deposition patterns to reconstruct the initial radioactivity of the particles in the nuclear cloud by using winds to assign a point on the ground to a given particle size and altitude of origin. There is a considerable body of local fallout information for the relatively low-yield explosions that occur in the Nevada Test Site, and predicting dosages in the unpopulated areas adjacent to the Nevada Test Site is reasonably competent. There is appreciably less fallout information for high-yield explosions in the Pacific, and one cannot be sure that the particle size distribution will be similar for bombs exploded over large cities as for the Pacific coral or the Nevada desert. On the positive side the meteorologist can provide certain information about local fallout. First, he can tell the area in which there may be some radioactivity based on the winds and crude knowledge of the explosion characteristics. Second, he can tell the approximate time of arrival of local fallout; therefore, for civil defense applications, warning to the down-wind population is possible. The Federal Civil Defense Administration (FCDA) has prepared a hypothetical bombing attack on the U. S., using some 2500 Mtof fallout in the form of 250 bomb drops. Thepicture for FCDA Operation Sentinel, which was shown publicly for the first time at the congressional hearings on fallout,” illustrates the fallout pattern as it would appear 24 hr after this wide- spread bombing, using Nov. 20, 1956, winds (Fig. 1). The code in the lower left-hand corner of the figure indicates the dosages in the fallout pattern. It is quite evident that there is very little area east of the Mississippi Valley free of fallout. Secondly, I should like to make it clear that some of the uncontaminated areas in the western part of the country would be covered with fallout if another set of winds or ground zeros were used. The main purpose for including Fig. 1 is to put any remarks on test fallout in perspective. Nuclear tests provide milliroentgens of radioactivity in populated regions® in comparison with tens, hundreds, or thousands of roentgens to be found during a nuclear war. 3. INTERMEDIATE FALLOUT Raine tesat Fallout particles come in a continuous spectrum of sizes. It is believed that fallout that occurs more than a few days after an explosion consists of particles with negligible settling speeds. Probably the bulk of the fallout that occurs during the period from 1 to about 60 days originates in the troposphere, even for explosions whose clouds gointo the stratosphere. The justification for saying that most of the intermediate fallout is deposited within 30 to 60 days is shown in Fig. 2. The ordinate shows the amountof radioactivity measured by air filtration in the lower atmosphere on a logarithmic scale plotted against time in weeks. The concentra- tion decreases rapidly in time with a half time of about 20 days for Nevada tests whose nu- clear clouds do not enter the stratosphere. It can be seen that the delayed fallout from the megaton or thermonuclear tests do not show this rapid decrease with time. It has been known for many years that the prevailing winds blow west to east or in few instances, east to west. Thus, transport in a north-south direction is much slower than ina west-east direction. This results in a fallout band,‘ which lies almost entirely near the latitude of the source. Figure 3 showsthis for a Nevada test series in the spring of 1953. Intermediate or tropospheric fallout from tests conducted in the Pacific Proving Grounds is similarly distributed in‘a band aroundlatitude 11°N. The technique for sampling fallout by using gummedfilm collectors is quite uncertain, but, if it is accepted as correct, then 25 per centof all the fission products in a test appears as intermediate fallout for Nevada tests.! The fraction of the fission products deposited in intermediate fallout from Pacific tests is even less well known because of the vast unsampled ocean areas in the tropics, but the fraction is considerably smaller than 25 per cent.! In terms of evaluating hazard from intermediate fallout, two points should be noted: first, although all the nonlocal fallout fission products formed in kiloton-sized explosions contribute to intermediate fallout, the amount produced by such an explosion is very small compared to that of a high-yield explosion. Thus, 500 nominal (20 kt) bombs provide the same amountof fission products, roughly speaking, as one 10-Mt fission bomb. On the other hand, the deposition per unit area is clearly higher in those regions in which tropospheric fallout occurs than would be the case if the radioactivity were distributed uniformly over the entire globe. Thus, the average concentration at latitude 40°N in Fig. 3 is about five times higher than would 311