THE MUTATED GENE 



139 



clear that we are facing here a series of facts showing that a 

 definite gene acts through controlling reaction velocities pro- 

 portional to its quantity. We have discussed (see page 104) the 

 manifold facts regarding this gene that enhance the safety of this 

 conclusion. We shall add one significant fact: Hersh (1934c) 

 calculated from Driver's data on temperature effect that the 

 curve of reaction controlled by the Bar gene is a sigmoid curve, 

 which would mean that the differences between the members 

 of the series in regard to this reaction do not represent merely 

 developmental arrests of the process at a greater or lesser distance 

 from a common upper asymptote but that the termination of the 

 process is approached asymptotically. This seems to point more 

 to a process of the type realized in the v^-wing than to an influence 

 upon cell divisions. 



The Bar series has furnished additional important material 

 upon the problem discussed here, which is derived from the fact 

 that another mutation, Infrabar B l , was found which may be 

 combined with all the others and may also be obtained as double 

 Infrabar by unequal crossing over. The facet counts of all these 

 combinations are available, and many have been tested at 

 different temperatures. 



Infrabar, which in usual terminology is one of the multiple 

 alleles of Bar (see, however, page 143), differs from the rest of the 

 series in some properties. The eye facets are not only fewer than 



Table 15 

 (From Sturtevant) 



