The identity crisis of speciation genomics

Abstract Understanding how reproductive isolation (RI) evolves remains a central goal of speciation research. Quantifying RI is challenging, and researchers can rely on either organismal proxies (e.g. hybrid fertility or fitness) or genetic proxies (e.g. gene flow or genomic divergence). The increasing availability of genomic data has transformed the field, enabling inference of genomic divergence and patterns of gene flow across diverse taxa. As a result, speciation research has become increasingly dominated by genomic approaches. Here, we combine a literature survey with a reanalysis of publicly available datasets to evaluate the consequences of this shift. We show that genetic proxies commonly used to infer RI can be unreliable, as genomic divergence does not consistently predict the strength of reproductive barriers. We recognize that genetic proxies can, in principle, be applied broadly across taxa, and organismal proxies can be logistically challenging or infeasible to measure. We conclude by recommending: (i) the distinction between studying the mechanisms underlying RI (speciation) and the patterns associated with speciation (diversification); (ii) moving ‘back to basics’ by reinvigorating classic experimental designs, fieldwork, and population studies with genomic data; (iii) renewed tests of chromosomal models of speciation, (iv) stronger links between estimates of gene flow and both intrinsic and extrinsic reproductive barriers.