THE REDUCTION DIVISION 131 



217. The hypothesis that ancestral contributions are halved in 

 every generation is a deduction from Galton's (and other) statistics, 

 and from the circumstance that a ' reduction ' division occurs in 

 the germ-tract. So much has been made of the reduction division 

 that it is necessary to consider it. It commonly happens that when 

 cells are about to multiply by self-division the ' chromatin ' of each 

 nucleus gathers into * rods ' or masses termed chromosomes, the 

 number of which is definite for each species. " Thus in man we 

 find 32, in a mouse 24, in a donkey 36, in a cockroach 32, in Ascaris 

 megalocephala, var. univalens, 2 ; var. bivalens,4, and so on in different 

 organisms." 1 In ordinary cell division the chromosomes split, each 

 half from each chromosome forming a new chromosome which passes 

 to one of the daughter cells. The division of the cells thus extends 

 to a minute division of the structures within them. In this way the 

 normal number of the chromosomes is preserved. But when the 

 cells of the germ-tract are about to produce actual germs fit for 

 conjugation, the chromosomes do not split. In the case of sperms, 

 half of them pass as entire bodies to each daughter cell, which thus 

 receives only half the normal number of chromosomes. This is 

 the 'reduction' division. A second division in which the chromo- 

 somes split, results in four cells, each of which, possessing half the 

 usual number of chromosomes, becomes a spermatozoon. In the 

 case of the ovum, four cells are not formed ; when the reduction 

 division occurs, one-half of the total number of chromosomes are 

 cast out of the cell in a ' polar ' body, leaving almost the whole of 

 the cell-body to the remaining chromosomes. At the next division 

 the reduced number of chromosomes split as in ordinary cell- 

 division ; but one-half of the product is again cast out of the cell 

 as a second polar body, leaving the ripened ovum, like the ripened 

 sperm, with only one-half of the normal number of chromosomes. 

 Therefore the divisions which result in four cells in the case of the 

 sperms, result only in one in the case of the ova. 



218. An obvious interpretation of these phenomena is that 

 reduction implies nothing more than a preparation for conjugation ; 

 for, if the number of chromosomes were not reduced to half, they 

 would be doubled in each generation when sperm united with ovum, 



for example albino, ancestor more remote than the eleventh may be reproduced 

 by offspring to the exclusion of the character of more recent ancestors. His 

 ' unit,' therefore, must be strong enough to direct development. Must we 

 assume, therefore, that the germ-plasm is compounded of millions or billions of 

 units, the more modern of which are millions or billions of times more potent than 

 is necessary for the reproduction of a character ? 

 1 C. E. Walker, The Essentials of Cytology, p. 21. 



