The results of these two experiments show that green is a sex-linked melanistic character like sable, but the somatic difference produced is much less than in the case of sable, so that the new mutation, although genetically definite, is of little practical value. We have found several eye-colors which differed from the red color of the wild fly by very small differences. With some of these we have worked successfully by using another eye-color as a developer. For example, the double recessive vermilion "clear" is far more easily distinguished from vermilion than is clear from red. But it is no small task to make up the stocks

necessary for such a special study. In the case of green we might perhaps have employed a similar method, performing all experiments with a common difference from the gray in all flies used.

CHROME.

In a stock of forked fused there appeared, September 15, 1913, three males of a brownish-yellow body-color. They were uniform in color, without any of the abdominal banding so striking in other body-colors. Even the tip of the abdomen lacked the heavy pigmentation which is a marked secondary sexual character of the male. About 20 or more of these males appeared in the same culture. This appearance of many males showing a mutant character and the non-appearance of corresponding females is usual for sex-linked characters. In such cases females appear in the next generation, as they did in this case when the chrome males were mated to their sisters in mass cultures. Since both females and males of chrome were on hand, it should have been an easy matter to continue the stock, but many matings failed, and it was necessary to resort to breeding in heterozygous form. The chrome, however, gradually disappeared from the stock. Such a difficult sex-linked mutation as this could be successfully handled (like a lethal) if it could be mated to a double recessive whose members lie one on each side of the mutant, but in the case of chrome this was not attempted soon enough to save the stock.

LETHAL 3.

In the repetition of a cross between a white miniature male and a vermilion pink male (December 1913), the F2 ratios among the males were seen to be very much distorted because of the partial absence of certain classes (Morgan 1914c). While it was suspected that the disturbance was due to a lethal, the data were useless for determining the position of such a lethal, from the fact that more than one mother had been used in each culture. From an F2 culture that gave practically a 2:1 sex-ratio, vermilion females were bred to club males. Several such females gave sex-ratios. Their daughters were again mated to vermilion males. Half of these daughters gave high female sex-ratios and showed the linkage relations given in table 55.

Table 55.—Linkage data on club, lethal 3, and vermilion, from Morgan, 1914c.

Females. Males.

Club. Wild-type. Club vermilion. Vermilion.
588 182 28 11 1

Lethal 3 proved to lie between club and vermilion, 13 units from club and 5 from vermilion. The same locus was indicated by the data from the cross of vermilion lethal-bearing females by eosin miniature males. The complete data bearing on the position of lethal 3 is summarized in table 56. On the basis of this data lethal 3 is located at 26.5.