OF MUTANT CHARACTERS. 169 



PURPLE (p.). 



(Plate 5, figure 8.) 



ORIGIN OF PURPLE. 



In a stock kept in the stock-room and supposed to be simply vesti- 

 gial, there was found, February 20, 1912, a single male which had an 

 eye-color much like that of the well-known double recessive vermilion 

 pink. The color of the vermilion pink eye is about that of the pulp of 

 an orange, and the early papers accordingly referred to this double 

 recessive as ''orange." The new color was seen to differ slightly from 

 vermilion pink in that it Was of a brilliant ruby-like transparency 

 and lacked the flocculent or slightly cloudy appearance of vermilion 

 pink. This difference seems to arise partly from a difference in the 

 distribution of the pigment. In vermihon pink the pigment looks as 

 though it were mainly in the spaces between the radially arranged 

 ommatidia with a clearer zone just under the surface of the eye. One 

 sees in the vermilion pink eye a light fleck which travels over the eye 

 as it is turned. This seems to be due to a deficiency of pigment in the 

 deeper parts of the eye and the light fleck is this light center seen 

 through the small group of facets whose axes are in hne with our eye. 

 The pigment in the case of the new eye-color gave the appearance one 

 would expect if it were uniformly distributed or even in solution 

 throughout the eye. 



INHERITANCE OF PURPLE. 



This single male with the orange-like eye-color was out-crossed to a 

 wild female, and in Fi gave only wild-type males and females (wild- 

 type 9 32, cf 33; reference No., B 1) which showed that the color 

 was recessive. In F2 the orange-like color reappeared, but in addition 

 the sex-linked eye-color vermilion emerged, and also a new eye-color 

 "purple" which appeared equally among the F2 females and males and 

 was therefore known to be an autosomal (not sex-linked) character. 

 It was now evident that the orange-like color resembled the old 

 "orange" (vermihon pink) genetically as well as somatically, for it 

 was proved by this F2 to be a double recessive, vermilion purple, in 

 which purple corresponds to pink. 



It seems probable that the two eye-color mutations, vermihon and 

 purple, present in the male first found were not of simultaneous or 

 related origin. There was a vague rumor that the vestigial stock had 

 contained vermihon at some time previous to this discovery. No 

 vermihon or purple was found in it subsequently, however. 



DESCRIPTION OF PURPLE. 



The eye-color of purple flies passes, in its development, through an 

 interesting cycle of changes closely parallel to those seen in the ripen- 

 ing of a sweet cherry. In the pupa the eye is at first colorless, then it 



