NATURE OF THE GENETIC EFFECTS 411 



have provided some interesting points, both of agreement and of contrast. 

 To note first some agreements, it is to be observed that lethals (as pre- 

 viously defined) appear among organisms in general to greatly outnumber 

 the externally visible mutations, both in the case of spontaneous and of 

 radiation-induced mutations. As for comparative frecjuencies in different 

 organisms, the attempt to estimate total mutation frequency directly has 

 been made only in Drosophila, for either spontaneous or induced muta- 

 tions. However, certain spontaneous frequencies for specific visible loci 

 have been determined for a number of organisms. From this work it has 

 turned out, rather surprisingly, that the median frequencies per locus per 

 generation for spontaneous gene mutations seem to be of the same order 

 of magnitude in maize (Stadler, 1942), Drosophila (Muller, J. I. Valencia, 

 and R. M. Valencia, 1950), mice (Russell, 1952 and unpublished), and 

 human beings (Haldane, 1949). Nevertheless, in maize immense varia- 

 tions in spontaneous frequency between one locus and another appear to 

 be very common, and the range covers three orders of magnitude, from 

 the order of 1 in 1000 to 1 in 1,000,000 as found by Stadler, while in the 

 three animal types mentioned the variation in frecjuency seems on the 

 whole much smaller, although in certain cases still large, e.g., of one or 

 occasionally even two orders of magnitude. 



It is, however, when the mutations induced by ionizing radiation are 

 examined that the most startling apparent differences in frequency are 

 found. Most notable here is the fact that in the work on such mutations 

 conducted by Stadler on maize since 1928, it has not yet been possible to 

 obtain convincing evidence of the production of any gene mutations at all, 

 either in the investigations of over-all mutation frequency or in those 

 more critical experiments which deal with specific loci. It is true that 

 so-called "point mutations," resembling gene mutations and inherited 

 like them, were produced in abundance. Yet the results of analyses of 

 mutations involving certain particular loci indicate that in the great 

 majority of these cases, at least, the changes consist of minute deletions 

 (Stadler, 1941; Stadler and Roman, 1948). This was evidenced by the 

 facts that (1) the mutation induced at the given locus was always the 

 most extreme, apparently amorphic, allele of the normal gene, unlike 

 what was true of the spontaneous and ultraviolet mutations of the same 

 gene, and (2) pollen containing the mutant genetic condition was always 

 more or less deleteriously affected in their development or growth, like 

 those with deficiencies (see p. 371) and unlike those with spontaneous or 

 ultraviolet mutations of the given gene. 



One important reason for this apparent dearth of induced gene muta- 

 tions probably lies in the fact that, for a given dose of radiation, gross 

 structural changes of chromosomes (including unrestituted single breaks) 

 arise with far greater frequency, per cell, than in Drosophila. As these 

 cause a high frequency of lethal effects, thus lowering the effective 



