This is the first of a series of posts that will review a number of papers that examined species richness patterns in anoles, starting with Rand’s 1969 paper described below and moving towards the present day understanding. Read on!

Species richness predictions from the Theory of Island Biogeography

 Rand, A.S. 1969. Competitive exclusion among anoles (Sauria: Iguanidae) on small islands in the West Indies. Breviora 319: 1-16.

The empirically observed species-area relationship (SAR) is one of the closest things to a law that we have in evolutionary ecology. All else equal, the larger the area, the more species will be in it. In a paper in 1963, and a book in 1967, MacArthur and Wilson (M&W) put forward the Theory of Island Biogeography (TIB) to explain the species-area relationship on islands. They observed that the number of species that inhabit an island scales positively with island area and negatively with distance of an island from the mainland species source. M&W argued that the SAR is governed by two ecological processes: colonization and extinction. Colonization probability increases with the size of an island and decreases with island distance from the mainland. Extinction probability increases as island size decreases because small islands support smaller population sizes and leave species are more vulnerable to fluctuations in abiotic and biotic environmental factors. Thus, the TIB predicts that large islands close to the mainland will be species rich, while small islands far from the mainland will be species poor. The TIB inspired thousands of papers: according to Google Scholar, the 1967 book has been cited 11,148 times and the 1963 paper has been cited 1,416 times.

In 1969, Stan Rand applied TIB to the radiation of anoles in the Caribbean. Anoles occupy nearly every piece of land in the Caribbean, from the largest islands down to teeny cays and spits of rock no larger than a basketball court. Because the four islands of the Greater Antilles (Cuba, Hispaniola, Jamaica, and Puerto Rico) are so large (all >9000 km2) and species rich**, Rand treated these as his “mainland” species source and focused on the smaller Caribbean islands as the “islands” in the TIB framework. Rand tallied species counts for 122 smaller islands from 44 island banks. The largest non-Greater Antillean island was Isle of Pines (3000 km2), the smallest was Emerald Island (0.16 km2), and Rand did not include the countless islands smaller than Emerald Island that likely contain anoles but for which the data weren’t available. The exclusion of these smallest islands, however, wouldn’t affect his conclusions.

Rand found that the species-area relationship didn’t hold for anoles. Instead, there were many large islands (i.e. in the Lesser Antilles) that had only one or two species, whereas smaller islands (i.e. the Great Bahamas Bank and satellite islands of the Greater Antilles) had up to 5 species inhabiting them. How did Rand explain this?

First, Rand noted that the islands could be divided into four classes: (1) satellite islands connected to one of the Greater Antillean islands by a shallow underwater bank; (2) the islands of the Great Bahama Bank; (3) the isolated islands of the Lesser Antilles; and (4) oceanic islands surrounded, as you might expect, by deep ocean.

The first two classes of island – those that would be connected to a large body of land during sea level mimima – actually do match the predicted TIB species-area relationship, where larger islands have more species. During sea level minima, the satellite islands were connected to their Greater Antillean “mainlands” and likely were colonized simultaneously by many lowland species along these connections. When the sea level rose, the largest islands were able to maintain their 3-5 five species while smaller islands lost species through extinction. The same explanation holds for the Great Bahamas Bank. During the sea level minima, the many islands of this bank would have been joined into one very large island close to Cuba that could be colonized easily over water. As sea levels rose, the species-area relationship in the Great Bahama Bank was created by the extinction process on smaller islands.

The Lesser Antillean and oceanic islands, on the other hand, do not show the species-area relationship.  Instead, they have one or two species, without any obvious relation to island area or distance from source islands.

Fig. 2 from Rand, showing species-area relationships for his four island groups.

Rand argued that the difference in species richness between the oceanic islands and satellite islands cannot be explained by differences in island ecological diversity or carrying capacity. Indeed, many of the islands in the Lesser Antilles have more ecological diversity than the satellite islands mentioned above. Moreover, Rand also that the depauperate nature of oceanic islands weren’t caused by a lack of propagule pressure. Anoles are excellent colonizers (Williams, 1969) and there is no relationship between wind and ocean current patterns and species richness in Rand’s data.

Instead, Rand argues that species interactions and competitive exclusion are what truly matter when explaining the deviation from the species-area relationship for oceanic islands. On the Lesser Antilles, a generalist species arriving on an empty island would have time to occupy the island completely by the time one or several colonizers of a second species landed. Rand argues that the chances of the second species successfully establishing a population would be small due to competition with the first species. The chances of establishment of a third species would be even smaller, and so on.

In contrast, the satellite islands of the Greater Antilles and the islands of the Great Bahama Bank would have been colonized simultaneously under heavy propagule pressure from species that had already evolved to coexist with one another on a Greater Antillean “mainland,” allowing islands that are much smaller and less ecologically diverse to maintain a higher species richness than their oceanic counterparts.

So, it seems that anoles match species-area predictions in those places where the colonizing anoles were pre-adapted to coexist with other colonizing species (i.e. competition had already been reduced and was no longer important in the evolution of a community). In those places where the colonizing anoles were generalists, as is likely the case in the Lesser Antilles, competition may have played a much more important role in structuring communities. Indeed, Losos and Roughgarden exchanged a series of papers in the early 1990s discussing the role of competitively driven character displacement and ecological sorting for the evolution of Lesser Antillean Anolis communities (see Losos 1992, Roughgarden 1992).

**The Greater Antilles islands are also the only islands in the Caribbean on which anoles have undergone in situ speciation. Although MacArthur and Wilson’s theory is generally thought to be purely ecological, the authors did address the potential for evolutionary processes like speciation to affect the species-area relationship in chapter 7 of their book. We’ll return to this idea in the next post.

Latest posts by Yoel Stuart (see all)