Abstract
Island colonization is often assumed to trigger extreme levels of phenotypic diversification. Yet, empirical evidence suggests that it does not always so. In this study we test this hypothesis using a completely sampled mainland-island system, the arid clade of Hemidactylus, a group of geckos mainly distributed across Africa, Arabia and the Socotra Archipelago. To such purpose, we generated a new molecular phylogeny of the group on which we mapped body size and head proportions. We then explored whether island and continental taxa shared the same morphospace and differed in their disparities and tempos of evolution. Insular species produced the most extreme sizes of the radiation, involving accelerated rates of evolution and higher disparities compared with most (but not all) of the continental groups. In contrast, head proportions exhibited constant evolutionary rates across the radiation and similar disparities in islands compared with the continent. These results, although generally consistent with the notion that islands promote high morphological disparity, reveal at the same time a complex scenario in which different traits may experience different evolutionary patterns in the same mainland-island system and continental groups do not always present low levels of morphological diversification compared to insular groups.
Highlights
While more studies are needed to address the generality of the island effect on empirical grounds, a major stumbling block is the need of well-sampled phylogenies in both island and continental domains
Our final dataset included the sequences for all 31 species known to occur in Arabia, the Socotra Archipelago, the Levant, and adjoining areas of the Persian region, and 17 species occurring in the Horn of Africa
Our results provide strong evidence supporting an acceleration in the rates of body size evolution that produced great disparities in Abd al Kuri and, to a lesser extent, in Socotra
Summary
The present study uses a completely sampled mainland-island system, the so-called arid clade of Hemidactylus geckos[14] to test whether the patterns of phenotypic diversification in islands differ from those occurring in the continent. The arid clade represents a well-studied monophyletic radiation of 48 species distributed across the arid regions of northeast Africa, the Levant, Arabia, the adjoining areas of southwest Asia and, interestingly for our purposes, in the Socotra Archipelago[14,15,16,17,18] (Fig. 1) This archipelago is located in the northwestern Indian Ocean and originated as a continental fragment that detached from Arabia around 30 Ma19. We applied a variety of comparative methods to pose three main questions: 1) Do island and continental species significantly differ in their morphologies, with the most divergent phenotypes occurring on islands? 2) Do insular species assemblages exhibit higher phenotypic disparities compared to continental species assemblages? And if so, 3) Are these higher disparities driven by faster evolutionary rates? Our results yield support for the island effect, and suggest that the phenomenon is more complicated than often assumed
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
