How new species evolve is one of the most fundamental questions in biology. In particular, there has been considerable debate about the relative roles of genetic drift and natural selection in speciation. A salient example is the controversy surrounding the origins of Amazonian biodiversity. The Pleistocene has long considered one of the most important periods of diversification in the Amazon. According to the Pleistocene refugia hypothesis rainforest contracted into small refugia and genetic drift played the dominate role in driving speciation. However, the centers of diversity that would have provided evidence for these supposed refugia were never found. In addition, phylogenies and fossil records show that most lineages are older than the Pleistocene, refuting this simple allopatric model of speciation. Although multiple mechanisms such as geographical barriers, marine incursions, and ecological factors have been hypothesized to explain the Amazon's hyperdiversity there have been relatively few tests of these hypotheses.
Here I examined the importance of ecological factors in driving population divergence in the lesser tree frog (D. minutus) between the Amazon and Cerrado and along the gradient between these two biomes. Substantial temperature and precipitation differences between these biomes raise the possibility that environmental factors drive population differentiation, potentially resulting in speciation.
In chapter one, I examined the importance of environmental variables in driving reproductive divergence in the lesser tree frog. Water availability and temperature emerged as important drivers of body size variation, with body size found to be negatively correlated with vocal traits. For frogs, vocalization is one of the most important traits for mating success. Thus, these results suggest that the contrasting selective pressures on body size between the Amazon and Cerrado influence vocalization, which may result in reproductive divergence and suggests how reproductive isolation may evolve as a result of divergent natural selection pressures in these distinct habitats.
In chapter two, I evaluated the relative roles of genetic drift and natural selection in driving divergence in the lesser tree frog along the Amazon-Cerrado gradient. I showed that environmental variables are more important than geographical distance for driving divergence in reproductive and morphological traits. This suggests that ecological differences along the gradient result in adaptive divergence. Such divergence in phenotypic traits indicates the dominant role of natural selection over neutral processes for speciation events in this region.
Finally, in the third chapter, I investigated the sources of phenotypic variation in response to environmental variation between the Amazon and Cerrado biomes. The goal was to understand the contributions of genetic and environment to D. minutus phenotypic variation in larval development between these two biomes. I conducted a common garden experiment to test whether phenotypic variation in D. minutus has a genetic component, whether tadpoles from the Cerrado develop faster due to a more seasonal habitat, and whether D. minutus shows plastic responses to different thermal conditions. This experimental work showed that both environment and genetics play a role in phenotypic variation. Moreover, tadpoles from Cerrado populations developed faster than those from the Amazon populations. This is indicative of local adaptation to a more seasonal habitat by the Cerrado populations.