The effects of genotype, age, and social environment on male ornamentation, mating behavior, and attractiveness.

The traits thought to advertise genetic quality are often highly susceptible to environmental variation and prone to change with age. These factors may either undermine or reinforce the potential for advertisement traits to signal quality depending on the magnitude of age-dependent expression, environmental variation, and genotype-age and genotype-environment interaction. Measurements of the magnitude of these effects are thus a necessary step toward assessing the implications of age dependence and environmental variability for the evolution of signals of quality. We conducted a longitudinal study of male guppies (Poecilia reticulata) from 22 full-sibling families. Each fish was assigned at maturity to one of three treatments in order to manipulate his allocation of resources to reproduction: a control in which the male was kept alone, a courtship-only treatment in which he could see and court a female across a clear partition, and a mating treatment in which he interacted freely with a female. We measured each male's size, ornamental color patterns, courtship, attractiveness to females, and mating success at three ages. Size was influenced by treatment and age-treatment interactions, indicating that courtship and mating may impose costs on growth. Tail size and color patterns were influenced by age but not by treatment, suggesting fixed age-dependent trajectories in these advertisement traits. By contrast, display rate and attempted sneak copulation rate differed among treatments but not among ages, suggesting greater plasticity of these behavioral traits. As a result of the different patterns of variation in ornamentation and behavior, male attractiveness and mating success responded to male age, treatment, and the interaction between age and treatment. Neither age nor treatment obscured the presence of genetic variation, and the genetic relationship between male ornamentation and attractiveness remained the same among treatments. Our findings suggest that neither age-dependent variation nor environmentally induced variation in reproductive effort is likely to undermine the reliability of male signaling.