134 PROBLEMS OF EVOLUTION 



variations and the environment? The environment does not 

 decide what the variation is to be, nor does the organism 

 fashion its environment. And yet the one dovetails into the 

 other. To explain this wonderful fact, I hope to show that 

 there is a limit to the variations possible and a corresponding 

 limit to the environment. To illustrate by a simile. Suppose 

 that there is a breach in a wall through which men hope to 

 force their way in spite of a hail of bullets. The men and the 

 bullets make equally for the breach. Therefore it is no wonder 

 if in the narrow space there are many hits : yet each particular 

 hit is a coincidence. 



Before going further, I must discuss Professor Weldon's 

 theory dealing with chance in evolution. He tries to show 

 that, according to the law of chance, similar variations will 

 arise in plenty, so that a favourable variation is not likely to 

 be lost (through crossing of the favoured individuals, I sup- 

 pose, with others not thus favoured or through other acci- 

 dents). But he does not explain why there should be any 

 favourable variations at all (i.e. any that prove to be adapta- 

 tions), but only why if one individual effects, so to speak, an 

 adaptation, there are likely to be many that do so. 



Professor I will first try to explain the law of chance and Professor 



W theo n ry Weldon's application of it. 



When a great number of chances can be noted and surveyed, 

 they are found to group themselves : in fact, a definite tendency 

 shows itself. This was proved by throwing dice thousands of 

 times and noting the numbers in each case. There was an 

 average or mean throw, and on either side of this the numbers 

 arranged themselves with a close approximation to symmetry 

 the greater the deviation either way, the more rarely it 

 appeared. To apply this principle to evolution, Professor 

 Weldon has taken particular features and made most careful 

 measurements. He has shown that in shore crabs the dimen- 

 sions of the carapace vary in accordance with the law of chance, 

 and that the variations group themselves nearly symmetrically on 

 either side of a mean. There is a large number of individuals 

 slightly above the mean, a large number slightly below. 



