Why are estimates of the strength and direction of natural selection from wild populations not congruent with observed rates of phenotypic change?

Observing adaptive evolution is difficult. In the fossil record, phenotypic evolution happens much more slowly than in artificial selection experiments or in experimental evolution. Yet measures of selection on phenotypic traits, with high heritabilities, suggest that phenotypic evolution should also be rapid in the wild, and this discrepancy often remains even after accounting for correlations between different traits (i.e. making predictions using the multivariate version of the breeder's equation). Are fitness correlations with quantitative traits adequate measures of selection in the wild? We should instead view fitnesses as average properties of genotypes, while acknowledging that they can be environment-dependent. Populations will tend to remain at fitness equilibria, once these are attained, and phenotypes will then be stable. Thus, studying the causes of adaptive change at a genotypic rather than phenotypic level may reveal that, typically, it is occurring too slowly to be easily observed.