A post about lizards on islands. But not the ones you’re probably thinking of.

Yemen Socotra Felletti 48_00

Socotran Adenium obesum

Evolutionary biologists are fascinated by islands. There are a number of reasons for this. Islands systems can act as natural evolutionary experiments. They are small, less biodiverse, and isolated, so their biota can often be treated as simplified models of more complex mainland ecosystems (e.g. Darwin’s finches on the island Daphne Major). Ecologically similar islands can also act as replicates, with related taxa playing out the same evolutionary scenarios over and over again in isolation (e.g. Caribbean Anolis). Or they can act as life preservers, providing isolated strongholds for ancient evolutionary lineages that have long been extinct in the rest of the world (e.g. the Tuatara of New Zealand).

The Socotra archipelago is a particularly interesting, but poorly studied island system. Socotra consists of four islands in the Indian Ocean. It is extremely isolated (150 miles from the horn of Africa, 240 miles from the Arabian Peninsula) yet it has a continental origin. That means it was once part of the supercontinent Gondwana and suggests that some species may have lived there since it first became an island (~17.6 million years ago). Socotra has a very high level of endemism, with 37% of its plant species and 90% of its reptiles occurring nowhere else. As the islands are very remote and in a politically unstable part of the world, most of this unique biodiversity has not been studied using modern techniques. The islands are rugged and mountainous, reaching 1500m elevation, and primarily classified as tropical desert, making for a fairly fantastical landscape. A recent paper by Goméz-Diaz et al. (2012) takes a broad-brush approach to characterizing a chunk of Socotra’s obscure diversity: the Hemidactylus geckos.

A Socotran Gecko, Hemidactylus inintellectus

Goméz-Diaz et al. looked at the evolutionary history of Socotran Hemidactylus at multiple time scales. They wanted to know where the island Geckos originated (Africa or the Arabian peninsula?), if they dispersed to the islands, or surfed the continental fragments out to sea, how the various species on the islands were related (do close relatives inhabit the same islands? How did speciation happen?) and how historical climatic changes might have impacted the evolution of the group.

Going into the study, there were seven previously described species. They sampled these species intensively from all four islands as well as species thought to be their closest relatives in the Arabian Peninsula and the horn of Africa. They collected DNA sequence data from each of over 200 specimens and using those data they used various statistical methods to estimate time calibrated evolutionary trees. These are the fundamental quantities they use to try to answer their questions.

The results indicate that Socotran Hemidactylus have at least two, and possibly three origins. A group of four species branched off from the rest of Hemidactylus at approximately the same time Socotra became an archipelago (17.6 million years ago). These four species are all restricted to the largest island (Socotra). The remaining three species, which are scattered over the four islands, diverged from an Arabian ancestor between 3 and 8 million years ago. This is recent enough that their ancestors must have survived a brutal transoceanic crossing to establish themselves on the islands.

When the authors examined the genetic data for each of these previously named species more closely, they found several of them had high levels of geographically structured genetic differentiation, which suggests that many of these species may actually be cryptic species complexes. Their analyses suggest that the real number of species may be as high as 14.

As for the question of how and when speciation occurs in Hemidactylus, these data are essentially fuel for speculation. The authors find that closely related species are often ecologically divergent (on these islands there are rock-dwelling, ground-dwelling and arboreal species), and that intraspecific genetic variation is sometimes associated with ecological factors rather than geographic location, suggesting that divergent natural selection may be influencing the speciation process. Interestingly, there appear to be many cases of within-island speciation and ecological divergence, the causes of which, on other islands, have been the source of much discussion. The timing of divergences within the islands varies greatly and given the data, is highly uncertain, but it appears that most of the divergences are older than the Pleistocene, a period of repeated climatic shifts that biologists have long thought could spur speciation.

Hemidactylus granti

On the whole, this study provides an intriguing window into the origins of diversity in Socotra, but inspires as many questions as it answers. We can look forward to learning more as future work will undoubtedly take a more precise genomic approach, which will provide more precise dating and clarify which populations deserve species status, and involve more in-depth studies on each species ecology, which will allow a greater understanding of the forces that caused diversification in this group.

About these ads

One comment on “A post about lizards on islands. But not the ones you’re probably thinking of.

  1. jeffollerton says:

    Nice blog, thanks. Would be interesting to know whether these lizards are pollinators and/or seed dispersers. Lizard pollination has been documented on a number of oceanic islands, including the Canary archipelago and some of the Indian Ocean islands.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s