OTHER THEORIES OF SPECIES-FORMING 273 



successful mating. Such conditions are readily observable between 

 closely allied species. Again, the prevention of intercrossing may 

 result from the appearance of a lowered interfertility between the 

 variant individuals and those of the parent-stock. If individuals 

 varying in the same direction were even slightly more fertile inter se 

 than those varying in different directions there would be a progressive 

 tendency in a series of generations for the varying individuals to 

 diverge more and more markedly, and ultimately to become practi- 

 cally sterile except with members of their own group. 



That environmental changes do frequently affect the fertility of 

 animals is seen when wild animals are kept in confinement. Rela- 

 tively few wild animals breed in captivity. Such a lowering of fer- 

 tility as the result of environmental changes might restrict crossing 

 between unlike forms, while permitting it among the like ones. 



Summary on isolation theories. — ^There is a great divergence of 

 opinion as to the importance of isolation as a causal factor in species- 

 forming. Some writers, such as D. S. Jordan and V. L. Kellogg, con- 

 sider isolation an indispensable, and therefore primary, factor; others, 

 especially geneticists, almost ignore it as an effective factor. Still 

 others, hke the present writer, take a middle ground and conclude 

 that isolation, especially geographic isolation, has helped greatly in the 

 segregation and establ'shment of well-defined groups such as species 

 or varieties, the latter developing into the former after prolonged 

 isolation and the addition of new variations. Isolation theories, how- 

 ever, have no light to shed upon the difficult problem of adaptation, 

 and it is here that isolation is auxiliary to natural selection. / 



THEORIES ALTERNATIVE TO NATURAL SELECTION 



The three theories that have been offered by their authors as sub- 

 stitutes for natural selection are: 



1. Theory of the inheritance of acquired characters commonly called 

 Lamarckism: This theory has been outlined in the chapter on the 

 history of evolution (pp. 19 ff.). It will again be dealt with in con- 

 siderable detail in chapter xxii. For the present, then, we may pass 

 by this theory without further comment. 



2. The orthogenesis theories: These theories have already been 

 presented in sufficient detail for our purposes in chapter ii (pp. t^t^ ff.). 



3. The mutation theory of Hugo De Vries: This theory has been 

 dealt with in chapter ii, and will be discussed in further detail in 

 chapter xxiv. 



