SELECTION 447 



stituents or determinants, may result in an intra-f^jerminal 

 struggle and selection. But we need not go further, since 

 our point is simply that the selective processes are probably 

 more manifold than even Darwin realised. 



(3) Whenever we turn from expositors of Darwin to 

 Darwin himself we discover afresh how subtle was his idea 

 of the process of Natural Selection. We realise, for instance, 

 that the selection need not imply a sudden elimination of 

 the relatively less fit, for a persistently shortened life and 

 a consistently unsuccessful family will work to the same 

 result in the long run as lopping off heads. As Professor 

 Punnett puts it: " If a population contains .001 per cent, of 

 a new variety, and if that variety has even a 5 per cent. 

 Selection advantage over the original form, the latter will 

 almost completely disappear in less than a hundred genera- 

 tions." In human affairs we may be thus encouraged in 

 patience. It has also to be realised that the web of life 

 has so fine a texture that apparently trivial differences in 

 organisms may be of critical moment in determining the 

 survival of those who possess them. And just as in animal 

 courtship what determines the female's preference for one 

 suitor out of many is very probably an irresistible tout 

 ensemble of gifts and graces, rather than excellence in one 

 particular decoration or quality, so in natural selection it 

 may be that what gives survival value is often a general 

 stability of constitution and efficiency of behaviour. In a 

 well-known instance when 136 storm-spent sparrows were 

 brought into shelter, 72 revived and 64 died. Careful meas- 

 urements showed Professor Bumpus that the eliminated birds 

 were less near the normal than those which sun-ived. Ex- 

 cept in one measured character, the range of variation was 

 greater in those that succumbed. Thus while natural selec- 



