Despite the title of Charles Darwin\'s most famous work, the mechanisms whereby species originate are poorly known. The textbook view is that populations of a species become physically separated, so that individuals are denied the opportunity of interbreeding. Over time, the genetic idiosyncrasies of each population will increase until they become separate species. However, it is increasingly clear that many species diverge from parent populations without any kind of intervening barrier. Understanding how this \'sympatric speciation\' works is an abiding challenge for evolutionary theorists. Two models of sympatric speciation appear in this week\'s issue.
In very simple models, it is easy to produce separate species if fitness and mate choice are determined by the same, or linked, genetic loci. This leads to selection for distinct forms, or phenotypes, with the rapid extinction of intermediates. However, this may not be a very realistic simulation. On p351, Alexey S. Kondrashov of the National Institutes of Health, Bethesda, Maryland, and Fyodor A. Kondrashov of Simon\'s Rock College, Great Barrington, Massachusetts show that sympatric speciation should still be possible even if the loci for fitness and mate choice vary separately. In these circumstances, the increase in the number of variable genetic loci affecting fitness facilitates sympatric speciation, whereas the increase in those influencing mate choice has the opposite effect. These predictions could be used to identify real-life cases of sympatric speciation.

Ulf Dieckmann and Michael Doebeli of the Institute of Systems Analysis, Laxenberg, Austria, and the University of Basel, Switzerland, tackle the same issue in a separate report on p354. They show that sympatric speciation is a likely outcome of competition for resources. Their model describes assortative mating (in which individuals tend to choose mates that look more like themselves) dependent on either an ecological character affecting resource use, or on a selectively neutral trait. In both cases, the evolution of assortative mating often leads to reproductive isolation between ecologically diverging subpopulations.

Tom Tregenza and Roger K. Butlin of the University of Leeds, UK, discuss the research in an accompanying News and Views article.