77/ A' M ( 'T.\ TED GENE 1 95 



studied in a later chapter.) With increasing intersexuality, 

 these patches increase on both wings without being symmetrical 

 and finally cover a whole wing. If we look at such a series of 

 wings of intersexual males of increasing femaleness, we are struck 

 by the fact that the arrangement of the dark male and white 

 female parts, irregular as it is, is of such a type that one might 

 describe it in terms of a stream of pigment flowing from the wing 

 base over the wing along the veins, covering the surface in an 

 irregular way, leaving islands in between until the stream is 

 stopped when all available pigment has flowed out (more in lower, 

 less in higher intersexuality). Obviously, the quantity of 

 "pigment" is a definite one for each stage of intersexuality and 

 available simultaneously to all four wings, to which it is dis- 

 tributed by chance. An exact study of numerous individuals 

 (Minami, 1925) bears out the general correctness of this picture. 

 Now, within dark male areas the form and arrangement of the 

 scales is of the male type; and within the light areas, of the female 

 type. The substance that has flowed across the wing therefore 

 determines color, shape, and arrangement of scales of male type. 

 As the amount of this substance is proportional to the degree of 

 intersexuality, and as the degree of intersexuality is proportional 

 to the time of occurrence of the turning point, it follows that the 

 simplest picture that we can make of the process is the following. 

 The differentiation of the male-type wing scales, in all their 

 features, is controlled by a determining stuff, which enters the 

 wing Anlage at its base and slowly flows across the wing, the 

 channels being determined by the actual structure of the wing 

 at this time. If the male is intersexual, this flow ceases at a 

 definite time which is controlled by the degree of intersexuality; 

 and the result is that all parts of the wing that have not yet been 

 reached by the flow will develop as female parts. (The details 

 may be conceived in different ways; the problem, however, 

 belongs to the problems of sex differentiation). From this it 

 follows that the visible patterns of male color and structure in 

 a series of male intersexes indicate the progress and type of flow 

 of an underlying determining substance. Adopting a conception 

 introduced by Spemann, we call it the determination stream 

 which thus is made actually visible; and inasmuch as the link 

 with the specific genie combinations that produce intersexuality 

 is known, we see that genes actually controlling color and struc- 



