SOME ASPECTS OF GEOLOGIC TIME 269 



what exactly were species and what were merely varieties. 

 Other naturalists have been involved in the same difficulty. 

 But with regard to the higher forms of animal life we have an 

 independent criterion. It is generally recognised that the mutual 

 infertility of nearly allied animals is a test of species difference. 

 In the rare exceptional cases, such as the horse and the donkey, 

 when hybrids can be formed, the hybrids are infertile. 



We shall, therefore, do well to leave the morphological side 

 and to pay more attention to the aspect of physiological fertility. 

 It is hopeless to attempt to decide what degree of morphological 

 change does or does not constitute species difference. The 

 difference in shape between the horse and the donkey is 

 comparatively small, yet a fertile cross cannot be obtained. On 

 the other hand, notwithstanding the enormous differences 

 between the varieties of domestic dogs, differences of size, shape, 

 proportion, colour, character of coat, these varieties are mutually 

 fertile. 1 The variegated types of domestic pigeons, notwith- 

 standing enormous differences, are not only mutually fertile, but, if 

 left to themselves, revert to the ordinary rock pigeon from which 

 they are descended. Yet the differences, were they found in fossil 

 forms, would probably be classed as greater than species difference. 



Such facts as these throw some light on the course of organic 

 evolution. Physiological infertility is evidently not correlated 

 with accidental differences in shape, colour, or form, but connotes 

 an essential, deep-seated organic change. It seems probable, 

 therefore, that this may not be obtainable by artificial breeding, 

 but that it may be a natural process, which, for its accomplish- 

 ment, requires a prolonged time. It has certainly not been 

 found among the multitudinous races of human-kind. If this 

 theory were actually proved (as yet it is only a speculation), it 

 might give us a minor limit for the time required for the pro- 

 duction of species. 



It is interesting to note that the discoveries of Mendel can, 

 without undue straining, be made to fit into the same hypothesis. 

 It has not yet been proved that all inheritance can be described 

 in Mendelian terms. Mendelism may account for inheritance in 

 mixed races, such as the Caucasian and Negro half-breeds, but 

 even this is doubtful. Certainly, in ordinary human inheritance, 



1 For obvious reasons, it would hardly be possible to obtain a first cross when 

 there was more than a certain difference in size, but this is not true physiological 

 infertility. 



