Search:

Speciation in the Pelagic Zone by George F. Turner

Speciation is the process where new species are formed. For the last 60 years, most biologists have believed that speciation in animals took place when populations were split apart by some kind of geographic barrier. This would allow the genetic make-up of the isolated populations to change independently, either in response to the different conditions in their local environments, or by sheer chance. Later, when the barrier disappeared, and these populations met again, they might have diverged enough that they would no longer interbreed. If so, they would have become separate species. This kind of speciation is known as ‘allopatric’, meaning ‘other homeland’.

During the last 10 years or so, many people have begun to doubt that all speciation in animals had to be allopatric. Partly, this has come about because computer models have shown many circumstances in which speciation might take place without geographic isolation. But, also evidence from studies of the DNA of wild animal populations has come up with a few surprising results. The form of speciation without previous geographic isolation is known as ‘sympatric’, meaning ‘same homeland’.

Along with colleagues at the University of Hull, my research group has recently been working on this problem, investigating the pelagic cichlids of Lake Malawi. Some of our results have recently been published (Shaw et al. 2000).

The pelagic cichlids are not terribly well-known in the aquarium trade. Occasionally, small specimens of Rhamphochromis ferox are offered for sale, almost always wrongly labelled as Rhamphochromis macrophthalmus. The eight species of sleek silvery Rhamphochromis are active predators of fish or zooplankton. There are also at least 12 species of the genus Diplotaxodon, which are mostly zooplankton feeders or more sluggish predators. And there is the bug-eyed, huge-mouthed Pallidochromis tokolosh, which I described in 1994. So, there are at least 21 species of cichlid specialised for living, hunting and probably breeding in the open waters of Lake Malawi.

I first became interested in these fish when I was working on the fisheries of Lake Malawi in the early 1990s, and discovered that there were a great many species of these fish, and that they were very important in local fisheries. I was concerned that they were being heavily exploited for food, but yet virtually nothing was known about them. The UK Department for International Development agreed to fund a research project to investigate these fishes, so that we could assist the local fisheries departments in drawing up plans to help ensure that these fishes could be exploited on a sustainable basis. That can be the subject of a later article.

In the course of these studies, we collected many thousands of these fishes. All 21 species seemed to be found all around the lake, wherever there was suitable habitat for them. This is a complete contrast with the well-known mbuna, where many species are only found around a single island or area of rocky coastline.The mbuna hate to leave the rocky habitat, and rocky habitats are found in patches separated by areas of sand or deep water that mbuna rarely cross. So, it has always seemed likely that mbuna could have speciated by the usual allopatric mechanism. But what about the pelagic cichlids?

I can think of four possible ways they could have evolved by allopatric speciation. Rule these out and we are left with sympatric speciation.

First of all, we all know that water levels of the African Great Lakes have changed dramatically over the last few hundred thousand years. Lake Victoria was completely dry 12,500 years ago, and the level of Lake Tanganyika fell so far that it was split into 3 separate basins. Could something similar have happened in Lake Malawi, and isolated the pelagic cichlids into separate basins? A quick look at the contour map of the bottom of Lake Malawi shows this is very unlikely. It would have taken a drop of almost 500m to split the lake into separate water bodies. There is no evidence that this has ever happened. Even if it had, it would need to have happened at least 6 times times to allow 21 species to evolve from one, and even then we’d have to make some pretty implausible assumptions. Ever time the lake level fell, every species would have been split in two. And every time a species was split in two, it became 2 new species- this just doesn’t happen in the real world. Every species in the UK is isolated from every species on the European mainland and NONE of them have evolved into new species. Usually, this kind of speciation takes a long time, if it happens at all. Not only that, but this estimate of 6 isolation periods is assuming that none of these new species has ever gone extinct. So, realistically, we’d have to imagine many, many more massive water level falls. So, I think we can probably rule this one out.

What about lake level rises? Higher water levels could have flooded the surrounding area and as the floods retreated, small populations of cichlids in pools or small lakes could have been cut off from the main lake. These could then evolve into new species. This might seem more plausible, because many pools could be created with a single rise in water levels, so we might not have to imagine too many different floods. And the rises in water level could have been fairly small. Well, there are small pools and lagoons around the main lake at the moment, and I’ve visited some of them. Babies of two of the Rhamphochromis species occasionally turn up in shallow bays or even in lagoons which are still connected to the main lake. But, there are no signs of Diplotaxodon in any of these lagoons. Indeed, in all the hundreds of hours I have spent diving in the lake, I have never seen a single Diplotaxodon underwater, and I don’t know anyone else who has either. They just never come into shallow waters along the shores. So, the idea of them being repeatedly stuck in pools around the lake seems pretty far-fetched. Even more far-fetched is the idea that they would survive in these pools and evolve into new species. And even more bizarrely, we'd have to imagine that when they eventually got back into the main lake, they re-evolved into offshore living deep-water species that never go anywhere near swampy pools!

The third possibility is that these fish see habitat barriers that we don’t. Maybe, water currents keep them separated. We can test this, by looking at the the fish’s DNA. We have used bits of DNA called 'microsatellites', that mutate very quickly. If you look at a bunch of different microsatellites every individual has a different combination. This makes it a very useful tool for comparing species and populations. Take samples of a lot of individuals, say 50, from 2 different places in the lake. Look at a few different microsatellites, say 6-10. If the fish in the two places are regularly interbreeding with each other, there will be no statistical differences between the samples. What's more, if they are interbreeding with other populations in between, which are in turn interbreeding with more distant populations, then there will still be no differences between them. When we collected samples of three Diplotaxodon species from all over the lake, that’s what we found- no differences. So the fish at the far north end of the lake are exchanging genes regularly with the fish at the far southern end. There are no habitat barriers to these Diplotaxodon.

The last possible option for allopatric speciation is that the pelagic cichlids are not really related to each other. Cichlids are famous for their convergent evolution. Petrochromis in Lake Tanganyika look like Petrotilapia from Lake Malawi. There is a Lake Victoria cichlid with big rubbery lips like the Malawian Cheilochromis euchilus or the Tanganyika Lobochilotes labiatus. These fish have evolved similar body structures because they have adopted similar ways of life. Could the same be true for the pelagic cichlids? If so, each one of the 21 species might have evolved from an ancestor living on the shore. We already know that inshore cichlids are split up into geographically isolated populations. It is easy to imagine allopatric speciation producing these species, which then each independentlybecame adapted to live offshore. We can test this by making a phylogeny of the Malawian cichlids. A phylogeny is a kind of ‘family tree’ of the evolutionary relationships of species. The difference is that in family tree, it takes two parents to make one (or more) offspring. In a phylogeny, one species splits into two ‘daugher’ species. If the convergent evolution theory is true, we should find that each pelagic cichlid species will have an inshore-living species as its ‘sister’, although we couldn’t rule out the occasional case where both sister species separately evolve from an inshore to an offshore lifestyle. If the sympatric theory is true, most pelagic species will have other pelagic species as their sisters. So, how do we build a phylogeny? Again, the answer is DNA. The best bit to use depends on how recently the species have evolved. In this case, we used two different bits of the mitochondrial DNA. The results were clear-cut. All of the pelagic cichlids have other pelagic cichlids as their sister species. All of them were descended from a single species, and all of the surviving descendents of that species are pelagic cichlids. The last allopatric option had been ruled out, and we are now pretty confident that these fishes must have evolved by sympatric speciation.

So, what’s the fuss all about? First of all, there are very few convincing examples of sympatric speciation in any kinds of animals. So, this result will be a surprise to a lot of biologists, who thought that sympatric speciation hardly ever happened, if it happened at all. Second, it means that we can rule out some kinds of processes that might have operated in allopatric speciation. As I said at the beginning of this article, geographically isolated populations might evolve into new species by chance. This can’t possibly happen in sympatric speciation. So we have to look to Darwin’s mechanisms of natural and sexual selection. This means we might get closer to finding out just exactly why there are so many species of cichlids in the African Great Lakes. Finally, if Malawi cichlid speciation was caused by some of the possible allopatric mechanisms, like dramatic rises and falls of water levels, then it all happened a long time ago, and we can only guess at how it all happened. But, if it is has been happening by sympatric speciation within the lake, then speciation is almost certainly still going on. This is a tremendously exciting idea, because it means we might just strike it lucky and see new species at the moment of their birth.

Reference:

Shaw, P.W. Turner, G.F., Idid, M.R., Robinson, R.L. & Carvalho, G.R. (2000). Genetic population structure indicates sympatric speciation of Lake Malawi pelagic cichlids. Proceedings of the Royal Society of London (Biological Sciences). 267, 2273-2280.