..••.•.......•.....•.••.•.•...•••.....•.•••..•.•••.••...••••••...••....•.•..••.•••.•.•...•....•••...•..•••••.•..•••••..•••.••.•••.•...••..... 2 RESUMEN ............................................................................................................................................... 2 INTRODUCTION ................................................................................................................................... 3 ACKNOWLEDGMENTS •.••.•.•.••..•.•....•...••....••••......•••••.....•••..•.•.....••..••••.•.....••••.....•.••.....•.••••••.•••.••••.....••.... 6 MATERIALS AND METHODS ............................................................................................................ 6 RELIABILITY OF DIETARY DATA ............................................................................................................. 9 DIETARY ANALYSES ........................................................................... : ................................................ 11 MORPHOLOGICAL ANALYSES ..................................................................................................... : ......... 11 COMMUNITY ANALYSES ...................................................................................................................... 12 RESULTS ................................................................................................................................................. 13 DIETS OF ANuRA.Ns ............................................................................................................................ 13 MORPHOLOGICAL CORRELATES OF DIETS ............................................................................................ 21 COMMUNITY ANALYSES ...................................................................................................................... 26 DISCUSSION ......................................................................................................................................... 38 CORRELATIONS OF DIET WITH PHYLOGENY, MORPHOLOGY, MICROHABITAT, AND FORAGING MODE ..... 38 COMMUNITY ANALYSES AND FEEDING GUILDS .................................................................................... 42 COMPARISONS WITH OTHER ASSEMBLAGES .......................................................................................... 43 CONCLUSIONS ...••.•.....•.•.......•......•...•••••.•.•...••....•..••..••••.•......•.•••....•...•...•...••••..••....•...••...••...•••.....•...•. 47 LITERATURE CITED ........................................................................................................................... 48 APPENDIX 1. SPECIMENS EXAMINED····································································································· 52 APPENDIX 2. DIETARY DATA ................................................................................................................ 53 © Natural History Museum, The University of Kansas ISSN No. 1094-0782 2 SCIENTIFIC PAPERS, NATURAL HISTORY MUSEUM, THE UNIVERSITY OF KANSAS ABSTRACT I examined the diets of 867 anurans of 58 species from Cuzco Amaz6nico, Peru. A total of 6393 prey items in 62 prey categories was identified from the 610 anurans (70%) with prey in their gastrointestinal tracts-4316 (77%) from the stomach, 2077 from the intestines. Anuran species differ greatly in average number of prey per individual and the relative size of prey consumed. Bufonids, microhylids, and dendrobatids eat large numbers of relatively small prey (a high percentage of ants). Most hylids eat a few large prey, and leptodactylids are intermediate in number and size of prey eaten. Larger hylid species eat primarily orthopterans, roaches, and moths, whereas smaller hylids eat primarily spiders, beetles, and larvae. Most leptodactylids have large niche breadths and eat a great diversity of prey. Beetles, orthopterans, and millipedes are important prey items volumetrically, and ants and beetles are most important numerically. Most morphological variation (corrected for size) among species (71 % ) was accounted for by two principal component axes and seems to be associated with phylogeny, and to a lesser extent, diet. The two hylids that differ most from other members of their family, Sphaenorhynchus lacteus (differs in diet), and Phyllomedusa atelopoides (differs in microhabitat and diet) differ in morphology as well. Microhylids and dendrobatids have narrower heads and shorter jaws than hylids or leptodactylids. Maximum, and to a lesser extent, minimum prey size is correlated with frog size, but different families exhibit different relationships. Head shape is important in the number and size of prey consumed regardless of overall size; anurans with narrower heads and shorter jaws eat more, and smaller prey items. Most diet overlaps are low and terrestrial species have lower diet overlap values than arboreal species. The anurans exhibit guild structure in their diet. The terrestrial species are distributed in two distinctive feeding guilds-an ant/termite guild and a larger-prey guild. The arboreal community has only one ant specialist; many of the other species of hylids do not include ants in their diets.I examined the diets of 867 anurans of 58 species from Cuzco Amaz6nico, Peru. A total of 6393 prey items in 62 prey categories was identified from the 610 anurans (70%) with prey in their gastrointestinal tracts-4316 (77%) from the stomach, 2077 from the intestines. Anuran species differ greatly in average number of prey per individual and the relative size of prey consumed. Bufonids, microhylids, and dendrobatids eat large numbers of relatively small prey (a high percentage of ants). Most hylids eat a few large prey, and leptodactylids are intermediate in number and size of prey eaten. Larger hylid species eat primarily orthopterans, roaches, and moths, whereas smaller hylids eat primarily spiders, beetles, and larvae. Most leptodactylids have large niche breadths and eat a great diversity of prey. Beetles, orthopterans, and millipedes are important prey items volumetrically, and ants and beetles are most important numerically. Most morphological variation (corrected for size) among species (71 % ) was accounted for by two principal component axes and seems to be associated with phylogeny, and to a lesser extent, diet. The two hylids that differ most from other members of their family, Sphaenorhynchus lacteus (differs in diet), and Phyllomedusa atelopoides (differs in microhabitat and diet) differ in morphology as well. Microhylids and dendrobatids have narrower heads and shorter jaws than hylids or leptodactylids. Maximum, and to a lesser extent, minimum prey size is correlated with frog size, but different families exhibit different relationships. Head shape is important in the number and size of prey consumed regardless of overall size; anurans with narrower heads and shorter jaws eat more, and smaller prey items. Most diet overlaps are low and terrestrial species have lower diet overlap values than arboreal species. The anurans exhibit guild structure in their diet. The terrestrial species are distributed in two distinctive feeding guilds-an ant/termite guild and a larger-prey guild. The arboreal community has only one ant specialist; many of the other species of hylids do not include ants in their diets.