318 



INHERITED LINKAGE VARIATIONS 



in this same direction (Bridges, 1915; Plough, 1917), the decrease is 

 probably significant. More exact methods (see Plough, 1917) are 

 necessary for obtaining clear-cut data on this point, as has already 

 been stated. 



Here, as also in the case of C IIr , wherever reliable information 

 regarding coincidence is available, the value is not far from the one 

 found in females that contain neither Cm nor C// r (see Bridges 



TABLE 15. Cl1 ' 



TABLE 16. 



C IIr 



and Morgan, 1919). But in no case does this include a region in the 

 " sphere of influence" of the cross-over gene present; for in all such 

 regions the percentage of crossing-over is too small to give statistically 

 reliable results. 



C ni 



Cultures 677 and 678 both contained females, 



6 p r 



'in 



(678 was 



also heterozygous for m r ), mated to b p r c males. A cross-over female, 



