respect to Mo" for various radionuclides with volatile precursors indicates changes in nuclide composition with time, which are in rough agreement with the observed values for Sr®¥ and Ba!*° on Wahoo and Umbrella. If this hypothesis is accepted, it would be reasonable to expect that, because of the later surfacing time on Wahoo, a larger proportion of Ba'#°, which has shorter- lived gaseous precursors than Sr°*, would remain with the ocean water. The early surfacing of Umbrella would permit a larger proportion of the volatile precursors of Ba'** to escape. These would then be rapidly scavenged by airborne droplets. effects have been observed (Reference 105). It would appear that both these The speculations offered here cannot be consid- ered conclusive; they do, however, appear to justify the limited observations made during Operation Hardtack and suggest that more precise investigation into such possible mechanisms might further illucidate the fractionation phenomeaon. Other miscellaneous samples collected by various simple means gave further evidence of fractionation. These samples included ocean-water samples, funnel samples, bottom samples, and cloud samples. The ocean-water samples were simply collected in polyethylene bottles. The funnel samplers (FS) described in Sections 1.33 and 2.2.6 were placed at a number of positions within the array in a special bracket that permitted helicopter recovery immediately after the shot; thus, decay information from H+4 hours was obtained for both Wahoo and Umbrella. Cloud samples were obtained for Project 2.3 by the LASL cloud-sampling aircraft. Cloud Samples were collected on 2-inch-square filter paper patches placed upon the large filter paper normally used by LASL for aircraft sampling. A single bottom sample was obtained from the Umbrella crater on 13 June 1958 by means of an improvised bottom trawl. The locations and times of collection for all these miscellaneous samples are summarized in Table 3.35. The crystal decay curves obtained for some of these samples are presented in Figures 3.222 and 3.223. All these decays were counted on Shelf 1 of End-Window Gamma Counter 2 (described in Section 3.3.1). The distance from the support of Shelf 1 to the bottom of the crystal is 21 mm (conversion factor from Shelf 1 to Shelf § is 0.173). The cloud sample filter patches were digested by vigorous stirring in nitric acid, and the resulting suspension wasfiltered and made up to 50 ml. A 2-ml aliquot of this filtrate was distributed over the collecting surface of a blank IC tray, which was then counted. All FS’s were treated in a similar manner. These samples were made up to a known volume, and an aliquot of the resulting solution was distributed over an IC tray as summarized in Table 3.36. Similarly a 3-ml aliquot of che H+ Z hour ucean-water Sampie for Umbrella was distributed over an IC tray and counted. Tle Umbrella crater sample was also counted by first disSolving 2.4 grams of the air-dried mud in nitric acid and then distributing the solution over an IC tray. The treatment of these miscellaneous samples permits comparison with decays determined for the IC collections if correction for shelf geometry is made. Funnel Sample 1 and Cloud Sample 1 for Wahoo were also counted in a 47 gammaionization chamber (Reference 110). The relative ionization readings versus times are presented in Figure 3.224. 3.5.2 Physical and Chemical Properties. The majority of project instrumentation was spe- cifically designed for the measurement of the gammafields resulting from an underwater detonation. Only those instruments installed on the platforms aboard each target destroyer (Sections 1.3.3 and 2.2.6) were designed to provide samples from which physical and chemical properties could be determined. Thus, the conclusions presented in this report are obtained from a limited number of samples collected at no more than three downwind positions. During Wahoo, samples were obtained from the DD-593 only because of the power fatlure aboard the DD-474 and DD-582 prior to the shot. During Umbrella, nearly complete sets of samples were obtained from the DD-592 and DD-593, those aboard the DD-474 having been destroyed by what appeared to be the impact of the water-laden base surge moving at its initially high velocity. The AFI installed on the platiorm aboard the DD-592 (Figures 1.13, 1.16, and 2.8) is fully described in Section 2.2.6. A series of samples are collected at a rate of 10 ft/minon DMT filters, each backed by reservoirs that act as liquid traps. Thus, the sample is separated at the time of collection into liquid and solid fractions. A total collection by the AFI is composed 326