model, therefore, we consider a future test programmein which the amountof Sr®” generated per three-year period is equal to that created between the spring of 1954 and the spring of 1957. If the deposition rate is assumed to be 25% per year, this quantity, from the last para- emete yet The last model to be considered is one in which the future rate of injection of Sr’’ into the stratosphere is assumed to remain constant at the level of the past few years. The actual deposition at Milford Haven during these years has not been consistent with this assumption which would lead to a greater increase from year to year than is actually observed. For this graph and Fig. 2, will be approximately 16.8 mc/km’, correspondingto an effective creation rate of 5.6 mc/km?/year. With these assumptions, an equilibrium value of about 200 mc/km? FoF PG ET Poy SUMMER fl PG WINTER SS ok 40 bw ae will be reached in about 100 years. If the deposition rate be assumed to be 12% per year, the ultimate equilibrium level will be approximately 300 mc/km’. at 30 }— I A na 0. igh = x & 20 FoO li x _ 1Of—- Len — ee mee -_——a a T= a > “ee” ope_t 30° 1 | 60° — i | | ti 30° Te ML fy oO LATITUDE ft | ft 30° tt dt) 60° 90° Fig. 7—- Atmospheric circulation model (after Dobson and Brewer). 10 CONCLUSIONS The programme of measurements described has revealed some interesting and important features of the exchange processes between the stratosphere and the troposphere which give of Cs’ measurement has been started with the gamma-ray spectrometer on air filters which have been collected over the past few years. The object is to study the simultaneous variation of the Cs!" concentrations in rain water and in tropospheric and stratospheric air. It is hoped to acquire more data to test the theory which has been advanced for the transfer of fission products from the stratosphere to ground. In this same connection, measurements of the meridional distribution of fission products in the stratosphere would be of interest and, in 244 et bbe more information could be gleaned about the size of the stratospheric reservoir of activity. A programmeof stratospheric sampling up to 14,000 metres, comparable with that described in HP/R 2017' but more sustained, is now being planned in the U. K. In the meantime a programme we eOP-el cy ever, such a programme would limit the number of sampling stations which could be operated, and the present arrangement is a fair compromise. Forecasting would also be improved if tes pling points it might be possible to obtain reasonable values for the integral of sr*° deposition over the surface of the earth. A summary of the main results is given in the abstract at the front of the report. In order to forecast future ground-level concentrations from existing data, it is important to distinguish between tropospheric and stratospheric fallout and the differentiation could be improved by reducing the present sampling period from three months and by including one or more short-lived isotopes in the radiochemical analysis. For a given analytical effort, how- ae weee rise to a non-uniform deposition pattern of long-lived fission products over the surface of the earth. The relative smoothness of this pattern suggests that with a sufficient number of sam-