Chromosomes and Genes 159 



containing a scute inversion was irradiated, yellow mutants were 

 produced in the same varying proportions. According to Belgovsky 

 (1938), these mutants are position effects of small rearrangements. 

 Nevertheless, other supposed point mutants and lethals did not show 

 the same proportionality, though they were still more frequent in the 

 second period, but not as frequent as the breaks. From this evidence 

 it is concluded that the point mutations must be intragenic and not 

 intergenic. (Similar results of D'Amata and Gustafsson, 1948, and 

 Kaplan, 1951, for cereals are quoted, but it would be difficult to 

 distinguish cytologically in cereals small rearrangements from point 

 .mutants.) I cannot see that the experiments prove the conclusion. The 

 yellow "rearrangements" v/ere not checked in salivary chromosomes 

 and neither were the lethals and putative point mutants, many of 

 which would probably have turned out to be rearrangements. Even 

 if the facts should be found, after a check, to be what they are sup- 

 posed to be, they do not prove the point. If it were certain that 

 invisible changes were less frequent at a definite chromosomal stage 

 of irradiation than gross breaks, this might have many reasons, for 

 example, the smaller size of the breakable material within a band than 

 between the bands. Many other interpretations of such a type could 

 be thought out, and there is no reason to assume that the difference of 

 effect in regard to aging, even if it were more reliably established, has 

 any bearing at all upon the problem of gene mutation versus invisible 

 pattern change by breaks (e.g., deficiencies). 



bbb. The chromosomal sections of action. — This leads us to 

 inquire in more detail into the relation between point mutants and 

 allelic position effects. The results are of the greatest importance for 

 the theory of the genie material. Demerec (1943) and I, with the 

 collaboration of Kodani ( Goldschmidt, 1944), independently made 

 such a study with similar results. Since those of Demerec were based 

 upon a much larger material, we illustrate them with his diagram 

 (fig. 11). The salient points are these. The position of the assumed 

 locus or gene within a given band or group of bands is derived mainly 

 from the location of deficiencies giving the recessive effects. The map 

 shows that the position effect (e.g., for yellow and scute) can be 

 produced by rearrangement breaks at very different locations to the 

 right and left of the assumed locus. For the yellow effect, breaks 

 within a section of about six bands are responsible; for the scute effect 

 the section is composed of five bands; about seven for the white 

 effect; about five for Notch; and four for Cut. But, as figure 11 shows, 

 these regions were considerably extended when the rearrangement 



