C. P, SWANSON 



spectrum of initial events. Since there is little or no oxygen dependence 

 with alpha rays, and since the probability of a single alpha particle produc- 

 ing a break is close to unity (Gray 2, 1953), the initial events induced by 

 alpha rays would probably lie far to the right of the spectrum, i.e. they are 

 principally primary breaks. Their transformation into actual breaks with 

 separated broken ends may be immediate or delayed— this we cannot 

 establish, and the time factor may be different with different organisms — but 

 complete rupture of the chromatids takes place regardless of the oxygen 

 tension (Phase II). 



The less densely ionizing gamma and X-rays would produce changes 

 more to the left of the spectrum — neutrons would be intermediate in this 

 respect — although some primary breaks would still be produced. The fate 

 of the potential breaks is determined -by the available oxygen in the cell in 

 such a fashion as to favour their complete disruption in the presence or their 

 repair in its absence. Also, the less the damage, the greater the dependence 

 on oxygen for complete transformation into actual breaks. If these changes 

 induced by radiation are in single-strand chromosomes, only a reduction in 

 total frequency of aberrations can be expected if anoxia exists during 

 irradiation, and rings and dicentrics, the customary types of aberrations 

 scored, would be reduced to the same degree. Giles and his co-workers^ 

 have shown that this expectation is realized. However, if double-strand 

 chromosomes are irradiated in the absence of oxygen, repair of one or two 

 damaged chromatids becomes possible, and a situation is therefore available 

 for the transformation of latent isochromatid deletions and exchanges into 

 actual chromatid deletions. Variable air/nitrogen ratios among chromatid 

 aberrations are therefore possible, and the extent of variability will be 

 determined not only by the ratio of potential to primary breaks in Phase I, 

 but also by the ratio of single- to double-strand lesions. Since both ratios 

 are dependent on the ion density of the radiation, a complex situation exists, 

 but there can be no doubt that there is a correlation between oxygen 

 dependence and ion density. 



The data obtained to date provide no reason for believing that oxvgen 

 tension exerts an influence once the actual breaks are formed (Phase III). 



The scheme presented appears to be consistent with the biological data 

 derived from the study of irradiated chromosomes of Tradescantia as w ell as 

 with the physical facts of radiation. To what degree it corresponds to actual 

 events taking place in irradiated chromosomes remains to be demonstrated 

 for it is becoming increasingly evident that the final answers lie in the area of 

 radiochemistry (Gray 2, 1953). The work of Lea and others has provided 

 strong support for the belief that individual chromosome breaks are pro- 

 duced by individual ionizing particles, but the role of oxygen suggests that 

 both direct and indirect energy transfer to the molecular bonds within the 

 chromosome can lead to breakage. Presumably, the indirect energy transfer 

 is through the medium of reactive substance produced by the radiation in 

 or near the chromosome. With this in mind, the scheme in Figure 2 provides 

 a more general approach to the problem of how oxygen may aflfect breakage. 



Several possibilities exist, and for convenience they may be listed as follows : 



(i) Phase /—At this level the physiological conditions of the cell or the 

 surrounding medium can have an influence on the amount of indirect 



259 



