Locomotor Performance and Sexual Selection: Individual Variation in Sprint Speed of Collared Lizards (Crotaphytus Collaris)
Whole-animal performance traits, such as locomotor performance, are central to current concepts of phenotypic adaptation, yet the possible evolution of such traits via sexual selection is an underexplored hypothesis. We studied a cursorial, polygynous, territorial lizard to test two predictions of sexual selection theory applied to whole-animal performance: a secondary sexual performance characteristic may be sexually dimorphic, and variation in performance among males should be correlated to mating success. Maximal sprinting performance was measured in a series of wild-caught adult collared lizards (Crotaphytus collaris) from a single Oklahoma population. Sprint speed varied repeatably among individuals, but did not scale to body size among adults. Contrary to our first prediction, sprint speed did not differ between sexes. However, among 11 adult males sprint speed was strongly correlated to territory size and a spatial index of potential mating success (independent of body size), which indicates phenotypic intrasexual selection, whether direct or indirect, for whole-animal locomotor performance. The lack of sexual dimorphism in sprint speed may reflect trade-offs with other sexually selected traits (e.g., head size) and/or condition-dependence of running capability. Sexual selection of social behavior may underlie more generally the evolution of physiological performance, and therefore of suborganismal physiology and morphology.Abstract

Allometry of maximal sprint speed in adult collared lizards. Axes are logarithmic. Least-squares regression line is for an entire ontogenetic series (including unpublished data not shown; see text)

Relationships between maximal sprint speed and (A) home range or territory size and (B) number of adult female home ranges overlapped by adult male territories, for collared lizards on Sooner Lake Dam, Oklahoma, during the breeding season of 2000.Axes are logarithmic except for number of females (B). Lines shown are reduced major axis regressions (for males only), r values are Pearson correlation coefficients, and the P value in (B) is for a one-tailed test