210 THE PHYSIOLOGY OF REPRODUCTION 



animal to conjugate with dissimilar rather than with related 

 spermatozoa. 



Professor Ewart has informed the writer of a case in which 

 a Dandie Dinmont bitch in his possession copulated with a 

 dog belonging to the same breed, and two days subsequently 

 had intercourse with a Scotch terrier. In due time the bitch 

 littered three pups, and of these only one was a pure-bred 

 Dandie Dinmont, while the other two were half-bred Scotch 

 terriers, in spite of the fact that the Dandie Dinmont dog 

 copulated two days earlier than the Scotch terrier. This case 

 may be regarded as to some extent confirmatory of the experi- 

 ment described above. ^ 



Doncaster,^ in describing his experiments on Echinoid 

 hybridisation, states " that cross-fertihsation is assisted by 

 conditions which tend to reduce the vitality of the eggs." This 

 artificial reduction of vitality could be accomplished either by 

 warming the eggs, or by shaking them, or by keeping them for 

 several hours, or by placing them for from one to two hours in 

 diluted sea-water, the last method being the most uniformly 

 conducive to the occurrence of cross-fertihsation. There is 

 some evidence, therefore, that a reduction of vigour among the 

 gametes, whether occurring naturally as a consequence of in- 

 breeding or produced artificially as in Doncaster's experi- 

 ments, may lead to a similar result, since both- conditions 

 may bring about an increased tendency towards the union of 

 dissimilar gametes. On another view, the tendency towards 



' Seeing that an assortative mating of gametes can probably occur 

 between the ova of one individual and the spermatozoa derived from different 

 individuals, whether as a result of gametic similarity or reduction of vitality, 

 it is not improbable that gametic selection also sometimes takes place when 

 various gametes of a single individual are bearers of different characters, 

 in the manner postulated by the Mendelian theory. Such a preferential 

 mating, if it exists, would of course obscure the evidence of that very 

 gametic segregation, the probable existence of which, in other cases, is in- 

 ferred from the numerical proportions in which the different sorts of zygotes 

 or offspring are produced ; for if there is an assortative mating among the 

 gametes, it is obvious that the offspring would no longer be produced in 

 definite Mendelian proportions, since these depend upon the chance unions of 

 gametes. According to this view, prepotency may perhaps be interpreted 

 as the tendency of the gametes of an individual to conjugate with other 

 gametes bearing similar hereditary characters. 



* Doncaster, loc. cit. 



