These amber mutants had been known, and what was known is that they grew on one strain, didn't grow on the other, and I had… and you just said, 'well, what happens if you don't have an amino acid – what do you do? Do you just stop? I mean, does the thing get stuck?' In fact, we began to wonder that there could be dominant nonsense, because if the ribosome theory was correct, then tapes with these mutants that got stuck in it would just jam the entire system. So we wondered whether in fact… and I did an experiment actually, to try and test whether I could jam a system, but of course it's very hard to isolate such a dominant mutant by definition. But the experiment was a failure. And so I had the idea that there were two kinds of nonsense, one got stuck and one actually chain-terminated. One actually chain-terminated. Right? And I said, 'Well, maybe this nonsense is chain-terminating; maybe there is a mechanism for terminating a chain – it isn't that you just fall off at the end. And if this is the case, then when you suppress it, then we think you just carry on the chain.' So these were the nonsense suppressors. The chain-terminating thing did not find much favour in the lab, because it was very hard to prove. However, we decided we would have a go at it. And what we had is we knew the bacteriophage head protein, our old friend, has 70% of the protein synthesis, and so we decided we'd go and have a look to see if fragments of this protein were made. And the first experiment was done by a, a graduate student of mine called Anand Sarabhai, from which... from his success came the slogan the luck of the Sarabhais. And Anand was a fearless experimenter. He believed in having a go. So although he was judiciously advised even by Francis, not to do this experiment because it was bound not to work, he decided to have a go, and of course it worked. And then it became clear we didn't have to go through everything he did. We could use a little thing that Fred Sanger had been using and which we had started to use, which is radioactive labelling of the protein and fingerprinting. And of course suddenly we realised that we had all these amber mutants in this gene; that we could give a topological proof of co-linearity – we wouldn't have to do any protein sequencing. The only assumption we would have to make is the protein is read from the same end in all… at all times, which seemed a very reasonable one. And that was our paper, which we published in 1964, which proved that the gene and the protein were co-linear but by an argument which was totally unexpected at the time. All we showed is that as the mutant moved to the right we got more and more of the protein. And we didn't have to do any sequencing. And… and that then opened up the whole idea of chain-termination.