Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Biological Sciences

First Advisor

James V. Remsen, Jr


This study investigates an example of extreme foraging specialization by tropical birds, namely foraging for arthropods in suspended aerial leaf-litter in lowland tropical rainforest. Up to 16 species at two southwestern Amazonian sites constitute a guild of specialized dead-leaf foragers that make up roughly 11% of the region's insectivorous bird species. Most dead-leaf specialists are ovenbirds (Furnariidae) or antbirds (Formicariidae) that are characteristic members of mixed-species foraging flocks. Individual dead leaves represent an abundant, seasonally stable resource that supports higher prey densities (number per leaf) and a greater proportion of preferred prey than adjacent live foliage. The arthropod fauna of aerial leaf-litter (dominated by spiders, roaches, other orthopterans, and small beetles) was distinct from that on live foliage. All guild members differed significantly from each other in either foraging height, size or type of leaves searched, diet composition, or prey size, although overlaps between species pairs were usually high ($\leq$0.900). All species selected substrates (leaf types) and prey nonrandomly compared with their availability. Some species segregated by habitat, but individuals apparently joined mixed-species flocks in each habitat independently of the other species present. I further investigated the dietary consequences of substrate specialization in five species of antwrens (Myrmotherula); two dead-leaf specialists, two live-leaf foragers, and one generalist. In a series of outdoor cage experiments with wild-caught birds, all foraging groups showed a similar degree of selectivity of prey types, and each species ate a wider range of prey than seen in natural diets. In additional experiments, live-leaf foraging and generalist species showed little interest in dead- or live-leaf substrates, whereas all dead-leaf foragers repeatedly inspected and manipulated dead and curled leaves in the absence of food. I conclude that substrate specialization in these birds involves fundamental differences in search behavior, but is not accompanied by equivalent changes in prey selectively or preference. Dead-leaf specialization evolved independently in several bird families but shows strong phylogenetic constraints among genera. Genetic relationships among Myrmotherula antwrens suggests that foraging specialization arose before the radiation of modern species, raising questions about the relevance of present-day ecology to the evolution of such specialization.