Evolution
Darwin's rotifers
Mar 22nd 2007
From The Economist print edition
An obscure group of invertebrates casts light on how new species form
WHEN Charles Darwin opened his first notebook on the subject of how organisms change over time, the field was not even referred to as “evolution”. It was, rather, “the species problem”—in other words, how did life's variety arise? Darwin showed in detail how life changes over the course of time by the process of natural selection, but failed to explain how those changes can take different courses, dividing a species in two and thus multiplying the number of species.
The two main schools of thought about speciation that emerged in the 20th century to fill this intellectual vacuum can be summed up as “breeding defined” and “niche defined”. The former argues that the reason members of a species resemble one another is that they are constantly swapping genes. The latter says it is that they are all adapted to live in the same ecological niche. The two explanations are not mutually exclusive. But it is only by looking at an asexual group that the second can be tested independently of the first.
Few creatures—even bacteria—do without sex entirely, but bdelloid rotifers are among that minority. As far as anyone can tell, these microscopic freshwater animals have not exchanged genes, one with another, for 100m years. This asexuality seems to have arisen only once, yet biology recognises about 380 species of bdelloid. If these species are not defined by interfertility, which obviously they are not, are they real, or just the convenient conventions of taxonomists? If they are real, it provides one clear answer to the species problem. Which is why Tim Barraclough and his colleagues at Imperial College, London, spent three years collecting bdelloids from ponds and streams in Britain and Italy. The results of this international small-game hunt have just been published in Public Library of Science Biology.
Dr Barraclough and his team classified their trophies in three ways. One relied on the DNA sequence of a gene called cytochrome oxidase. The second applied the same method to another gene called 28S ribosomal DNA. The third used an anatomical technique—the dimensions of the mouthparts—that is also applied to the finches of the Galapagos Islands which, tradition has it, set Darwin thinking about the species problem in the first place.
One advantage of these three ways of classifying things is that the results of each can be quantified and processed statistically using a technique called cluster analysis. Doing so showed Dr Barraclough three things. The first was that the three variables clustered similarly, suggesting that they were accurate markers of ancestry. The second was that they clustered in a way that more or less agreed with the traditional taxonomy of the bdelloids. The third, and most significant, was that they clustered in a particular way. The clusters had what are known in the argot of the trade as deep roots. In other words, individuals within a cluster were very much more similar to each other than they were to those of other clusters. That, in turn, suggests something is acting to keep them that way, by eliminating intermediate forms.
Specific information
The only plausible candidate for that something is natural selection to fit particular niches. Different niches, having different requirements, result in different genes and different shaped mouthparts. Indeed, the team has one excellent example of this. Just as three sorts of body lice, specialised to be able to grasp different sorts of hair, live on humans, so water lice are host to two sorts of bdelloid. One sort live on the creature's legs, the other on its body. Dr Barraclough's classification shows that these two apparent species are, indeed, real and evolutionarily distinct, even though they share a host.
The niche model of speciation thus looks proved. That does not rule out a gradual drift to infertility as an additional explanation, but makes it possible that infertility is a consequence, rather than a cause of speciation. In that case, there is probably active selection against hybridisation, since hybrids will be neither one thing nor the other, and thus not fit for the niche. But that, best beloved, is a subject for a whole different research project.
Advertisement
Advertisement
