Do events unfold in a predictable sequence when organisms undergo adaptive radiation? Anoles have diversified in many ecologically important characteristics as they have radiated both in the Caribbean and on the mainland. As one of our best-understood cases of extraordinary evolutionary diversification, they make a great system in which to ask how ecological diversity builds up during adaptive radiation.
The idea that anoles radiated in stages dates to at least 1972, when Ernest Williams derived some hypotheses from his observations of Puerto Rican Anolis in particular, drawing upon earlier work by Stanley Rand and Rodolfo Ruibal. Williams noticed that the most closely related species on Puerto Rico tend to belong to the same ecomorph class and occupy similar structural habitats (e.g. branches, trunks or twigs), but occur in different thermal habitats (e.g. closed forests or hot open areas). He proposed that anoles on Puerto Rico diversified first in structural habitat, and later in thermal habitat, a pattern that might scale up to the entire adaptive radiation of Anolis. While this idea has been discussed many times, and helped to inspire more general hypotheses about stages of radiation (e.g. Streelman and Danly 2003), until now it had not been tested using modern analytic techniques that incorporate phylogenetic information for many species.
Enter Paul Hertz and several of your favorite Anole Annals contributors, who test the “stages of radiation” hypothesis in a forthcoming paper in Evolution. They fit evolutionary models using an anole phylogenetic tree and two types of data. First, they used measures of morphology such as body size and limb length, whose relationship to Anolis structural habitat is well established. As a measure of thermal habitat, they compiled measures of body temperature taken in the field, which is available for a good sampling of species and is generally representative of species’ temperature preferences.
Their results were consistent with Williams’ hypothesis that structural habitat usually diversifies before thermal habitat. Morphological traits, especially body size and limb length, were best fit by an “Early Burst” model where the evolutionary rate is initially high, then slows down over time. This is not the case for body temperature, which instead is better explained by a “Late Burst” model where the rate is slow early, then speeds up. It therefore looks a lot like structural habitat diversified before thermal habitat, and that the stages of radiation are real. One caveat to this result is that when we analyze evolution on phylogenetic trees that only include modern species, sometimes very different processes can lead to similar signals in the data. In fact, the Late Burst model is mathematically difficult to distinguish from another common model (“OU”), in which evolution occurs at a constant and high rate, but where some type of constraint prevents species from diverging too far from each other. This is a bit tricky, because these models have very different implications – the Late Burst model predicts that there was low trait diversity early on, whereas the OU model predicts that there has been plenty of trait diversity all along, as in the Early Burst model.
How can we resolve this dilemma? Ideally with fossils – we could look at whether trait diversity was high or low early in the radiation. Unfortunately, while we can measure morphological traits in fossil anoles, taking their temperature probably wouldn’t be especially informative. To sum up, we can be pretty confident that the rate of evolution of anole structural habitat was initially high then slowed down, whether the slowdown was uniform or dependent on the buildup of diversity on each island (Mahler et al. 2010). What is harder to say is whether the evolution of thermal habitat has been speeding up, or whether it also diversified early but for whatever reason did not stabilize over time. If I had to guess, I would say that ancient anoles probably shared their descendants’ ability to adapt to different thermal environments whenever they were available, but at the moment we can’t say for sure. Whether we can develop tests to distinguish between these scenarios remains to be seen, as does the bigger question of why some elements of anole niches continue to diversify, while others stabilize as the adaptive radiation plays out.
I found the most satisfying aspect of this paper to be how clearly the authors place it within the tradition of anole natural history. The predictions made by Williams and his contemporaries, long before we had fancy phylogenetic comparative methods, have proven largely correct. This highlights how the tools of the naturalist – a keen understanding of the study species and an ability to recognize the implications of patterns in the field – continue to be of immense importance.
Hertz, P.E., Arima, Y., Harrison, A., Huey, R.B., Losos, J.B. and Glor, R.E. 2013. Asynchronous evolution of physiology and morphology in Anolis lizards. Evolution, in press. doi: 10.1111/evo.12072
Mahler, D.L., Revell, L.J., Glor, R.E. and Losos, J.B. 2010. Ecological opportunity and the rate of morphological evolution in the diversification of Greater Antillean anoles. Evolution 64: 2731-2745.
Rand, A.S. Ecological distribution in anoline lizards of Puerto Rico. Ecology 45: 745-752.
Ruibal, R. 1961. Thermal relations of five species of tropical lizards. Evolution 15: 98-111.
Streelman, J.T. and Danly, P.D. 2003. The stages of vertebrate evolutionary radiation. Trends Ecol. Evol. 18: 126-131.
Williams, E.E. 1972. The origin of faunas. Evolution of lizard congeners in a complex island fauna: a trial analysis. Evol. Biol. 6: 47-89.