as fast as 1 cm/sec or a half-mile per day, which is roughly what I believe necessary, then a heating of 10°C per day is needed. Present theory calls for warming of only two or three degrees centigrade per day. Second, it does seem fortuitous, but not impossible, that two diverse atmospheric processes, cooling by expansion and radiational heating should produce as good a daily balance. A second objection is a dynamic one and is shown on the right-hand side of Fig. 8. A ring of air at the equator will, if transported poleward, maintain its absolute angular momentum. Thus, if there were no west or east wind at the equator, this ring when brought to, say 40° would rotate at a speed corresponding to a 250-mph west to east wind. These tremendous speeds are rarely if ever observed. If there is a poleward circulation, there must also be some mixing to dilute the high resulting winds. Thus not all meteorological reasoning favors the circulation model. Let us summarize the predictions that Fig. 7 offers for the problem of the motion of stratospheric radioactive particles: the large amount of debris that originates in the Pacific Proving Grounds will be carried poleward and then be subjected to descending motion. This subsidence has its peak value in the winter and spring. As air is brought to the lower stratosphere, certain processes in the tropopause region can then carryit into the troposphere. Ordinary downward movement through the tropopause may be helped by several other special processes in the temperate and polar latitudes. The air that enters the troposphere brings radioactive particles. These are then removed from the atmosphere ina short time in much the same way as the intermediate fallout is removed. Stratospheric debris from USSR tests should, by this picture, remain in the temperate latitude, or move even further poleward, but in any case it should have a shorter stratospheric residence time. The model does not predict whether there should be greater deposition of delayed fallout in temperate or polarlatitudes. Climatological statistics on precipitation would dictate more fallout in the rainier temperate latitudes, other things being equal. 7 THE OBSERVED FALLOUT It is now proposed to compare this meteorological model with the observed distribution of fallout. Figure 9 shows a meridional cross section in which the Sr® deposition per unit area in soil is plotted on a linear scale as the ordinate and sine of the latitude as the abscissa.‘ The data show a marked peak in the temperate latitudes of the Northern Hemisphere, a minimum in the equatorial region, and a secondary and uncertain maximum in the temperate latitudes of the Southern Hemisphere. It also shows great variability among samples collected at the same latitude. Part of the variability is due to the difficulties in the analysis of the soil samples, and part is due to meteorological conditions such as raininess. Soil analyses provide the cumulative fallout since the beginning of the atomic era. Figure 10 showsthefallout in rain during a given year, 1956, by some 11 stations in the U. K.8 and two stations in the U. S. rainfall network.® Again the ordinate is millicuries of sr® per unit area on a linear scale, but the abscissa is latitude on a linear scale. The same general distribution is evident. Everyone agrees that the data show more Sr™® deposition in temperate latitudes of the Northern Hemisphere than elsewhere in the world—a picture that would be used to recommendthereality of our meteorological model if it were not for one fact. The Sr® comes not only from the stratospheric deposition but also from tropospheric fallout from the smaller tests in Nevada and in the USSR test areas, both of which are located in the temperate or polar latitudes. The critical question is: “What part of the nonuniformity is from stratospheric fallout?” Fortunately, fission-product analysis is able to shed some light on this question. Both rain water and air filters have been analyzed for shorter-lived fission products as well as the long-lived Sr®, The contribution of the Sr® from tropospheric fallout may therefore be assessed by finding the age of the Sr. If it can be shown that the age of the fallout is appreciably greater than 30 to 60 days, then it is very unlikely that much of the Srcould have originated from tropospheric fallout, irrespective of whether there was a recent atomic test. Several short-lived fission products, such as Sr®®, Ce’! (references 8, 10, and 11) and others, as well as dating by gross fission-product decay,° indicate average ages greatly in excess of 60 days. This evidence suggests that mostof the fallout in the temperate latitudes must have been stratospheric fallout. The conclusion is further supported by estimates of the amount of tropospheric fallout from 319