CHROMOSOME ABERRATIONS IN ANIMALS 



669 



taneously by the densely ionizing track of a single recoil proton. Com- 

 parison with X rays indicated that neutrons produce a higher frequency 

 of chromosomal aberrations for the same dose. Studies on Drosophila 

 have not afforded such convincing data as those obtained from the 

 Tradescantia experiments. The frequency of viable types of aberrations 

 induced in D. melanogaster spermatozoa was determined by salivary- 

 gland-chromosome examination at two levels of neutron dosage (Demerec, 

 Kaufmann, and Sutton, 1942). The frequency at the lower dose did not 

 differ appreciably from that induced by an equivalent dose of X rays; 

 but the higher dose of neutrons was less efficient than X rays, as deter- 

 mined by the frequency of rearrangements and by the total number of 

 breaks. The types of rearrangements induced by neutrons and X rays 

 seemed to differ only with respect to the proportion of intercalary defi- 

 ciencies, which was greater in the neutron than in the X-ray series 

 (Kaufmann, 1941b). Genetic tests for translocations made by Catsch, 

 Peter, and Welt (1944) also indicated that neutrons are less effective 

 than X rays (Table 9-4), whereas similar tests by Dempster (1941b) sug- 



Table 9-4. Frequencies of Translocations between the Second and Third 

 Chromosomes of D. melanogaster Induced by X Rays and Fast Neutrons 



(Catsch, Peter, and Welt, 1944) 



gested that neutrons are slightly more effective. Differences in results 

 may stem in part from the difficulty of equilibrating doses of X rays and 

 neutrons. Marshak (1942) has presented data which suggest that neu- 

 trons are more effective than X rays in inducing chromosomal changes 

 in mouse tumors, but diagnosis was based on abnormal anaphase figures, 

 and presumably included single-event as well as two-break processes. 



The frequency of two-break rearrangements induced by X rays, in 

 which each break would be caused by a separate ionizing particle, would 

 be expected to increase as the square of the dose in accordance with the 

 formula N/Nq = k'D n , in which n indicates the number of independent 

 events. The observational data for two-break rearrangements conform 



