GENE-MUTATION RATE KNOWLEDGE INEXACT 321 



It is true that we can tell fairly exactly, for standard conditions, what 

 total frequency of recessive lethals will be found for a given dose applied 

 to Drosophila spermatozoa. But there is still a wide margin of error in 

 determining what proportions of these lethals represent gene mutations, 

 deficiencies, and position effects, respectively. And there is much 

 greater doubt in the estimation of what the ratio is of these recessive 

 lethal mutations to gene mutations in general, chiefly because very little 

 is known about the frequency of mutations of very minute effect, despite 

 their great importance for the population (see p. 322). There is still 

 considerable uncertainty also concerning the ratio of gene mutations in- 

 duced in Drosophila spermatozoa to those induced in ordinary interphase 

 cells, by the same dose, and the latter is the more important quantity in 

 practice. Moreover, estimates of the ratio of the mutation rate induced 

 by a given dose to the spontaneous mutation rate sometimes differ by a 

 factor of more than 5, if, for instance, lethals as a class have been fol- 

 lowed in one experiment and visible mutations at specific loci (the 

 determination of which is subject to far greater uncertainties) in another 

 experiment. And yet, in still other work, these two categories of mu- 

 tations have appeared to vary pari passu. Work on the spontaneous 

 mutation rate is a far larger undertaking than that on the induced rate, 

 yet it is equally important for gauging the significance of the results ob- 

 tained concerning the induced rate. Thus there remains room for far 

 more exact and larger-scale work along these lines than has yet been car- 

 ried out. Moreover, this work should be carried out under conditions 

 that will make possible the minimizing of so-called ''personal equations." 



The above is the situation even in Drosophila, although this has been, 

 for most purposes of detailed genetic analysis, the pilot form. This 

 situation reflects lack of recognition not only of the needs of such work, 

 but also of its implications. It is fortunate that it has been possible to 

 establish some of the more important principles at work in the field of 

 radiation mutations without having to ascertain the absolute values of 

 the factors involved. However, if estimates are to be made of the ge- 

 netic effects of radiation in organisms of intrinsic importance, as in man 

 or (as a clue to man) in other mammals, then it will be essential to gain 

 exact data on these quantitative features. But, as a guide to such stu- 

 pendous investigations, and to avoid a repetition of such costly missteps 

 as have already been made in the work that has been done on mammals 

 by persons with too limited genetic outlook, it is important that such 

 data be obtained first in lower-level pilot forms suitable for exact ge- 

 netic analysis. At the present day, this still means, in the first place, 

 Drosophila. 



Now, although so little is yet known of the magnitude of the effect of 



