Tropical Larval Anuran Fitness in the Absence of Direct Effects of Predation and Competition

I compare patterns of larval anuran species occurrences in aquatic habitats to results from exclosure experiments of tadpoles across all available habitat types. The purpose of this study was to investigate habitat effects on larval anuran fitness measures in the absence of direct effects of predators and competitors in a central Amazonian rain forest. In this sense, this study tests species—habitat associations established in an earlier study (Gascon 1991). Fifty—three primary forest aquatic sites of four distinct types (stream, streamside, peccary wallows, and terra firme sites) were surveyed for 2 yr for larval anuran species presence. In separate transplant experiments, I monitored growth, survival, and development of four larval anuran species reared in mesh cages as a function of habitat type. Tadpoles of Osteocephalus taurinus occurred in all types of habitats, but had a significantly higher mean number of cohorts in high—lying sites (terra firme and wallows) compared to low—lying sites (streams and streamside pools). Hyla geographica was restricted to streamside sites. Epipedobates femoralis and Phyllomedusa tomopterna occurred only in high—lying natural sites (wallows and terra firm sites). When transplanted, O. taurinus did significantly better in high—lying sites than in low—lying ones. Hyla geographica did not perform significantly better in sites where it naturally occurred. Phyllomedusa tomopterna showed significantly lower survival in low—lying sites compared to high—lying ones. Finally, E. fermoralis performed as well in streamside sites (where it never occurred naturally) as in high—lying sites. Occurrence patterns for P. tomopterna and O. taurinus were consistent with the transplant results. This suggests that these species occupy all habitats within their scope of physiological tolerance. Other factors, such as direct predation on larvae or interference competition, do not appear to account for the distribution of these species. For H. geographica and E. femoralis, results of transplants indicate that they do not perform better in their "preferred" habitat compared to unoccupied habitats.