After five generations of selection, Eknath et al. (1998) reported an annual genetic gain of 12-17%, which is considerably higher than our estimate for nine generations. Eknath et al. (1998) did not establish a separate control population and maintain it throughout these five generations, but rather, they recreated a new control in each generation by mating a sample of average individuals for that generation. There may have been inadvertent selection of the fish to be mated as controls, and sampling problems accumulated during the selection could have caused an over estimation of the genetic gain that reported by Eknath et al. (1998).