102 HEREDITY AND VARIATION 



continuity of the germ-plasm has to some extent restricted. The 

 stream of life passing through a continuous lineage of germ-cells 

 is, so to speak, hemmed in, but it continually tends to deviate 

 from this course, and there are not a few opportunities some 

 normally recurrent, some more accidental which allow of this 

 or even prompt it. In some cases, as we have said, it is impossible 

 to distinguish offspring from parent, or brother from brother, 

 or cousin from cousin. On what does this completeness of heredi- 

 tary resemblance (i.e. the absence of variation) depend ? 



It means, in the case of unicellular organisms, that the sepa- 

 rated parts are identical in substance and carry on the complete 

 organisation of the parent cell in absolute integrity. In the case 

 of multicellular organisms it depends on the same thing. The cell 

 which in the embryo begins the germ-cell lineage may be identical 

 with the fertilised ovum, and the complete heritage may be con- 

 tinued intact through successive cell-divisions until the next 

 generation is started, and the process begins anew. The com- 

 pleteness of hereditary resemblances depends, in Bateson's 

 phrase, on " that qualitative symmetry characteristic of all non- 

 differentiating cell-divisions." 



It seems, therefore, useful to say that variation is " the expres- 

 sion of a qualitative asymmetry beginning in gametogenesis. 

 Variation is a novel cell-division." But to tell what specific cause 

 induces this novelty is still beyond our power. Yet we can point 

 to certain conditions which may induce novelty or qualitative 

 asymmetry in gametogenesis. Thus, there is the complex change- 

 ful environment of the developing germ-cells, there is the possible 

 struggle of analogous hereditary units or determinants for sus- 

 tenance, there is the complex process of reduction which occurs 

 during the maturation of the germ-cells, and there are the chances 

 of new combinations and permutations in fertilisation. 



Results of A mphimixis. That amphimixis is one of the provocatives 

 of variations is strongly suggested by what results when two breeds 



