226 SPECULATIVE SCIENCE 



the offspring on the whole resemble their parents, the descend- 

 ants from any given generation will on the whole resemble the 

 more favoured rather than the less favoured individuals of the 

 species. So much of the theory of natural selection will hardly 

 be denied ; but it will be worth while to consider how far this 

 process can tend to cause a variation in some one direction. It 

 is clear that it will frequently, and indeed generally, tend to 

 prevent any deviation from the common type. The mere ex- 

 istence of a species is a proof that it is tolerably well adapted 

 to the life it must lead ; many of the variations which may 

 occur will be variations for the worse, and natural selection will 

 assuredly stamp these out. A white grouse in the heather, or a 

 white hare on a fallow, would be sooner detected by its enemies 

 than one of the usual plumage or colour. Even so, any favour- 

 able deviation must, according to the very terms of the state- 

 ment, give its fortunate possessor a better chance of life ; but 

 this conclusion differs widely from the supposed consequence 

 that a whole species may or will gradually acquire some one 

 new quality, or wholly change in one direction, and in the same 

 manner. In arguing this point, two distinct kinds of possible 

 variation must be separately considered : first, that kind of 

 common variation which must be conceived as not only possible, 

 but inevitable, in each individual of the species, such as longer 

 and shorter legs, better or worse hearing, &c. ; and, secondly, 

 that kind of variation which only occurs rarely, and may be 

 called a sport of nature, or more briefly a ' sport,' as when a 

 child is born with six fingers on each hand. The common 

 variation is not limited to one part of any animal, but occurs in 

 all ; and when we say that on the whole the stronger live longer 

 than the weaker, we mean that in some cases long life will have 

 been due to good lungs, in others to good ears, in others to good 

 legs. There are few cases in which one faculty is pre-eminently 

 useful to an animal beyond all other faculties, and where that is 

 not so, the effect of natural selection will simply be to kill the 

 weakly, and insure a sound, healthy, well -developed breed. 

 If we could admit the principle of a gradual accumulation 01 

 improvements, natural selection would gradually improve the 

 breed of everything, making the hare of the present generation 

 run faster, hear better, digest better, than his ancestors ; his 



