Our paper reviews an extremely active area of evolutionary research, the mechanisms by which new species arise. We relate recent theoretical and empirical studies—on the order of 100 papers published in the last 20 years—to ideas developed since the late 19th century. We focus on two issues: the role of geographical isolation in speciation and the burgeoning literature on mathematical models of speciation. Important new empirical and theoretical studies have revived interest in species originating without any geographical isolation ("sympatric speciation"). In the mid-19th century, both Darwin and Wallace conjectured that new species could arise when natural selection simultaneously favors alternative types in a given location. However, many subsequent naturalists argued against this on the grounds that most closely related species pairs have disjunct ranges, suggesting that geographical isolation typically contributes to the development of reproductive isolation ("allopatric speciation"). The empirical evidence suggesting allopatric speciation was synthesized by Ernst Mayr in his highly influential 1942 and 1963 books, and the near-universality of allopatric speciation became the reigning paradigm. Recent data documenting apparent sympatric speciation, adaptive divergence of mating preferences between incipient species that create relatively unfit hybrids ("reinforcement"), the genetics and patterns of hybrid inviability and sterility, divergent selection driven by competition for resources, and the importance of sexual selection in speciation have led to a renaissance of both theoretical and empirical speciation research. Our paper attempts to synthesize the verbal and mathematical theories of speciation and present them in the context of both new and old empirical findings. We hope it provides a convenient entry into this literature and indicates areas of theoretical and empirical research that are likely to be productive.

  Does it describe a new discovery or new methodology that's useful to others?

We do not present new ideas but instead provide: a modern perspective on the continuing importance of the biogeography of speciation, an overview of a large mathematical literature addressing speciation, and some suggestions for future research.

  What were some of the circumstances that led you to do this research?

One of the great pleasures of being a theoretician is the ability to change research areas regularly by collaborating with a wide range of colleagues. Coyne, Barton and I (known collectively as "The Three Bears") wrote our first paper together in 1996, a critical synthetic review of the lack of empirical or theoretical support for Sewall Wright’s "shifting balance theory" of adaptation (which requires an implausible, i.e., "just right" balance of forces). Since the early 1980s, Coyne and Barton have played important roles in rejuvenating empirical and theoretical speciation research. In contrast, I began working in this area only in the mid 1990s. Hence, when I was asked to review mathematical work on speciation for a special issue of Trends in Ecology and Evolution, it was natural to ask my more experienced pals for help. We quickly realized that the mathematical theories could be usefully evaluated only in the context of both verbal theories and empirical results.

  Could you summarize the significance of your paper in layman's terms?

Although essentially all evolutionists believe that natural selection is the dominant process underlying the origin of adaptations, the relative importance of various processes that can produce new species remains uncertain. For instance, the relative importance of natural versus sexual selection and the role played by geographical isolation remain topics of lively debate. In contrast, considerable progress has recently been made in understanding the genetics of hybrid inviability and sterility and the forces that produce repeated patterns of hybrid breakdown. Moreover, we now have compelling empirical evidence for both sympatric speciation and reinforcement. Modern analyses of speciation are greatly facilitated by the increasing availability of complete species-level phylogenies based on DNA sequence data. Data on complete clades (i.e., collections of all descendant species from a common ancestor) allow us to replace anecdotal accounts of specific speciation events with large-scale comparative analyses that are likely to indicate more reliably the relative importance of different factors in different groups. Such comparative analyses, complemented by field and laboratory work and mathematical analyses, are likely to greatly expand our understanding of speciation over the next decade.

Michael Turelli, Ph.D., Prof. of Genetics,
Chairperson, Section of Evolution And Ecology (EVE),
Division of Biological Sciences, U.C. Davis