Progressive levels of trait divergence along a ‘speciation transect’ in the Lake Victoria cichlid fish Pundamilia

Abstract

Introduction and outline Identifying mechanisms of speciation has proven one of the most challenging problems in evolutionary biology, perhaps mainly for two reasons, speciation is not readily accessible to experimental approaches, and rarely to time series analyses. Any one case of speciation can usually be investigated only at a single stage of completion. Cases of parallel ecological speciation driven repeatedly within the same taxon by divergent selection along replicate environmental gradients, have therefore received considerable attention (Schluter & Nagel 1995; Rundle et al. 2000). Several such systems have become major model systems in evolutionary ecology research, including sticklebacks in postglacial lakes (Rundle et al. 2000), Heliconius butterflies (Mallet et al. 1998), leaf beetles (Funk 1998) and Timemawalking sticks (Nosil et al. 2002). They provide powerful means of identifying causes of divergence and may lend themselves to examining associations between variation in the environments and variation in the progress towards speciation (Nosil & Harmon, this volume). However, variation in the progress towards speciation among disconnected populations undergoing parallel speciation may be due to different contingency as much as different environments (Taylor & McPhail 2000). Ideally, to trace the correlates of the transition from panmixis to incipient speciation, one would want to study variation in the progress towards speciation in exactly the same pair of species, and along a continuous progress series, to minimize the potential confounding effect of variable historical contingency. The cichlid fish species Pundamilia pundamilia, Pundamilia nyererei, their hybrids and intermediate colour morphs occur at various stages of differentiation at different islands placed along an environmental gradient ofwater clarity in Lake Victoria. I will subsequently refer to this as a ‘speciation transect’. At the same time, populations occurring at different islands do exchange genes as evident from microsatellite-allele frequencies, and also from several observed cases of invasion of, and establishment in, previously unoccupied islands within the 15 years time window that I have collected community data for. Hence, I