110 HEREDITY AND SELECTION IN SOCIOLOGY 



namely, those of the mother and of the father. By this continuous 

 minghng of individual germ-plasms amphimixis is perpetually 

 efiecting a renewal of the germ-plasm of the species ; eliminating 

 those fortuitous variations which are out of harmony with the con- 

 stitution of that germ-plasm, and disseminating those variations 

 which are beneficial. Thus the value of such continuous mingling 

 is at once a negative and a positive one. In the case of con- 

 sanguinity, when both the parents belong to the same family, 

 amphimixis does not bring about a renewal of the germ-plasm 

 by the importation of new elements. It is true that the germ- 

 plasm is perpetually renewed, but with the same kinds of ele- 

 ments ; and as every fresh amphimixis brings in a fresh lot of 

 ids homogeneous with those already existing, in the course of 

 time a certain monotony must set in in the germ-plasm, due to 

 lack of variety in the nature of the ids composing it. Eventually 

 the germ-plasm will be exclusively composed of identical ids. 



Now, it may very well happen that some of the ids composing 

 the germ-plasm of the family possess a majority of determinants 

 which have entered on a period of unfavourable variation. In 

 the ordinary course of events such ids would be counterbalanced 

 by the introduction of foreign elements. In the case of an endo- 

 gamous race, supposing the variation to have set in among a 

 considerable number of individuals, and supposing only a hmited 

 number of identical kinds of id complexes to compose the germ- 

 plasm, this unfavourable variation will be inherited from both 

 sides ; and the degree of unfavourable variation will continue to 

 increase as the number of identical ids with unfavourable variation 

 determinants increases. It is obvious that in the germ-plasm of 

 a family where constant intermarriage has been long the custom 

 the ids must become increasingly identical ; and in such a family 

 unfavourable variations have an infinitely greater chance of pro- 

 pagating themselves than in a family practising exogamy. 



If we take a family which has long practised consanguinity, 

 and suppose the number of heterogeneous ids in the germ-plasm 



