486 * RADIATION BIOLOGY 



this that broken ends tend to lose their capacity to unite with ends from 

 other breaks before they become incapable of restitutional union. How- 

 ever, the effect could, alternatively, be interpreted by applying Lane's 

 explanation in reverse to this case, i.e., on the assumption that, in this 

 material with the given interval (48 hours), one irradiation causes a 

 heightened breakage susceptibility to a later irradiation. 



Regardless of the way these questions, which are still under dispute 

 (Sax and Luippold, 1952; Lane, 1952), are settled, the fact that the fre- 

 quency of observed exchange aberrations in Tradescantia microspores 

 approximates the % power of the dose instead of its square, under the 

 given conditions, is caused by other factors than in Drosophila sperma- 

 tozoa; yet the fundamental mechanism involved, that of the union of 

 ends derived from independently produced breaks, the chance of produc- 

 tion of each of which varies linearly with the dose, is the same in both 

 cases. Moreover, in both cases the union of broken ends must occur 

 when the chromosomes are in a relatively uncondensed condition. In 

 Tradescantia and Chortophaga this is shown by the fact that irradiation 

 before a certain stage of interphase results in exchanges of pieces of chro- 

 mosomes, both chromatids from which are thereby affected alike, and 

 not in exchanges of separately participating chromatids, whereas later 

 irradiation does give exchanges that affect individual chromatids differ- 

 ently from their sisters. For further details of the effects produced on 

 the incidence of different types of structural changes by varying the stage 

 and the conditions of irradiation, as determined by direct cytological 

 observation, and for deductions thereby made concerning chromosome 

 behavior and the manner of action of radiation, the reader is referred to 

 the chapters on this subject. 



It is deducible on physical grounds that, despite gross structural 

 changes varying in their frequency with the square of the dose or with 

 an exponent which approaches the square as the dose is reduced, the 

 exponent must, nevertheless, at still lower doses become smaller again 

 until the relation becomes a simple linear one. Within any given cell 

 or within that "critical volume" of any cell which is small enough for 

 the ends derived from different breaks to be able to meet and unite with 

 one another, the amount of radiation received at very low doses, over a 

 period long enough to allow union of ends, will not be affected except in 

 an all-or-none way by alteration of the dose applied from outside. For 

 each cell or critical volume which receives any radiation at all within this 

 period will at such doses be traversed by only one electron track, and 

 that cell or volume will then be subject to a minute but constant and cal- 

 culable chance, no longer related to the size of the external dose, of having 

 a structural change produced in it. Changing the dose, at these levels, 

 will not alter the amount of radiation received within a critical volume 

 that receives any radiation at all but only the number of units of volume 



