27G 



RADIATION BIOLOGY 



chiofly ill relation to the error involved in wave-length comparisons, rather 

 Ihaii to the nature of the reaction of the chromosomes to increasing dose. 

 The results of dosage trials at wave lengths 254, 297, and 302 mn are 

 shown in Table 7-5. 



Tablk 7-5. l{ia,.\TioN of Do.sk to FHWiiENrY OK Endosperm Deficiency 



(Stndlcr ;in(l t'l.cr, I'.tt2.) 



The relatively low yield of endosperm deficiencies at the lowest doses 

 tested for wave lengths 297 and 302 suggests the possibility of a multiple- 

 hit curve for the dosage relation. This possibility was suggested also by 

 the results of treatments at nine wave lengths in the range 238-293 m/z, 

 comparing the frequency of endosperm deficiencies from doses of approxi- 

 mately 0.5 X 10^ ergs/mm- with that from doses of approximately 

 2X10^ ergs/mm-. In every case the yield from the lower dose was less 

 than one-fourth of that from the higher dose. Since these treatments 

 were not seriated for the control of daily variation, a separate trial was 

 made with wave length 2G5 m/x, using doses of 0.25, 0.50, 1.0, and 2.0 X 10^ 

 ergs/mm'-. The frequency of induced endosperm deficiencies was again 

 disproportionately low at the lower doses, but the deviation from linearity 

 was not statistically significant. 



Indications of nonlinearity of ultraviolet effect on the frequency of 

 mutations at low doses have also been found in lower organisms. The 

 data of Emmons and Ilollaender (1939, Table 2; see also Hollaender and 

 Emmons, 1941) on induced mutation frequencies in Trichophyton suggest 

 that low doses are disproportionately less effective than higher doses, 

 although again the departure from linearity is not always obvious or 

 consistent. Similar results have been obtained in AspfrgiUns (Swanson 



