Abstract:
Madagascar is renowned for its exceptionally diverse, unique, and threatened biota, yet much of the island's flora and fauna remains undescribed, and the underlying drivers of diversification and endemism are poorly understood. The family Tenrecidae is one of four extant terrestrial mammal lineages to have colonized and diversified on Madagascar from continental Africa. The goal of this dissertation is to elucidate the evolutionary history of tenrecs at both deep and shallow time scales, and to use tenrecs as a proxy to understand the drivers of diversification on Madagascar. In Chapter 1 I generate the first rigorously inferred phylogeny of tenrecs to include every currently recognized species, revealing that they colonized Madagascar 30-56 million years ago. I also demonstrate that speciation rates have been higher in humid habitats compared to arid habitats - a finding that sets the groundwork for my next three chapters. To better understand the patterns and processes of diversification in the humid forest, I next explore the phylogeography of a species endemic to that region, Oryzorictes hova, in Chapter 2. Using genetic and morphometric data, I identify three populations (later identified as cryptic species) within O. hova that correspond to northern, central, and southern regions of the island. The same phylogeographic pattern has been observed in some of Madagascar's other humid-forest taxa, and it had been hypothesized that population structure is driven by low-elevation breaks between Madagascar's northern, central, and southern highlands. In Chapter 3, using genetic data from 20 small mammals and five reptiles codistributed along the island's humid-forest belt, I find this structure is directly related to the distribution of high-elevation areas and is congruent (spatially and temporally) across many species. This result demonstrates that the highlands have played an important role in recent diversification on Madagascar, most likely by functioning as refugia during Quaternary climate cycles. Finally, in Chapter 4 I continue to explore diversification in Madagascar's humid forests by studying species limits and patterns of gene flow in a clade of shrew tenrecs endemic to that region (M. fotsifotsy, M. soricoides, and M. nasoloi). Using a massively multi-locus genetic dataset, I demonstrate that M. soricoides and M. fotsifotsy (which are broadly sympatric) have hybridized in the past, and that this has caused conflicting phylogenetic results between different genetic datasets. I also recover two distinct clades of M. fotsifotsy: one that occurs only in the far north of Madagascar, and one that is widespread and broadly sympatric with M. soricoides. Evidence of reproductive isolation, plus subtle but significant morphometric differences between these clades, lead me to recognize them as distinct species. While I accomplished my primary aim of clarifying the phylogeny and taxonomy of Madagascar's tenrecs, my findings will also be important to scientists outside that initial scope. This research illuminates one of the mechanisms by which Madagascar's flora and fauna became so diverse -namely, that diversification has been driven by latitudinally segregated high-elevation refugia along the humid forests that historically spanned the island's eastern escarpment - and it reaffirms the need for continued collection and preservation of specimens in one of the world's hottest biodiversity hotspots as these forests face unprecedented rates of anthropogenic alteration.
