The implications of pelvic reduction in threespine stickleback for long-term persistence of populations

摘要

Questions: Do threespine stickleback populations that exhibit pelvic reduction exist at lower densities than populations expressing the full pelvic girdle? Does population size fluctuate more in pelvic-reduced populations? Does trout stocking contribute to fluctuations, if they are present? Organisms: Twenty-three populations of lake-dwelling freshwater threespine stickleback (Gasterosteus aculeatus). Populations are of three types: 12 full-pelvic populations, all but two with predatory rainbow trout; five pelvic-reduced populations lacking trout; and six pelvic-reduced populations originally lacking trout, but into which trout are now stocked. Time and place: Annual field collections within the Cook Inlet region of Alaska, 1993-2010. Analytical methods: Nested ANOVAs (lake nested within population type) of logCPUE (catch-per-unit-effort), pooling all samples from each lake category in each year. ANOVA on coefficient of variation (CV) across populations and types. ANOVAs followed by planned comparisons. Results: Pelvic-reduced populations of both types exist at about one-third the density of full-pelvic populations. However, the coefficient of variations did not differ significantly. No effect of trout stocking was evident. Conclusions: Although catch-per-unit-effort values are only relative rather than absolute population size estimates, if our values of CPUE are comparable across populations, then populations that have evolved pelvic reduction have lower population densities than do full-pelvicpopulations. The small size of lakes supporting these unique populations and their lower fish density suggest that this rare type of stickleback population may be at elevated risk of extinction.