OF MUTANT CHARACTERS. 



141 



Table 9. — The offspring 

 given by pairs of cut 

 snub flies from the slock 

 of cut snub. 



cases a lethal was present in the homologous chromosome. Autosomal 

 lethals are very plentiful and have been clearly demonstrated in many 

 other cases, though in this case no further test has yet been made of the 

 correctness of this explanation. 



That the same kind of modifiers were present in the snub stock as 

 were resopnsible for the short-winged types of truncate appeared 

 certain from the rather sharp differences between the cut snubs which 

 occurred in the inbred stock. While most of the cut snubs were of the 

 type described, in which the wing is nearly as long as the normal cut 

 but differently shaped at the tip, certain ones were much shorter and 

 the oblique truncation was very marked. These shorter ones, just as 

 in the case of the selected truncate stock, 

 were most numerous in those cultures in 

 which the expected 2 : 1 ratio was most closely 

 approached. Pair 10 of table 9 showed a 

 ratio of 31 short snubs to 63 of medium or 

 slight truncation to 46 which showed no 

 truncation; this rather close approach to a 

 1:2:1 ratio suggested that probably only 

 one such modifier was present in the cross. 



All of the characteristics of snub thus far 

 found have agreed exactly with those of 

 "specific" truncate. If it should be found 

 that the chromosome locus of snub were the 

 same as for the original truncate, then we 

 should conclude that the mutation is specifi- 

 cally truncate. 



Because of the fluctuating nature of the 

 dominance of snub and its easy modification, 

 a direct linkage experiment offered diffi- 

 culties. A more exact method would be to 



establish the lethal nature of the truncate-snub compound. This 

 could be done by showing that the Fi ratio obtained by crossing 

 truncate by snub was a derivative of a 2:1 instead of a 3:1 ratio. 

 The observed ratio of 174 truncate to 132 not-truncate in the Fi from 

 this cross would somewhat favor the view that the ratio is 2 : 1 rather 

 than 3:1, and that snub is therefore truncate (table 10, Morgan). 

 But here again the uncertainty that the number actually showing 

 the truncate character would be a close enough approach to the 

 number heterozygous for truncate, so that we could decide whether 

 we were really dealing with a 2 : 1 or a 3 : 1 ratio made the results of 

 such experiments of doubtful value. 



It was recalled that cut had acted as an int^nsifier of snub, so that a 

 larger proportion of cut flies showed the snub character than was the 

 case among the not-cut flies. Advantage of this fact was taken by 



