Ancient species flocks and recent speciation events: what can rockfish teach us about cichlids (and vice versa)?

The large assemblage of northeastern Pacific species of rockfish (genus Sebastes) has been proposed to be the modern result of an ancient explosive speciation event (Johns and Avise 1998). This radiation has been compared to the more recent radiation of cichlids in the Great African lakes (Greenwood 1991), and by analogy, the Sebastes assemblage was termed the “ancient species flock” (Johns and Avise 1998). An increasing number of molecular evolutionary studies focusing on cichlids from the Great African Lakes have uncovered unique modes of speciation. Sexual selection has been shown to play a crucial role in all three Lakes. In the oldest lake, Lake Tanganyka, sexual selection was evoked to explain the comparatively large genetic divergence observed between congeneric Tropheus species (Sturmbauer and Meyer 1992). Indeed, these species did not show any morphological differences besides coloration, although there was significant genetic divergence at the mitochondrial control region level [14.5% sequence divergence (Sturmbauer and Meyer 1992)]. On the other hand, haplochromine species from the most recent lake, Lake Victoria, and the species flock from the evolutionary intermediate lake, Lake Malawi, were found to exhibit a vast array of color patterns but little genetic divergence (Crapon de Caprona and Fritzsch 1984; Meyer et al. 1990; Seehausen et al. 1997; McElroy and Kornfield 1990; McKaye et al. 1993; Deutsch 1997). Furthermore, lineage sorting was shown to be incomplete in the Malawi species flock, locally called mbuna, using mitochondrial control region sequences (Bowers et al. 1994; Kornfield and Parker 1997; Parker and Kornfield 1997). Recent microsatellite studies, however, show that mbuna species could be distinguished using fast-evolving molecular markers (Kornfield and Parker 1997; van Oppen et al. 1997; Markert 1998; Arnegard et al. 1999). Thus, the relationship between speciation and divergence at the morphological and genetic level in cichlids (as in most organisms) is not simple (Meyer 1993). While few morphological differences besides coloration patterns were found in Victoria haplochromines, Malawi mbunas (within genera) and Tanganyka Tropheus species, levels of genetic divergence at the mitochondrial control region level were either large—14.5% sequence divergence for the genus Tropheus—or very limited— 2.3 and 8.2% sequence divergence for the whole Victoria and Malawi species flocks, respectively. A similar situation appear to exits in the Sebastes species flock, where examples of species groups that exhibit little or no morphological difference also exist [i.e., Sebastes rubrivinctus and S. babcocki; S. rosenblatti, S. eos, and S. chlorosticus; S. carnatus and S. chrysomelas; S. helvomaculatus and S. simulator (Rocha-Olivares et al. 1999a,b)]. What levels of genetic divergence should we expect in these species? If these species groups have been maintained for an extensive period of time, as the idea of an ancient species flock would suggest, we Correspondence to: Giacomo Bernardi; e-mail: bernardi@biology. ucsc.edu J Mol Evol (1999) 49:814–818