 125, 228, 236 marmoreus 228, 236 mazatlanensis 236 microscaphus mexicanus 236 occidentalis 230, 232, 236 punctatus 231, 232, 236 woodhousei 1 26, 1 28 bulleri, Sceloporus 232, 238 Bungarus caeruleus 1 5 1 Bursera 223 Buteo jamaicensis 2 1 2 butleri. Thamnophis 36 Cacophis 177, 178 harrietae 178 squamulosus 178 cactorum, Hylactophryne augusti 236 caeruleus, Bungarus 1 5 1 Caesalpinia 223 calamarina, Tantilla 240 caliginosa, Allolobophora 1 66 callicephalus, Eumeces 232, 238 calligaster Lampropeltis... 152, 164-166, 168, 169, 171, 197 La mpropeltis calligaster 1 9 7 calligaster calligaster, Lampropeltis 197 calligaster rhombomaculata, Lampropeltis 197 Callisaurus 137, 237 draconoides 232, 238 draconoides bogerti 237 calonotus, Neelaps 173, 174, 176, 179, 180 calva, Tithonia 223 cana, Pseudaspis 199 canescens, Thamnophis melanogaster 24 1 caninus, Dermestes 138 capensis, Aparallactus 5 7 carinata, Cyclura 117, 119 carolinensis, A nolis 1 48 Carphophis amoenus 148 vermis 46, 52, 1 1 1 carpicinctus, Scaphiodontophis 185 Cassia 223 catenatus, Sistrurus 58, 164, 166-168, 197 catenifer, Pituophis melanoleucus 41,51 catesbeiana, Rana 123-129, 131, 166, 167 catesbvi, Dipsas 189, 190 Ceiba 223 cembroides, Pinus 220 Cemophora 180 coccinea 179 cenchoa, Imantodes 189, 190 Cephalanthus 220 occidentalis 43 cerastes, Crotalus 199 Cercocarpus macrophyllus 223 cerulea, Betula 205, 206 Chamaedorea 223 Chaulognathus pennsylvanicus 138 Chelydra serpentina 98 cherriei, Sphenomorphus 1 92 Chilomeniscus 173, 179, 181 Chionactis 173,179,181 chiricahuensis, Rana 229 Chrvsemvs 80 picta.. 80, 83, 85 chrysoscelis, Hyla 1 66 ciliaris, Barisia imbricata 239 Cladma alpestris 205, 206 rangiferina 205, 206 Cladonia mitis 205, 206 clamitans, Rana 124-129, 131 clarki, Sceloporus 228, 232 clarki boulengeri, Sceloporus 238 Clelia clelia 1 86 clelia, Clelia 186 Clethrionomys gapperi 2 1 2 cliftoni, Dryadophis 231-233, 239 climacophora, Elaphe 51, 52 VERTEBRATE ECOLOGY AND SYSTEMATICS 269 Cnemidophorus 137, 141. 142, 144. 156 costatus 239 costatus huico 239 costatus mazatlanensis 239 exsanguis 137-144 gularis 1 5 2 murinus 141, 142, 144 scalaris scalaris 239 tigris 190 coccinea, Cemophora 1 79 Cocos nucifera 223 cognatus, Bufo 236 Coleonyx variegatus 232 variegatus fasciatus 237 colimensis, Eumeces 232, 238 collaris, Thamnophis cvrtopsis 230, 240 Coluber 14-16, 18, 20, 22, 24, 25. 27, 29-33, 51 constrictor. 13, 35-37, 58-60, 148, 152, 164- 166, 168, 169 constrictor flaviventris 13, 33-38, 52 constrictor mormon 13, 15-20, 22, 24-38, 51, 52 constrictor oaxaca 229 viridiflavus 199 commune, Polytrichium 205 compactilis, Bufo 236 Coniophanes fissidens 189 lateritius 228 lateritius lateritius 239 Conolophus 121 subcristatus 1 19 Conopsis nasus 230 nasus nasus 239 constrictor Boa 228 Coluber 13, 35-37, 58-60. 148. 152, 164-166, 168, 169 constrictor flaviventris. Coluber 13, 33-38, 52 constrictor imperator, Boa 239 constrictor mormon. Coluber 13, 15-20, 22, 24-38, 51, 52 constrictor oaxaca. Coluber 229 contortrix, Agkistrodon 36, 58, 153, 164, 166-168, 198 corais rubidus, Drymarchon 239 coronata, Tantilla 1 5 2 Coronella austriaca 1 99 coronoides, Drysdalia 1 78 coronuta stejnegeri, Cyclura 117, 119 costatus, Cnemidophorus 239 costatus huico, Cnemidophorus 239 costatus mazatlanensis, Cnemidophorus 239 couchi, Scaphiopus 231, 232, 235 crepitans, Acris 89, 90, 92, 93, 95-98, 101, 150, 166 cristatus, Amblyrhynchus 1 19 Crocodylus acutus 228 Crotalus 33, 196 adamanteus 199 atrox 58, 197, 199 basiliscus basiliscus 24 1 cerastes 199 horridus 57, 58,61 horridus atricaudatus 1 99 lepidus 229, 230 lepidus klauberi 199, 241 lepidus maculosus 24 1 molossus 197, 198, 228, 230 molossus nigrescens 24 1 pricei 230 pricei pricei 241 ruber 1 99 scutulatus scutulatus 24 1 stejnegeri 241 triseriatus 198 viridis 36, 46, 57, 58, 61, 68, 164, 167, 168, 196-198, 200 viridis helleri 41 viridis lutosus 30 viridis oreganus 4! viridis viridis 1 98, 1 99 willardi meridionalis 24 1 Ctenosaura 1 1 7, 237 pectinata 228, 237 similis 117, 119. 121 Ctenotus 1 76, 1 78 cucullatus, Stegonotus 1 78 cyclocarpum, Enterolobium 223 Cyclura 117, 121 carinata 117, 119 coronuta stejnegeri 117, 119 cymbispina, Acacia 223 cvrtopsis Thamnophis 230 Thamnophis cvrtopsis 240 cvrtopsis collaris, Thamnophis 230. 240 cvrtopsis cvrtopsis, Thamnophis 240 cvrtopsis pulchrilatus, Thamnophis 230, 240 D dacnicolor, Pachymedusa 228, 236 dactyloides, Tripsacum 43 Dasylirion 220 debilis insidior, Bufo 236 decipiens, Rhadinaea 189, 190 decurtatus. Phyllorhynchus 228 dekavi, Storeria 58, 148. 149, 152-155, 163-166. 168. 169 delicatissima. Iguana 1 1 9 Demansia 178 olivacea 178 psammophis 178 reticulata 57 textilis 199 Dendrophidion dendrophis 189, 190 deppei Pituophis 228, 230, 232, 233 Pituophis deppei 240 deppei deppei, Pituophis 240 dendrophis, Dendrophidion 189. 190 Dermestes caninus 138 Deschampsia jlexuosa 205, 206 270 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY deserticola Pituophis melanoleucus 30, 41, 44, 48. 49, 51, 52 Salvadora 232, 240 Diadophis 154 punctatus.... 34,46, 111, 148, 152, 163-166, 168, 169, 230, 233, 239 punctatus arnyi 52 punctatus dugesi 239 punctatus regalis 239 differentialis, Melanoplus 1 66 diplotropis, Leptophis 240 Dipsas catesbyi 1 89, 1 90 sancti-joannis 263 Dipsosaurus 1 17 distans, Micrurus distans 241 distans distans, Micrurus 241 divaricata, Lysiloma 223 Do/omedes sexpunctatus 98 douglassi brachycercum, Phrynosoma 238 draconoides, Callisaurus 232, 238 draconoides bogerti, Callisaurus 237 Drvadophis cliftoni 231, 232, 233, 239 melanolomus 232 melanolomus stuarti 239 Drymarchon corais rubidus 239 Drymobius margaritiferus ftstulosus 239 Drysdalia 177 coronoides 1 7 8 dugesi Diadophis punctatus 239 Geophis 231 Geophis dugesi 239 Leptotyphlops humilis 239 dugesi dugesi, Geophis 239 dulcis, Leptotyphlops 148, 150, 152 durangensis, Pinus 220 E Elaphe 51, 154, 170 climacophora 51,52 guttata 152 obsoleta 51, 52, 58, 148, 152 obsoleta bairdi 1 99 obsoleta obsoleta 1 99 quadrivirgata 36, 51, 52 triaspis 231 triaspis intermedia 239 vulpina 169 elapoides, Pliocercus 190 Elapsoidea 1 7 3 sundevalli 179 elegans Arizona 1 52, 230-232 Holbrookia 232, 238 Holbrookia elegans 238 Lerista 176 Thamnophis 58, 60, 69, 71, 72, 167 elegans elegans, Holbrookia 238 elegans errans, Thamnophis 240 elegans expolita, Arizona 239 elegans noctivaga, Arizona 239 Eleocharis 220 Eleutherodactylus 260 hobartsmithi 232, 235 occidentalis 235 peruvianus 260 vocalis 236 Elgaria kingi ferruginea 239 Enhydrina 153 Enterolobium cyclocarpum 223 ephippiata, Leptodeira splendida 240 eques, Thamnophis 230 eques megalops, Thamnophis 240 eques virgatenuis, Thamnophis 240 errans, Thamnophis elegans 240 erythrogasier, Nerodia 1 48 Etheostoma 243 grahami 243-245, 247-254 lepidogenys 244 lepidum 243-254 lepidum lepidogenys 244 pottsi 245 spectabile 244-246, 25 1 spectabile pulchellum 245 eulaemus, Anolis 257, 261-264 Eumeces 152, 154, 163, 170, 171, 185 brevirostris 228, 230 brevirostris bilineatus 238 callicephalus 232, 238 colimensis 232, 238 fasciatus 148, 149, 151, 152, 154, 164-166, 168, 170 inexpectatus 1 52, 1 70 lynxe belli 239 parvulus 228 tetragrammus 148, 152, 154 euryxanthus, Micruroides 189, 232 euryxanthus neglectus, Micruroides 241 eximia, Hyla 236 expolita, Arizona elegans 239 exsanguis, Cnemidophorus 1 37- 1 44 exserta, Tillandsia 223 extenuatum, Stilosoma 1 52 Fagus grandifolia 204, 205 Farancia abacura 1 52 farnesiana. Acacia 220 fasciatus A nolis 26 1 Brachylophus 1 19 Coleonvx variegatus 237 Eumeces 148, 149, 151, 152, 154, 164-166, 168, 170 fasciolatus, Simoselaps 173, 174, 176, 179, 180, 182 ferruginea, Elgaria kingi 239 Festuca rubra 205 Ficimia 154, 173, 179, 181 olivacea 152 Ficus 223 ftssidens, Coniophanes 1 89 ftstulosus, Drymobius margaritiferus 239 VERTEBRATE ECOLOGY AND SYSTEMATICS 271 fitchi, Anolis frontispiece, 257-263, 265 flagellum, Masticophis Jlagellum 1 99 Jlagellum flagellum, Masticophis 199 Jlagellum lineatulus, Masticophis 240 Jlavescens, Perognathus 43 Jlavilata, Rhadinaea 1 89 flaviventris, Coluber constrictor 13, 33-38, 52 Jlexuosa, Deschampsia 205, 206 Jloridana, Rhineura 152 floridanus, Sylvilagus 43 fodiens, Pternohyla 237 forreri, Rana 229 foxi, Adelophis 239 Fraxinus americanus 89 frontalis Micrurus 153 Pseudoficimia 240 fulva, Quercus 223 fulvius, Micrurus 147, 150-158, 189 Furina 177, 178 Galium 204 gapperi, Clethrionomys 2 1 2 Gastrophryne olivacea 232, 237 usta 232,237 Gastrotheca 265 Gaultheria procumbens 204 Geagras redimitus 229 Gehyra mutilata 228, 229 gemmistratus latistratus, Imantodes 239 gemmosus, Anolis 260-265 Geophis dugesi 231 dugesi dugesi 239 germanica, Blaltella 138 Gerrhonotus 163 liocephalus liocephalus 239 getulus, Lampropeltis 163, 169 getulus holbrooki, Lampropeltis 197-200 getulus splendida, Lampropeltis 239 gibbosus, Lepomis 1 3 1 girardi. Masticophis taeniatus 240 glandulosa. Arbutus 223 Gleditsia triacanthos 89 godmani, Bothrops 198 gracilis Peromvscus maniculatus 203-205, 210-213 Tantilla 148, 152 grahami, Etheostoma 243-245, 247-254 grahamiae, Salvadora 152 grahamiae lineata, Salvadora 240 grammicus microlepidotus, Sceloporus 238 grandifolia, Fagus 204, 205 groenlandicum. Ledum 204 gryllus, Acris 89 Guazuma ulmifolia 223 gularis, Cnemidophorus 152 guttata, Elaphe 1 5 2 Gyalopion 173, 179, 181,232 quadrangularis 229, 232, 239 Gymnophthalmus 185 H hallii, Andropogon 43 Haematoxylum 223 hammondi, Scaphiopus 130 harrietae, Cacophis 178 hawni, Physa 166 helleri, Crotalus viridis 4 1 Heloderma horridum horridum 239 hespena, Rhadinaea 232, 240 Heterodon 179 nasicus 52 nasicus kennerlyi 239 platyrhinos 52, 148 heterolepis, Sceloporus 232 heterolepis shannonorum, Sceloporus 238 Hieracium aurantiacum 204, 205 hirtipes murrayi, Kinosternon 237 hispidus Sauromalus 1 19 Sigmodon 43 hobartsmithi, Eleutherodactylus 232, 235 holbrooki, Lampropeltis getulus 197-200 Holbrookia 137 approximans 238 elegans 232, 238 elegans elegans 238 horridum, Heloderma horridum 239 horridum horridum, Heloderma 239 horridus, Crotalus 57, 58, 61 horridus albiventris, Sceloporus 238 horridus atricaudatus, Crotalus 1 99 huico, Cnemidophorus costatus 239 humilis dugesi, Leptotyphlops 239 Hura polyandra 223 Hvdrophis 153 Hyla 125 arenicolor 230, 236 bistincta 232, 236 chrysoscelis 166 eximia 236 smaragdina 232, 236 smithi 236 Hylactophryne august i 230,231 augusti cactorum 236 august i latrans 236 tarahumaraensis 228, 230, 236 Hypopachus oxyrrhinus oxyrrhinus 237 variolosus 237 Hypsiglena torquata 231, 239 I Iguana 1 17 delicatissima 1 19 iguana 115-121, 238 iguana. Iguana 115-121, 238 Imantodes cenchoa 189, 1 90 gemmistratus latistratus 239 imbricata ciliaris, Barisia 239 imperator. Boa constrictor 239 272 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY incinctus, Simoselaps 173, 174 inexpectatus, Eumeces 152, 170 insidior, Bufo debilis 236 integrum, Kinosternon 230, 232, 237 intermedia, Elaphe triaspis 239 interorbitalis, Syrrhophus 228 Ipomoea arborescens 223 jamaicensis, Buteo 212 jarrovi Sceloporus 137-144, 229, 230 Sceloporus jarrovi 238 jarrovi jarrovi, Sceloporus 238 Jatropha 220 Juniperus 220 kelloggi, Bufo 228, 232, 236 kennerlyi, Heterodon nasicus 239 kingi ferruginea, Elgaria 239 Kinosternon hirtipes murrayi 237 integrum 230, 232, 237 klauberi, Crotalus lepidus 199, 241 laevis, Xenopus 1 2 7 lambda paucimaculata, Trimorphodon 24 1 Lampropeltis 51, 154, 196 calligaster 152, 164-166, 168, 169, 171, 197 calligaster calligaster 1 9 7 calligaster rhombomaculata 1 9 7 getulus 163, 169 getulus holbrooki 1 97-200 getulus splendida 239 mexicana 239 mexicana alterna 1 99 pyromelana 199 triangulum 51, 52, 199 triangulum nelsoni 240 triangulum sinaloae 240 lateralis Scincella 148-150, 152, 154, 155, 185, 190 Urosaurus ornatus 229 lateristriga, Rhadinaea 189 lateritius Coniophanes 228 Coniophanes lateritius 239 lateritius lateritius, Coniophanes 239 latistratus, Imantodes gemmistratus 239 latrans, Hylactophryne augusti 236 laureata, Rhadinaea 240 lebentina, Vipera 198 lecontei, Rhinocheilus 232 lecontei antonii, Rhinocheilus 240 lecontei tesselatus, Rhinocheilus 179 Ledum groenlandicum 204 Leiobunum vittatum 166 leiophylla, Pinus 220 lemniscatus, Micrurus 1 53 leonensis, Oligocephalus 244 lepida, Boleosoma 243 Lepidochelys 229 olivacea 229 lepidogenys Etheostoma 244 Etheostoma lepidum 244 lepidum, Etheostoma 243-254 lepidum lepidogenys, Etheostoma 244 lepidus Crotalus 229, 230 Poecilichthys 244 lepidus klauberi, Crotalus 199, 241 lepidus maculosus, Crotalus 241 Lepomis gibbosus 1 3 1 Leptodactylus melanonotus 236 occidentalis 236 Leptodeira maculata 232, 240 punctata 240 septentrionalis polysticta 240 splendida ephippiata 240 Leptognathus sancti-joannis 263 Leptophis diplotropis 240 Leptotyphlops 154 dulcis 148, 150, 152 humilis dugesi 239 Lerista 1 76, 1 78 elegans 176 lineata 176 picturata 176 praepedita 176 leucopus, Peromyscus 43, 167 leucostoma, Agkistrodon piscivorus 199 Limnodynastes 1 2 5 Limonia 236 lineata Lerista 176 Salvadora 179 Salvadora grahamiae 240 lineatum, Tropidoclonion 148, 152 lineatulus, Masticophis flagellum 240 liocephalus, Gerrhonotus liocephalus 239 liocephalus liocephalus, Gerrhonotus 239 lippiens, Sympholis 240 littoralis Malaclemys terrapin 77, 82 Simoselaps 173, 174, 176, 179, 180 lumholtzi, Pinus 222 lutosus, Crotalus viridis 30 lutrensis, Notropis 1 6 6 Lycosa rabida 1 66 lynxe belli, Eumeces 239 Lysiloma divaricata 223 Lytorhynchus 179 M macrolepis, Sceloporus poinsetti 238 macrophylla, Quercus 223 macrophyllus, Cercocarpus 223 maculata, Leptodeira 232, 240 maculosus, Crotalus lepidus 24 1 madagascariensis, Sanzinia 1 99 VERTEBRATE ECOLOGY AND SYSTEMATICS 273 magnaocularis, Rana 229 Magnolia shicdcana 223 Malaclemys 77, 79-85 terrapin 77, 82 terrapin littoralis 77, 82 terrapin pileata 77, 82 terrapin tequesta 77, 81-85 terrapin terrapin.. 77, 82 mangle. Rhizophora 223 maniculatus bairdii, Peromyscus 203, 212 maniculatus gracilis, Peromyscus 203-205, 210-213 marcianus, Thamnophis 1 52 margaritiferus fistulosus, Drymobius 239 mariana, Picea 204 marinus, Bufo 123, 125, 228, 236 marmoreus, Bufo 228, 236 Masticophis 33, 51, 60, 68, 195 bilineatus 240 flagellum flagellum 1 99 flagellum lineatulus 240 mentovarius 231 mentovarius striolatus 240 taeniatus 20, 34, 52, 58, 60, 61, 111 taeniatus girardi 240 taeniatus taeniatus 30 mazatlanesis Bufo 236 Cnemidophorus costatus 239 megacystis, Zeugorchis 93 megalops, Thamnophis eques 240 megalotis, Reithrodontomys 43 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 melanolomus, Dryadophis 232 melanolomus stuarti, Dryadophis 239 melanonotus, Leptodactylus 236 Melanoplus differentialis 166 Menetia 176, 178 mentovarius, Masticophis 23 1 mentovarius striolatus, Masticophis 240 meridionalis, Crotalus willardi 241 mexicana, Lampropeltis 239 mexicana alterna, Lampropeltis 199 mexicanus, Bufo microscaphus 236 microlepidotus, Sceloporus grammicus 238 Micropterus salmoides 98 microscaphus mexicanus, Bufo 236 Microtus 43 ochrogaster 43, 1 67 pennsylvanicus 212 Micruroides euryxanthus 1 89, 232 euryxanthus neglectus 24 1 Micrurus 151, 153. 156, 158 distans distans 241 frontalis 153 fulvius 147, 150-158. 189 lemniscatus 153 nigrocinclus 188 miliarias, Sistrurus 1 97, 1 99 Mimosa 223 minima, Simoselaps 173, 174, 176 minis, A nolis 26 1 mississipiensis, Alligator 84 mitis, Cladonia 205, 206 modeslus, Syrrhophus 236 molilor, Tenebrio 1 38, 1 66 molossus, Crotalus 197, 198, 228, 230 molossus nigrescens, Crotalus 241 molurus. Python 195, 199, 200 Morethia 1 76, 1 78 mormon. Coluber constrictor 1 3, 1 5-20, 22, 24-38, 51. 52 mucosus, Ptyas 1 99 mucronata, Pseudotsuga 220 multiplicand. Scaphiopus 235 murinus, Cnemidophorus 141. 142, 144 murrayi, Kinosternon hirtipes 237 Mus musculus 166, 167 musculus, Mus 166. 167 muticus, Trionyx 85 mutilata, Gehvra 228, 229 N nasicus, Heterodon 52 nasicus kennerlyi, Heterodon 239 nasus Conopsis 230 Conopsis nasus 239 nasus nasus, Conopsis 239 Natrix tigrina 46, 52 nebulosus, Anolis 231, 237 Neelaps 173, 174, 176-182 bimaculatus 173, 174. 176, 177, 179. 180 calonotus 173, 174. 176, 179. 180 neglectus, Micruroides euryxanthus 241 nelsoni Lampropeltis triangulum 240 Sceloporus 228, 238 Terrapene 228 nemoralis, Poa 204 Neoseps revnoldsi 152 Nerodia 51. 150. 154. 170 erythrogaster 148 rhombifera 148 sipedon 34-36. 51. 52. 98. 164, 166-169. 171 valida valida 240 nigra, Salix 43 nigrescens, Crotalus molossus 241 nigriceps, Tantilla 148 nigrocinclus, Micrurus 1 88 nigronuchalis, Thamnophis 24 1 nitidus, Tomodactylus 232 nitidus petersi, Tomodactylus 236 noctivaga, Arizona elegans 239 Norops 257. 261 Notechis scutatus 1 1 1 nothus, Scaphiodontophis zeteki 1 90 Notropis lutrensis 1 6 6 nucifera, Cocos 223 Nymphaea 220 274 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY O oaxaca, Coluber constrictor 229 obesus, Sauromalus 117, 119 obsoleta Elaphe 51, 52, 58, 148, 152 Elaphe obsoleta 1 99 obsoleta bairdi, Elaphe 199 obsoleta obsoleta, Elaphe 199 occidentalis Bufo 230, 232, 236 Cephalanthus 43 Eleutherodactylus 235 Leptodactvlus 236 Thuja 204 occipitomaculata, Storeria 58, 152 ochrogaster, Microtus 43, 167 Oligocephalus 243, 245, 246 leonensis 244 Oligodon 179-181 olivacea Demansia 1 7 8 Ficimia 152 Gastrophryne 232, 237 Lepidochelys 229 oocarpa, Pinus 222 Opheodrvs aestivus 105-112, 148, 152, 154, 155 vernalis 41, 109, 110 Ophisaurus 152 attenuatus 164-166, 168, 170, 171 Opuntia 220 orbiculare bradti, Phrynosoma 238 oreganus, Crotalus viridis 41 Oreopanax peltatum 223 ornata, Pseudemys scripta 237 ornatus lateralis, Urosaurus 229 Ostrya virginiana 223 Oxybelis aeneus 228 aeneus auratus 240 oxyrrhinus, Hypopachus oxyrrhinus 237 oxyrrhinus oxyrrhinus, Hypopachus 237 Pachycereus pecten-arboriginum 223 Pachymedusa dacnicolor 228, 236 Panicum 43 virgatum 43 parietalis, Thamnophis sirtalis 52 parilis, Anolis 260, 26 1 parvulus, Eumeces 228 paucimaculata, Tnmorphodon lambda 24 1 pecten-arboriginum, Pachycereus 223 pectinata Ctenosaura 228, 237 Spartina 43 Pelamis 153 platurus 228 Pelobates 124, l>b, i28 peltatum, Oreopanax 223 pennsylvanicus Chaulognathus 1 38 Microtus 2 1 2 pensylvanicum, Acer 204 Perognathus jlavescens 43 Peromyscus 127, 209 leucopus 43, 167 maniculatus bairdii 203, 212 maniculatus gracilis 203-205, 2 1 0-2 1 3 peruvianas, Eleutherodactylus 260 petersi, Tomodactylus nitidus 236 philippi, Tropidodipsas 24 1 Phrynohyas venulosa 229 Phrynosoma douglassi brachycercum 238 orbiculare bradti 238 Phvllodactylus tuberculosus saxatilis 237 Phyllophaga 1 3 8 Phvllorhynchus 179 browni 232, 240 decurtatus 228 Physa hawni 1 66 Picea mariana 204 picta, Chrysemys 80, 83, 85 picturata, Lerista 1 7 6 pileata, Malaclemys terrapin 77, 82 Pinus cembroides 220 durangensis 220 leiophylla 220 lumholtzi 222 oocarpa 222 strobiformis 220 strobus 205, 206 teocote 222 pipiens, Rana 123, 126, 128, 166, 167, 229 piscivorus, Agkistrodon 1 98 piscivorus leucostoma, Agkistrodon 199 Pithecollobium sonorae 223 Pituophis 33, 5 1 deppei 228, 230, 232, 233 deppei deppei 240 melanoleucus 34, 43-46, 48, 49, 58, 232 melanoleucus afflnis 240 melanoleucus catenifer 41,51 melanoleucus deserticola 30, 41, 44, 48, 49, 51, 52 melanoleucus sayi 41, 44, 48, 50, 52, 54 plamceps, Tantilla 1 52 platurus, Pelamis 228 platyrhinos, Heterodon 52, 148 Pliocercus 190 elapoides 190 Poa nemoralis 204 Poecilichthys lepidus 244 poinsetti macrolepis, Sceloporus 238 polyandra, Hura 223 Polygonum 220 polysticta, Leptodeira septentrionalis 240 Polytrichum commune 205 Populus 89 porphyriacus, Pseudechis 57, 111, 178, 199 pottsi, Etheostoma 245 praepedita, Lerista 1 76 pretiosa, Rana 97 pricei Crotalus 230 Crotalus pricei 24 1 VERTEBRATE ECOLOGY AND SYSTEMATICS 275 pricei pricei, Crotalus 24 1 procumbens, Gaulthcria 204 Prosopis 220 Prosymna 1 73, 1 79-1 8 1 proximus, Thamnophis 52, 98, 148, 152 Prunus angustifolia 43 psammophis, Demansia 1 78 Pseudaspis cana 1 99 Pseudechis porphyriacus 57, 111, 178, 199 Pseudemys 80 scripta 79 scripta ornata 237 Pseudoeurycea belli 228 Pseudoficimia frontalis 240 Pseudonaja textHis 178 Pseudothelphusa 223 Pseudotsuga mucronata 220 Psidium 223 Psittacanthus 223 Pteridium 223 aquilinum 204, 205 Pternohyla fodiens 237 Ptyas mucosus 199 pubens, Sambucus 204 pulchellum, Etheosloma spectabile 245 pulcherrima rogerbarbouri, Rhinoclemmys 237 pulchrilatus, Thamnophis cyrtopsis 230, 240 punctata, Leptodeira 240 punctatus Anolis 264 Bufo 231, 232, 236 Diadophis 34,46, 111, 148, 152, 163-166. 168, 169, 230. 233, 239 punctatus aryni, Diadophis 52 punctatus dugesi, Diadophis 239 punctatus regalis, Diadophis 239 pustulosa, Rana 237 pygaea, Seminatrix 152 pygmaea, Umbra 1 3 1 pvromelana, Lampropeltis 1 99 Python molurus 195, 199, 200 quadrangularis. Gyalopion 229, 232, 239 quadrivirgata, Elaphe 36, 51, 52 Quercus 223' fulva 223 macrophylla 223 velutina 89 viminea 223 R rabida, Lycosa 166 radix, Thamnophis 41, 164, 166-169 Ramphotvphlops braminus 228 Rana 124-128 berlandieri 229 catesbeiana 123-129. 131. 166. 167 chiricahuensis 229 clamitans 124-129. 131 forreri 229 magnaocularis 229 pipiens 123, 126, 128, 166, 167, 229 pretiosa 97 pustulosa 237 sinaloae 237 sphenocephala 126 tarahumarae 237 temporaria 125, 128, 129 tigrina 126 rangiferina, Cladina 205, 206 redimitus, Geagras 229 regalis, Diadophis punctatus 239 relicta, Tantilla 1 5 2 religiosa, Abies 220 Reithrodontomys megalotis 43 repens, Aprasia 1 76 reticulata, Demansia 5 7 reynoldsi, Neoseps 152 Rhabdophis tigrinus 36 Rhadinaea 1 89. 1 90 decipiens 189 flavilata 1 8 9 hesperia 232, 240 lateristriga 189 laureata 240 Rhineura floridana 1 52 Rhinocheilus lecontei 232 lecontei antonii 240 lecontei tesselatus 1 79 Rhinoclemmys pulcherrima rogerbarbouri 237 Rhizophora mangle 223 rhombifera, Nerodia 1 48 rhombomaculata, Lampropeltis calligasier 197 Rhus terebinthifolia 223 rogerbarbouri, Rhinoclemmys pulcherrima 237 roperi, Simoselaps 173, 174, 176. 178. 180 rosaceum, Ambystoma 235 ruber, Crotalus 199 rubidus, Drymarchon corals 239 rubra Festuca 205 Tantilla 1 5 2 rubrum, Acer 205. 206 Rubus 223 rufipunctatus, Thamnophis 24 1 rugosa, Alnus 204 Rumex 205 saccharum, Acer 204 Salix 89,220 nigra 43 salmoides, Micropterus 98 Salvadora 1 54, 1 79 bairdi 233. 240 desert icc'.a 232, 240 gi ahamiae 1 5 2 grahamiae lineata 240 lineata 179 Sambucus pubens 204 sancti-joannis Dipsas 263 Leptognathus 263 Sanzinia madagascariensis 1 99 sartorii, Tropidodipsas 1 52 sauritus, Thamnophis 52 276 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY Sauromalus 117, 121 hispidus 1 1 9 obesus 117, 119 variits 1 1 9 saxatilis Phyilodactylus tuberculosus 237 Tomodactylus 232, 236 sayi, Pituophis tnelanoleucus .... 41, 44, 48, 50, 52, 54 scalaris Cnemidophorus scalaris 239 Sceloporns 238 scalaris scalaris, Cnemidophorus 239 Scaphwdontophis 185, 187, 189-192 anmdatus 185-192 carpicinctus 1 8 5 venustissimus 185-192 zeteki 185 zeteki nothus 190 Scaphiopus couchi 231, 232, 235 hammondi 130 nudtiplicatus 235 Sceloporus 1 52, 1 56 bulled 232,238 clarki 228,232 clarki boulengeri 238 grammicus microlepidotus 238 heterolepis 232 heterolepis shannonorum 238 horridus albiventris 238 jarrovi 137-144, 229, 230 jarrovi jarrovi 238 nelsoni 228, 238 poinsetti macrolepis 238 scalaris 238 spinosus 232 spinosus spinosus 238 undulatus 112, 152, 167 utiformis 232, 238 Scincella 154 lateralis 148-150, 152, 154, 155, 185, 190 scoparius, Andropogon 43 scripta. Pseudemys 79 scripta ornata, Pseudemys 237 scutatus, Notechis 1 1 1 scutulatus, Crotalus scutulatus 24 1 scutulatus scutulatus, Crotalus 24 1 semiannulata, Sonora 148, 152, 153 senufasciatus, Simoselaps 1 73-1 82 Seminatrix 155 pygaea 152 septentrionalis polysticta, Leptodeira 240 serpentina, Chelydra 98 sexpunctatus, Dolomedes 98 shannonorum, Sceloporus heterolepis 238 shiedeana. Magnolia 223 Sigmodon 43 hispidus 43 similis, Ctenosaura 117, 119, 121 Simophis 188 Simoselaps 173, 174, 176-181 anomala 173, 174, 176, 180 approximans 173, 174, 176, 180 australis 173. 174, 176, 177, 179, 180, 182 bertholdi 173-182 fasciolatus 173, 174, 176, 179, 180, 182 incinctus 173, 1 74 littoralis 173, 174, 176, 179, 180 minima 173, 174, 176 roperi 173, 174, 176, 178, 180 semifasciatus 173-182 warm 173, 174, 176, 177, 180 Simulium 223 sinaloae Lampropeltis triangulum 240 Rana 237 sipedon, Nerodia 34-36, 51, 52, 98, 164, 166-169, 171 sirtalis Thamnophis 58-70, 72, 98, 164-169 Thamnophis sirtalis 52 sirtalis parietalis, Thamnophis 52 sirtalis sirtalis, Thamnophis 52 Sistrurus 196 catenatus 58, 164, 167, 168, 197 miliarius 1 97, 1 99 smaragdina, Hyla 232, 236 Smilisca baudini 228, 237 smithi, Hyla 236 Sonora 154 aemula 232, 240 semiannulata 148, 152, 153 sonorae, Pithecollobium 223 Sorghastrum avenaceum 43 Spartina peclinaia 43 spatulatus, Triprion spatulatus 237 spatulatus spatulatus. Triprion 237 spectabile, Etheostoma 244-246, 251 spectabile pulchellum, Etheostoma 245 Sphaerodactylus torqualus 229 sphenocephala, Rana 1 26 Sphenomorphus 1 8 5 cherriei 192 spinosus Sceloporus 232 Sceloporus spinosus 238 spinosus spinosus, Sceloporus 238 splendida, Lampropeltis getulus 239 splendida ephippiata, Leptodeira 240 squamosa, Annona 223 squamulosus, Cacophis 1 78 Stegonotus 180 cucullatus 178 stejnegeri Crotalus 241 Cyclura coronuta 117, 119 Stevia 223 StUosoma 155 extenuatum 152 Storeria 154, 155 dekavi 58, 148, 149, 152-155, 163-166, 168. 169 occipitomaculata 58, 1 52 storerioides 230, 240 storerioides, Storeria 230, 240 striatula, Virginia 52, 148, 149, 153, 155 striolatus, Masticophis mentovarius 240 strobiformis, Pinus 220 VERTEBRATE ECOLOGY AND SYSTEMATICS 277 strobus, Pinus 205, 206 stuarti, Dryadophis mclanolomus 239 subcristatus, Conolophus I 1 9 sundevalli, Elapsoidea 179 superbus, Austrelaps 178 Syhilagus Jloridanus 43 Sympholis lippiens 240 Syrrhophus mterorbitalis 228 modestus 236 teretistes 232, 236 taeniatus Masticophis 20, 34, 52, 58, 60, 61. 68, 111 Masticophis taeniatus 30 taeniatus girardi, Masticophis 240 taeniatus taeniatus, Masticophis 30 Tantilla 148, 152-155 bocourti 228 calamarina 240 coronata 152 gracilis 148, 152 nigriceps 148 planiceps 152 relicta 152 rubra 152 wilcoxi wilcoxi 240 yaquia 232, 240 tarahumarae, Rana 237 tarahumaraensis, Hylactophryne 228, 230, 236 tau, Trimorphodon 228, 231 temporaria, Rana 125, 128, 129 Tenebrio molitor 138, 166 teocote, Pinus 222 tepidariorum, Achaearanea 1 66 tequesta, Malaclemys terrapin 77, 81-85 terebinthifolia, Rhus 223 teretistes, Syrrhophus 232, 236 Terrapene nelsoni 228 terrapin Malaclemys 77,82 Malaclemys terrapin 77, 82 terrapin littoralis, Malaclemys 77, 82 terrapin pileata, Malaclemys 77, 82 terrapin tequesta, Malaclemys 77, 81-85 terrapin terrapin, Malaclemys 77, 82 tesselatus, Rhinocheilus lecontei 1 79 tetragrammus, Eumeces 148, 152, 154 textilis Demansia 199 Pseudonaja 178 Thamnophis 46, 51, 57, 60, 62, 152, 154, 170 butleri 36 cyrtopsis 230 cyrtopsis collaris 230, 240 crytopsis cyrtopsis 240 cyrtopsis pulchrilatus 230, 240 elegans 58, 60, 69, 71, 72, 167 elegans errans 240 eques 230 eques megalops 240 eques virgatenuis 240 maraanus 152 melanogaster canescens 24 1 nigronuchalis 241 proximus 52,98, 148, 152 radix 41, 164, 166-169 rufipunctatus 241 sauntus 52 sirtalis 58-70, 72,98, 164-169 sirtalis parietalis 52 sirtalis sirtalis 52 Thuja occidentalis 204 tigrina Natrix 46,52 Rana 126 tigrinum, Ambystoma 235 tigrinus, Rhabdophis 36 tigris, Cnemidophorus 190 Tillandsia benthamiana 223 exserta 223 Tithonia calva 223 Tomodactylus nitidus 232 nitidus petersi 236 saxatilis 232, 236 torquata, Hypsiglena 231, 239 iorquatus, Sphaerodactylus 229 Toxicodendron 223 triacanthos, Gleditsia 89 triangulum, Lampropeltis 51, 52, 199 triangulum nelsoni, Lampropeltis 240 triangidum sinaloae, Lampropeltis 240 triaspis, Elaphe 231 triaspis intermedia, Elaphe 239 Trimorphodon biscutatus biscutatus 24 1 lambda paucimaculata 24 1 tau 228, 231 Trionyx muticus 8 5 Triplasis 43 Triprion spatulatus spatulatus 237 Tripsacum dactyloides 43 triseriatus, Crotalus 198 Triturus 124-128 Tropidoclonion 154 lineatum 1 48, 1 52 Tropidodipsas 155 annulifera 241 philippi 241 sartorii 1 5 2 tuberculatus, Urosaurus bicarinatus 238 tuberculosus saxatilis, Phyllodactylus 237 U ulmi folia, Gauzuma 223 Ulmus americana 89 Umbra pygmaea 1 3 1 undulatus, Sceloporus 112. 152. 167 Unechis 1 77. 1 78 Uromacer 189 Urosaurus bicarinatus tuberculatus 238 ornatus lateralis 229 278 SPECIAL PUBLICATION-MUSEUM OF NATURAL HISTORY usta. Gastrophryne 232, 237 utiformis, Sceloporus 232, 238 utowanae, Anolis 229, 237 V Vaccinium... 205 valeriae, Virginia 1 5 3 valida, Nerodia valida 240 valida valida. Nerodia 240 variegatus, Coleonyx 232 variegatusfasciatus, Coleonyx 237 variolosus, Hypopachus 237 varius, Sauromalus 1 1 9 velutina, Quercus 89 ventrimaculatus, Anolis 257, 261-265 venulosa, Phrynohyas 229 venustissimus, Scaphiodontophis 185-192 Vermicella 173, 177 annulara 173 vermis, Carphophis 46, 52, 111 vernalis, Opheodrys 41, 109, 110 viminea, Quercus 223 Vipera 60, 199 aspis 199 bents 58,60,68, 195, 199,200 lebentina 198 virgatenuis, Thamnophis eques 240 virgatum. Panicum 43 I 'irginia 1 54, 155 striatula 52, 148, 149, 153, 155 valeriae 153 virginiana, Ostrya 223 virginianus. Bubo 2 1 2 viridiflavus. Coluber 199 viridis Crotalus 36, 46, 57, 58, 61, 68, 164, 168, 196-198, 200 Crotalus viridis 198, 199 viridis helleri, Crotalus 4 1 viridis lutosus. Crotalus 30 viridis oreganus, Crotalus 4 1 viridis viridis, Crotalus 108, 199 vittatum, Leiobunum 1 66 vocalis, Eleutherodactylus 236 vulpina, Elaphe 169 W warro, Simoselaps 173, 174, 176, 177, 180 wilcoxi, Tanti/la wilcoxi 240 wilcoxi wilcoxi, Tantilla 240 willardi mendionalis, Crotalus 241 woodhousei, Bufo 126, 128 Xenopus 1 2 7 laevis 1 2 7 yaquia, Tantilla 232, 240 Yucca 220 zeteki, Scaphiodontophis 185 zeteki nothus, Scaphiodontophis 190 Zeugorchis megacystis 93 - AVAILABLE SPECIAL PUBLICATIONS MUSEUM OF NATURAL HISTORY, UNIVERSITY OF KANSAS 1. Catalogue of publications in herpetology published by the University of Kan- sas Museum of Natural History. By Linda Trueb. Pp. 1-15. December 1976. $0.25. 2. Catalogue of publications in mammalogy published by the University of Kansas Museum of Natural History. By Robert S. Hoffmann. Pp. 1-19. 15 February 1977. $0.25. 3. Maintenance of rattlesnakes in captivity. By James B. Murphy and Barry L. Armstrong. Pp. 1-40. 29 December 1978. $3.00. 5. The natural history of Mexican rattlesnakes. By Barry L. Armstrong and James B. Murphy. Pp. 1-88. 14 December 1979. $6.00. 7. A diapsid reptile from the Pennsylvanian of Kansas. By Robert R. Reisz. Pp. 1-74. 18 February 1981. $5.00. 8. 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. 9. 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 V48 1984 \ crtcbrati ccologi and systematic Harvard MCZ Library \KI5464 I I II II 2044 062 369 020'