Sex-limited inheritance in Lychnis dioica L. 289 



prior symbols were themselves chosen, it being required only that they 

 conform to the conventional method stated above under (1) (a), most of 

 the symbols would soon become as familiar to the geneticist as are the 

 characteristics they represent. 



(3) The use of subscripts, exponents, accents and different styles 

 of type should be abandoned, except as temporary didactic expedients. 

 The positive homozygote is by some writers represented by a subscript- two, 

 as A 2 , instead of A A, but if such a scheme were generally adopted the 

 homozygous purple -flowered individual would be Pa and a homozygous 

 female F 2 , but these symbols already have general genetic usage with 

 other meanings. Duplicate determiners might be represented by suc- 

 cessive alphabetic symbols, the first being the initial of the character 

 which is represented in duplicate; thus instead of Ri, R 2 , Rs or R', R", R'" 

 for NILSSCXN-EHLE'S red pericarp-colors of wheat, we would have R, S, T; 

 for the ligula in oats, L, M, N, 0; and for EAST'S two determiners for 

 yellow endosperm in maize, Y and Z. 



(4) It sometimes happens that a character is not easily described 

 or suggested by any single word. In such a case any unoccupied alpha- 

 betical S3 r mbol may be used, and except for good and sufficient reasons 

 such a symbol once adopted should be accepted by other writers who 

 deal with the same character. Several writers have intentionally aban- 

 doned the use of suggestive symbols, for the laudable purpose of em- 

 phasizing the fact that no unit -character is the product of the single 

 gene represented by the symbol. It appears to me that the true rela- 

 tion between the symbol and the unit- character can be made sufficiently 

 clear without sacrificing the didactic advantages of the suggestive symbol. 



While none of these suggestions, except perhaps the first, should 

 be made a hard and fast rule, a general adherence to them would greatly 

 aid in making genetic literature more lucid. At present the number 

 of synonymous symbols and formulae involving any given character are 

 roughly proportional to the number of investigators who have studied 

 that character. This is well illustrated by the genetic formulae of the 

 sexes, for the sex-phenomena are more universally interesting than any 

 other single unit- character-complex with which geneticists deal and the 

 number of different formulations has become surprisingly large. 



In this case the confusion has been more than doubled by the fact 

 that cytology and experimental breeding have each developed a formu- 

 lation of the sexes, and that the formulae derived from these two fields 

 of investigation have not only had considerable influence upon each 

 other, but that they have often been more or less intermingled. 



