involved, but the incorporation of nuclides into the human system through the usual biological channels introduces the possibility of long-term effects whose seriousness is not easily ‘ determined. : The local fallout from the tests at Eniwetok, as defined earlier, will settle out in the Pacific Ocean and hence will be of only indirect concern. However, the tropospheric and stratospheric fallout will come down over land areas. Careful consideration of the nucildes present in global fallout has indicated that Sr® is the one to be most feared because of its possible accumulation in the human skeleton and subsequent long-term irradiation of the hematopoetic tissues (Reference 27).. Consequently, a major part of the work done on worldwide fallout has been directed toward the estimation of Sr. Measurements have been made to determine the existing levels at the earth’s surface, the quantity stored in the stratosphere, and the deposition rate. Samples of fallout have been taken from the soil and vegetation, by gummed tape and pot-type collectors on the ground and by air-filter samplers at the surface and in the troposphere and stratosphere (References 8, 24, 25, and 43 through 56). ’ Based on this work, it was estimated thatin the fall of 1956 the Sr™ levels were about 22 mc/mi? in the midwestern section of the United States, 15 to 17 mc/ mi’ for similar latitudes elsewhere, and perhaps 3 to 4 mc/mi? for the rest of the world (References 43 and 57). The total amount in the stratospheric reservoir, if uniformly distributed over the area of the globe, would increase these figures by acout 12 mc/ mi”, The deposition rate of the stored material was considered to be around 10 percent per annum. It was further estimated that, if these levels were maintained for 15 years, the concentration in the human skeleton would be about 1 percent of the maximum permissible (Reference 27). The quantity of radioactivity in the stratospheric reservoir was estimated by summation of the contributions of all the bursts through Operation Redwing that have deposited debris itn the stratosphere. The available fraction of the device was determined by subtracting the local and intermediate fallout from the total. The intermediate fallout is thought to contain 1 to 5 percent of the weapon for megaton-range detonations (References 17,58, and 59). Determinations of this quantity by a worldwide network of stations for Shots Mike and King of Operation Ivy gave a figure of 2 percent (Reference 59). Much information on Sr™ concentrations in the stratosphere has been obtained bythe extensive high-altitude sampling program (HASP) of the Defense Atomic Support Agency. In addition, other data was gathered from filter samples collected on high-altitude balloons. The latter work was part of a continuing program for sampling the stratosphere along the 80th meridian (References 50 through 54, and 60). 1.2.7 Fractionation Effects-—— Observations at Other Tests. The occurrence of fractionation is manifested by differences in radiochemical composition, decay rate, or energy spectra among various samples of failout taken at different times or locations in the contaminated region. Observations of some degree of fractionation have been made at many different detona- tions. As expected, fission product nuclides such as Sr®?, Sr®, cs!9’, or Ba’, which have rare-gas ancestors with half-lives of a fraction of a minute or longer, are frequently found among the products that are most severely fractionated with respect to the bulk matrix material (always a refractory substance). The location of the burst is also an important factor. Separation of thenuclides from one another appears to be most pronounced in underground or surface shots (References 61 and 62), generally less for a water surface (Reference 63) and still smaller for balloon, high tower, and air detonations (References 63 and 64). Relatively little fractionation was found in water samples for one device detonated in deep water (Refer- ence 37). During Operation Greenhouse, it was noted that the exponent of the beta decay curve in-. creased from 0.95 to 1.3 with median particle size for samples taken from the clouds of Shots Dog, Easy, and Able. This indicated that the close-in particles were enriched in fast-decaying components with respect to the more distant fallout (Reference 65). For surface shots during Operation Jangle, pronounced depletion of chains 89, 115, 111, and 140 referred to Mo’® was obaerved in comparing long-range with local fallout samples. 17