So I'd like to contrast that with my own experience when I entered science – if I might use myself; I hope, as a humble example. We had to memorize the Krebs cycle and the glycolytic cycle, the cycles having to do with metabolic energy and ATP so-called, its origins of biological energy, and they sort of had a neat – how shall I say – causal ring to them. But now, when we look at so-called signal transduction, what happens when something binds to a molecule on the surface of the cell, which starts up a whole bunch of chemical events inside the cell, which finally sends something to the nucleus, which finally turns on transcription and that goes back to the cytoplasm and does translation; things are no longer neat. The number of molecules just is mounting enormously, and there's a terrible danger which I believe, Ralph, you've mentioned, by saying everything seems to be connected with everything. Well, if it were that way, that would be cool because then it would be N2 and you know the answer. But in fact everything isn't connected to everything completely, but to enough somethings to make things very confusing, number one; number two degenerate – namely, we already know there are many different ways of achieving the same thing in a cell; and third of all, we don't have the methods as good as our methodologies to look at the cell as a dynamic processor within the 200 milliseconds... a fifth of a second at a time.

So we have an immense challenge in front of us, and the challenge is posed by what Darwin already knew – namely, you select, and it isn't the case that there's one gene, one protein, one disease. In fact, Ralph, you've worked on this issue. It's that that the gene networks are not fully understood and they are perhaps maybe not as complex as the neural ones I've talked about, but they're pretty darn complex and we do not understand all the possible equivalence. We don't know which gene in the context of which other gene is leading to a particular effect. Now it's very important, I think, to understand our predicament because if you don't understand what the question is, you'll never get the answer. So the question is: well, what is necessary and what do we have to give up? What notions do we have to quit by looking at this problem of complexity?

Well, one of them is, I think, the one gene, one protein, one disease idea, even though that may be extremely lucrative for... for a big pharma, or whatever... pharmaceutical companies. There's going to be a point at which we must do some reflection. My own... my own reflection on the subject – which is purely a question of an opinion, not as an expert – is that we are in a stage of molecular natural history. We're really not doing fundamental work in the full sense of what's possible; we are, because our techniques are so easy, just generating vast amounts of data in a system that is sort of recognizing that as if it were an end. But it isn't an end, and I have to remind you that Wallace was off in Malaysia collecting samples out of the jungle with endless numbers of variants. In just this way that finally led to what he did with the natural selection... what Darwin did. In the same way I'm hoping that it will be recognised that, in biology, the problem of the 21st Century is going to be: how do we cut through the complexity? Now, there is a concept that people have pushed, the philosophers of science and historians of science called emergence, and this is the idea that at a certain level of interaction in ensembles, new properties emerge that you don't see in the elements below. There's another related element which says that, even if you know what the element below is contributing, it's not the language to use. For example, protons and neutrons are made of quarks. Well, you don't do chemistry in terms of quarks. That would be just too complicated and... and wouldn't really yield anything. So there are really immense problems of taste and challenge which I don't believe are being explicitly met and maybe there's nothing we can do about it now. But eventually I think we are going to have to address these problems if we're going to complete Darwin's program. It's already clear, for example, in consciousness work that the... a complexity of the integration involved in the dynamic core, the numbers involved, the hyperastronomical numbers involved, in integration, and then in differentiation from minute to minute, are so large that it's not very likely we'll ever have the opportunity to itemize every single thing.