THE QUANTITATIVE GENETICS OF WING DIMORPHISM UNDER LABORATORY AND FIELD CONDITIONS IN THE CRICKET GRYLLUS-PENNSYLVANICUS

Laboratory estimates of the heritability of threshold traits indicate a typically large additive genetic component. In natural populations, it has been suggested that the heritability might be considerably redu ced owing to environmental variation. No test of this hypothesis for t hreshold traits has been undertaken. In this paper, such a test is rep orted using wing dimorphism in the cricket, Gryllus pennsylvanicus. A full-sib, split-family design was used, one half of each family being raised under constant conditions in the laboratory, and the second rai sed in cages outside at the site from which the parents were originall y collected. Pitfall-trap data indicated that nymphs in the field cage s grew similarly to individuals in the natural population. Heritabilit y of wing dimorphism in the laboratory was larger than heritability in the 'field' (0.70 vs. 0.21). However, analysis of variation across en vironments suggested that the genetic correlation was close to or even 1.00: this is supported by the estimate of the genetic correlation of 0.94. Thus selection in the laboratory environment will mirror the re sponses expected in the field. Laboratory estimates of heritability ca n therefore be useful indices of field values. These results do not su pport the hypothesis that heritabilities of threshold traits are reduc ed to insignificant values under natural conditions. More experiments are required to test the generality of this result and its consequence s for the evolution of threshold traits in natural populations.