Sexual imprinting leads to speciation in locally adapted populations

Abstract

Sexual imprinting is a process whereby individuals choose mates that resemble other individuals, usually one of their parents. Sexual imprinting seems to be a general feature of birds, shown to exist in over 100 species belonging to 15 different orders, and in both sexes (ten Cate & Vos, 1999). ten Cate and Vos (1999) noted that theoretical models for evolutionary processes had generally ignored the fact that mate preferences in many birds are acquired by sexual imprinting. Sexual imprinting has been found in other taxa too, including mammals, fish (cichlids and stickleback) and frogs (Verzijden et al., 2012;Yang et al., 2019). Sexual imprinting results in phenotype

Times birds need to evolve mating cues under allopatry and parapatry

“…In (Sibly and Curnow, 2022)'s model generations are discrete and individuals die after mating. At the start of each generation individuals in each niche mate with others of the same mating-cue phenotype, and all mating individuals obtain the same number of offspring.…”

“…2). During the exponential phase the mating cue selection coefficient on C alleles was obtained by simulations carried out as in (Sibly and Curnow, 2022). By repeating this procedure over a range of values of s1, s2 and migration rate m, a picture was built up of the dependence of s on s1, s2 and m. The frequency of the CCQQ and CDQQ genotypes in the two niches.…”

“…Speciation occurs in (Sibly and Curnow, 2022)’s model because individuals in each niche avoid mating with incomers who predominantly carry disadvantageous alleles. The model incorporates a (Felsenstein, 1981) “two-allele mechanism” sensu (Butlin et al, 2021), with the key difference that Felsenstein’s two-allele models assume linkage between a locus controlling assortative mating and local adaptation loci, whereas (Sibly and Curnow, 2022)’s does not. In (Sibly and Curnow, 2022)’s model the loci are not close to each other in the genome—a situation allowing much greater evolutionary potential.…”

“…The model incorporates a (Felsenstein, 1981) “two-allele mechanism” sensu (Butlin et al, 2021), with the key difference that Felsenstein’s two-allele models assume linkage between a locus controlling assortative mating and local adaptation loci, whereas (Sibly and Curnow, 2022)’s does not. In (Sibly and Curnow, 2022)’s model the loci are not close to each other in the genome—a situation allowing much greater evolutionary potential. Whether the population will evolve towards speciation depends in (Sibly and Curnow, 2022)’s model as in Felsenstein’s on the balance between migration and selection: some selection at the ecological locus is necessary.…”

“…In the simplest case of parapatric speciation the mating cue is an adaptive trait which is acted on by selection directly, but this does not in general result in species-specific songs or plumage coloration. Alternatively, (Sibly and Curnow, 2022) showed that, with phenotype matching, local adaptation induces selection on a mating cue even if the locus coding for the mating cue is not linked to those producing local adaptation. Detailed analysis of two cases shows how mating cue selection then depends on the migration rates between the locally-adapted populations and the selective conditions within them.…”