Intraspecific variation in the information content of an ornament: why relative dewlap size signals bite force in some, but not all island populations of Anolis sagrei

Abstract

In many animals, male secondary sexual traits advertise reliable information on fighting capacity in a male-male context. The iconic sexual signaling device of anole lizards, the dewlap, has been extensively studied in this respect. For several territorial anole species (experiencing strong intrasexual selection), there is evidence for a positive association between dewlap size and bite capacity, which is an important determinant of combat outcome in lizards. Intriguingly, earlier studies did not find this expected correlation (relative dewlap size–relative bite force) in the highly territorial brown anole lizard, Anolis sagrei. We hypothesize that the dewlap size-bite force relationship can differ among populations of the same species due to interpopulation variation in the degree of male-male competition. In line with this thought, we expect dewlap size to serve as a reliable predictor of bite performance only in those populations where the level of intrasexual selection is high. To tackle this hypothesis, we examined the relationship between male dewlap size and bite force on the intraspecific level in Anolis sagrei, using an extensive dataset encompassing information from 17 island populations distributed throughout the Caribbean. First, we assessed and compared the relationship between both variables in the 17 populations under study. Second, we linked the relative dewlap size–bite force relationship within each population to variation in the degree of intrasexual selection among populations, using sexual size dimorphism (SSD) and dewlap display intensity as surrogate measures. Our results showed that absolute dewlap size is an excellent predictor of maximum bite force in nearly all Anolis sagrei populations. However, relative dewlap size is only an honest signal of bite performance in 4 out of the 17 populations. Surprisingly, the level of signal honesty did not correlate with the strength of intrasexual selection. We offer a number of conceptual and methodological explanations for this unexpected finding.