~24- AECD-3446 standard deviations to background plus three standard deviations. The result of these changes is that fewer samples of marginal significance are reported in cine $Are ee te tree peaen 1949, Geometry, when defined as "the fraction subtended by the detector of the total solid angle as seen by the radioactive source", is about 50 per’ cent for the internal counting chamber. Therefore, the factor for correcting a sample count for geometry is 2.0. Backscatter was determined by adapting the counter so that a sample placed on a very thin film could be counted without backing other than the film and air, To make this determination a hole was drilled through the bottom of one | plate holder, the counter was supported so that the bottom of the counting chamber o :s “he gg? Weg? was backed only by air, the chamber was sealed to prevent gas leaks, and the film was coated with aquadag to insure electrical conductance. Using a dried crop of pe? a pure beta emitter of 1.7 Mev, on a Formvarfilm that weighed .07 mg/ em’, acount was made. Immediately following, part of the film includ- 6 ing all of the p22 was transferred to a regular 2-inch counting plate and counted in . the conventional manner. The difference between the count made on the stainless on steel plate and on the thin film divided by the count on the film was the per cent backscatter, assuming there is no backscatter from the film. The backscatter of p*? from a .005-inch stainless steel plate was determined by this method as ~~ being 30 per cent, Therefore the factor for correcting a sample count for back- ~ scatter is about 0.77 and for geometry and backscatter combined, 1.54. , The me eae CESPRCW oe : value used was 1.5. Be No correction was made for self absorption but an effort was made to keep Ee the ash on the plate relatively thin and as stated above, theaverage thickness 7 & was 4+0.7 mg/ cm” - too thick for complete alpha counting but thin enough so , that beta of approximately 0.05 Mev or greater should be counted. If a single Z. known isotope were being counted a suitable correction could be madefor self absorption but since a mixture of unknown isotopes of low activity was probably * te being counted, the best solution seemed to be to prepare a relatively thin sample and to neglect making a mathematical correction. Weight of ash on the counting a 2 average, ; a plates was not determined in 1948 but from inspection, the 1949 plates appear to be considerably thinner, a guess being that they are about one-half as thick on an = . . # UNNERSITy Lo ormeg UNV. OF Sask t uniuEse