36 



pletely in harmony with Mendel's conception. The principal of these 

 views are the continuity of the individual nuclear chromosomes, their 

 existence in pairs, the separation of the pairs in the maturation 

 process, the derivation of one of each pair from the male and the 

 other from the female parent, and the different but complementary 

 roles which the various pairs play in the development of the whole 

 organism. More work is needed, however, to put these ideas about 

 the chromosomes on a really firm basis. 



Having now passed in review a few of the principal facts and 

 theories brought out by the work of Mendel and his twentieth century 

 disciples, it will be well to consider briefly what effect the acceptance 

 of Mendel's law must have upon current views of heredity and 

 evolution. In the first place it gives us quite a new conception of 

 what constitutes a pure race. This is no longer a question of pure 

 ancestry for numerous generations, but entirely a question of the 

 union of pure gametes. It is now seen that a pure strain may 

 originate from a hybrid stock, and that w r hen once it appears it will 

 remain pure indefinitely, so far as merely hereditary tendencies are 

 concerned. On the other hand, the direct ancestry of a particular 

 form may have appeared pure for any number of generations, and yet 

 it may produce progeny of a different character. 



Next we are forced to a new analysis of variation, for it is evident 

 that under this term widely divergent phenomena have been included 

 in the past. The results of the experiments made in connection 

 with the testing and amplifying of Mendel's law have rendered it per- 

 fectly clear that at the very least the variations in gametic constitution 

 will have to be carefully separated from those due to responses made 

 by the body of the organism to its environment, for the former will 

 be transmissible and the latter will not. The outcome of this is that, 

 owing to the close approximation of these two kinds of variation in 

 some cases, or occasionally even a possible overlapping, the usual 

 biometric methods based on the measurement of a number of indi- 

 viduals taken at random without consideration of their gametic 

 constitution may sometimes lead to quite erroneous ideas as to the 

 course of evolution and other matters in relation to the group of 

 organisms dealt with. 



Then again, the idea of independent alternative unit characters 

 throws a new light on our notions about specific differences. For 

 these to be valid they must rest on deep-seated differences in the 

 structure of the germ-plasm. In a strict sense, therefore, there is 

 probably no merging of one species into another by the selection of 

 infinitely small variations, but rather a saltatory passage from one to 

 the other, each step being marked by a specific change in the germ- 

 plasm of such a character that it produces either a factor in distinct 

 opposition to some previously existing factor of the same order, or 

 an altogether new factor. In other words, Mendel's law is decidedly 

 favourable to the idea of what is known as the discontinuous origin of 

 species. Needless to say, this latter doctrine does not in any way 



