My good friend and colleague Seymour Benzer had also decided to drop phage and go into this. He decided to work with drosophila. And he had the idea that we would go from gene to function. He also had this view of genetic dissection which is take the phenotype, find out how many ways there are of breaking it, and now you have all the genes specifying this. And then what you could do was in fact try to relate these genes to the final function, so you could get a classification of the genotype. Of course, our theoretical views there were extremely primitive, but I felt you could not relate the genes to function that simply. Because I think the mapping of the genetic material onto this external phenotype would have to be quite complex in order to understand this. And so I thought to… to say… as we used to say, I… I remember these pictures of Lorenz being followed around by his ducks, and I read that ducks had this fixed behaviour, that they… they did this. And of course I wanted to know, you know, was there a gene, so to speak, to put the left foot forward, followed by another gene that said, 'And having done that put the right foot forward'? In other words, what were the units of behaviour and did they map onto the genetic program? And indeed if you take the whole of complex organisms, what were the units of a complex organism that mapped? I mean, did you have genes for liver? Did you have separate from the genes, let us say, for kidney? Or did you have genes for the cardiovascular system separate from the genes for the whole of the central nervous system? And those ideas of mapping, what we do to analyse an organism, onto its program became very important theoretically, at least for me, to decide. And I decided that at least for a nervous system, the kind of program Seymour was suggesting would not work.