This essay comments on recent research on Darwinian fisheries science and on the future development of this field. From a practical point of view, the key question is: how fast are evolutionary changes caused by fishing happening? To answer this question, there is a need to understand intensities of selection generated by fishing, heritabilities and genetic correlations of the traits under selection, and whether the rates of change in traits predicted from this information are consistent with the changes observed. Although there is little doubt about the existence of phenotypic change in life-history traits of exploited fish stocks, there are few direct estimates of selection differentials caused by fishing. Results that are available, together with the relatively low heritabilities of life-history traits, suggest that the evolution caused by fishing occurs at a modest rate, and is likely to need a decadal time scale to be clearly observable. Given the pressing need for attention to fisheries in the short term, measures to control the longer-term evolutionary impact of fishing are most likely to be adopted if they also help to meet short-term objectives of management. With this in mind, the essay mounts a defence of large, old fish, the presence of which would be beneficial to stocks in the short term, and the conservation of which would set in motion selection for improved growth in the longer term.