11/9 THE FULLY -JOIN ED SYSTEM 



11/8. A well-known physical example illustrating the difference 

 is the crystallisation of a solid from solution. When in solution, 

 the molecules of the solute move at random so that in any given 

 interval of time there is a definite probability that a given molecule 

 will possess a motion and position suitable for its adherence to 

 the crystal. Now the smallest visible crystal contains billions of 

 molecules : if a visible crystal could form only when all its mole- 

 cules happened simultaneously to be properly related in position 

 and motion to one another (Case 1), then crystallisation could 

 never occur: it would be too improbable. But in fact crystallisa- 

 tion can occur by succession (Case 2), for once a crystal has begun 

 to form, a single molecule which happens to possess the right 

 position and motion can join the crystal regardless of the positions 

 and motions of the other molecules in the solution. So the crystal- 

 lisation can proceed by stages, and the time taken resembles T 2 

 rather than 7\. 



We may draw, then, the following conclusion. A compound 

 event that is impossible if the components have to occur simul- 

 taneously may be readily achievable if they can occur in sequence 

 or independently. 



11/9. It is now becoming apparent in what way the 1,000-unit 

 Homeostat, which takes such an excessively long time to adapt, 

 differs from the living organism, which usually gets adapted in a 

 fraction of its generation-time. The organism, of course, does not 

 reach its full adult adaptation by making trial after trial, all of 

 which count for nothing until suddenly everything comes right ! 

 On the contrary, it conforms more to the rules of Cases 2 or 3, 

 achieving partial successes and then retaining them while improving 

 what is still unsatisfactory. 



However, before we turn to consider the latter cases we should 

 notice that the 1,000-unit Homeostat is not wholly atypical, for 

 environments do exist, though they are rare, that demand that 

 all successes occur simultaneously and against which partial 

 successes count for nothing. When they occur, it is notorious 

 that they give the living organism great difficulty. A trifling 

 example occurs when a trout would take a fly on the surface; the 

 trout must both break the surface at the correct place and must 

 also close its jaws at the right moment; two variables here are 

 essential in the sense that both their values must be within proper 



153 



