Genetic and demographic responses of mercury-exposed mosquitofish (Gambusia holbrooki) populations: temporal stability and reproductive components of fitness.

Two previous mesocosm studies showed changes in glucosephosphate isomerase-2 (Gpi-2) allele frequencies in mosquitofish populations exposed to mercury for 111 d or two years. A previous selection component analysis of single-generation populations exposed for 111 d to 18 microg/L Hg suggested that female sexual selection and fecundity selection could contribute to changes in Gpi-2 allele frequencies. The present multigeneration study was conducted to determine the stability of Gpi-2 allele frequencies over four years of mercury exposure, measure the reproductive fitness of Gpi-2 genotypes inhabiting control and mercury-contaminated mesocosms to determine a mechanism explaining changes in Gpi-2 allele frequencies, investigate differences in the demographic characteristics of mercury-exposed and control populations, and investigate the water quality of the mesocosms to determine if variables other than mercury show concordant patterns among mesocosms. Differences in Gpi-2 allele frequencies between control and mercury-exposed populations were stable over four years (approximately eight generations) of mercury exposure. Mercury-exposed female mosquitofish had a lower probability of being gravid than control females (p = 0.001). Mercury-exposed females also had lower fecundity (total number of eggs and embryos) than control females (p = 0.036). Unlike the results of the more intense mercury exposures in the single generation study, no strong evidence was found that Gpi-2 genotype influenced fecundity or the probability of being gravid in both control and mercury-exposed females. The quantification of fitness components is difficult but has the potential to enhance our understanding of how toxicants alter allele frequencies in exposed populations.