OF MUTANT CHARACTERS. 283 



parable to that found here. It is, however, possible that the one 

 found by Bridges (C Us) may on further investigation be found 

 comparable. 



On the whole, the evidence is in entire agreement with the assump- 

 tion of dachs-deficiency, but at the same time is not in such disagree- 

 ment with the linked-lethal view as to disprove it. Before either 

 alternative can be dropped it is necessary to repeat the two tests 

 which offer definite solution the search for dachs flies from an inbred 

 line and far more rigid tests of the system of linkage that is present. 

 Meanwhile, the ambiguous term " dachs-lethal " will be retained to 

 cover both possibilities.. 



SQUAT (5,). 



(Text-fig. 85.) 



ORIGIN OF SQUAT. 



In tracing the course of "high" non-disjunction (Bridges, 1916) a 

 female heterozygous for the sex-linked genes vermilion, sable, garnet, 



v * 



and forked ( - =- } was tested for the occurrence and percentage of 

 \+ s g +/ 



secondary exceptions by out-crossing to a male from the bar (sex- 

 linked dominant) stock. One of the regular vermilion forked sons of 

 this pair (culture No. 2480) was found (November 29, 1915) whose 

 wings, legs, and body were considerably shorter than normal, giving 

 a "squat' ' appearance to the fly. Only one such squat fly was found 

 among 372 offspring from this pair, which in such cases is always a 

 strong indication that the mutant is either dominant or sex-linked. 



In order that a non-sex-linked recessive should appear in a culture, 

 both parents must be heterozygous for the gene, and in such cases the 

 recessive character appears as a quarter of the individuals and not as 

 a single individual, as was here observed. 



INHERITANCE OF SQUAT. 



The squat male was out-crossed to a wild female, and in FI produced 

 (culture 2635) a total of 53 squat flies to 47 not-squats. This means 

 that the mutant is a dominant and the original male was likewise a 

 heterozygous dominant. That it is an autosomal dominant, rather 

 than a sex-linked dominant, was proved by the fact that half (22) of 

 the FI squats were males; had the squat been sex-linked none of the 

 sons could have shown the mutant, since their single X chromosome 

 comes from their mother and not from their father. 



A squat male and female from FI were inbred, and gave in F 2 121 

 squats to 53 not-squats (No. 2728). This seemed to be an approxi- 

 mation of a 3 : 1 ratio rather than of a 2 : 1 ratio, and was thought to 

 indicate that the dominant is probably not lethal when homozygous. 



