THE RAW MATERIALS OF EVOLUTION 111 



" If, as I must think, external characters produce 

 little direct effect, what the devil determines each 

 particular variation ? " 



Having made this confession of ignorance, we 

 venture to discuss the possibilities of answer to a 

 question which can never be far from any thought- 

 ful mind. 



There are variations and variations. There are 

 variations that mean nothing more than an aug- 

 mentation or diminution of an already existing 

 quaUty. The hair may be very long or the tail 

 very short. Or a variation may mean that a 

 character present in parents and ancestry is absent 

 from the offspring : the entail has been broken. 

 An albino expresses such a variation, or a hornless 

 calf, or a tailless kitten. Or, again, a variation 

 may be interpretable as a novel arrangement of 

 characters or qualities which were present in the 

 ancestry. A piebald pony may serve as a diagram. 



Now, in regard to variations of this sort — 

 which may be described as permutations and 

 combinations of the already existing unit char- 

 acters — the modern knowledge of the conditions 

 of heredity has made it plain that there are many 

 opportunities for their occurrence before, during, 

 and after fertihsation. We know that each germ- 

 cell contains a definite number of stainable bodies, 

 or chromosomes, which appear to be the bearers 

 of the heritable qualities. We may compare these 

 to a microscopic pack of cards, and we may say 

 that there is an extraordinarily elaborate shuffling 

 before development begins. HaH of the pack is 

 ejected from the egg-cell in what is known as a 

 "polar body," and the number is raised to the 

 normal again (constant throughout the body of the 



