VARIATION AND HEREDITY 213 



nature. On the other hand, variations in the ger- 

 minal substance, in the nature of things, will affect 

 future generations. Such are called, therefore, 

 germinal or blastogenic variations. 



There have been numerous theories dealing with 

 the architecture or organization of the germ-plasm, 

 which account for the spontaneous appearance of 

 new characters, but all such hypotheses are founded 

 upon abstract speculations and yield no verifiable 

 explanation. Nevertheless the knowledge that we 

 have gleaned through observation and experiment, 

 of what goes on in the germ-cells just prior to the 

 initiation of the ontogenetic process, is sufficient 

 to afford us a good clue to what may be the source, 

 or at least one source, of germinal variations. 



It will be recalled that both gametes undergo a 

 process of maturation just before zygosis takes place, 

 in the course of which the amount of the chromatin 

 in the animal egg is reduced and the number of chro- 

 mosomes in both gametes is halved. This phenom- 

 enon is known as reduction. By it there is 

 frequently, perhaps always, produced two kinds of 

 male gametes and two of female gametes. The 

 possibilities of fusion between these two sorts of 

 gametes are fourfold. But more than this; in the 

 extremely complicated process of mitosis it is very 

 unlikely that the chromosomes should in every way 

 be exactly halved, and since we believe the nature 

 of the organism to be determined in large measure 

 by the chromatin substance which is passed on in 

 the germ-cells, the potentialities of the zygote may 



