No. 557] SHORTER ARTICLES AND DISCUSSION 309 



Drosophila (if we adapt Morgan's earlier scheme) would then 

 become PVO, VO, PO, PV and 0, instead of PVO, pVO, PvO, 

 PVo and pvO, for red, vermilion, pink, eosin and orange re- 

 spectively. The great difficulty in thus leaving out the small 

 letters comes in distinguishing the heterozygous from the homo- 

 zygous condition. True we can let PVO stand for the hetero- 

 zygous condition of the three factors and PPVVOO for the 

 homozygous condition. Then PPVO would indicate what is 

 now commonly expressed by PPVvOo. But we now use the 

 single letters when we wish merely to designate phenotypic 

 differences or to indicate the factors in gametes, where of course 

 all factors are simplex, and employ duplicate letters only when 

 we desire to indicate genotypic differences. If then the small 

 letters are discarded, we shall need to use some arbitrary sign 

 to distinguish phenotypes from genotypes, else PVO might as 

 now stand for a group of phenotypically like individuals or for a 

 class having the genotypic constitution now commonly indicated 

 by PpVvOo. 



But Castle's suggestion is far from what is outlined above. 

 He would use no letter to represent red eye color in Drosophila 

 but merely write normal. For vermilion he would use v, for 

 pink p, for pink-vermilion pv, etc. My first notion on reading 

 the list of designations for eye color in fruit flies was that 

 Castle used them only as abbreviations for the names of the 

 colors, and v is really a better abbreviation for vermilion than 

 is say Verm, or V'r'm'l'n. Now why, I thought, should one sug- 

 gest such character abbreviations as a revised MendeUan termi- 

 nology when Mendelism is concerned fundamentally with gametic 

 factors and only incidentally with the zygotic characters that 

 happen to develop through the interaction of particular com- 

 binations of gametic factors in a particular environment. But 

 Castle's terminology is not concerned with mere abbreviations 

 for characters, as witness: 



The revised terminology is more convenient than Morgan's in calcu- 

 lating the expected result of any mating, and is equally reliable. The 

 results of every possible mating within the series can be readily com- 

 puted without the confusing presence of the large letters. 



Here I must frankly admit that I have experienced great diffi- 

 culty in using Castle's terminology in calculating the expected 

 results of matings in case of the eye colors in Drosophila, though 

 this is probably due to some mi>nml' -rsr.-imlinir of just how 

 Castle's formulas are to be used. For instance, a cross of v 

 (vermilion) with p (pink) should, if ordinary usage were fol- 



