One method for measuring sex-biased dispersal, which we have already referred to in earlier chapters, is the comparison of population differentiation estimates that are based on biparentally (autosomal nuclear) versus uniparentally (mtDNA, Y-chromosome) inherited markers. When males disperse and females are philopa-tric, mitochondrial markers should show higher levels of population differentiation than autosomal nuclear markers; conversely, at least in mammalian species, if females disperse and males are philopatric then Y-chromosome data should show higher levels of differentiation than either autosomal or mitochondrial DNA.
This method has been used to compare the dispersal of males and females from the nurseries of several bat species. Female bats often form maternity colonies in which they raise their young, and therefore they tend to be highly philopatric despite being proficient fliers that could easily move between sites. In Bechstein's bat (Myotis bechsteinii), colonies are closed to non-native females and yet one study found that overall relatedness within colonies was only 0.02 (Kerth, Safi and Konig, 2002b). In the absence of female-mediated gene flow, these low levels of relatedness must mean that males are regularly dispersing among colonies, a suggestion that was supported by a comparison of mitochondrial and nuclear differentiation among multiple colonies. Genetic differentiation based on nuclear alleles was much lower (FST = 0.003-0.031) than that based on mtDNA (Fst = 0.658-0.961), a pattern that reflects extremely rare female dispersal in conjunction with widespread male dispersal, with the latter preventing the colonies from becoming inbred (Kerth, Mayer and Petit, 2002a).
In the previous example of Bechstein's bats, the genetic data were supplemented by field observations and an understanding of the species' ecology, and therefore the conclusions were well-supported. However, studies of sex-biased dispersal that are based solely on comparisons between mitochondrial and nuclear DNA should be interpreted with caution for two reasons. First, the different mutation rates in the two sets of markers can influence the levels of observed genetic differentiation. Second, as we know from Chapter 2, the effective population size of mtDNA is expected to be approximately a quarter of that of nuclear DNA, although this is true only if mating is random, which often is not the case. In a strongly polygynous mating system, for example, many more females than males will reproduce each breeding season, and as a result the effective population size of maternally inherited genes can be larger than that of biparentally inherited genes (Chesser and Baker, 1996). It can be difficult, therefore, to anticipate how the population sizes of different genomes will influence observed levels of genetic differentiation.
Was this article helpful?