No. 518] 



VARIATION 



71 



showing segregation of yellows and whites at the ratio of 

 3:1, were obtained, 



One other case should be mentioned. One ear of a dent 

 variety of unknown parentage obtained for another 

 purpose was found to have some apparently hetero- 

 zygous yellow kernels. Seven selfed ears were obtained 

 from them, of which two were pure yellow. The other 

 five ears each gave the di-hybrid ratio. There was a 

 total of 1906 yellow seeds to 181 white seeds, which is 

 reasonably close to the expected ratio, 1956 yellow to 131 

 white. 



It is to be regretted that I can present no other case of 

 this class that has been fully worked out, although several 

 other characters which I have under observation in both 

 maize and tobacco seem likely to be included ultimately. 

 Nevertheless, the fact that we have to deal with conditions 

 of this kind in studying inheritance is established; grant- 

 ing only that they will be somewhat numerous, it opens up 

 an entirely new outlook in the field of genetics. 



In certain cases it would appear that we may have 

 several allelomorpliic pairs each of which is inherited in- 

 dependently of the others, and each of which is separately 

 capable of forming the same character. When present in 

 different numbers in different individuals, these units 

 simply form quantitative differences. It may be objected 

 that we do not know that two colors that appear the same 

 physically are exactly the same chemically. That is true; 

 but Nilsson-Ehle's case of several unit characters for 

 presence of ligule in oats is certainly one where each of 

 several Mendelian units forms exactly the same char- 

 acter. It may be that there is a kind of biological 

 isomerism, in which, instead of molecules of the same 

 formula having different physical properties, there are 

 isomers capable of forming the same character, although, 

 through difference in construction, they are not allelo- 

 morpliic to each other. At least it is quite a probable 

 supposition that through imperfections in the mechanism 

 of heredity an individual possessing a certain character 



