l86 AMPHIMIXIS OR ESSENTIAL MEANING OF [XII. 



two conjugating animals invariably gives rise to the reproductive 

 nuclei. This is alone capable of further development, while the 

 three which occupy more remote positions are destined to 

 disintegrate and disappear. It is only the accident of position 

 which fixes upon that one of the four which shall undergo 

 development. 



If this be true, the causa effia'ens which decides upon that one 

 of the grand-daughter-nuclei which shall give rise to the repro- 

 ductive nuclei must be sought for in some influence which is 

 exercised by the corresponding nucleus of the other animal, 

 and which naturally affects most strongly that nucleus which 

 lies nearest to it. 



At any rate we are justified in assuming that the idioplasm 

 of the four grand-daughter-nuclei of the micronucleus is, apart 

 from individual differences, essentially similar, i.e. that each 

 contains the same number of idants in the same stage of 

 development, and this number will be half that which is normal 

 for the species in question. Thus nine would be the number in 

 Urostyla grandis, which would be reached in the following 

 manner. According to my supposition, during the growth of 

 the micronucleus from A^ to A- (see Fig. XII), the i8 idants are 

 doubled by longitudinal fission, becoming 36 ; the two following 

 ' reducing divisions ' not only diminish the idants from 36 to 18 

 in stage B, and from 18 to 9 in stage C, but lead to afresh 

 grouping of the idants Just as in the analogous ' reducing divisions ' 

 of the egg- and sperm-cell. Since the 18 idants are doubled, it is 

 clear that each one of them will be represented by two idants in 

 the enlarged micronucleus of stage y^'^, and hence the two 

 ' reducing divisions ' can originate a number of difterent com- 

 binations of 9 idants, just as in the egg- and sperm-cell, described 

 in the first part of this essay. 



Although in any single individual, only four out of the 

 numerous possible combinations would become actual, we may 

 perhaps perceive, — in this very fact that there are always at least 

 four different possibilities to select from, — the reason why all 

 four grand-daughter-nuclei of the micronucleus are formed, 

 and why both the daughter-nuclei undergo the second ' reduc- 

 ing division,' while the division of one of them alone would 

 suffice to ensure the origin of two reproductive nuclei. 



I 



