DEVELOPMENT BY NATURAL SELECTION. 123 



the germ-cells carries with it the corollary that all true 

 or heritable variation is primary variation of the germ- 

 cell. But as different individuals of a species vary in 

 spite of this supposed physiological isolation, how does 

 the variation occur ? To help in this dilemma, the large 

 and very important class of facts associated with the 

 name of the Abbe Mendel has been invoked. He made 

 the original definite discovery that in sexual reproduction 

 the characters inherited are not an evenly-blended mix- 

 ture of those of the parents, but that some, at least, are 

 exclusively derived from one parent, or may even be 

 missing ; and that the proportion in which these " Men- 

 delian " characters, or " factors," are distributed among 

 the progeny can often be calculated on the abstract 

 theory of probabilities that is to say, as a mere matter 

 of chance. " Factors " may also influence one another 

 in different ways, and may thus be either more or less 

 capable of combination in one individual. In these 

 ways there is great scope for variation in progeny, apart 

 altogether from immediate influence on the developed 

 organism of varying environment. On the theory that 

 the germ-cells are physiologically isolated from the 

 parent organism and its environment, we can thus 

 account for variation, provided we assume that variation 

 is due to a mere re-shuffling, or dropping out, of original 

 " factors " ; and this is the substance of the position 

 taken up by Professor Bateson in his Presidential 

 Address at the meeting of the British Association in 

 1914. This position, as is recognised by Bateson, tends, 

 logically, towards the view that all the " factors " were 

 present in the original amoebae, or whatever other lower 

 organisms we are all descended from. On the mechan- 



