OF MUTANT CHARACTERS. 



223 



DESCRIPTION OF STREAK. 



The principal characteristic of streak flies is the hand of pifjmont 

 along the thorax and scutellum. This band seems to he rather deep- 

 lying, and is possibly situated in a different layer from that in which 

 the other pigment characters develop. There is considerable variation 

 in both the intensity and the extent of the dark color. In its greatest 

 development it is a solid band, like that of colored figure 5, filling in 

 the entire region between the dorso-central bristles and extending over 

 the entire scutellum. In less-developed types the weakening starts in 

 the region ahead of the dorso-central bristles and is most pronounced 

 between the prongs of the trident pattern, so that an appearance much 

 like typical band is given. (For figures of "with," band, and trefoil, 

 see Alechanism of Mendelian Heredity, p. 206). 



The intensity of color is never very great and the color may nearly 

 vanish. However, there are other accessory characteristics that aid 

 in the classification. Chief of these is a flattening of the thorax and 

 the appearance of bubbles. Both of these effects seem to be due to an 

 ill development of the underlying muscles. There appear to be present 

 in the thorax large spaces or sinuses filled only with blood and large 

 bubbles. Where there are no bubbles present this condition is not so 

 easy to distinguish, though it may sometimes be made out by slightly 

 pressing the thorax. The wings are apt to droop and to diverge 

 slightly, probably also on account of the muscular condition. 



DOMINANCE AND LETHAL EFFECT OF STREAK. PARALLEL TO 



YELLOW MOUSE. 



The occurrence of the mutation as a single individual — a female — 

 in a pair culture, its immediate reappearance in about half the Fi flies 

 after crossing to normal males, and the failure of these Fi flies to breed 



Table 68. — Pi, streak 9 X wild cT. 



true, found an explanation (rather delayed) in the assumption that 

 streak was an autosomal dominant. Moreover, the fact that the 

 stock could not be made to breed true (continually producing at least 

 a third of the offspring wild-type) and that repeated pair mat ings as 

 well gave this same result, led to the assumption tliat homozygous 

 individuals were not produced. This case was seen to be a parallel to 



