Genetic compatibility and sexual selection.

In a recent review in TREE [1], Mays and Hill discuss the interface between sexual selection for good genes (i.e. female choice based on traits indicating heritable fitness) and sexual selection for genetic compatibility (i.e. how well the genes of the parents function together in their offspring).We feel that the scopeof their contribution is somewhat limited, primarily because they implicitly equate genetic compatibility with genetic dissimilarity. Compatibility does, however, not equal dissimilarity. The relationship between genetic dissimilarity and offspring fitness is such that the highest level of fitness (compatibility) is achieved at intermediate levels of genetic dissimilarity [2]; in the low dissimilarity end of the continuum, reproduction between close relatives yields low fitness, as does reproductionbetweenspecies in theotherendof the continuum.Even within populations, maximal genetic dissimilarity between parents might not yield the fittest offspring [3,4]. We suggest that genetic compatibility is best understood as non-additive genetic variance in fitness, and good genes, correspondingly, as additive genetic variance in fitness [5]. The genetic mechanisms underlying non-additive genetic variance are diverse, but all include interactions between genetic elements [6,7]. For example, dominance effects are responsible for inbreeding depression [8], which can be considered a special case of genetic incompatibility. Outbreeding depression, however, reflects epistasis (interactions between alleles residing in different loci), caused by the disruption of co-adapted gene complexes [9]. Because a largepart of thestandinggenetic variation forfitnessoften seems to be due to epistatic genetic effects [10], genetic compatibility might be important in mate choice. Mays and Hill suggest that females face a dilemma when choosing males with ‘good’ versus ‘compatible’ genes [1]. We suggest that this is not necessarily so. If good genes are understood to be additive genetic variance in fitness, and compatibility is understood to be non-additive genetic variance in fitness, good genes and genetic compatibility are, by definition, uncorrelated. Indeed, females might use both criteria in mate choice [11,12]. By combining information about additive genetic quality and about compatibility of potential mates, a female might have fitter offspring than by using either criterion alone. Discussion in this field began with the problem of polyandry and the concept of genetic incompatibility [6]. Genetic incompatibility then referred to negative interactions between genetic elements inherited from the parents that cause inviability in the zygote. Thus, the original idea was that there are costs of incompatibility (and selection to avoid these costs). More recently, however, there has been a