Position Effect and Allelism in Drosophila 



193 



takes. This phenomenon is therefore called a 

 cis-trans position effect. In order to detect 

 such an effect it was necessary to separate two 

 very closely linked genes. The genes used in 

 the experiment had previously been con- 

 sidered alleles because of their closeness on 

 the genetic map and their similar pheno- 

 typic effects. But the fact that the cis and 

 trans positions of them gave different pheno- 

 typic effects made it possible to prove they 

 are nonalleles, occupying different loci. You 

 may next ask about the other genes making 

 up the "white multiple allelic series" (Chap- 

 ter 9). Are some allelic to vr and others 

 allelic to aprl The answer is yes. More- 

 over, some are allelic to neither, and appro- 

 priate crossing over studies have shown that 

 the "white region" on the X is a nest of four 

 (perhaps five) separate, linearly arranged loci 

 with similar effects. 



In view of the result obtained with the 

 white region, you may correctly ask next 

 whether there are other regions in the ge- 

 nome where two or more genie alternatives 

 have been considered allelic by the same 

 criteria as were originally used in the white 

 region, but which prove to be pseudoallclic, 

 that is, prove to be nonallelic when subjected 

 to the cis-trans test. Again the answer is yes. 

 Numerous examples of such pseudoallelism 

 have been found in diverse organisms includ- 

 ing, for example, cases involving color in 

 cotton, taillessness in mice, lozenge and also 

 vermilion eye colors in Drosophila, and other 

 cases found in Aspergillus, other microorgan- 

 isms, and corn. 



Another case of pseudoallelism in Drosoph- 

 ila will be discussed ^ briefly, in which the 

 nonalleles differ in their functioning some- 

 what more than do apr and \v. The normal 

 (wild-type) fly (Figure 22-6A) has small club- 

 shaped balancers (halteres) located on the 

 posterior part of the thorax. One of the 

 pseudoalleles, bithorax (bx), converts the 

 haltere into a large wing-like structure 

 ^ Based upon work of E. B. Lewis. 



(Figure 22-6 B), another called postbithorax 

 ipbx) appears to do much the same thing 

 (Figure 22-6C). But close examination re- 

 veals that these two recessive pseudoalleles 

 really do different things, bithorax convert- 

 ing the front portion, and postbithorax the 

 hind portion, of the haltere into wing-like 

 structure. This is verified by obtaining the 

 double mutant combination {bx pbx, hence 

 the cis form) by crossing over (at a rate of 

 .02%), and observing the phenotypes of flies 

 made homozygous for the double mutant 

 combination. Such flies (Figure 22-6D) have 

 a fully developed second pair of wings. 



What are the cis-trans effects for bx and 

 pbxl The cis form (+ -\-/bx pbx) has nor- 

 mal balancers, while the trans form 

 {bx -\- / -[- pbx) shows a slight postbithorax 

 effect, providing another example of cis-trans 

 position effect and demonstrating the non- 

 allelism of these genes. 



You may have already tried to visualize 

 how cis-trans position effects are produced. 

 One model, but not necessarily the only one 

 possible, compares the two homologous 

 chromosomes to assembly lines, each of which 

 makes its products independently. The cis 

 form can make all the products in turn, in the 

 strand containing the two normal alleles. 

 (The strand with both mutants makes less or 

 no end product.) So, over all, much end 

 product is produced by the cis form. In the 

 trans form, however, because each strand 

 contains a mutant (defective machine), the 

 total end product produced is zero or rela- 

 tively little. (It should be noted, however, 

 that detection of a cis-trans position effect 

 requires only that the phenotypes produced 

 by the two arrangements be different — it 

 does not require or specify that cis appear 

 phenotypically normal.) 



Why should cis-trans position effects be 

 produced? It is reasonable to think that the 

 products of genes which are linear neighbors 

 would be more likely to depend upon each 

 other than they would be to depend upon the 



