94 PROBLEMS OF GENETICS 



cannot surmise any source from which they may have been 

 derived. Just as when typhoid fever breaks out in his district 

 the medical officer of health knows for certain that the bacillus of 

 typhoid fever has by some means been brought into that district 

 so do we know that when first dominant white fowls arose in the 

 evolution of the domestic breeds, by some means the factor for 

 dominant whiteness got into a bird, or into at least one of its 

 germ-cells. Whence it came we cannot surmise. 



Whether we look to the outer world or to some rearrangement 

 within the organism itself, the prospect of finding a source of 

 such new elements is equally hopeless. 



Leaving this fundamental question aside as one which it is 

 as yet quite unprofitable to discuss, we are on safe ground in 

 foreseeing that the future classification of substantive variations, 

 which genetic research must before long make possible, will be 

 based on a reference to the modes of action of the several factors. 

 Some will be seen to produce their effects by oxidation, some by 

 reduction, some by generating substances of various types, 

 sugars, enzymes, activators, and so forth. It may thus be 

 anticipated that the relation of varieties to each other and to 

 types from which they are derived will be expressible in terms 

 of definite synthetical formulae. Clearly it will not for an in- 

 definite time be possible to do this in practice for more than a 

 few species and for characters especially amenable to experi- 

 mental tests, but as soon as the applicability of such treatment 

 is generally understood the influence on systematics must be 

 immediate and profound, for the nature of the problem will at 

 length be clear and, though the ideal may be unattainable, its 

 significance cannot be gainsaid. 



Note. — With hesitation I allow this chapter to appear in 

 the form in which it was printed a year ago, but in passing it for 

 the press after that interval I feel it necessary to call attention 

 to a possible line of argument not hitherto introduced. 



In all our discussions we have felt justified in declaring that 

 the dominance of any character indicates that some factor is 



