At the moment, we just don't know which features of life are like that because of the accidents of common origin and which of them are like that because they have to be. And I know you have speculations about what things have to be, and I do, but it would be nice to have another case of an independent origin of life and see whether our hunches about, for example, inheritance have to be Weismannian rather than Lamarckian, or like, inheritance has to be digital and not analogue, and so on. Are these speculations correct?

[Q] Double our sample size from one to two.

Well, three. Haldane had these three theorems which would enable you to do the whole of science. One was Pangloss's Theorem, which is 'All is for the best, in this best of all possible worlds.' But the one that is relevant to the question you've just asked, was the Bellman's Theorem, which is 'What I've told you three times is true.' The third one, just for completion, was Aunt Jobisca's Theorem, which is a very powerful theorem if you get into an argument; 'It's a fact the whole world knows, that the Pobble is happier without its toes.' And indeed, these theorems do get... three cases would be enough, but actually two would be great. Two would be marvellous.