XII.] CONJUGATION AND SEXUAL REPRODUCTION. 173 



I hold that this is not only possible, but even probable. The 

 difference between the embryogenies of two allied species not 

 only depends upon the characteristic differentiation of the single 

 cells which compose the body, but also equally upon their 

 number, both relatively and absolutely, in all parts of the body. 

 One and the same part of the body may be long in one species, 

 and short in another : more cells will be required for the con- 

 struction of the former than for the latter, or, in other words, 

 the earliest embryonic cells of this part of the body must 

 multiply more rapidly in one species than the other. If now 

 this mode of cell-division is determmed by the specific nature 

 of the above-named centrosomata of such cells, it follows that 

 embyrogeny must be essentially controlled by the centrosoma, 

 i. e. by a part w^hich lies in the cell-body, and which we have 

 hitherto regarded as a part of it. 



We do not however know that this is really the case ; pos- 

 sibly the centrosoma may have been originally derived from 

 the nucleus. But even if we admit that it is, not only in posi- 

 tion but also in origin, a part of the cell-body, we must never- 

 theless believe that its activity is dependent on the nucleus and 

 nuclear substance. The centrosomata form the active, and thus 

 the chief part of that remarkable mechanism which controls 

 nuclear division. If this mechanism is once set in motion, it 

 completes the division in the manner described above, just as a 

 spinning machine twists its numerous threads, but that the 

 apparatus is put in motion, does not depend upon itself, but 

 obviously upon the internal conditions of the cell, which react 

 upon the mechanism for division, so that it is compelled to 

 enter upon activity. How can we otherwise understand Flem- 

 ming's recent discovery that the centrosoma is always present 

 in the cell-body, but only periodically initiates the nuclear 

 division ? Now the internal condition of the cell is, as we are 

 aware, primarily determined, in all its qualities, by the nuclear 

 substance, and consequently the centrosoma and the dependent 

 mechanism for division are ultimately controlled by the nuclear 

 substance, which regulates the rhythm of cell-division and 

 dominates the whole structure of the organism. If it were 

 otherwise, this nuclear material could not be the hereditary 

 substance— the material basis of hereditary qualities '. 



^ Fol's recent observation that the centrosomata of ovum and spermato- 



