28 THE ORIGIN OF GYNANDROMORPHS. 



the germ-tract and to the soma. Through the former, inheritance 

 becomes possible, through the latter the effects of the mutation be- 

 come visible only on the plant in which the mutation took place. 

 There are other mutative changes in corn that Emerson describes in 

 which the effect is only in the epidermal cells; hence, while it becomes 

 visible in the plant in which it has taken place, it is not inherited, since 

 the germ-tract does not come from this part of the plant. 



In the course of our work on Drosophila a few flies have appeared 

 with characters which seem to have arisen by somatic mutation. If, 

 as there is reason to suppose, the mutation changes that gave rise to 

 them appeared in only one chromosome, the change must either have 

 been dominant or, if recessive, in the single X chromosome of the 

 male. Since visible mutations in the sex chromosome have been 

 shown to be at least four times as frequent as dominants in all of the 

 chromosomes together, the chance that these sporting characters are 

 dominants is smaller than that they are recessive and in the sex 

 chromosome. In support of the latter is the fact that nine out of ten 

 of the sporting characters look like known sex-linked genetic char- 

 acters, and more important still is the fact that all the cases so far 

 found are males. 



(1) One of these somatic sports is shown in plate 1, figure 4. The 

 right side of the body is pale, almost white. The history of this fly 

 is as follows '} 



One of the X chromosomes of the mother contained the genes for lethal 7 

 and for forked, the other X the genes for yellow and for white. The X 

 chromosome of the father carried the genes for yellow and for white. The 

 fly was a yellow white forked male throughout, but the right side of the 

 thorax, the right wing, and the right side of abdomen were pale, almost white, 

 as shown in the drawing. Testes were present, with sperm. The pale light 

 side is clearly due to somatic mutation, since no such pale body-color was 

 present in the cross or was known elsewhere. Whether the mutation oc- 

 curred in the X (if recessive) or in an autosome (if dominant) is undeter- 

 minable, since the fly was not bred. 



(2) In another case (II 108, Oct. 21, 1913), the left side of the body, 

 at least for a middle section, is brown in color, looking like the double 

 recessive yellow black (text-fig. 8). The fly had the following history: 



Some F2 blacks from the cross of black by jaunty (both second-chromo- 

 some) were inbred in an attempt to secure the double recessive black jaunty. 

 One of the F3 black males had the left side of its thorax and abdomen, left wing, 

 and left legs colored like the double recessive yellow black. It was at once 

 assumed that mutation to yellow had occurred in the early embryo in the 

 cells which gave rise to the left side. A test was made to see whether the 

 germ-cells carried the mutant gene. The mosaic male was outcrossed to 

 a black female and gave only black offspring (M69, black 9 27, black cf 23). 

 Three pairs and a mass-culture of these Fi flies were inbred and gave a total 



^No. 2493; November 20, 1915. 



