664 RADIATION BIOLOGY 



only a few points between blocks that are virtually unbreakable (Muller 

 and Gershenson, 1935; Muller, Raff el, Gershenson, and Prokofyeva- 

 Belgovskaya, 1937). On this assumption the close correlation between 

 break frequency and mitotic chromosome length revealed in the cyto- 

 logical studies would be attributable to mere coincidence (Muller, 1945). 



In order to evaluate these two alternatives more adequately, an exten- 

 sive study was made of the distribution of breaks along the X chromo- 

 some of D. melanogaster (Kaufmann, 1939a, 1946a). The proximal third 

 of this chromosome is heterochromatic in mitotic cells of the stock 

 studied, being represented in salivary-gland chromosomes by only a few 

 discs, whereas the distal two-thirds is primarily euchromatic and con- 

 stitutes the major portion of the salivary-gland chromosome. The loca- 

 tions of about 1400 breaks were determined with reference to the lettered 

 subdivisions of Bridges' salivary-gland map of this chromosome. About 

 one-fourth of these were in the proximal heterochromatin, but rearrange- 

 ments confined to this region were not detected. When the estimated 

 number of breaks involved in such rearrangements was added to the 

 observed number, the proportion in the proximal region was increased to 

 about one-third of the total for the entire chromosome. This value cor- 

 responds closely to that expected on the basis of length in late prophase 

 stages of mitosis. A similar close correlation between mitotic chromo- 

 some length and break frequency in the proximal heterochromatic regions 

 of the second and third chromosomes was presented by Bauer (1939b). 

 The frequency of breaks induced in the second and third chromosomes by 

 treatment of spermatozoa with the chemical mutagen, nitrogen mustard, 

 also showed close correlation with mitotic chromosome length (Kauf- 

 mann, Gay, and Rothberg, 1949). Approximately equal numbers of 

 breaks occurred in each of the four arms, and the proportion in the 

 proximal heterochromatic regions was similar to that obtained in X-ray 

 experiments (e.g., 17.6 and 13.6 per cent in the second and third chromo- 

 somes, respectively, after nitrogen mustard treatment of males, as com- 

 pared with 18.3 and 12.5 after X-ray treatment). 



In the light of these findings it seems highly probable that break dis- 

 tribution is a function of chromosome length at the time of irradiation. 

 Breakage disrupts the fundamental structural units that give the chromo- 

 some its linear continuity; and there is no evidence that the chromatids 

 and their subsidiary strands have different basic patterns of organization 

 in heterochromatin and in euchromatin, although there may be con- 

 siderable differences in the types and proportions of constituent nucleic 

 acids and proteins. The discrepancy in expression of euchromatin and 

 heterochromatin as the mitotic-type chromosome becomes transformed 

 into the salivary-gland type of structure is a problem of differentiation 

 (see Krivshenko, 1950), and does not necessarily reflect any dissimilarity 

 in structure at the time of irradiation. All these considerations indicate 



