Heredity and Natural Selection. 289 



when one bone is removed in the leg of the dog, the 

 other bone, which has to carry the whole weight of the 

 body, increases in thickness. But from the facts given 

 with respect to horned and hornless cattle, it is probable 

 that the horns and skull would immediately act on each 

 other through the principle of correlation. Lastly, the 

 growth and subsequent wear and tear of the augmented 

 muscles and bones would require an increased supply of 

 blood, and consequently an increased supply of food; 

 and this again would require increased power of mas- 

 tication, digestion, respiration and excretion." 



It will be seen by a careful examination of this ex- 

 tract that Darwin is compelled, by cases of this kind, to 

 believe that other influences have played a part equal to 

 or more important than that of natural selection, and he 

 is compelledto attribute the co-ordinated modification of 

 related parts to the action of the law of correlated vari- 

 ability. 



I have already called attention to the fact that this law 

 of correlated variation is a necessary result of our view 

 of the nature of heredity, for a change in one part must 

 cause variation in co-ordinated parts; and gemmules 

 thrown off by a certain organ of the body may cause 

 co-ordinated variation in all the homologous parts of a 

 descendant. I believe that it will be clear to every one, 

 without further explanation, that the acceptance of our 

 theory will greatly simplify our conception of the action 

 of natural selection, and will enable us to understand 

 the rapid evolution of co-ordinated structures, without 

 being compelled to attribute them to other influences. 



Darwin appears to have felt the need of something of 

 the kind, for we find evidence that he has hunted long 

 and faithfully, but in vain, for something to show that 

 changed conditions produce, directly, the proper modi- 



