SPONTANEOUS GENERATION AND EVOLUTION : CONCLUSION 377 



regard to the adaptations of the parts which develop from the ovum, 

 especially during the course of development, and it is also in- 

 dispensable all through life in maintaining the equilibrium of the 

 parts, and their adaptation to the varj-ing degree of function 

 required from them (use and disuse). But its influence does not 

 reach directly beyond the life of the individual, since it can only 

 give rise to ' transient ' modifications, that is, to changes which cease 

 with the individual life. 



In contrast to this is Germinal Selection, which depends upon the 

 struggle of the parts of the germ-plasm, and thus only occurs in 

 organisms with differentiation of somatoplasm and germ-plasm, 

 especially in all Metazoa and Metaphyta — forming in these 

 the basis of all hereditary variations. But not every individual 

 variation to which germinal selection gives rise persists and spreads 

 gradually over the whole species, for, apart from the cases we have 

 already mentioned, in which indiflerent variations favoured by external 

 circumstances gain the victory, this happens only if the variations 

 in question are of use to their bearer, the individual. Any variation 

 whatever may arise in a particular individual purely through 

 germinal selection, but it is only the higher form of selection — 

 Personal Selection— that decides whether the variation is to persist 

 and to spread to many descendants so that it ultimately becomes 

 the common property of the species. Germinal and personal selection 

 are thus continually interacting, so that germinal selection contiuuall}^ 

 presents hereditarj^ variations, and personal selection rejects those 

 that are detrimental and accepts those that are useful. I will not 

 repeat any exposition of the marvellous way in which personal 

 selection reacts upon germinal selection, and prevents it from 

 continuing to offer unfavourable variations, and compels it to give 

 rise to what is favourable in ever-increasing potency. Although it 

 apparently selects only the best-adapted i^ersons for breeding, it 

 really selects the favourable id-combinations of the germ-plasm, that 

 is, those which contain the greatest number of favourably varying 

 determinants. We saw that this depends upon the multiplicity of 

 ids in the germ-plasm, since every primary constituent of the body 

 is represented in the germ-plasm, not once only, but many times, 

 and it is always half of the homologous determinants contained in 

 the germ-plasm of an individual which reach each of its germ-cells, 

 always, moreover, in a difierent combination. Thus, with the 

 rejection of an individual by personal selection, a particular combination 

 of ids, a particular kind of germ-plasm is in reality removed, and thus 

 prevented from having any further influence upon the evolution of 



