per gram of soil was estimated. Corrections to the soil activity were made on the basis of results for soil sample blanks, comparison sample blanks, and method yield. A listing of the gamma-ray spectroscopy results and !291 results for soil samples is shown in Table 6. The earliest surface soils dated back to 1955, about one and one-half years post~detonation of BRAVO. The 60¢o, 137¢5, and '55gu activities per gram of soil appeared to have declined slowly over the years, while 1291 soil results declined at a much more rapid rate. Individual counting errors were normally less than 5%, although a few samples approached 20%. 3. Estimate of Initial Surface Activity. Positive 125s, results (see Table 6) were too Sparse for inferences to be drawn. For !291, results were plentiful and, therefore, a least squares fitting was performed on results for nuclide soil activity per unit mass of dry soil vs days post-detonation using linear, exponential, logarithmic, and power function models. Sample results varied from their best-fit value by as much as a factor of 9 and by an average factor of 2.5 over the period 1955 to 1977. The best-fitting function was determined from a comparison of the co-efficient of determination for each model. Functions used with !291 results for Rongelap soil were plotted in Figure 3. Only the 1955 to 1957 results were plotted to illustrate the following points. For I soil results, the bestfitting function was exponential. All four fitting functions were generally useful in predicting soil activity per gram at times after 600 days post-BRAVO for all nuclides.- A significant divergence between functions occurs during the period several hours out to one year post-BRAVO. For example, at 0.5 day the difference between the exponential and power function estimate spans 5 orders of magnitude for 1297, Single exponential fitting gave the best coefficient of determination for 1291. the exponential fit of the 1297 soil results at Rongelap led us to estimate a mean residence time in surface soil of about 5.4 years. The descriptive on these samples indicated that they were soils originally located at depths no greater than 5 cm (2 in.) beneath the surface. 4. Ratio of Nuclide Activity to Total Fallout Activity in Archival Soils. The ratio of '29T activity to total fallout activity would help to determine whether the archival soil anaylsis for radioiodine corresponded to other meas- ured or hypothetical ratios. The surface-soil activity of each nuclide measured by 3rauer was estimated for 0.5 and 1.5 days post-detonation. These were the assumed times of cessation of fallout at Rongelap and Utirik, respectively. The value for the nuclide activity per unit mass of soil at the cessation of fallout was estimated from the best fit of archival soil results. For !291, the estimate was recorded in Table 7, columA. At Rongelap, the total fallout activity per unit soil, based on four soil samples taken March 8, 1954 (0C68), was 3.0 x 109 + 4.1 x 10° Bq gt (8.2 +11 uci g!) at 0.5 day. The total fallout activity per unit mass of Utirik soil was 11% of the Rongelap result based on a ratio of exposure rate at the two -~ 12+