I will break up my discussion of the paper below into two parts that will appear today and tomorrow. This is because I want to avoid a single long post that may put off readers. I give references to all the papers mentioned at the bottom of the post.
All evolutionists agree, and the data show, that nearly all new species form as descendants of what were populations of a single ancestral species. (Occasionally new species, especially in plants, form after hybridization of two pre-existing species.) One of the biggest controversies in my own field of speciation is this: can new species form in one area without any geographic isolation of populations (“sympatric speciation”), or is a period of partial or full geographic geographic isolation necessary (“parapatric” or “allopatric” speciation, respectively)? (I’ve simplified the meaning of these terms a bit.) While theory shows that geographic isolation facilitates the development of reproductive isolating barriers between populations (sexual isolation, hybrid sterility, and so on) that are the sine qua non of speciation for most biologists, some theory also suggests that geographic isolation is not necessary: under special conditions, new species can form in one area in situ.
The data, summarized in my book Speciation with Allen Orr (now a decade old), suggest that geographic isolation is usually necessary, but there are a few cases implying speciation without any geographic isolation. These are hard to demonstrate unequivocally, largely because closely related species that now live in the same area could have speciated in allopatry (different areas), and then come into secondary contact after the reproductive barriers evolved in isolation. Since it’s harder to form species sympatrically, to demonstrate this process one must rule out that ancestral populations were ever isolated geographically. Since that’s hard to do (speciation takes thousands to millions of years to complete), convincing cases are rare.
In Speciation (pp. 142-143), Allen and I laid out criteria for showing convincing cases of sympatric speciation. They include the presence of sister species (each other’s closest relatives) in the same area; the demonstration that these are indeed “good” species (i.e., they are distinct groups that never or rarely exchange genes); the demonstration that their status as each other’s closest relatives does not come from hybridization between more distantly-related species (this would homogenize their genomes and make them look closely related when they really aren’t); and the hardest bit: showing that these those sisters species descend from populations that were never geographically separated. That’s the biggest issue, because when you see two closely related species living in the same area, how can you convincingly show that their ancestors always lived in the same area?
Allen and I decided that one of the best situations for meeting these criteria occur on islands: either oceanic islands (islands like Lord Howe or St. Helena that were formed without life on them, usually as volcanoes that rose above the sea), or “habitat islands”: isolated patches of habitat that have existed for a long time. (Landlocked lakes are one example.) If you could show, for instance, that on one such island you find two or more sister species that do not occur elsewhere (i.e., are endemic to that island), then that would be pretty strong evidence that those species had formed sympatrically on the island, descending from a common ancestor that invaded the area long ago.
Trevor Price and I tested this theory by looking for endemic sister species of birds on oceanic islands. In a paper published in Evolution in 2000, we found not a single such case on 46 isolated oceanic islands, implying that sympatric speciation was rare in birds. (If it didn’t occur on islands, it is unlikely to occur on continents.) Further work by Yael Kissel and Tim Barraclough in 2010 showed the same situation in several other groups, including lizards, mammals, and flowering plants. Sister species on islands were observed only when the islands were so large that geographic barriers were likely to have been present. This further implies that sympatric speciation is rare, at least in those groups studied.
In contrast, though, work by Papadopulos et al. on the flora of Lord Howe Island (a small oceanic island between Australia and New Zealand, with an area of about 15 square kilometers) shows the existence of sister species in about nine groups of plants, most notably two species of endemic palm trees that are wind pollinated. Since the sister species in these groups are found nowhere else, they likely formed on the island. This, too, seems a pretty good case of sympatric speciation.
But in vertebrates we have only a couple of cases—all involving fish—that point to sympatric speciation. These cases involve species living in small lakes that fill the craters of extinct volcanoes—”crater lakes”. In 1994, Ulrich Schliewen and his colleagues described groups of closely related cichlid species, each group descending from a single common ancestor, that inhabited crater lakes in Cameroon. (Such cases have since been described in lakes in Nicaragua as well). Lake Barombi Mbo, only 2.3 km across, contains a group of 11 “monophyletic” cichlid species (descended from a single invader), while Lake Bermin, only 0.7 km across, has a monophyletic group of 9 cichlids. Here are the two lakes:
And here are the putatively monophyletic species flocks, shown in the paper of Martin et al. mentioned below. (There are only 10 species shown for Lake Barombi Mbo because the authors sampled only 10 of the 11 for genetic markers.)
Because the lakes are no longer connected to rivers that could carry in fish (lake Bermin has an outflow but no inflow); because they are small and uniform (so that raising or lowering the lake levels would not create isolated pools that could facilitate allopatric speciation); and because each small lake harbors a group of species from one putative invading ancestral species, this situation fulfills all four criteria we proposed for sympatric speciation. When we wrote our book, Allen and I considered this perhaps the best case of sympatric speciation in nature.
But that’s now in question. A new paper in Evolution by Christopher H. Martin et al. (reference and link below) genetically examined the radiations in these lakes and finds that sympatric speciation isn’t that likely after all. Today I gave you the background; tomorrow I’ll show you the results.
Coyne, J. A. and H. A. Orr (2004). Speciation. Sunderland, MA, Sinauer Associates.
Coyne, J. A. and T. D. Price (2000). Little evidence for sympatric speciation in island birds. Evolution 54(6): 2166-2171.
Kisel, Y. and T. G. Barraclough (2009). Speciation has a spatial scale that depends on levels of gene flow. Amer. Natur. 175: 316-334.
Martin, C. H., et al. (2015). Complex histories of repeated gene flow in Cameroon crater lake cichlids cast doubt on one of the clearest examples of sympatric speciation. Evolution 69(6): 1406-1422.
Papadopulos, A. S., et al. (2011). Speciation with gene flow on Lord Howe Island. Proc Natl Acad Sci U S A 108(32): 13188-13193.
Schliewen, U. K., et al. (1994). Sympatric speciation suggested by monophyly of crater lake cichlids. Nature 368: 629-632.