Chromosomal rearrangements, speciation and reproductive isolation: The example of two karyotypic species of the genus Sorex

During the last twenty years, growing evidence appeared in favour of the role of chromosome rearrangements as a factor that may promote speciation (White, 1978). Several theoretical models have been proposed (Sites and Moritz, 1987) many of which are based on empirical data from various taxa, including, among others, the genus Drosophila (Wallace, 1953), grasshoppers (Shaw, 1981), lizards (King, 1981) and mice (Capanna, 1982). However, Sites and Moritz ( 1987) still find this evidence for a chromosomal involvement in speciation too fragmentary. Moreover, chromosomal speciation models are often taken as sympatric speciation models and thus assume that when pre-mating barriers have evolved to reduce the production of unfit hybrids, this should have occurred by reinforcement of characters increasing assortative mating, a process that still lacks good theoretical and empirical support (Butlin, 1987; Coyne and Barton, 1988). One further problem is that, in many natural cases, hybrid zones are found between karyotypic races, sometimes with appreciable stability and persistence that does not necessarily suggest current speciation (Barton and Hewitt, 1985). For example, a study on British karyotypic races of the shrews of the Sorex araneus complex has shown that between some races, each characterized by different sets of metacentric chromosomes, natural hybridization may occur without any evidence for strong preor post-mating barriers. On the contrary, in the hybrid zone, a karyotype with a higher number of acrocentric chromosomes and that may hybridize with other karyotypes appears to be favourably selected (Searle, 1984a, 1986). In this paper, we present results obtained by biochemical and karyological analyses of two other taxa of the Sorex araneus complex, Sorex coronatus and the Vaud race of Sorex araneus (Hausser et al., 1986). These taxa are considered to be