The rate and pattern of tail autotomy in five species of Puerto Rican anoles

Background: In lizards, tail autotomy is used in defence against predators. Question: Can we infer predation regime from the frequency and pattern of tail autotomy in five lizard species? Organisms: Five species of common Puerto Rican anoles: Anolis cristatellus, A. evermanni, A. gundlachi, A. krugi, and A. pulchellus. Methods: Monte Carlo simulations. Our Monte Carlo models incorporated the probability of tail loss (as opposed to mortality) during a predatory attack, the strength of the tail over its length (tail strength modelled as a heterogeneous probability of breakage among caudal vertebrae in the tail), and age-biased sampling (young lizards are more likely to have intact tails, but less likely to be included in our sample). Results: Our models exhibited good fit to the data, with the best fitting model showing a significant lack of fit in only one species. Our parameter estimates had biologically reasonable values. Our estimated rate of mortality from predatory attacks resulting in either mortality or tail injury was quite high (> 0.4 in the best fitting model) for all species. In three of the five species, the best fitting model included heterogeneity in the strength (probability of breakage) of the tail over its length, with the tail much more likely to break towards the tip than towards the base. The remaining two species (A. krugi and A. pulchellus), for which heterogeneous tail strength was not part of the best fitting model, are known from other studies to be ecologically and morphologically similar. These two also had the most similar estimated mortality rates. © 2010 Liam J. Revell.