as percent) square inches and gives 2.90 X 10° when multiplied by the proper I/A from Fig. 9. The latter intensity is from 2.5to 7.6 percent lower than the intensity determined from the calibration of the standard, but agrees within the experimental error. The comparison of the He@03 photon intensities proceeds similarly to the Na@? comparison, The Hg°°3 source strength was given in actual disintegrations per second but only 78.7 of these disintegrations result in 279-kev gamma rays because of the internal conversion in He 03, After the appropriate corrections are made the photon intensities obtained from the extrapolated total absorption peak area and calibrated source strength are 4.15 X 10° (+5 percent) and 4,22 X 10© (+5 percent) photons per second respectively. The intensi- ties obtained by the two methods differ by less than 2 percent, well within experimental error. With 54 recordings of the 145 pe He 203 stangard and 94 recordings of the 75.6 yc Na°* standard made over a period of several months already available, it was decided to make a@ least squares fit of the decay curve to get a quantitative idea of the machine's stability. It was assumed that the functional form of the equations relating area to time was A = exp(-At) or a polynomial of the form y = a + a,t (a straight line) where y = 1nA. 31 i. ~