370 



AN INTRODUCTION TO MODERN GENETICS 



reduced the frequency of the bristles a^ c^ and /; another affected a, h^ 

 h and i, while a third aifected e^ f and g. No regularity is apparent in 

 this until we hit on the idea of writing the bristles in the order e g f c a 

 hib d, when we find that each allelomorph affects a compact group 

 (i.e. fca, ahih and egf). Allelomorphs which behave in this way have 

 been called step-allelomorphs. 



From this it was argued that the empirical linear order of bristles 

 corresponded to a real linear arrangement of sub-genes within the 

 scute locus, so that each of the allelomorphs was produced by a change 



Fig. U9. Step Allelomorphs of the Scute Locus in Drosophila melanogaster. — 

 The upper row of letters symbolize the names of the bristles on the back of the 

 fly. When they are arranged in this particular order, it is found that each allelo- 

 morph, whose numbers are given on the left, affects a set of bristles which lie 

 next to one another. The order in which the bristles have to be arranged has 

 little obvious relation to their pattern on the body of the fly. /jo, I3, and 1^ are 

 lethals associated with some of the allelomorphs. 



(From Child, after Dubinin and Friesen.) 



in a contiguous set of sub-genes, and affected only the bristles. con- 

 trolled by those sub-genes. In confirmation of this, it was found that in 

 compounds of two scute allelomorphs, only those bristles were affected 

 which were common to both; which was interpreted to mean that only 

 those mutated sub-genes were effective which were included in both 

 allelomorphs and therefore homozygous. 



This theory is obviously in a sense an extension of the theory of the 

 linear order of the genes. But it is arrived at in a totally different w^y, 

 by a direct reflecting back into the zygote of the order and system 

 which can be found in the final product of development. It has all the 

 simple logicality which is characteristic of such preformationist -hypo- 

 theses; but it depends essentially on the statement that each scute 

 allelomorph affects only a certain definite group of bristles, and it is 

 just this most fundamental point which seems to be most doubtful. 



