222 THE AMEBIC AN NATURALIST [Vol. XLIV 



which has white flowers and is like the pure white parent 

 used in the cross. 



De Vries states that in the reds and the Delilahs found 

 in F 2 there w r as a good deal of variation, which is what 

 we might expect from the relations shown in Table II, 

 but he was able to group them, as shown at the bottom 

 of Table II, and get the usual Mendelian ratios. 



I have assumed here that when the enzyme is present in 

 proportions from 82 per cent, to 84 per cent, of that in the 

 pure red race, we get the Delilah type. This merely means 

 that in the Delilah types the amount of enzyme present is 

 near the critical point for the production of red color. 

 There seems to be no question that the environment in 

 the organism itself is a very important determining factor 

 in the development of any character. It is not therefore 

 fanciful to assume that in view of the fact that we get 

 color development only in the petals the tendency to color 

 development might be greater in one part of the petal than 

 in another. We thus see that our assumptions are in 

 agreement with the facts of inheritance made out by de 

 Vries. 



Cryptomeric Characters 



In one of Bateson's Matthiola crosses, 10 between a 

 cream-colored and a white race, he found that the first 

 cross was purple and that amongst its progeny it pro- 

 duced purples, reds, creams and whites. He explained 

 these phenomena by assuming that red is due to two 

 cryptomeric factors which are ineffective unless present 

 in the same cell, while purple was due to a third factor 

 which converted red into purple and which was found in 

 the white variety. These phenomena may be brought in 

 line with our hypothesis, as shown in Table III. 



In this case we evidently have to deal with more than 

 two chemical substances. For the production of the red 

 character itself two of these substances are necessary. 

 It is, of course, possible that in the simpler case consid- 

 ered earlier the same two substances, possibly others also, 



10 See Rep. Ill, Evolu. Com., Roy. Soc, 1906. 



