Puerto Rican Anoles Are Chilling In Florida – New Research By Jason Kolbe And Colleagues

A male Anolis cristatellus dewlaps on a tree in Miami, Florida. Picture reproduced with permission from Kolbe et al. (2012).

Anoles are remarkably adaptable creatures. You can find anoles in hostile environments, such as the tops of mountains in the Dominican Republic, in near-desert environments, and in places with over-winter freezing. Anoles are also a model system for rapid evolution; in response to strong selective pressure, an equally strong evolutionary response occurs within a few generations. It is perhaps unsurprising, then, that anoles are also one of the most invasive reptiles in the World. Although they are endemic to the tropical and subtropical regions of the New World, today anoles can also be found in such remote places as Guam, Hawaii, Taiwan, and Hong Kong.

One of the major questions surrounding anole invasions is how the organisms will respond to the challenges of a new environment. When anoles invade new environments they inevitably encounter new thermal and hydric conditions – how do these anoles adapt to a different environment? Jason Kolbe has spent many years exploring the ecology and genetics of Anolis invasions, and has focused especially on invasions in Florida (1, 2, 3). The Puerto Rican trunk-ground, A. cristatellus, has been found in Key Biscayne and South Miami since the mid-1970s. Ambient temperature is important for A. cristatellus and other anoles have been documented to acclimate to low temperatures. In this study Jason Kolbe and colleagues addressed two questions: (1) To what extent does the thermal environment change from Puerto Rico to Florida? and (2) Is there a phenotypic response in tolerance to cold?

To address the first question the authors used species distribution modeling (SDM) to model the thermal niche shift from Puerto Rico to Florida experienced by A. cristatellus. They gathered locality data for this species from museum databases and extracted relevant temperature variables (mean annual temperature, maximum temperature, minimum temperature, seasonality, etc.) from the WORLDCLIM data set. They then generated niche models using Maxent, a widely used program that uses the environmental conditions of known localities to predict habitat suitability over large geographic areas. They ran two models – one with the entire Caribbean basin as the background and one with just Puerto Rico as the background.

Figure 1. Suitability of the Florida using the Caribbean basin as a background.

The discrimination ability of the Caribbean model, which refers to how well it can predict occurrences compared to a random selection of points, was greater than the model using just Puerto Rico as a background. Both models were similar, however, in that they gave low suitability scores to the Florida habitat (Fig. 1 and Fig. 2). In fact, all of Florida received a suitability score of zero from the Caribbean model. A strong thermal niche shift was detected in both runs, but the inability of the models to detect suitable habitat in Florida, despite the presence of A. cristatellus there, suggests that locality data alone do not predict distributions well. There is a growing literature, in fact, arguing that the inclusion of organismal data will improve distribution models (i.e. ‘mechanistic niche modeling’; 1, 2, 3, 4).

Figure 1. Suitability of the Florida using Puerto Rico as a background.

To address the second question the authors assessed acclimation response in temperature tolerance in various native and invasive populations of A. cristatellus. The ability to acclimate thermal tolerance to ambient temperature conditions is potentially instrumental in facilitating invasion in a cooler environment in this species, and so the authors hypothesized that invasive populations of A. cristatellus should exhibit more plasticity in their tolerance as compared to native populations. The metric used in this study is CTmin, which refers to the low temperature at which a lizard loses the ability to right itself when flipped onto its back. Because performance is tightly dependent on body temperature in ectotherms such as anoles, the CTmin is a good metric for understanding the thermal limits to performance. In a previous study Kolbe and colleagues found that populations of A. cristatellus in Florida derive from two distinct invasions. These two genetic sources came from different regions of Puerto Rico, permitting a natural replicate of the CTmin acclimation experiment.

To this end, they maintained invasive populations of A. cristatellus (Key Biscayne and South Miami) and their source populations from Puerto Rico (Fajardo and San Juan) in the laboratory under winterizing conditions (22.5◦C) for four weeks. This temperature falls within the typical range of winter temperatures in south Florida, and so it accurately reflects the thermal conditions experienced by the invasive populations. The authors also tested a population of A. sagrei, the invasive brown anole from Cuba, and the native green anole, A. carolinensis.

Surprisingly, the results of the acclimation experiment varied among populations of A. cristatellus (Fig. 5 above). Although they experience similar winter conditions, only the Miami population of A. cristatellus exhibited plasticity in CTmin. The population from Key Biscayne showed no appreciable change in cold tolerance – in fact, it increased between weeks 2 and 4. Neither source population exhibited an acclimation response in CTmin. Both A. sagrei and A. carolinensis showed plasticity in thermal tolerance, and their final mean CTmin was similar to that of the Key Biscayne population.

We know that animals chilled beyond their CTmin lose mobility and can certainly die. In a previous post, I discussed this possibility in Dominican anoles from cool pine forests at high elevation. Thus, seasonal adjustment of CTmin to track environmental conditions is likely adaptive, and so it is puzzling why the Key Biscayne population does not exhibit tolerance plasticity. Although the invasions are equally young, Kolbe notes that the invasion in Miami is more genetically diverse than the Key Biscayne population, which suggests that more additive genetic variation in the Miami population may be involved in the acquisition of thermal acclimation. Moreover, CTmin acclimation is potentially sensitive to many factors, and so the experimental conditions used here may not trigger an acclimation response in the Key Biscayne population. Perhaps a different thermal treatment, such as acute or chronic exposure to progressively lower temperatures, may elicit a response that exposure to mean winter temperatures does not. It is also possible that animals in the Key Biscayne population (but not the Miami population) use retreat behavior to evade thermal conditions that approach the thermal limit, and so acclimation in cold tolerance may not be ecologically relevant in this population.

The contingency in thermal acclimation in different populations of A. cristatellus highlights that understanding invasions requires studying organismal variation at the population level. While it is difficult to project how differences in thermal plasticity will translate into invasion success, these results do show that similar thermal environments do not always yield the same phenotypic outcome, making this paper an informative and enjoyable read.


Jason J. Kolbe, Paul S. VanMiddlesworth, Neil Losin, Nathan Dappen & Jonathan B. Losos (2012). Climatic niche shift predicts thermal trait response in one
but not both introductions of the Puerto Rican lizard
Anolis cristatellus to Miami, Florida, USA Ecology and Evolution DOI: 10.1002/ece3.263

About Martha Muñoz

Martha is a postdoctoral researcher in Sheila Patek's laboratory at Duke University. She received her Ph.D. at Harvard University, where she studyied the evolutionary ecology and thermal physiology of anoles, focusing on the cybotoid anoles from the Dominican Republic. Martha serves as Conference Editor for the Anole Annals. Website: www.marthamunoz.weebly.com

2 thoughts on “Puerto Rican Anoles Are Chilling In Florida – New Research By Jason Kolbe And Colleagues

  1. Nice summary Martha. It occurred to me upon reading your review (and the authors note) that differences in thermal environments between South Miami & Key Biscayne could also be contributing to this pattern. The two sites have basically the same latitude, but Key Biscayne is an island, which may buffer it against extreme temperature fluctuations. Apparently there is some evidence of this (see Figure 3 of the study).

Leave a Reply

Your email address will not be published. Required fields are marked *

Optionally add an image (JPEG only)