ORGANIC EVOLUTION — THE FACTORS 87 



birth mean of the new generation is below the survival 

 mean of the preceding generation; but in a species 

 that is undergoing evolution the survival mean of the 

 new generation, by reason of the great elmiination of its 

 unfit, is somewhat beyond the survival mean of the 

 preceding generation, and thereby small variations 

 shade in the progress of time into great differences. 

 But never, or very rarely, — so rarely that if they ever 

 occur examples of it are altogether unknown to us in 

 nature, — does a specific change of type result from a 

 great individual variation. Never do we see a wild 

 animal possessed of an abnormality — i. e. a great varia- 

 tion — which affords to it such an advantage that it far 

 transcends the rest of its species in the perfection of its 

 adaptation to the environment ; and most certainly never 

 do we see a wild animal possessed of an abnormality so 

 great, so favourable to survival, so advantageous in tlie 

 struggle for existence, that as a result all competitors 

 are exterminated, and the abnormality is imprinted on 

 the race. In fact, as I have already said, " evolution 

 proceeds not on lines of traits, Jiowevcr favourable, 

 which occur infrequently or abnormally, but on lines of 

 traits common to all the individuals of the whole 

 species, that is, traits in which every individual rises or 

 falls below the specific average." The Past Evolution 

 of the races of men surely affords conclusive proof of 

 this.^ 



^ In the November number of the Nineteenth Century Review 

 Mr. Herbert Spencer has some remarks on Lord Salisbury's 

 Oxford speech which are curiously like my own. ]\Ir. Spencer 

 had of course not seen mj manuscript, while, on the other hand, at 

 the time his article appeared this book had long been in the 

 hands of the publishers, and the passages in c[uestion had already 

 been " set up " in type. 



