SIZE DISTRIBUTIONS OF RADIOACTIVE AEROSOLS 175 collections. The errors in these ratios are based ona 1o counting error and the estimated error in the decay-curve resolution as men- tioned earlier. Data for ‘Ce, Sr, and *Zr are the most complete and will bé considered in detail. Figure 6 depicts ‘4Ce/*Sr activity ratios as a function of particle size and altitude as observed for the 1962 collections, and Figs. 7 and 8 show “4Cce/"Sr, ®Zr/"Sr, and Zr/'4Ce activity ratios from the 1963 collections. It should be noted that the respective production ratios estimated from the data of Katcoff'! are approximately 40, 220, and 4.8. 14400/ 9s, Activity Ratios For the March 1962 collection (see Fig. 6), approximately 130 days after the most recent atmospheric nuclear test with a fission yield greater than about 1 Mt, there appears to be a wide variation in 44Ce@ and “Sr distributions with particle size. Within the altitude bands at 9 to 15 and 15 to 21 km, however, the “4Ce/*Sr activity ratio is constant among particles with 0.02 u< r< 0.15 uy and r< 0.02 yu. With the exception of a relatively high “Sr concentration in the 27- to 30-km altitude band, these two particle size ranges contained about 90% of the “Ce and Sr activities observed in the March 1962 collections. We find a more systematic pattern in the M4Ce and “Sr distributions with particle size for the September 1962 collection (see Fig. 6), which was made during a period of nuclear testing. Although the errors in the '“4ce/"Sr activity ratios are admittedly large because of the low “Sr concentrations encountered, the observed variation in size distribu- tion of ‘Ce and "Sr at each altitude band is never greater than about a factor of 2. One is tempted to conclude that, within the estimated accuracy of the observed ratios, there is no chemical fractionation between ‘44Ce and “Sr in this collection. There is a slight tendency, however, toward enrichment of Mace, relative to Sr, in the largest particles in the 9- to 27-km altitude region. Also, the *Ce/*Sr ratios in September are about a factor of 10 higher than those observed in March 1962. Both of these obServations support the fact that relatively fresh debris was collected inSeptember 1962. The decrease in “4Ce/*Sr activity ratios with increasing altitude is probably a result of the con- tinuing influence of older debris at 21 to 30 km. In Fig. 7 a most striking result is observed in the plot of “4ce/*"Sr activity ratios as a function of particle size and altitude for the May 1963 collection when At is equal to about 150 days. With the exception of only the largest size fraction at the highest altitude sampled, which showed a very low “Sr concentration, the “4Ce and "Sr are distributed in an identical manner among the three particle size ranges. Less than 5% of the total “4Ce and Sr collected within each of the four altitude bands for this flight is found on particles with r > 0.15 up. Also,