OF MUTANT CHARACTERS. 163 



number of flies tested and proved to be non-cross-overs, it was shown 

 that these loci were probably within 2 or 3 units of each other. I^ter 

 a more refined method was used, whereby only such flies were tested 

 as were known to be cross-overs very close to black, that is, between 

 dachs and black or between black and purple. The results of these 

 tests only served to bring out more clearly that the locus of jaunty 

 was very close to that of black. 



When the black X jaunty cross failed to give the double recessive, 

 a more roundabout method was started by crossing jaunty to curved. 

 The F2 jaunties (table 24) were inbred instead of being crossed to the 

 F2 curved. This was to avoid possible doubt as to the double form 

 being separable from curved. Any curved flies descended from the 

 inbred F2 jaunties must be the double recessive form jaunty curved. 

 Fortunately, jaunty curved flies appeared in F3 and a stock was made 

 up. The wings of the jaunty curved flies were easily classified as 

 curved, and, surprisingly enough, the jauntiness was usually detectable 

 in the wing-tips. The intended experiments were not carried out 

 with this stock, but Muller (Muller, 1916) used it in building up the 

 multiple heterozygotes which he tested. The tests of Muller furnished 

 462 flies, of which possibly one was a cross-over between black and 

 jaunty. If this questioned cross-over were genuine, jaunty would be 

 mapped to the right of black and 0.2 unit away. If the apparent cross- 

 over were due to error, then we are only certain that jaunty is excep- 

 tionally close to black. 



A large series of crosses between jaunty and the third-chromosome 

 mutant pink, carried out by a graduate student, Mrs. Binkley, gave 

 most interesting and puzzling results, the data for which was unfor- 

 tunately lost with the second-chromosome summaries. As we recall the 

 case, the F2 ratios were more nearly 4:2:2:1 than 9:3:3:1. The 

 back-crosses likewise gave aberrant ratios which we could not explain 

 as due to viability effects and which were not due to linkage disturb- 

 ances in the relation of jaunty and pink, since the back-cross tests of 

 Fi males and Fi females gave like results, in which the percentage of 

 recombination was approximately 50, as expected. Besides these 

 peculiarities of the ratios there was a curious appearance of white- 

 eyed flies which in inheritance reminded us somewhat of the case of 

 "whiting," an eosin modifier worked out by Bridges, but which was 

 far more elusive in the manner of the alternate appearance and dis- 

 appearance of white and in the relation of pink to the case. 



VALUATION OF JAUNTY. 



Though jaunty is easily separable from wild-type and is of good 

 viability and behavior, it is not much used. The main drawback is 

 the closeness of its locus to another locus already satisfactorily filled 

 by black. This region of the chromosome could be represented 



