Chapter 22 



POSITION EFFECT 



AND ALLELISM IN DROSOPHILA 



WE HAVE already mentioned a 

 number of discoveries made 

 possible by the abundance of 

 structural changes in chromosomes induced 

 by ionizing radiations (Chapter 21). In the 

 course of irradiation studies with Drosophila, 

 the same kind of two-break rearrangement 

 has been produced time and again, in which 

 both of the breakage points have occurred at 

 the same, or nearly the same, positions. In 

 many of these cases it is found that there is a 

 phenotypic change which occurs simultane- 

 ously with the rearrangement, and that the 

 same kind of phenotypic effect is produced 

 by each nearly identical rearrangement. Fur- 

 ther study shows that the phenotypic effect is 

 transmitted whenever the rearrangement is, 

 and that, in many cases, this effect is similar to 

 that of a known allele of a gene located at, or 

 very near, one of the points of breakage. In 

 these cases, then, the change in phenotype 

 seems to be directly connected with the muta- 

 tion of a gene known to be located at or near 

 a point of breakage. 



To explain this effect, you might at first 

 think that the very fact that a break took 

 place in or adjacent to this gene automatically 

 changed it to this allelic form. But this can- 

 not be the explanation since other breaks at 

 this locus, which go into different kinds of 

 rearrangement, do not produce such a change 

 in phenotype. For the same reason, it is not 

 tenable to presume that the track of ioniza- 

 tions which produced the break simultane- 

 ously produced a minute deficiency or duph- 

 185 



cation of the affected locus. We may con- 

 clude, therefore, that an important feature of 

 this phenotypic change is that it is not based 

 upon what occurs at the time of breakage; 

 this suggests that the change is due to the 

 nature of the broken ends that join. More- 

 over, this change takes place only when the 

 broken end, which joins to the broken end 

 carrying the affected locus, comes from cer- 

 tain specific loci in the genome. So, the 

 gene affected at the broken end is not mutant 

 and will not change its effect if it joins to 

 some broken ends, but will do so, and in the 

 same way, if it joins to certain others. In 

 other words, we can hypothesize that the 

 phenotypic effect, the functioning, of the 

 same gene may be modified when its linear 

 neighbors are changed. This type of pheno- 

 typic change is said to be due to position 

 effect. Would we be justified in calling posi- 

 tion effect a mutation? Mutation was defined 

 on page 5 as a change in the genes, not in 

 their phenotypic expression. Note that all 

 the types of mutation discussed hitherto, 

 from Chapter 18 on, have been based upon 

 material changes in the genome, such as 

 gains, losses, or rearrangements of chromo- 

 somal material, and all novel phenotypic 

 changes so far discussed could be attributed 

 to such mutations. However, in the case of 

 position effect we are presumably dealing with 

 a change in gene effect in which the gene itself 

 has been unchanged! Position effect, there- 

 fore, can be one of the phenotypic conse- 

 quences of mutations involving structural 

 rearrangements, but is not in itself a muta- 

 tion. Position effect is no more to be called 

 mutation than is dominance, although both 

 require a prior mutation for their detection. 

 You may be dissatisfied with our presump- 

 tion that the gene showing position effect is 

 chemically and physically unchanged in any 

 permanent way. You might suppose that 

 when the gene in question joins to certain 

 genes it does so by means of one kind of 

 chemical or physical connection, leading to 



