I think it's important to distinguish two very different processes. The one that I and my colleagues have worked on mainly is a process called transformation, which only goes on in some kinds of bacteria. It does go on in both neisseria and streptococcus, which are the ones we're working with. This really does look like a highly evolved process for transferring bits of genes between closely related bacteria. You have really to be a pretty close relative before you can do it, and it's very common. Gonococcus, the organism that causes gonorrhoea, was shown by a young colleague of mine, Maria O'Rourke, to be essentially random mating. I mean, the amount of frequency of exchange of bits of DNA between gonococci is so frequent that their genes are in what we call linkage equilibrium, it's complete random assortment from what genes you have at different loci. It says something about not only their sexual habits but ours. Of course, you'll appreciate that recombination only matters if you've got gonococcus from two different sources, otherwise it's just having sex with yourself, and that doesn't have any...

[Q] So they constitute what you might call a gene pool, in a real sense?

That's right, that really do. But that's one kind of recombination. It's one I'm very interested in, because I'm interested in how did it evolve and so on. The other is a much more accidental thing. What happens is that there are these what are called bacteriophages, they're sort of viruses that live inside, say E. coli, as a disease. And they may destroy the E. coli. And when they destroy the E. coli they make a new protein coat, and they package their DNA in the protein coat and then they get into another bacterium. Occasionally they make a mistake, and instead of packaging their own DNA, they package E. coli DNA, and if that happens, then they carry E. coli DNA into whatever the next cell they attack is. And this is a much more accidental procedure, it's much less common. It transfers great big hunks of DNA sometimes. And the... the donor and the recipient don't even have to be particularly closely related for it to work. And both these processes are going on, and other ones as well, it's all very complicated.

[Q] So could you almost say that the chunks of DNA that are available to all the bacteria in the world constitute a kind of giant gene pool that they're constantly swapping in and out of combinations with each other?

Almost you can. And this was sort of the football team analogy, the idea of E. coli is that it just takes in DNA somewhere when it needs it. And it can buy it from a team in Germany if it wants to, you know, it doesn't have to stick to another English team. Indeed, this led some people... there's a... I think he's a Belgian microbiologist called Sonea to say that we really ought to think of the bacterial world as a sort of super-organism, and what was evolving is not the individual bacterial species but the organism as a whole. I have to say, I think this is complete nonsense, because it's, you know, evolution by natural selection requires that you have a population of things competing with one another, and you can't talk about natural selection operating on a single whole. But nevertheless, he was on to a point that there is this whole field of creatures which are exchanging genetic material.