Historical Biogeography of endemic seed plant genera in the Caribbean: Did GAARlandia play a role?

Abstract

The Caribbean archipelago is a region with an extremely complex geological history and an outstanding plant diversity with high levels of endemism. The aim of this study was to better understand the historical assembly and evolution of endemic seed plant genera in the Caribbean, by first determining divergence times of endemic genera to test whether the hypothesized Greater Antilles and Aves Ridge (GAARlandia) land bridge played a role in the archipelago colonization and second by testing South America as the main colonization source as expected by the position of landmasses and recent evidence of an asymmetrical biotic interchange. We reconstructed a dated molecular phylogenetic tree for 625 seed plants including 32 Caribbean endemic genera using Bayesian inference and ten calibrations. To estimate the geographic range of the ancestors of endemic genera, we performed a model selection between a null and two complex biogeographic models that included timeframes based on geological information, dispersal probabilities, and directionality among regions. Crown ages for endemic genera ranged from Early Eocene (53.1 Ma) to Late Pliocene (3.4 Ma). Confidence intervals for divergence times (crown and/or stem ages) of 22 endemic genera occurred within the GAARlandia time frame. Contrary to expectations, the Antilles appears as the main ancestral area for endemic seed plant genera and only five genera had a South American origin. In contrast to patterns shown for vertebrates and other organisms and based on our sampling, we conclude that GAARlandia did not act as a colonization route for plants between South America and the Antilles. Further studies on Caribbean plant dispersal at the species and population levels will be required to reveal finer‐scale biogeographic patterns and mechanisms.

Global biogeography of Tetragnatha spiders reveals multiple colonization of the Caribbean

“…While some research find support for colonization of the Caribbean via GAARlandia for diverse group of lineages e.g. fishes (Říčan et al, 2013), frogs (Alonso et al, 2012), mammals (Dávalos, 2004), invertebrates (Binford et al, 2008;Dziki et al, 2015;Matos-Maraví et al, 2014) and even plants (Fritsch, 2003;van Ee et al, 2008) but see (Nieto-Blázquez et al, 2017), we do not find evidence that would support the use of GAARlandia land bridge by Tetragnatha. Both, our chronogram and scattered phylogenetic pattern of Caribbean Tetragnatha, disagree with such scenario.…”

Parallel ecological filtering of ultramafic soils in three distant island floras

“…The island of Cuba emerged 30–35 Myr (Iturralde‐Vinent, ; Iturralde‐Vinent & MacPhee, ), and according to the GAARlandia hypothesis, ancestors of a portion of the terrestrial biota may have colonized the island through a land bridge with South America (Ali, ; Iturralde‐Vinent & MacPhee, ). Nevertheless, long‐distance dispersal seems to have been the dominant source of colonizers among seed plants (Nieto‐Blázquez, Antonelli, & Roncal, ). Ultramafic substrates, recorded on Cuba by Humboldt (), cover c. 7% (2,700 km 2 ) of the island, including areas of ancient origin (10–30 Myr) in the Western and Eastern regions (Finko, Korin, & Formell, ), while Cuba's central ultramafic area was exposed more recently (<1 Myr, Zonn, ).…”

“…The island of Cuba emerged 30-35 Myr (Iturralde-Vinent, 2006;Iturralde-Vinent & MacPhee, 1999), and according to the GAARlandia hypothesis, ancestors of a portion of the terrestrial biota may have colonized the island through a land bridge with South America (Ali, 2012;Iturralde-Vinent & MacPhee, 1999). Nevertheless, long-distance dispersal seems to have been the dominant source of colonizers among seed plants (Nieto-Blázquez, Antonelli, & Roncal, 2017).…”

Evidence for Miocene overwater colonization in CaribbeanCyrtognathaspiders

“…As a means of biotic evolution on the Greater Antilles, the GAARlandia hypothesis allows for a combination of overland dispersal and subsequent vicariance and can be tested with the help of time calibrated phylogenies and fossils. While patterns of relationships that are consistent with predictions based on GAARlandia have been found in some lineages 44–48 it is not a good model for explaining the biogeographical history of others 27,29,49,50 .…”