352 THE CELL 



length on p. 235, under the title of division with reduction, is so 

 distributed amongst the four grand-daughter cells, that each of 

 them only contains half the nuclear mass of an ordinary cell, and 

 hence only half the normal number of nuclear segments. 



The second course occurs during the process of fertilisation in 

 Closterium. Here, according to the observations of Klebahn (VII. 

 27), the germinal nucleus, formed by the fusion of two nuclei, 

 divides consecutively twice without entering into a state of rest, 

 just as when pole-cells are formed. Of the four vesicular nuclei, 

 two disintegrate, so that each half of the original mother-cell 

 contains only one nucleus, which possesses only a fourth part of 

 the germ-nucleus, instead of one half, as in normal division (see 

 the description and figures on pp. 280, 281). 



If, according to our assumption, the nuclear mass is identical 

 with the hereditary mass, we must conclude, arguing from the 

 process of division with reduction, that the hereditary mass may be 

 divided up to a certain point, without losing its power of reproducing 

 the whole out of itself. The question then arises, as to how far this 

 conception is admissible. 



Weismann and I both lay emphasis upon the necessity of a 

 reduction of mass, but we have arrived at different conclusions as 

 regards particulars. 



In his ancestral germ-plasm theory, Weismann starts with the 

 supposition, that in the hereditary mass the paternal and maternal 

 portions having kept themselves apart, form units, which he calls 

 ancestral germ plasms. He assumes that these are very compli- 

 cated in structure, being composed of extremely numerous biological 

 units. At each new act of fertilisation still more numerous ancestral 

 germ-plasms come together. Supposing that we revert to the ' 

 beginning o f the whole process of fertilisation, then in the tenth 

 generation 1024 different ancestral plasms must have taken part 

 in the formation of the hereditary mass. But since the total mass 

 of the latter does not double itself with each act of fertilisation, 

 Weismann makes the ancestral plasms divisible in the first stages 

 of the process, and supposes that they are transmitted to the 

 following generation, reduced each time by one half; " at last, 

 however," he continues, " the limit of this constant diminution of 

 the ancestral plasms must be reached, and this must occur when 

 the mass of substance, which is necessary in order that all 

 elemental germs of the individual may be contained therein, has 

 reached its minimum." 



