2.2 HIGH-ALTITUDE SAMPLES A series of high-altitude samples, starting in late 1956, has been taken for radiochemical analysis. The samplers are carried to altitude on balloons, and an attempt is made to obtain total volumes of approximately 1000 std. cuft. Four sampling sites are used: Minneapolis, Minn., San Angelo, Texas, the Panama Canal Zone, and Sao Paulo, Brazil. An attempt is made to obtain monthly samples at four nominalaltitudes: 50,000, 65,000, 80,000, and 90,000 ft. In addition to the difficulty of controlling sample flights at altitude for the required length of time, a number of samples are not recoverable or are otherwise lost. Therefore, fewer than 16 samples per month are usually available. The detailed data on the completed monthly samples taken during 1957 are given in Table 15. Although the complete interpretation of this type of data requires meteorological knowledge, there are some interesting points that can be made using average values for various groups of samples. The average values for Sr®° and Cs'*" at each altitude and at the three Stations submitting sufficient samples are given in Table 14. The averages in Table 14 will not necessarily agree with the averages taken from Table 15 since some additional data on Sr* and Cs!3" were available from incomplete samples. The over-all average values for Sr® and Cs!"" for all stations at the four nominalaltitudes are plotted in Fig. 7a. The distribution with altitude of both isotopes shows a maximum atthe nominal 65,000 ft regardless of whether the activity is expressed on the basis of standard cubic feet of air or cubic feet of space. The same distribution holds for the individual stations. The curve may be integrated by using this set of mean values in terms of cubic feet of space to show the presence of a mean of 0.25 megacurie of Sr®in the stratosphere during 1957. If the estimated efficiency factor of the stratospheric filters of 25 per cent is assumed to be correct, a mean stratospheric content of 1 megacurie of Sr®*’ would be obtained. The cesium to strontium ratio is considerably higher than would be expected from thermal neutron data, and the ratio is sufficiently constant to make it appear that this is a real dif- ference. , If the mean values at the various altitudes are plotted against latitude, there is no indication of any particular trend. The over-all mean, for example, at Minneapolis is 29; Texas, 28; and Sao Paulo, 23 dis/min/1000 cu ft. The values for the individual latitudes are plotted in Fig. Tb. This sampling program is continuing, and with detailed interpretation it should be of considerable assistance in material balance studies for Sr®”. Certain data have been presented for lower altitudes by the United Kingdom, but no direct comparison is presently possible for tropospheric and stratospheric air concentrations at the same location. Surface air filter data at Washington, D. C., showed a mean of 70 dis/min/1000 cu ft for total mixed fission products for 1957. This may be compared with the maximum of 3000 dis/ min/1000 cu ft found as a mean for the 65,000-ft stratosphere samples. Surface concentrations are hence much lower than stratosphere concentrations, and it is expected that the general tropospheric activity would be intermediate. 38 topless