The processes that give rise to ecological patterns of body size and diversification remain poorly understood. The recently proposed Cope-Bergmann hypothesis provides a causal association between interspecific body size evolution and environmental temperature. Using the phylogenetic comparative method within a Bayesian Markov Chain Monte Carlo (BMCMC) approach we assessed this hypothesis of body size evolution, using Sigmodontinae rodents as a study model. The BMCMC approach constitutes an important tool for evaluating historical hypotheses, and permits an understanding of the origin and diversification of body size in South American rodents.

We evaluated Cope’s Rule using two principal hypotheses: (a)- The common ancestor of Sigmodontine rodents has a greater probability of presenting small body size compared with body size variation in extant species; (b)- Since Sigmodontines are generally small to medium sized rodents, and greater body sizes are proposed to have greater biological fitness, the distribution of forces and/or constraints that affect the tendency of body size to increase over time should be homogenous, and generate a non-random directional process. If Sigmodontine rodents have a positive directional trend in body size evolution, and global temperatures continue the decreasing trend of the last 13 million years, we can suggest the Cope-Bergmann relationship occurs.

Results/Conclusions

Body size was significantly influenced by phylogeny (λ=0.91: Bayes-factor=503, strong evidence), and presented changes consistent with a gradual model of evolution, in other words changes in the trait were proportional to branch lengths (δ=1: Bayes-factor=3976, strong evidence). Furthermore, results indicated a greater probability that the most recent common ancestor of Sigmodontine rodents presented a small body size (α=93.1 mm). Subsequent radiation occurred towards larger body size, given that a directional change model was found to be the most appropriate model for explaining the evolution of body size in Sigmodontine rodents (Directional model v/s Random-walk model: Bayes-factor=1.8, supporting evidence for Directional model).

These results suggest that, after the ancestral colonization, diversification in Sigmodontines began in northern latitudes of South America, in tropical warm environmental temperatures, with some small ancestors. Later, they advanced and diversified towards the colder southern latitudes. This, together with a decrease in global temperature, caused an increase in the body size of Sigmodontines, consistent with the Cope-Bergmann hypothesis. This hypothesis suggests that over evolutionary time in South America, larger Sigmodontine rodents presented a surface-to-volume ratio that allowed them to conserve more heat, and, thus, to diversify in colder global environments and austral latitudes. (FONDECYT grant-11080110)