for natural uranium before these estimates of device-related uranium can be accepted as accurate. Four profile samples from A Site are currently available for study. . 235 238 239-240 Figs. 5, 6, 7, and 8 are plots of the U, U, and Pu results from the 10 levels (2.5-cm increments) from each profile (the data from two other profiles of A Site are not yet available). Figs. 5 and 6 show the decrease with depth in atom percent 2354 235 and concentration of U (ug/g dry soil), respectively. The device used in the safety shot at A Site was largely 2351, and the decrease in concentration with increasing distance from GZ is reflected in Figs. 5 and 6. Concerning stake 3 in stratum 1 (210 feet from GZ), the atom percent averages 0.720 over the 10 depths and plots as a horizontal line in Fig. of natura 5. As noted above, 1 *35y is 0.72, the 7° U since the atom percent concentrations for this profile (plotted in Fig. 6) have been used to approximate the inventory of 238 U for this profile natural U (Table 3). The concentrations for were used in a similar manner. Fig. 7 is a plot of tions in these same profile samples. 238 U concentra- The slightly elevated concentra- tions in the first 2.5 cm of soil for those three profiles nearest ground zero suggest a relatively small percentage increase of 238y to the surface soil from the device. This contrasts with the rather 235 U concentrations in surface soil for large percentage increase in profiles near ground zero (Fig. profile data for 239-240, u. 6). Fig. 8 gives the available Most of the plutonium appears to be in the top 2.5 cm, with perhaps a gradual decline with depth below 5 cm. The plutonium present here could have come in part from the safetyshot device at B Site which is some 1,600 feet or so away. in Figs. 5 and 8 were fit by eye. In Fig. 6, The lines the data for two profiles were fit by the sum of two exponential terms (parameter estimates statistically significant at 9 = 0.05 or less). CONCENTRATIONS IN DIFFERENT SOIL FRACTIONS Analysis of plutonium and americium concentrations in ball-mill, mesh, and >100 mesh soil fractions was approached by asking, and attempting to answer, the following three questions: 131 <100