238 



Sturtevant. 



giving 39 •! per cent of crossing over in rather small numbers (399), 

 about 33 being the expected result. 



All possible combinations involving two of the six loci at a time 

 have been made. As before I have used in calculating the percent of 

 cross -overs (so far as possible) those matings in which no other sex- 

 linked genes are involved, since data without such complications are less 

 likely to be inaccurate because of the effects of differential viability. 



Table I. 



YW 



-»—I — 



V 



— I- 



M 



-1— 



R 



-t- 



— f- 



LI 



33.5 3t5 



Fig. 1. 



53.3 57.7 



Data of the desired sort are available for all combinations except YBr, 

 and a summary of them is given in table I. The number given for 

 YBr is that obtained in an experiment involving also V and v (see next 

 section), and is based on a count of only 191 fKes, so that it is not very 

 reliable. In tliis table the data for "coupling" and "repulsion" (Bateson) 

 are combined. All data involving the four allelomorphs in the locus W 

 (W, w, w% w"'') are also combined, since they show the same linkage. 

 The detailed data are tabulated at the end of this paper (tables XIV, XV). 

 On the basis of these results the genes fall into a linear series 

 in the relative positions shown in figure 1. 



