Sexual Selection and Assortative Mating in Subdivided Populations of the Thrips Elaphrothrips tuberculatus (Insecta: Thysanoptera)


  • Bernard J. Crespi

    Corresponding author
    1. Museum of Zoology and Department of Biology, University of Michigan, Ann Arbor
      Dept. of Zoology, South Parks Road, Oxford University, Oxford, OX1 3PS, England
    Search for more papers by this author

Dept. of Zoology, South Parks Road, Oxford University, Oxford, OX1 3PS, England


This study presents four years of data on size-related mating patterns in Elaphrothrips tuberculatus, a subsocial thrips inhabiting subdivided populations (clusters of hanging dead oak leaves) that vary in sex ratio, operational sex ratio (ratio of males to females guarding eggs), and number of adults. Laboratory observations, experimental removals of guarding males, and field collections showed that: (1) males guard mates after copulation, (2) guarding males are larger than non-guarding males, and (3) large male mating advantage is caused by male fighting. The populationwide intensity of sexual selection for large male body size varied substantially among years, reflecting yearly differences in the sex ratio and operational sex ratio. Among local populations (leaf clusters), the sexual selection intensity was correlated with the operational sex ratio in each year, and with the sex ratio in three of four years. Males and females mated assortively for size (fore-femora length) in 1986, but not in 1983, 1984, or 1985. Assortative mating in 1986 was attributed to Darwin's (1871) mechanism for sexual selection in monogamous species: in this year alone, there was a highly significant tendency for large females to initiate breeding earlier than smaller females and strong directional selection for large male body size. Population subdivision affects mating patterns of E. tuberculatus by: (1) creating single-male populations where male size does not affect mating success, (2) promoting high sexual selection intensities, relative to a continuous population, when the population-wide sex ratio is female biased, and (3) constraining the size covariation between paired males and females.