Butterfly ball: How species separate is still mysterious. Lepidoptera make things clearer

Butterfly ball: How species separate is still mysterious. Lepidoptera make things clearer

Nov 2nd 2013 |From the print edition

Pinning down the truth

BIOLOGISTS love Heliconius. In the 19th century, the mimicry by edible butterflies and moths of the brash warning colours sported by members of this genus (Heliconius butterflies themselves are usually toxic) was one of the first-noted and best examples of natural selection at work. And the genus is still doing duty in evolutionary biologists’ laboratories. Marcus Kronforst of the University of Chicago, for example, is using it to understand how speciation happens.Despite its ambitious title, Charles Darwin’s master work did not really explain “the origin of species”. Rather, it explained how species change, which is not quite the same thing. For new species to originate, that change must bifurcate. This can happen if a population gets physically divided—perhaps by an alteration in the climate making parts of its range uninhabitable. But that does not seem enough to explain all the biodiversity seen today, which suggests that undivided populations speciate too. Dr Kronforst and his colleagues have been trying to work out how this happens, at least in Heliconius. They have just published their results in Cell Reports.

There are about 45 species of Heliconius, and the evidence suggests they have diverged recently—not least because some of them can still interbreed. Though they rarely do so in the wild, rarely is not never. And that provided the key Dr Kronforst used to unlock what is going on.

If species cannot interbreed at all, it is hard to tell which genetic differences between them are substantial to their identities, and which are accidental. But interbreeding, even if occasional, homogenises the accidental differences, so that only the substantial ones are left.

Dr Kronforst and his colleagues looked at three species of Heliconius from Costa Rica, whose ranges overlap to varying degrees, and which are all known to be capable of interbreeding, again to varying degrees. Modern genetics, in the form of the Heliconius Genome Project (yes, there is one), has provided a reference genome to which others can be compared and contrasted, which Dr Kronforst did.

Two of the three are reckoned to have separated about 500,000 years ago, and it was with these that the study began. The researchers found 12 places where the genomes of these species differ in specific ways. All 12 are relatively short stretches of DNA, and eight of them are in places already established as carrying genes for wing colour and pattern, or mate recognition, or both.

The third species, which diverged from the ancestral stock 1.5m years ago, shows more differences. That is hardly surprising. What is surprising is the pattern of those differences. First, there are lots of them—over 100 times as many. That is far more than would be expected if differences accumulate at a steady rate. Second, the pattern of speciation seems to have begun in the same way, for the same dozen initial places are found in this species, too. Only after they had come about did evolution get to work on genes for vision, hearing and body-shape—all important for identifying who is and who is not a conspecific.

The upshot, at least in Heliconius, seems to be that speciation is a two-stage process. First—and slowly—enough differences accumulate in some crucial variable that serves to isolate part of the population behaviourally. Wing patterns are important for mate recognition, so if some butterflies have one sort and others another, they will tend to stop mating with each other.

Then, when that separation is reasonably well established, there is a tipping-point. The two groups go their own ways, evolutionarily speaking, and lots of other differences accumulate rapidly.

Colour coded

The question is, what causes the wing patterns to change in the first place? And for that, you have to go back to mimicry—but a different sort of mimicry from the one in which edible mimics hide among poisonous models. In this second type of mimicry, poisonous species come to resemble each other, so that their predators have fewer patterns of warning colouration to learn. This type of mimicry is also widespread inHeliconius, and Dr Kronforst thinks it is what is driving the initial changes in wing patterns in the species he has been studying.

How applicable all this is to speciation in other groups, where mimicry does not provide the impetus, is unclear. But history suggests it may be. Both the first type of mimicry, discovered by Henry Bates, a British naturalist, and the second type, discovered by Fritz Müller, a German, helped establish Darwin’s ideas. It would be a neat conclusion to the story if understanding precisely how mimicry evolves clarified what actually is the origin of species.