There has been an explosion of studies of selection and evolution on quantitative traits—traits including size, shape, etc.—in nature during the past two decades. The strength of selection is central to evolutionary change and adaptation: directional selection determines the rate of evolutionary change of a population; quadratic selection determines the strength of selection towards an optimal phenotype. We reviewed the published, peer-reviewed literature during the past 15 years that quantified selection in terms of two standard metrics-selection differentials and selection gradients. Using these standard measures of selection allowed us to compare across different study systems, kinds of traits, and aspects of fitness. We found more than 2500 estimates of the strength of selection in over 60 different study systems. We have a great deal of evidence for selection in nature. We used the resulting database, which is freely available on my lab web site, to address a variety of general questions about selection in nature: Is strong directional selection rare? Is optimizing selection common? Is selection on life history stronger than on morphology? How strong is natural versus sexual selection?

The strength of selection is key to understanding evolution and adaptation in nature. Our study has three findings of broad significance for our understanding of evolution. First, we now have an abundance of evidence documenting selection in nature in many different species. In many species selection has the potential to cause substantial evolution during our lifetimes, over the time scale of years to decades. So our understanding of selection and evolution goes far beyond familiar examples such as industrial melanism in moths and Darwin's finches. Second, our analyses indicated that sexual selection—e.g. selection resulting from differences in mating success—is typically stronger than natural selection with selection resulting from differences in survival. We sometimes associate selection and evolution with phrases such as 'the struggle for existence' and 'the survival of the fittest'. Our results suggest that it may be more appropriate talk about 'the struggle for mates' and 'the mating of the sexiest'. Third, we found little evidence for optimizing selection; we would expect abundant evidence of optimizing selection if organisms are optimally 'designed' in nature. This is surprising, and will require new field studies that are specifically designed to detect this type of selection.

This paper was also featured in Fast Breaking Papers for the month of June 2002.