Female choice and the consistency of courtship feeding in black-horned tree crickets Oecanthus nigricornis walker (Orthoptera: Gryllidae: Oecanthinae)

Courtship feeding in insects is often strongly correlated with insemination duration and therefore provides a potential postcopulatory episode of sexual selection. We tested whether courtship feeding and other courtship traits in the black-horned tree cricket Oecanthus nigricornis showed sufficient consistency potentially to respond to sexual selection by testing whether they differed significantly among males. Duration of courtship feeding differed among males when measured repeatedly and this caused significant differences in the duration of spermatophore attachment, a trait that determines the maximum duration of insemination and thus has important fitness consequences in crickets. We also partitioned variance in courtship behaviour between the sexes to test whether differences in courtship behaviour were attributable primarily to males, females or both sexes. Duration of spermatophore attachment was controlled by females and therefore represents a mechanism of female mate choice. Significant variation in duration of spermatophore attachment was associated with differences between individuals of both sexes. Differences among males indicate that females agree in their preference of certain males whereas differences among females indicate that females differ in their receptivity to postcopulatory courtship and insemination. The fact that differences among males in duration of spermatophore attachment were due to significant differences solely in the period of courtship feeding indicates that postcopulatory female choice was mediated through courtship feeding. Whether males manipulate female choices by allocating more or fewer resources requires further testing, but we found that males court some females more vigorously than others after females dismount. The number of previous mates had opposite effects on the duration of courtship feeding for the sexes, decreasing it for males but increasing it far females, and we discuss the possible causes of these results.