GENE CONTROLLED PROCESSES 183 



points are like intrusive masses which can divert the course of the 

 developmental processes down a side valley. 



This attempt to symbohze the developmental reactions may seem 

 unduly picturesque and too abstract to be of much value. Its abstract- 

 ness, however, must be blamed on the fact that we know so Uttle about 

 the actual processes concerned. The two important, but abstract, facts 



wild 



claret 



Fig. 86. Diagram of the Developmental Processes of Pigment Formation 

 in the Eyes of Drosophila. — The developmental process moves from left to 

 right along the branching tracks. The points marked ca, v, and en symbolize the 

 alternatives dependent on whether the claret, vermilion, and cinnabar substances 

 are produced. There must be very many more tracks which we cannot yet fit into 

 the picture. Thus it is known that carmine has no, or little, vermilion substance, 

 and mahogany little cinnabar substance; but we do not yet know whether the 

 developmental processes in flies homozygous for these genes branch off from 

 the normal track before the vermilion and cinnabar forks, or are secondary 

 branches, as indicated by the dotted lines. 



which are expressed in visual form by the valley model are, firstly, that 

 the course of any developmental process is determined by many genes, 

 and secondly that these genes often define alternative courses along 

 which the reactions may go. 



This same scheme may be used to describe the development of 

 characters which are not simple substances. For instance, a similar 

 history of successive reactions has been invoked to explain the develop- 

 mental effects of the Bar gene in Drosophila which has been worked out 

 in some detail by studying the effect of temperature on the number of 

 facets formed in the eye.^ Unfortunately, we only see the end-result 

 1 Cf. Margolis 1935, Margolis and Robertson 1937. 



