The mating system of steelhead and the effect of length and arrival date on steelhead reproductive success

Mating systems reflect the degree of sexual selection in the past and present. Sexual and natural selection operating before and during spawning and natural selection on offspring combine to determine patterns of individual reproductive success. The direction and strength of selection may change over the lifetime of parents and their offspring or over the course of years or seasons. In order to first, describe the genetic mating system of steelhead, and second, test hypotheses concerning the effects of body size and arrival date on individual reproductive success parents were genetically matched to young-of-the-year (YOY), yearling and smolt offspring in 4 brood years, and to returning adult offspring in 19 brood years in a small population of steelhead (Oncorhynchus mykiss). Male and female steelhead mated multiply in all brood years. No evidence was found of size-assortative mating, however males tended to arrive before the females with whom they mated. Mature parr were inferred from the excess of parental assignments of mothers over that of fathers. YOY offspring of early arriving females were larger than the offspring of later arriving females; large females had large YOY offspring in only one brood year. Large and early arriving males produced more smolt and adult offspring than small or late arriving males, and large females produced more smolt and adult offspring than small females. Large males and females had more adult offspring consistently across brood years, while selection for arrival date for both sexes varied in direction among years. Standardized variance in reproductive success was positively correlated with both the sex ratio and the spawning density for males and with only the spawning density for females. The strength and direction of selection on male length varied more when the sex ratio was female biased. In females, the strength of selection was unrelated to either the sex ratio or the spawning density. Freshwater survival of maternal and paternal half-sibling families appeared unrelated to young-of-the-year length, except for maternal half-sibling families in one brood year when families which were smaller survived in higher numbers. Marine survival of maternal and paternal half-sibling families appeared unrelated to either smolt length or date of migration, and instead appeared to be random or equal among families in the three years for which we had data. Extensive yearly variation in the direction and shape of selection on both traits in concert with overlapping generations will confound response to selection on both traits, thus making predictions of evolutionary change in response to changing environments difficult.