Where was immunology at the time all this was happening? I never looked on myself as an immunologist – in fact I didn't look on myself as anything – but it was more physical chemistry than it was immunology; immunology was the background knowledge, as was medicine. So what did they think? Well, what they thought was what a very brilliant and great chemist Linus Pauling thought –namely, the way the specificity problem was solved... remember the size problem might be solved by cleaving the molecule, the heterogeneity by assuming that you actually have cancer supervening on one cell that makes just one gamma globulin, each different for each patient, and third the specificity problem.

Well, Pauling suggested that the way... and he was a friend of Landsteiner... suggested that the way you recognized these molecules is the foreign molecule came in and then you folded the antibody chain around it and it's sort of like a cookie in a cookie cutter, and then the foreign molecule went away and you were ready for the next time, so it would go like that. That was called the instructional theory, or the instructive theory of antibody formation. It was very powerful in the sense that it was simple, elegant, the way Linus Pauling was in the way he did his theories... turned out it was wrong, but at the time it was the prevalent theory, except for one other gentleman: Macfarlane Burnet.

Macfarlane Burnet was a virologist and epidemiologist from Australia – remarkably broad mind. A friend by the way of René Dubos whom I mentioned before. And Macfarlane Burnet just persistently attempted to understand the specificity problem, and following upon some work of Nils Jerne, that maybe the antibodies all existed beforehand and what they did is brought the foreign molecule to the cell and then made more of the same backwards, which turned out to be a wrong idea. But the idea that it was a selectional system, a natural selectional system, got to Macfarlane Burnet, who finally came up with the idea of the clonal selection theory of antibody formation. So that was the selectional theory as against Pauling's instructional theory. Pauling wasn't the only instructivist but he was the most prominent. Here was Macfarlane Burnet's idea: that each cell – this was the cell that would make a myeloma protein if it got cancerous... each cell would have one antibody on it. He didn't call them antibodies, he called them recognizers. And each one would have a different one, and it would be a repertoire, and you'd make that independent of anything that happened in the outside world, but when you got a foreign molecule in there, the foreign molecule would go down and poll this library and finally hit one that stuck to it, and another that stuck to it, and those would be produced clonally – that is, the cells would divide asexually two to the nth: one, two, four, eight etc. And this idea, which was by no means accepted, was one that Joe Gally and I found enormously attractive and it fit exactly what we said about the myeloma, where instead of stimulating cell division by antigen binding, what happened is a cancer came onto this one cell and made huge amounts of the protein.