yet at the same time create little opportunity for spatial isolation between tenants of alternative peaks in the adaptive landscape, causing a situation in which a panmictic population experiences disruptive selection. Evolutionary response to such disruptive selection is constrained by homogenizing gene flow and indeed good examples of sympatric speciation are still rare. Yet, while most taxa fail to speciate altogether in such circumstances, some radiate into many new species with predictable regularity. Despite one and a half centuries of speciation research, surprisingly little is known about the causes of this differential response. Absence of radiation may be because populations of many taxa fail to adapt to disruptive selection in the face of gene flow, but alternatively, taxa that do not radiate may not experience disruptive selection to begin with, or they may adapt through mechanisms that do not require speciation. Few studies have explored these alternatives in the same system. In this project we are doing just that. While some haplochromine cichlid lineages have repeatedly radiated in lake environments, others consistently fail to diversify. This is an ideal system to study constraints to speciation. We investigate pairs of sister species that have diverged between contrasting habitats and other pairs of populations that have not speciatied but occupy the same or similar habitat contrasts. We estimate fitness functions in the wild for phenotypes and genotypes in contrasting habitats and adaptive landscapes for populations. We are using Next Generation Sequencing technology to sequence genome-wide markers for a large sample of these individuals. This will allow us to measure the total magnitude and genomic distribution of response to disruptive selection between habitats in cases with and cases without speciation. Third, are making laboratory crosses between speciated and non-speciated populations to map traits that are under selection. Combining these approaches will allow us to test predictions of alternative models of adaptation and speciation with gene flow.
