XII.] CONJUGATION AND SEXUAL REPRODUCTION. 1 23 



The Double Division of the Nuclear Substance in the Formation 

 of Germ-cells. 



With regard to the egg, the following question can be for- 

 mulated — What is the meaning of the first division in the 

 formation of polar bodies, since the second alone would suffice 

 to halve the nuclear substance? With regard to the sperm- 

 mother-cell however the question must run, — why should 

 division take place twice, when its single occurrence would have 

 sufficed to reduce the nuclear loops by one half? The simplest 

 answer to these two questions is to be found in the fact that the 

 number of nuclear rods is doubled at the beginning of the 

 reducing process, and must therefore be quartered if a diminution 

 to one half the normal number be the ultimate necessity. This 

 leads us to enquire why the preliminary doubling of the nuclear 

 rods is necessary. 



Regarding spermatogenesis only, it might be maintained that 

 here we are simply dealing with a process for increasing the 

 number of spermatozoa as far as possible, but if we attempt 

 hus to explain a fourfold instead of a twofold increase, com- 

 parison with the egg-mother-cell, producing four descendants 

 of which one only undergoes development, renders any further 

 discussion of this idea superfluous. 



In attempting to explain the phenomenon I start from the 

 conception which led me to the idea of a ' reducing division,' 

 i. e. the building up of the germ-plasm, that is the active sub- 

 stance of the nuclear rods, from innumerable ancestral units 

 As I explained on the first statement of this idea, it is a view 

 which is necessarily suggested if we accept certain premisses, 

 the chief of which is, that the hereditary substance from the two 

 parents does not altogether become one during the fusion 

 which occurs at fertihzation, but that each retains a certain 

 independence. This agrees with observed facts in so far that, 

 as a result of fertilization, the paternal and maternal rods come 

 to lie close to one another in the same nucleus, but undergo no 

 true fusion into a single mass. If we assume that this remains 

 true during the whole ontogeny, we can only suppose that half the 

 nuclear rods of every cell are paternal and half maternal and that 

 both these simultaneously influence the cell. We do not yet 

 understand how this takes place, and must for the present dis- 



