170 Nature of the Genetic Material 



to Pontecorvo), give a position effect, but not if they are in the same 

 chromosomes ("cis"). That is, they are position alleles, in Lewis' 

 terminology. We have stated that this unusual behavior is considered 

 to be the consequence of a, b, and c being repeats of the same 

 original gene which acquired somewhat different actions of the order 

 of multiple alleles but retained their function as alleles, though now 

 separable by crossover breaks. Therefore, according to one definition 

 of the gene, a, b, and c constitute different genes. The logic behind 

 describing this as an effect of position is not quite clear, since alleles 

 should be located in difiFerent chromosomes. It is based, I assume, on 

 the original explanation of the Bar case as due to unequal crossing 

 over of a pair of genes, which has since been explained as the 

 position effect of a duplication break (see above). My own explana- 

 tion, which does not need any of these unusual assumptions, has 

 already been presented. The three (or more yet undiscovered) 

 separable lozenge loci are part of a chromosomal section in which 

 every visible or non-visible change produces an allelic effect, as 

 shown in many instances. Recently, Chovnick and Fox (1953) tried 

 to obtain more information by studying the antigenic properties of 

 these lozenge alleles. They found differences but not pseudoallelism 

 of the different properties. Whatever the reason for this, it is clearly 

 not a problem of genie structure, as the authors seem to believe, but 

 of the chemistry of action involved, which we know is somewhat 

 different for antigenic properties in allelic condition (see blood 

 groups) from those found in other pairs of alleles, for example, inde- 

 pendent action of both partners. I think, therefore, that it is an 

 error to draw conclusions upon the underlying theory from such 

 facts. At this point in our analysis we shall not discuss possible 

 chemical explanations, but shall do so later. 



Another well-analyzed set of three "pseudoalleles" is Lewis' 

 (1951) work on the "bithorax" group of mutants, which affect the 

 metathoracic segment of Drosophila, in the extreme case making the 

 metathorax resemble a mesothorax. Each of the separated loci has its 

 specific, multiple allelic type effect, and in addition may itself have 

 a number of multiple alleles. Further, the pseudoallelic effects of the 

 compounds — the so-called position allelism — parallel those described 

 for the lozenge alleles. At the first locus (separable by crossing over 

 at 58.8 of the third chromosome) three recessive alleles are found; 

 at the second locus (0.02 unit to the right) a dominant Bx is found; 

 and at the third locus (0.01 unit to the right), another recessive. 



A third case with three members has been found by Roper 



