32 HERBERT SPENCER AND 



two varieties of sweet-pea both having white flowers 

 are crossed, the resulting generation is all purple. 

 The two white peas are apparently similar, differing 

 only in the shape of their pollen grains. Each kind 

 will breed true for generations, but when crossed 

 they produce the coloured sweet-pea. This coloured 

 form, when self-fertilized, produces several varieties 

 of purples, reds, and whites. I have not the time to 

 explain this phenomenon fully, but may state that 

 analysis showed that the facts are susceptible of 

 the following explanation : One of the original white 

 flowers contained a factor, which we may call C, which 

 by itself is incapable of producing colour. The other 

 contained a factor, which we may call D, which by itself 

 is incapable of producing colour. But when the two 

 factors are combined, as they are in crossing the two 

 varieties, colour is produced. The subsequent experiments 

 show that it is red colour, but the hybrid generation is 

 purple. It was shown that this was because one of the 

 original parent whites contained yet another factor B, 

 which in the presence of both C and R produces purple, but 

 does not produce any colour if it meets only C or only R. 



I am informed that it has since been shown that in 

 the case of these sap colours in plants, one at least of 

 the factors has been discovered to be a chemical sub- 

 stance diffused in the sap, but the results have not yet 

 been published. 



These discoveries seem to introduce a complication 

 into our conception of factors in the germ-cells, but it 

 seems probable that they will be found quite consistent 

 with the results of experimental work on developing 

 ova, such as I have described. For it has been shown 

 that certain necessary constituents for the production 

 of adult characters are contained in the cytoplasm of the 

 egg. From a large number of considerations we are 

 obliged to conclude that other constituents are con- 

 tained in the nucleus, and it . seems probable that for 



