June 27, 1913] 



SCIENCE 



991 



Except for disturbances due to difierences in 

 viability these are obviously 9:3:3:1 ratios, 

 except the third, wbicli is 1:1:1:1, as ex- 

 pected. Evidently pink is not linked either 

 to black or to vestigial. 



In the balloon-vcinged stock (see Morgan, 

 '11) there appeared a very dark mutant, which 

 has been named ebony. This fly is darker and 

 more shiny than the black fly, and has a 

 slightly greenish, smoky color when young. 

 It behaves as a simple Mendelian recessive to 

 the normal gray color, though the heterozy- 

 gous form is often somewhat darker than the 

 wild fly. The females of this race are often 

 sterile, and seldom produce many offspring, 

 but the males breed freely when mated with 

 other females, and produce many offspring. 

 That the gene concerned is not in the second 

 chromosome is indicated by the following ex- 

 periment : 



these flies were used for " coupling " exjjeri- 

 ments the result was as follows: 



Ebony pink 



X 

 Gray red 



^^ 



gray red 



{gray red 272 

 gray pink 12 

 ebony red 10 

 ebony pink 65 



The result indicates that the two genes in- 

 volved are really linked in the ordinary sense. 

 In order to get more exact data doubly hetero- 

 zygous flies have been crossed to double re- 

 cessives (ebony pink). Owing to the difficulty 

 of breeding ebony females it has not yet been 

 possible to make any extensive tests of the 

 amount of crossing over in males, but the 88 

 flies so far raised from doubly heterozygous 

 males by ebony pink females have none of 

 them been cross-overs. Similar tests of 

 doubly heterozygous females have disclosed 

 the existence of much variability in the 

 amount of crossing over, the extreme yalues 



Vestigial gray J 



X 

 Long ebony (^ 



(long gray 268 



long ebony 79 



vestigial gray 94 

 I vestigial ebony 24 



This is a simple 9:3:3:1 ratio, except that 

 the viability of the ebonies is not quite as 

 good as that of the grays. Ebony has also 

 been shovsm to be independent of two other 

 wing factors in the second chromosome — 

 curved and balloon. 



When pink-eyed flies were crossed to those 

 with ebony body color, the following result 

 was obtained: 



Gray pink 



X 

 Ebony red 



gray red s- 



{gray red 3764 

 gray pink 1369 

 ebony red 1112 

 ebony pink 



This count includes a few F, flies derived 

 from F. red grays mated together, but also 

 includes large numbers of F/s from each of 

 the reciprocal crosses — ebony ? X pivk 3 and 

 pink 9 X ebony c?. In the later generations 

 from these crosses the writer was able to ob- 

 tain a few ebony pink flies, and a pure stock 

 of this combination was then made up. When 



so far obtained being about 1 per cent., on the 

 one hand, and about 27 per cent., on the other. 

 This matter is being further investigated, 

 and it seems probable, from results already 

 obtained, that external conditions play a 

 large part in it. The detailed counts are re- 

 served until this side of the question is more 

 thoroughly worked out, but the totals are 

 given below, because they show beyond ques- 

 tion that linkage occurs. When females that 

 showed " coupling " (in Bateson's ('06) sense) 

 — i. e., in which red and gray, pink and 

 ebony were linked — were tested by mating to 

 ebony pink males, the result was as follows : 



Gray red (heterozygous) ? [gray red 880 



^-^ J gray pink 86 



^ ^^^ I ebony red 65 



Ebony pink t? " [ebony pink 691 



Tests of females showing Batesonian " re- 

 pulsion " (having red linked with ebony, pink 

 with gray — Bateson and Punnett, '11) gave 

 the following result: 



