526 RADIATION BIOLOGY 



approximate atomic composition dealt with, is, of course, known from 

 physical data, and m, the chance or frequency of detectable mutation of a 

 given gene, is obtained from genetic data. Finally, a, the number of 

 atoms thus arrived at, may readily be converted into a volume, from 

 considerations of the approximate density of the material. 



Blackwood calculated a sensitive volume of about 12 m^t diameter for 

 the gene which mutates to white eye. However, he used too high a value 

 for m since he apparently based it on the frequency of appearance of white 

 spots in the eye in Patterson's work on the irradiation of Drosophila 

 larvae heterozygous for white, and most of these spots are in fact caused 

 by somatic crossing over. Values ranging from about 1 to 4 uifj. were later 

 obtained by Timofeeff-Ressovsky, Zimmer, and Delbriick (1935), using 

 the frequency of production of various visible mutations by irradiation 

 of spermatozoa. Although at first stating that this volume was mit 

 einem. Gene zu identifizieren, these investigators in their later works, 

 accepting criticisms put forward by Muller at the Gene Conference held 

 in Copenhagen in 1936, adopted the position that the sensitive volume 

 must be distinguished from the gene. Lea and Catcheside (1945a) and 

 Lea (1946), by essentially the same method, arrived at a value of about 

 2-6 van oi^ the basis of the same data. In this case, however, in order to 

 allow for the nonrandom distribution of the ionizations arising from X-ray 

 treatment, the modification (Lea, 1940) had been introduced of using, 

 instead of the number of separate ionizations, the number of ultimate ion 

 clusters (each consisting, on the average, of about two ionizations). The 

 ions in such clusters are within only a few angstrom units of each other 

 and must therefore be within the same sensitive volume in the great 

 majority of cases; they should therefore be treated as units in the 

 calculation. 



Lea (1940, 1946), further taking into account the effect of differences in 

 specific ionization, worked out tables showing the sensitive volume which 

 would correspond to any given mutation frequency, for radiation of 

 different types, such as ordinary X rays, 4- and 8-A X rays, neutrons, and 

 a rays, on the assumption that the volume in cpestion was a compact, 

 approximately spherical one. He showed that the approximately 33 per 

 cent reduction in mutation frequency that had been found with a given 

 dose of neutrons as compared with ordinary X rays was to have been 

 expected if the sensitive volume had a diameter of about 4-8 m^ (there is 

 an inordinately large statistical error involved in this calculation), and he 

 offered this result as a confirmation of the value that had been obtained 

 from the X-ray calculation considered by itself. This seeming agree- 

 ment would, however, lose its significance for the ciuestion of the size of 

 the sensitive volume of an individual gene if the apparent reduction in 

 efficiency found with the neutron data was caused by a clustering of 

 mutations in different but nearby loci rather than by an overlapping of 



