382 EXPERIMENTAL STUDY OF INHERITANCE 



species " result invariably in sterile offspring is also erroneous ; 

 for the hybrids of American bison and European wild ox, of 

 Indian humped cattle and domesticated ox, of common goose 

 and Chinese goose, of common duck and pintail duck, of different 

 kinds of pheasants, and many more are certainly fertile. 



At the same time, it seems safe to say that the likelihood 

 of successful crossing and of the fertility of the hybrid offspring 

 is in inverse proportion to the distinctness of the species crossed. 

 It seems also safe to say that the characters of species-hybrids 

 do not conform to any general formula. They may be a blend 

 of the parental characters, they may be exclusive or particulate, 

 they may be reversionary — i.e. allowing expression of long-latent 

 ancestral characters — or they may be novel and peculiar. 



On the whole, the crossing of distinct species, while it may be 

 interesting physiologically, does not seem to have much interest 

 for the evolutionist. It does now and then occur in nature, but 

 it seems to be a mere by-play of little phylogenetic importance 

 — unless perhaps in very early days, of which we know nothing. 

 Diverse Results of Hybridising. — An inheritance is such a 

 complex integrate of items that no one can hope to predict 

 the result of mingling two more or less distinct inheritances. 

 We have two organisms, A and B, which can be crossed and 

 produce offspring : but, before the germ-cells of A and B are 

 ready for union, they have undergone a process of maturation 

 which may definitely affect the burden of hereditary qualities of 

 which each germ-cell is the vehicle ; by the process of amphi- 

 mixis or fertilisation a new integrate or zygote is formed — 

 the fertilised egg-cell — and in this integration the inheritance 

 may be affected by permutations and combinations, mutual 

 adjustments and new states of equilibrium, victories and defeats 

 of particular items, of all which we have no actual knowledge. 

 In the process of development, if there are several different sets 

 of primary constituents representative of a future structure — an 

 hypothesis from which we can see no escape— then the result 



