is present for reasons of time and experiment, there is no large variation in the stratospheric content of Sr*° between the latitude of 30°S and the Northern Hemisphere. Since mostof the megaton yield explosions have occurred in the northern latitudes, though the Pacific Testing Grounds are only 11° north of the equator, it appears that this evidence argues for rapid north and south mixing in the stratosphere. As we shall see later, other evidence in the dissemination of nonradioactive carbon dioxide derived from the combustion of fossil fuels*!~* and of the dis- semination of bomb derived radioactive C'* seemsto confirm this.*-** It is interesting to note also that the actual content of the stratosphere is not in disagreement with the estimates given earlier,* although the value of the filter efficiencies remains to be settled, and it is estimated at the efficiency of about 25 per cent on evidence assuming homogeneity of the particle size. Experiments are now underway to settle the point. In the model previously advanced,** it is proposed that material introduced into the stratosphere is mixed immediately horizontally to a uniform concentration and has a residence time of 10 years. Further, it is assumed that the latitudinal spread of tropospheric bomb clouds is only 10° with a sharp step function rather than a normal error curvedistribution. The bomb debris is arbitrarily assigned to the stratosphere except for 1 per cent tropospheric in the case of megaton yields. Local fallout is assumed to be 80 per cent for land surface shots, 20 per cent for surface water shots, and 10 per cent for air shots. All kilo yield shots are assigned to the troposphere. On these very simple bases we are then, from classified data about the magni- tudes and nature of the explosions, able to estimate the total fallout for any place on earth if the deposition from the troposphere is assumedto be proportional to the rain content at a given location. Figure 3 gives such a theoretical latitudinal fallout profile for world-wide fallout as of December 1957, neglecting rainfall variation, and Fig. 4 is the corresponding world map. Figure 5 gives the corresponding timewise variations in the northern latitudes and compares them with the rainfall fallout curves for Milford Haven in England.” Figure 6 gives a similar comparison for Chicago and Pittsburgh. Curves for other latitudes are given in Figs. 7 and 8. Figure 9 gives the estimated stratospheric reservoir and the expected composition in Sr® versus time. If a further assumption is made, namely that the proportion of the fallout ina given location is given by the ratio of the rainfall to the world-wide average, 0.77 m,“it is possible to compare the detailed fallout observed by the pot collection programs in various localities with the theoretical predicted values given in Table 1, Part 1. On the basis of these comparisons and in the absence of conclusive evidence as to the age of radioactive fallout, it appears that the simple theory outlined explains the known information within the experimental error. It may develop when more reliable data are available on the age of fallout through the use of short-lived, 12.8 day half-life Ba’ fission product, that a mecha- nism by which a sort of concentrated leaking from the stratosphere occurs ata latitude of about 40° or more may be proved or disproved. At the present time the observed extreme concentration may be explained as being due to coincidence of the tropospheric fallout from the U. S. and Russian tests. If this theory is correct, the Ba!’ content in periods of high fallout will show that the fallout is young. It is to be hoped that these data will be forthcoming soon. Machta,!*“! and Stewart, Osmond, Crooks and Fisher® have stated that meteorological con- siderations and likely stratospheric wind patterns, together with evidence that the Sr®/sr*® ratio of the fallout shows the fallout to be old, have led them to the conclusion that the heavier fallout observed in the 40° to 50°N latitude band is stratospheric and not troposphericin origin as proposed here. The issue still seems to be unsettled since the radiochemical difficulties of the determination of the Sr® /Sr*° ratio are large and may well have introduced sizeable errors into some of the reported values for this number andsince it apparently is possible to account reasonably weil for the observed fallout distribution on the present uniform stratospheric fall- out theory as shown in the present paper. The critical difference between the two theories is in the matter of the age of the fallout. Better and more significant results probably will be avail- able soon using the Ba‘*’/sr®* ratio which for both radiochemical and lifetime reasons is more suitable than Sr®/Sr®*®. Ba! has a half life of 12.8 days which is more appropriateto distin- guishing between an expected fallout age of perhaps 30 days on the one hand and of about 1 to 2 years on the other, than is the Sr®™ half life of 51 days. The radiochemical procedure for Ba!“ is very similar to that for Sr°’ and both are more sensitive and reliable than the Sr® procedure which is particularly susceptible to errors from radioactive impurities such as other fission products which may have been imperfectly separated. Both Ba‘? and Sr*° are measured by 259