NATURAL SELECTION SINCE DARWIN 77 



cestors of the modern swan there could have been in- 

 dividuals whose neck, having one extra vertebra, was 

 slightly longer, and that such a peculiarity had proved 

 beneficial to the species, so that only the individuals 

 thus favoured could have survived and transmitted 

 this character to their offspring. Their offspring 

 would then have had the same number of cervical 

 vertebrae, a number greater than the original number. 

 Why then should a following generation have two 

 more and a third generation three more vertebra?? 

 The fact is, on the contrary, that after countless gen- 

 erations, this character will remain constant and the 

 neck will retain (leaving aside possible causes of re- 

 version or the consequences of crossing with normal \ 

 individuals) the original number of vertebrae. 



This appears self-evident and it is hard to under- 

 stand why, among so many objections to natural se- 

 lection, this one should not have taken the first place. 

 There is only one explanation for it. From the very 

 beginning the theory ofLsele ction has dealt with ab- 

 stracti ons rather t han with facts. What was sup- 

 posed to be transmitted was not a c ertain struct ural 

 chara cter bu t a tendency to vary in a certain direc- 

 tion. In the case" of the "swan, it was a tendency 

 toward the elongation of the neck which caused the 

 number of vertebras to increase with each generation. 

 A tendency, however, has no existence of its own; 

 it is an abstraction which merely expresses the 



