218 



THE AMERICAN NATURALIST [Vol. XLIV 



ferent stages of a single process of oxidation. This, how- 

 ever, can hardly be true. The fact that we get two or 

 three kinds of pigment deposited in the same region of the 

 organism, apparently at the same time, would indicate 

 that there is more than one oxidation process involved. 

 Furthermore, Eiddle himself points out that, while in 

 some oxidation processes series of colors occur extending 

 from light yellow to black, with reds and browns as inter- 

 mediates, in others the final stages of the oxidation proc- 

 esses are red. Quite a number of different oxidation 

 processes are cited. Now the facts of color inheritance 

 indicate that there may be several oxidation processes, 

 and that in some of them a given color, for instance, red, 

 appears as soon as color appears at all. 



Assuming that the production of enzyme and chromo- 

 gen is a general function of protoplasm, and assuming 

 further that the relative amount of enzyme and chromo- 

 gen present have a determining influence on the stage 

 which the oxidizing process roaches in the organism, the 

 phenomena of Mendelian color inheritance are easily ex- 

 plained without recourse to the idea of unit characters at 

 all, as I shall now attempt to show. It must be under- 

 stood that the figures in the following tables are merely 

 illustrative and are not meant at all to indicate actual 

 amounts. 



TABLE I 



Let us consider first the cross between red and white 

 sweet peas, in which generation F x of the hybrid is red 

 and generation F 2 gives us the ordinary Mendelian ratio 



