PRESEIST PROBLEMS IN EVOLUTION AND HEREDITV. 365 



still left ill doubt. Why is tlie cliroiiiatiii sul)staii('e doubled iii the 

 mother-cells so that two sueeessive sul)-divisious are necessary to reduce 

 it to halt the ori_i;inal ([uaiitity? Ilertwiy; has not atteuipted to answer 

 this question, as he prefers to wait for further research. Weismaini, 

 however, who is iiiifortuiiately cut off from research hy failing eyesight, 

 has offered a spcu-ulative solution to this problem which he trusts may 

 guide future investigation. 



This leads me to say a few words in regard to his conception of the 

 relation of the chromatin to heredity. ( 1 ) His lirst i)remise is that in fer- 

 tilization there is not a fusion of chromatin, but that a certain independ- 

 ence is preserved between the maternal and paternal elements, based 

 upon the observed fact that the two pairs of rods do not fuse but lie side 

 by side, and upon the assumi)tion that these i)airs are kept distinct in 

 each cell through all the subsequent stages of embryonic and adult devel- 

 opment. If this is the case, the hereditary substance contributed by the 

 father would remain separate from that conti ibute<l l)y the mother, 

 throughout. (-) " Each of these ])airs would be made up of the col- 

 lective i)redispositioiis which are indispensable tV)r the building up of 

 an individual, l)ut each possesses an individual character, for they are 

 not entirely alike. I have called such units ' amu'stral plasms,' and 

 r coiicei\(' tiiat they are contained in iiunil>ers in the chromatin of the 

 mature germ cells of living organisms, also that the older nuclear rods 

 are made up of a certain number of these. - - 01>viously these 



units <'an not become infinitely minute; howe\er small they may be 

 they must always retain a certain si/.e. This follows from the extremely 

 complicated stiucture which we must without any doubt ascribe to 

 them." These units are not, liowe\ er, ultimate, they are in turn ex- 

 tremely complex, and are composed of countless biological units of the 

 kind conceived of by Njigeli and otlu^rs. (3) The rechiction of the chro- 

 matin only ac(piires a meaning when taken in connection with the 

 above supposition of distinct ancestial i^lasms, and has no meaning 

 if we accept llertwig's view that there is a complete fusion of nni- 

 ternal and paternal germ plasm. This meaning is that redu(;tion in 

 the maturation of germ cells is siii gcnerUs^ it does not divide the ances- 

 tial plasms into two similar groups, but one daughter-cell receives one 

 set of germ plasms or hereditary predis])ositioiis, and another daughter 

 cell receives another; reduction is thus ditferential. According to tiiis 

 view the four sperm and ovum daughter-cells would each contain a 

 different set of ancestral jdasms. (4) The fact that the chromatin sub- 

 stance is doubled in the sperm and o\uiii mother-cells, so that we ob- 

 serve double the number of rods characteristic of the species, is to be 

 explained as an adai)tation to the re(piiremeiits of natural selection, 

 ^or this doubling and subseiiuent double di\'ision render possible an 

 inhnite numl)er of combinations (as many, in fact, as there are individ- 

 uals) for selection to operate uixtii. 



This explanation of W'eismann's is an example of ids apotheosis of 



