CHAPTER 2 GROSS ACTIVITY IN PLANTS, SOIL, CORAL, ALGAE, AND WATER At the time of their collection in the field, the representative plant, soil, coral, and algae samples were placed in individual plastic bags for shipment to this laboratory. Water samples were collected in l-liter polyethylene bottles. 2.1 PROCEDURES AT THE LABORATORY The edible portions of the food plants were separated from the inedible portions with every precaution being taken to ensure a low probability of cross-contamination. Weighed samples of the food were dried at 110°C and then ashed in a muffle furnace at 500°C. The ash was transferred to tared aluminum planchets, weighed, and prepared for counting. Samples of supporting plant systems and grass were prepared in the same fashion. Fresh water samples were acidified and evaporated to dryness. The residue was taken up in water and theslurry transferred toa planchet for drying, weighing, and counting. The radioactivity was isolated from ocean water samples by (a) buffering with NH,C1 to hold magnesium in solution, and (b) precipitating the natural calcium with Na,CO,. The resultingflocculation was allowed to settle overnight and the bulk of the supernatant was removed by suction or decantation. The remaining slurry was transferred to lustroid tubes and centrifuged, The precipitate was washed once with water, transferred to an aluminum planchet, and dried at 110°C. Analysis of the supernatants from the more active samples showed that recovery of activity by precipitation ran from 80 to 90 per cent. Soil and lagoon-bottom silt samples were dried and mounted in aluminum planchets for counting. Coral and algae samples were dried overnight at 110°C, ashed for 24 hr at 500°C, pulverized, and mounted in aluminum planchets. After mounting, the samples were counted with a gas-flow proportional counter at 26 per cent geometry as determined witha U,30 standard or with a 1.9 mg/sq cm, end-window, G-Mtube and scaler at 14 per cent geometry as determined with a U 30, Standard. Absorption and scattering corrections were determined empirically by counting varying weights of individual samples and extrapolating the specific ot ancarves