potassium, and definitely showed that plant uptake of Cs'*’ was inversely proportional to ex- changeable and available soil potassium. Without considering exchangeable soil potassium, others*4’#5 have studied the ratio of Cs'*’ per g of dry plant materials to the concentration per g of soil and obtained values of 0.006 to 0.18 (average 0.07). These data suggest that the average Cs'*? concentration in the potassium of plants should be about 0.04 times the exchangeable Cs'8' concentration in exchangeable soil potassium. Exchangeable soil potassium, to a depth of 2.5 in., may vary from about 25 to 400 lbs/ acre. About 100 lbs/acre is a reasonable average value for the agricultural soils of the United States. This is equivalent to about 3 x 10’ g of exchangeable K/sq mile. Deposition and mixing to a depth of 2.5 in. of 1 me of Cs!?? per square mile gives a total concentration of about 30 pyc per gram of exchangeable soil potassium. Larson et al.‘® added Cs!" to three different types of soils and determined the amount that could be extracted with N NH,Ac. The ex- changeable Cs'*’ ranged from 13 to 33 per cent with an average of 25. Assuming that 75 per cent of the Cs!” is fixed in a form unavailable to plants, the discrimination factor (DF,) in going from soils-to-plants would be equal to 0.01 and the concentration of Cs'*" in plant potassium from fallout of 1 mc/sq mile would be about 0.3 yuc/g. The Cs'*? deposition level in the northern United States (mid-1957) is estimated at about 46 mc/sq mile, which suggests 15 puc Cs'*’ per gram of plant potassium, or a Cs*"/K*" gamma ratio of 0.18. The calculated ratio is in reasonable agreement with values measured in the Los Alamos large-volume liquid scintillation counter.’ Measured Cs!*"/K*" ratios in 1957 dried milk samples from the northern United States*’ averaged about 30 yuc/g, giving an estimated discrimination factor (DF,) of about 2 in favor of Cs'*’ in going from plants-to-milk. Tracer studies on mar*® show that Cs"! and K”, upon ingestion, are absorbed essentially 100 per cent and that they are excreted with mean times of about 150 and 50 days, respectively. These data suggest a discrimination factor of about 3 in favor of Cs'®" in going from diet (DF; and DF,) to man.* Since 50 per cent of the potassium in a western diet comes from milk and dairy products,’ the over-all ratio (OR) of Cs'3"/K is going from soils-to-man equals 0.5 (0.01 x 2 x 3) + 0.5 (0.01 x 3), or 0.045. In other words, the Cs'*? concentration per gram of body potassium should be about 4.5 per cent of the total Cs'*’ concentration per gram of exchangeable soil potassium. Andersonet al.’ suggested that Cs!” may be entering the biosphere and man largely through direct fallout on vegetation and not by plant uptake from the soil. This suggestion was based on the following considerations: (1) The high fixation of Cs'4? in soil and its very slow leaching rate makeit unlikely that the Cs'8’ can be in equilibrium with exchangeable soil potassium to the depth of the plant feeding zone. (2) The csi8t/K*" ratio of people does not seem to be increasing in relation to integrated Cs'*" fallout. (3) Cs'3"/K** ratios in milk show Sharp increases during periods of weapons testing, after which they rapidly return to near their previous levels, suggesting the possibility of a quasi-equilibrium condition with the rate of stratospheric fallout. The relatively small effect of a sharp increase in the Cs'*" content of foods during periods of tropospheric fallout on the Cs'*? content of people can be explained by the simple model shown in Fig. 6 (reference 7). A step function change in the foodstuff level will be followed by a (1 ~ e-4t) change in the population level (where A is the biological elimina- tion rate), and a new equilibrium value will be reached only after an elapsed time of about one year. If the foodstuffs return to their previous value before equilibrium is attained, the population level will cease rising and will return to its previous value with a half-time corresponding to the biological elimination rate. (b) Cs'¥? Levels in the Population. Concentrations of Cs'*" per gram of body potassium can be estimated from predicted average maximum surface deposition levels given in Tables 2, 3, and 4 and the ecological considerations discussed previously. One millicurie of Cs'*" per square mile gives a specific activity of 30 uuc per gram of exchangeable soil potassium. The specific activity times the surface deposition levels times the over-all discrimination ratio (0.045) gives the specific activity per gram of body potassium for the population of any fallout area, assuming no equalization between areas through food distribution channels. *A value of 3 for the discrimination factor from milk-to-man (DF3) is not confirmed by measurements on people and milk from the same areas.’ These data strongly suggest a discrimination factor of approximately one. 297 saath 36g