Now there is... there is a preconceived bias in... in accepting that idea: namely that indeed in the historical events, only a small array of possibilities was provided. But you can demonstrate... and I did demonstrate with a small mathematical... nothing very... very advanced, but anyway one can demonstrate that an event, any event, can be made to happen with a high probability provided you offer sufficiently... a sufficient number of opportunities for this event to happen. And you toss a coin; the chances that it falls on the head face is one in two, toss it ten times and the chances it will fall at least once on the head face is 99.9%. Play roulette, same thing. 250 times number 8 will come out with a 99.9% probability. Buy a lottery ticket, seven digits, your chance of winning is 1 in 10 million. You will win in 99.9% of the cases if there are 69 million drawings. So enough opportunities and the probability of winning increases. Well, obviously the lottery's not organised that way. There's only a single drawing. But the evolutionary lottery is something completely different. In the evolutionary lottery, it has been played on millions, billions, sometimes trillions of individuals, many generations over sometimes millions of years, and so one can... one can predict that in a number of cases, in fact, chance will have offered natural selection a more or less complete exhaustive array of possibilities so that the outcome is optimisation. And so selective optimisation is a new... is a new concept. I didn't invent... I didn't invent anything, but anyway I have come to that conclusion that selective optimisation may have been involved more often than is... than is generally believed... more often than is believed in the history of life, and therefore that a number of steps were more or less obligatory under the conditions under which they took place.