Sperm competition and the level of polyandry in a bushcricket with large nuptial gifts

Variation in paternity due to sperm competition or post-copulatory female choice has a major influence on animal mating system evolution and on the levels of genetic variability in natural populations. However, there are relatively few studies comparing the outcome of sperm-competition experiments in the laboratory with natural variation in polyandry among families from the field. In the bushcricket Ephippiger ephippiger, females mate multiply, and the males provide them with a large, nutritious, and probably expensive, donation at mating. We examined paternity in a series of laboratory matings, where females mated with two males, and amongst a series of families collected from a natural population. In the laboratory, paternity was highly bimodally distributed: 24% of families had offspring fathered by the first male to mate, 68% by the second male (in only 8% was paternity shared). In the field, paternity was more mixed: only 27% of families had a single father, 14% had more than two fathers, whilst 59% had two fathers. While unsuccessful matings may contribute to the highly biased paternity in the laboratory, they cannot fully explain the high incidence of complete P2 families. Nonrandom sperm utilisation is therefore likely. Greater sperm mixing in the field probably results from females mating with more males, but the distribution of paternity also reflects an active process of nonrandom spenn utilisation. Confidence of paternity due to last male advantage may be relatively high in this species, and therefore may have facilitated the evolution of the large spermatophore in E. ephippiger.