Spiny spiders and spiny trees: Molecular phylogenetics and biogeographic reconstruction reveal history of recurrent overwater dispersal events, single-island endemics, and new species of Caribbean Micrathena (Araneae: Araneidae)

Elucidating the processes responsible for the provenance of taxon biodiversity on islands can help us to understand macroevolutionary mechanics in addition to specific organismal histories. Current species distributions are a result of an admixture of both abiotic (i.e. geologic) and biotic (i.e. dispersal capability, speciation and extinction rates) factors. The heterogeneous tectonic histories and local ecological differentiation among islands within the Caribbean contribute to high levels of endemicity and diversity. Differential historical biogeographic processes, namely vicariance (physical separation of populations by barriers) or long-distance dispersal, represent hypotheses to explain current distributions of species on islands resulting from colonization followed by radiation and extinction. In the Caribbean, the GAARlandia landbridge, connecting northern South America with the Greater Antilles (emergent 35-33 mya) presents an appealing vicariant explanator for diversity among various animal groups. Spiders epitomize excellent study organisms for phylogeographic analyses due to their high biodiversity, occupation of a wide variety of ecological niches, and high variation in their ability to disperse. This study assesses the evolutionary and biogeographic history of the spider genus Micrathena within the Caribbean to discern the role of long-distance dispersal and vicariance in shaping their distribution and diversity. Through the process several new species were discovered that are here described, adding to our understanding of diversity within the group. Prior work on Micrathena has been limited by relatively poor specimen sampling and resulted in conflicting reports supporting opposing colonization processes and routes. A robust specimen set emphasizing increased sampling on continental mainland areas (Colombia, Florida, plus Genbank mining of data from Brazil and the largest Greater Antilles Islands (Hispaniola, Cuba) was included in our analysis. Multilocus sequence data, ancestral range reconstruction, and biogeographic model testing procedures were integrated to reconstruct an evolutionary history of Micrathena and test the roles of the alternative vicariant (GAARlandia) and long-distance dispersal hypotheses. The history of Micrathena in the Caribbean spans approximately 30 million years beginning in the mid- Oligocene. The genus originates in South America, but Caribbean groups have a newer North American origin and dispersed five times to the Caribbean, supporting a long-distance dispersal hypothesis for colonization and rejecting a GAARlandia colonization route. Our results suggest high endemicity among Caribbean islands. The novel phylogeny highlighted previously uncovered species diversity, and species delimitation along with the primary phylogenetic dataset was used to develop new species hypotheses and descriptions. Three new species are herein described with representatives from Costa Rica, Mexico, and Colombia, and we resurrect M. flavomaculata (Keyserling, 1863) from Hispaniola. As so few specimens (\u3c4) currently represent these species, future sampling will be necessarily undertaken at proximal collection localities to ensure diversity is adequately ascertained. These results spotlight Caribbean Micrathena among Caribbean arachnid groups in that they repeatedly dispersed to the Caribbean despite their profuse spininess, and as having originated in North America