When The Origin of Species was first published, biologists mostly thought species were easy to recognize - they looked different from each other, and they couldn't successfully interbreed with each other. This view was a problem for Darwin's ideas about gradual evolution by natural selection, since gradual divergence shouldn't give rise to nice, discrete species. In fact, as Darwin argued, different groups of organisms exhibit a whole spectrum of reproductive isolation, from complete interfertility to total isolation - and the degree of isolation is not easy to predict based on how similar organisms look. In Darwin's description, species are just labels that humans put on clusters of similar-looking organisms.
By the mid-Twentieth Century, evolutionary biologists favored what is commonly called the Biological Species Concept (BSC), defining species as non-interbreeding populations of living things. Research on speciation has accordingly focused on the ways that evolution creates reproductive isolation between populations. Mallet argues that this amounts to an abandonment of Darwin's insights, and that by focusing on isolating mechanisms, biologists have returned to viewing species as distinct, "real" entities, missing much of the evolutionary process as a result.
I'm not sure I believe the distinction that Mallet makes between Darwin's description of species and the BSC; they seem to me more different in their emphasis than in their fundamentals. Darwin was interested in demonstrating that species arise gradually, as accidents of adaptation to different environments - and, as Mallet says, he was trying to overcome the then-predominant view that species were real, discrete entities instead of the names that humans assign to clusters of similar organisms. Research motivated by the BSC generally takes this view as well, but it's interested in the processes that create such clusters, and can prevent them from merging into nearby clusters by interbreeding.
Research on the evolution of isolating mechanisms necessarily focuses on cases where isolation is incomplete, somewhere between complete speciation and free interbreeding. A prime example is my lab's research on the two pollinator-associated types of Joshua tree, Yucca brevifolia. It's not clear that the two types are reproductively isolated - preliminary genetic data suggests they're not [PDF] - even though they're pollinated by different moth species, and classified as separate subspecies, the taller Y. brevifolia brevifolia and the short, bushy Y. brevifolia jaegeriana. They may be on the way to becoming different species, but they're not there yet. Two other examples out of the endless forms available: marine snails that choose mates by their slime trails, and wildflowers that would interbreed if only they could survive each other's habitat.
Two types of Joshua tree
Photo by jby.
Two types of Joshua tree
Photo by jby.
As Mallet concludes in the more empirical part of his review, this is what we see across the diversity of life: a continuum of reproductive isolation between populations, not a granular world of neatly divided, obviously different species. Rather than over-simplifying this reality, the Biological Species Concept gives us a framework through which to understand it.
Darwin, C. 1859. On the Origin of Species by Means of Natural Selection. First ed. London: John Murray. Full text on Google Books.
Mallet, J. (2008). Hybridization, ecological races and the nature of species: empirical evidence for the ease of speciation Phil. Trans. R. Soc. B, 363 (1506), 2971-86 DOI: 10.1098/rstb.2008.0081
Smith, C., W. Godsoe, S. Tank, J. Yoder, & O. Pellmyr (2008). Distinguishing coevolution from covicariance in an obligate pollination mutualism: asynchronous divergence in Joshua tree and its pollinators. Evolution, 62 (10), 2676-87 DOI: 10.1111/j.1558-5646.2008.00500.x