504 RADIATION BIOLOGY 



pretreatment) distinctly increased the frequency of X-ray-induced muta- 

 tions, while it did not increase, in the more diagnostic "rising" part of 

 the dosage curve (see pp. 435 438), the frequency of ultraviolet-induced 

 mutations (which would include fewer structural changes). As a whole, 

 then, the evidence from infrared cannot be regarded as contrary to the 

 position-effect interpretation. If, however, this agent produces its effect, 

 as argued by Swanson (see pp. 570-572), by increasing the frequency of 

 breakage rather than that of exchange union, the observed rise in muta- 

 tion rate would be expected on the mutation-by-breakage as well as on the 

 position-effect interpretation. However this may be, it is evident that 

 the results with infrared cannot at this time be used in support of the 

 mutation-by-breakage hypothesis . 



Another condition differentially affecting the frequency with which 

 genetic changes of various kinds are induced by applying ionizing radia- 

 tion to male germ cells of Drosophila is the stage of the germ cells and the 

 age of the male containing them. It has recently been shown by Llining 

 (1952a, b, c, d), in a very extensive and searching investigation, that the 

 irradiation of immature but postmeiotic male germ cells, presumably 

 spermatids, results in several times as high a frequency of (nonrestituted) 

 chromosome breaks as does irradiation of fully mature spermatozoa. 

 This is evidenced both in the greater abundance of dominant lethals, 

 gross deletions, minute heterochromatic deletions, and losses of rings, 

 when the treated sperm were immature. Yet at the same time, the fre- 

 quency of visible mutations of a type which are very seldom associated 

 with structural changes (those to yellow body in an X chromosome of 

 normal structure) proved to be affected httle or not at all by this differ- 

 ence in germ-cell stage. Mutations to white eye, which are associated 

 with structural changes oftener than the yellows, were induced with sig- 

 nificantly higher frequency in the immature than in the mature germ 

 cells, but the increase was not nearly so great as for the four classes result- 

 ing almost exclusively from breakage. Similarly, recessive lethals also 

 occupied an intermediate position in this respect. These results gave 

 evidence that there is one class of point mutations, to which belong 

 yellows as well as some of the whites and lethals, which do not arise in 

 association with chromosome breaks. These may be regarded as the 

 true gene mutations. It is their frequency which varies linearly with 

 dose. At the same time, there are two other types of visible and lethal 

 mutations, associated with small deletions and with other structural 

 changes, respectively, and their properties evidently depend on the 

 arrangements formed by the joining of broken ends. 



The finding that the freciuency of induced recessive lethals, and even 

 more so of dominant ones, is strongly dependent on the age of the male 

 and on the stage of his germ cells at irradiation has another bearing on 

 the main problem at issue in this section. Unless these ages and stages 



