188 G. W. BARENDSEN 



two-hit mode of action on the chromosomal material. The theoretical 

 curve exj)ected from this assumption has the form 



— = e-BiA (1+D/.4) 



where n\n^ is the relative survival after a dose D and ^4 is a constant. 

 In Fig. 1. curve 3' is an example of this tyjje of curve in which ^4 is 

 chosen as 135 rad in order to obtain the best fit with the experimental 

 points. The accuracy of our experiments is sufficient to conclude that 

 the assumption of a two-hit mode of action as proposed by Puck is too 

 simple. It may be noted in passing that from the assumption of a two- 

 hit mode of action it may be inferred that the slope of the survival curve 

 at low doses a]Dproaches zero. This would imply that for this system the 

 RBE of a-radiation at low doses would approach infinity. However, 

 our results at low doses show an initial slope corresponding to an LD37 

 of about 450 rad. 



An hypothesis by which our survival curves may be explained starts 

 from the consideration that the single-hit survival curve obtained with 

 a-radiation suggests that deposition of a sufficiently large amount of 

 energy in a small volume anywhere in a relatively large part of the 

 nucleus inhibits clone formation. 



Consideration of the spatial distribution of the energy deposition by 

 X-radiation makes plausible that at least a small part of the damage 

 caused by ionizing radiations of low average LET will be due to the 

 same type of locally concentrated energy deposition i.e. part of the 

 X-ray damage may be considered to be caused by a single-event type of 

 action. The greater part of the energy, which is deposited in less con- 

 centrated form, may be assumed not to result in inhibition of clone 

 formation. The damage caused by this part of the energy deposited 

 might be reparable (see next section). At higher doses of X-radiation 

 two or more amounts of energy deposited in small volumes, each by 

 itself too small to inhibit clone formation, may be found sufficiently 

 close together to cause this effect. Thus with increasing dose a greater 

 part of the energy deposited is effective in causing the damage measured. 

 In this way the shape of the X-ray survival curve with increasing slope 

 at higher doses may be explained as a combined result of one-, two- and 

 multi-event types of action. 



Ejfecfs of fractionated doses 



A immber of experiments was carried out in which doses of X-radi- 

 ation up to 900 rad and doses of a-radiation up to 200 rad were frac- 

 tionated in such a way that the second half of the total dose was 



