The genetic effects of forest fragmentation on five species of Amazonian birds

Restriction fragment length polymorphisms of mtDNA were used to assess genetic structure at two geographic scales for five species of Neotropical forest-understory birds. At the local scale (in northeastern Bolivia), I studied populations of each species from six sites within 200 kin of one another. At this scale, I studied the effects of forest fragmentation on mtDNA genetic structure: three sites were in natural forest fragments separated by cerrado (savanna), and three sites were in continuous forest. Genetic variation did not appear to have be-en lost in the forest fragment populations of any species, However, for three antbirds (Thamnophilidae), patterns of haplotype distributions suggest fragmentation affected genetic structure in an unusual way. For these species, numerically dominant haplotypes in forest fragments did not occur in continuous forest, v whereas predominant haplotypes in continuous forest are widespread (occurring in fragments and continuous forest). These results suggest that forest fragmentation on a local geographic scale can affect genetic differentiation even in birds, a group that is considered to disperse well, The two other taxa studied were a woodcreeper (Dendrocolaptidae) and a tyrant-flycatcher (Tyrannidae). These two taxa did not show genetic effects of forest fragmentation, but they possessed notably different numbers of haplotypes per total individuals surveyed. The woodcreeper had few haplotypes (5 in 58 individuals), whereas the flycatcher had many (31 in 34 individuals). The numbers of haplotypes per individuals surveyed for the three antbirds were intermediate. Such variable levels of polymorphism can greatly influence analyses of genetic structure, At the regional geographic scale (across southwestern Amazonia). the flycatcher exhibited lower levels of differentiation than the other taxa, Levels of estimated sequence divergence within the other four taxa are similar to levels of differentiation between species from other avian studies, suggesting that genetic diversity is underestimated by current species-level taxonomy in these birds.