142 



THE SECOND-CHROMOSOME GROUP 



crossing a cut-snub female to truncate males, in which case all the sons 

 were cut. Among these cut sons the ratio of snub to normal was almost 

 exactly 2 : 1 (table 11). The fact that this experiment gave a close 

 approach to the ratio expected if the truncate-snub compound is lethal 

 could not be accepted as proving that theory, because there might still 

 be enough flies failing to show the truncation, so that the ratio is really 

 the normal 3 : 1 ratio. 



TABLE 10. F\ ratios obtained from crosses of truncate 9 X snub cf (cultures Ito4)> 

 and snub 9 X truncate d* (cultures 5 and 6). 



TABLE 11. FI ratios obtained from crosses of cut 

 snub 9 X truncate cf 1 (pairs). 



The scheme finally followed eliminated all classification of both 

 truncates and snubs and depended for identification upon the readily 

 classifiable "character star and upon the assumed lethal nature of the 

 truncate-snub compound. By mating star to truncate, flies can be 

 obtained carrying the star gene in one II-chromosome and truncate in 

 ( S' 4- 



the other 



V+ 



Two such flies mated together would give a 



2 star to 1 not-star ratio, unless homozygous truncate were lethal. But 

 since homozygous truncate is known to be lethal, this ratio becomes 

 modified by the death of most of the not-star flies. A few not- 

 star offspring will survive because of crossing-over in the female 



S' 



r 



r 



There is a precise relation between 



the amount of this crossing-over and the number of not-star flies which 



