128 AMPHIMIXIS OR ESSENTIAL MEANING OF [XII. 



as a rule ova and spermatozoa differ so enormously in size, 

 and above all because the ova of different species vary so 

 greatly in this respect. Moreover, Boveri has shown us 

 that in one and the same species two otherwise indistinguish- 

 able gerili-cells exist, one of which contains twice as many 

 nuclear rods as the other, and therefore as far as we can tell 

 twice the amount of nuclear substance. Hence the ' reducing 

 division ' cannot be a mere division of mass. 



There remain for consideration the ' chromatin elements ' of 

 O. Hertwig. What are these elements? Are they the smallest 

 possible portions of living matter, something like the pangenes 

 of de Vries ? This distinguished botanist in his highly sug- 

 gestive and thoughtful writings has developed the idea that the 

 nuclear substance of the fertilized ovum is composed of count- 

 less very minute particles, called by him pangenes. He thus 

 recalls Darwin's pangenesis, with which his theory has some- 

 thing in common. These pangenes however do not, like the 

 gemmules of Darwin, give rise to cells, but they are the bearers 

 of the various qualities of cells. If we now assume with de 

 Vries that the nuclear substance of germ-cells consists of innu- 

 merable kinds of such pangenes, we may regard these either as 

 uniformly mixed together without any kind of arrangement, 

 or as arranged in a definite order. In the first case, each 

 division (reduction) of the mass would only result in a diminu- 

 tion, and the components of both halves would remain the 

 same : the various kinds of ' chromatin elements ' would not by 

 this means be reduced to half, but all the elements would be 

 contained in each portion. But if these pangenes were arranged 

 in a regular order in the germ-plasm, and if with Hertwig we 

 designate the groups of these as predispositions, without ex- 

 pressing in any way how such predispositions can be conceived, 

 it follows that a halving of the mass of germ-plasm or nuclear 

 substance would give rise to two halves, neither of which 

 would contain all the predispositions necessary for the con- 

 struction of an individual, although both might contain many 

 double predispositions. Hertwig imagines that the predisposi- 

 tions which according to his view (loc. cit. p. no) are present 

 in the germ-plasm of the paternal and maternal germ-cells, 

 mingle together, and de Vries has also assumed this. Hertwig 

 states that 'it is not improbable,' that in the complete union 



f 



