XII.] CONJUGATION AND SEXUAL REPRODUCTION, 125 



it limits each such predisposition to being present only once, and 

 in only one variety. I conceive of this primitive germ-plasm, as 

 of one single ancestral unit of an existing species, only perhaps 

 relatively larger, its separate predispositions not having been 

 yet reduced to the present minimum. 



All this however was altered in the germ-cells of the first 

 sexually produced individual, in which the nuclear rods of the 

 two parents came together, and together composed the heredi- 

 tary substance of the child. If now, as has been argued above, 

 the paternal and maternal hereditary substances did not fuse 

 but only arranged themselves side by side, there will be found, 

 in the germ-cells of the child, two substances similar as regards 

 the species but dissimilar as regards the individual. If the 

 mass of nuclear substance cannot be increased, both kinds of 

 nuclear substance must be reduced by one half. If we imagine 

 the nuclear material of one such germ-cell to consist of a single 

 thread, one half of it would be made up of paternal and the 

 other half of maternal germ-plasm. 



I call to mind the diagram by which, in an earlier essay \ I 

 endeavoured to make intelligible how the number of ancestral 

 plasms of various kinds which meet together in the germ- 

 plasm are doubled in each successive generation, and how, 

 in the formation of the germ-cells of each generation, the 

 germ-plasms must be reduced to half their size, or their 

 united mass would be doubled in every generation. But in 

 time a limit to this continuous diminution of the ancestral 

 plasms must have been set, and this would occur when the 

 amount of substance necessary to contain all the predispositions 

 of the individual had reached its minimum. Obviously these 

 units cannot become infinitely minute ; however small they 

 may be they must always retain a certain size. This follows 

 from the extremely complicated structure which we must with- 

 out any doubt ascribe to them. These units which make up tlie 

 germ-plasm of living animals I have called ancestral plasms, 

 but my views about them have been misunderstood, and I have 

 been treated as though I had applied the term to the ultimate 

 biological units of idioplasm. Nothing was further from my 

 mind : I look upon the single ancestral plasms as extremely 

 complex, and built up of countless biological units. I have 



^ Vol. I. p. 369. 



