Causes and consequences of variation in the sex pheromone of the almond moth, Cadra cautella

Moth sex pheromones are usually female-produced blends of unsaturated olefinic chains with terminal functional moieties. An almost universal component of moth reproduction is the long-range attraction of males by female produced sex pheromones. Males are hypothesized to discriminate against incomplete and aberrant blends (i.e. male preference functions are stabilizing). Stabilizing preference functions would constrain pheromone evolution, reduce phenotypic variation, and eliminate heritable variation. Despite the fact that these predictions are implicit to models of pheromone evolution, little empirical support exists.;Gas-chromatographic analyses were used to determine the contents of individual female almond moth, Cadra cautella, sex pheromone glands. Abundant variation in the ratio of pheromone components among and within age-classes and periods of the L:D cycle were observed. A nested half-sibling breeding design was used to characterize the genetic basis of the observed variation in the ratio of sex pheromone components and a moderate to high narrow-sense heritability was observed.;The shape of the male C. cautella preference function was estimated using choice and no-choice wind tunnel trials to a range of naturally occurring pheromone blend ratios. Choice trials documented the expected unimodal (stabilizing selection) preference function whereas the no-choice trials observed a flat (no selection) male preference function. The opportunity to choose among equally receptive and available females may not occur in the field and therefore male mating preferences may not be stabilizing.;The observation of abundant heritable variation in the ratio of sex pheromone components in female C. cautella is not consistent with the existence of stabilizing male mating preferences. No-choice trials are likely more realistic of mating opportunities under natural conditions and suggest that male C. cautella mating preferences are not stabilizing and result in no selection on the pheromone blend.;Coordination of signal and response traits is predicted to be due to either a genetic correlation or coevolution. A bi-directional selection experiment was used to select for changes in the ratio of pheromone components. Choice trials were used to test for a correlated response in male preference. These experiments did not detect a genetic correlation between signal and response traits in C. cautella.