1226 BIOLOGICAL EFFECTS OF RADIATION 



tested for other mutations nearby. The detailed data are not as yet 

 published; but it appears that perhaps more than one-quarter of the 

 scute mutations found are lethal, or associated with a lethal. In a 

 number of cases MuUer was able to demonstrate that the lethal was at 

 a different locus from scute. Similar data were previously available in 

 a few scattered cases (Muller, 112; Patterson and Muller, 141). The 

 chance expectation of such an association of two mutants is very low; 

 the results suggest therefore that, although the mutation process is 

 highly localized, it is nonspecific in the sense that more than one gene in a 

 given neighborhood may be affected. Muller and Mott-Smith (Muller, 

 113) have examined and discarded the possibility that two hits by a 

 single electron could account for the observed results. It follows that 

 the phenomenon is a property of the biological system. 



For distances farther apart, there appears to be no correlation, if the 

 results of Gowen and Gay (52) on lethal mutations are to be accepted. 

 Yet Muller (112) has found a case in which of all three mutations in a 

 given experiment, two at widely separated loci occurred in a single indi- 

 vidual. Moreover, it is to be remarked that in these experiments one 

 chromosome only was followed; such correlations may extend to the 

 other chromosomes as well. 



The data of Patterson and Suche (143) show that in 20 cases, where 

 crossing over was induced in the male by X-rays, six lethals were found, 

 of which two were separable from the locus of crossing over. Shapiro 

 and Neuhaus (156) have obtained with a higher X-ray dosage 7.8 per 

 cent of lethals in the second chromosome. The result indicates a possible 

 correlation between the occurrence of crossing over in the male and the 

 production of lethal mutations. Contrariwise, these lethals showed no 

 tendency to accumulate close to the breakage points (or points of crossing 

 over), a result to be expected on the position-effect hypothesis. However, 

 the relation of the crossing-over process to the process of translocation is 

 still too unsettled to permit this to be used as evidence. 



The interpretation of the group effect is obviously bound up with the 

 interpretation of the mutational effects at breaks in chromosomes. It is, 

 as Muller has pointed out, possible to regard the cases which have been 

 discussed in connection with position effect as instances of the grouping 

 of mutational effects. This depends on an assumption which can now, 

 by the analysis of the chromosomes of the salivary glands, be verified — 

 that the point of mutation and break are not the same, but close together. 

 On the other hand the group effect itself is based for its demonstration on 

 data which involve lethals — possible deficiencies, and therefore capable 

 themselves of producing changes resulting in position effects. The chief 

 question then, is whether there are two i)henomena represented in these 

 data, or one. 



