I went to a colleague called John Smith, who was interested in RNA and RNA sequencing and so on, and I said, 'Look, I think I've got this suppresser on a phage, and if we induce it and then just do an assay which would just simply ask how much of the RNA that you extract after this cell has come up, how much of this would be… would be this tyrosine TRNA?' And I said, 'Well, you know, I mean, it should be an enormous amount, because if the tyrosine tRNA is one-fifth… one-twentieth, because there are 20 tRNAs for the 20 amino acids, and I'm going to increase its synthesis by, let us say, 10 or a hundred times – somewhere between that, because I have 100 copies of the phage growing in it – so let's say conservatively it should be half. And this means nearly everything should load up with tyrosine.' So we did the experiment on a Friday afternoon – I remember this – and it was all processed on the Monday, and of course what we didn't know then is that the gene for this was a minor transfer RNA, not the major one. It was something that was only about a tenth of that again. So absolutely would have had a lot of trouble finding it. And, of course, what we got was a threefold increase. Now, that means that that gene had increased more like 300-fold, more like 100-fold as it worked out, and there would have been no hope to have found that sequence, we would have only purified the major one. However, it was workable to do this from these phage infected things. And of course the first experiment was simply to take the normal allele of this and put it on the same phage and compare the two.