i150 errors. In addition, the only samples which snowed 239+240, Pu/-~ 238. ru catios mare than only slightly different from each other in absolute value (samples B-16, B-22, B-27) were those with the largest relative counting errors. 238 Pu Because the error term in these first tyvo sets of plutonium ratios arise mainly from the error associated with the 238 Pu concentrations inaasured, these different ratics can be assumed cue to normal variations in the low 2385, counting rates and/or an effect related to the addition of the 236 Pu tracer (in the first case). Since the 23€ Pu concentrations measured in these three samples was tower than in the unspiked sample, it is obvious that the tracer impurities were not resulting in calculeted 238 Pu concentrations which were greater tnan were present. Tne concentrations of 238 Pu and 23942405 measured in surface sediments are shown in Appenaix Table 5. The B0420C Fu concentrations an¢d measurement errors reported were cemputed for individual samples as previously samples for which the ratios The 2385, concentraticns in all but those 239+240,, 233 Pu were redetermined (stown in /ppendix table 4) were calculated in a similar manner. For these late” samples, the ratio reported is the ratio comouted from the two analyses by weighting each ratio ty its propogatad counting error (by the method c of Stevenson, Lop. cit.J; and a weighted mean ratio anc error was computed. -. ihe Z . : : . sey concentration in these few samnles wes then calculatec using the equation in fuot note | on the previous page. to this orocedure. There were two excestions No sample 5-19 or £-20 (Appenciix taole 4) remained for a second plutonium analysis so that a second sample previcusly storea as excess hed to he dissolved for the analysis, ana the semple was, in ae . 242 addition, spiked the ~“"Pe. . . . In the analysis, the fusion porticn of the dissolving the sample procedure was inadvertently omitted. Beceuse of weerenete explained (section 4.2-4).