One of the things I did was to write two papers which have... which my... only of my papers published by Her Majesty's Stationery Office. They published as parts of the report of GMAG [Genetic Manipulation Advisory Group]. And one paper is I think a very interesting paper because it is the paper that analyses theoretical pathogenesis, that's its subject. It says: ‘Why don't we try and make an estimate, a real estimate of potent... of conjectural danger?’. And get some numbers and with these numbers we can then begin to compare. And that system is still being used, because once you did this in a quantitative way, I mean, everything came out very simply. You got a lot of numbers, big numbers, you then divided them by a number almost as big, and so you got the... you got huge differentials of safety. And ultimately we changed the system here, and I think since that time experimentation in genetic manipulation is... is quite... it's quite acceptable now, and of course nothing has happened, nothing dangerous has happened. Because one of the things that in fact has... has happened as time has... time always brings rewards... I made a strong argument that the best case to take a dangerous virus is to take it and clone it; that would make it safe. And put it in E. coli [Escherichia coli], because it can't get around and infect people this way. And in fact, if you're worried, you put... you put... divide it into three pieces and put them in three different E. coli. And that I would prefer to work with clones of this virus rather than with the virus themselves. And I think that were it not for getting people to think sensibly about this we could not have done the work so rapidly. That happened with the AIDs virus. I mean, this technology has been invaluable. The virus was cloned, people could work with the DNA, they could sequence it, they could characterise it very rapidly. If you look at all the history of other infectious agents, we accomplished worldwide in a year what might have taken 20 years to do with standard methods. So I think that was a very good lesson.