A genic view of species integration

Major botanical advances often suffer the worst of fates ± the scienti®c community ignores them until they are independently `discovered' or con®rmed by zoologists. For example, the principles of inheritance, initially worked out in the garden pea, were not generally accepted until veri®ed in animals (Bateson, 1902). Likewise, the discovery of mobile genetic elements in maize by McClintock (1950) was largely ignored until similar observations were made in bacteria, yeast and ies (Bukhari et al., 1977). McClintock (1987) has commented on her disappointment upon receiving only three reprint requests for an important early paper reporting this discovery. More recently, botanical claims for a prominent role for major genes in adaptation (Hilu, 1983; Gottlieb, 1984) were received with scepticism (Coyne & Lande, 1985) until similar ®ndings in animals prompted a re-evaluation (Orr & Coyne, 1992). The `genic view of speciation' represents another example of a long and widely held view by botanists that is ®nally being given serious consideration by the zoological community. Obviously, we are grati®ed to see this occur and thank Chung-I Wu for acknowledging the botanical perspective. The genic view of speciation in plants had its roots in early observations that plant species often exchange genes without merging (Lotsy, 1925). Groups of species connected by hybridization were termed `syngameons' (Lotsy, 1925) and became a favourite topic of study by plant evolutionists (Anderson, 1949; Stebbins, 1959; Grant, 1981; Rieseberg, 1997). These observations, combined with evidence for ecologically mediated parapatric speciation (Antonovics, 1968), led botanists to emphasize the role of differential adaptation in the origin and maintenance of species differences (e.g. Clausen, 1951; Macnair et al., 1989; Levin, 2000). Selectively maintained species differences must directly or indirectly reduce gene ow (Barton & Hewitt, 1985), and even partial reproductive isolation (RI) may permit divergence at weakly selected loci that otherwise would be homogenized by gene ow. Although we have highlighted early botanical contributions to this discussion, zoologists have also questioned the requirement for genome-wide isolation. Students of hybrid zones, for example, have long recognized that species boundaries may be semipermeable to introgression (Key, 1968; Barton & Hewitt, 1985), with the movement of alleles dependent on their selection coef®cients and linkage relationships. What is perhaps most striking about the Wu review, then, is not its novelty, but rather its derivation from the Drosophila speciation community, which until very recently had seemed largely satis®ed with the concept of whole genome isolation. These comments are not intended to minimize the contributions of Wu's review, which represents an important synthesis of many new lines of evidence that call into question the whole genome view of isolation. Of particular interest are data from gene genealogies, which provide compelling evidence both for the mosaic nature of diverging genomes and for the small size of chromosomal segments in uenced by isolation genes. Two issues not addressed by Wu, but which have important implications for a genic view of speciation, concern (1) the potential role of factors that modify recombination rates, and (2) the mechanisms that hold species together.