and indicates to us that direct ingestion of fallout was a reasonable pathway to assume. Although BRAVO debris was not highly soluble in water, calcium carbonate and hydrated calcium oxide (the matrix in which BRAVO fallout was entrained) were both highly soluble in acid (Co72). Therefore, ingestion of BRAVO debris resulted in release of radioiodines trapped in the granules due to the acid enviroment of the stomach. The mass and volume of BRAVO fallout granules were insignificant relative to the normal amount of food eaten per meal, the normal amoung being about 400 g for adult meals (Ev66). The mass of BRAVO fallout per m° at Rongelap Island was 4.4 g. The volume corresponding to this mass was 1.9 cm’, about four tenths the volume of a teaspoon. The mass per m and corresponding volume af Island it was 1.5 g m Utirik Tslagd was 0.46 g and 0.20 em’. and 0.48 cm” m “. For Sifo These mass and volume per unit area estimates were for the point in time at which all fallout was on the ground. The values for Utirik and Sifo Islands were estimated by ratio of the exposure rate to that at Rongelap Island with the ratio being estimated after the cessation of fallout. The main water supplies at Rongelap, eight cisterns, each contained a height of 0.23 m of water during the later part of March and early April 1954 (Sh57). Water was drawn from six of these cisterns at Rongelap for gross beta analysis on March 2,.1954 (see report of the radsafe team, USS PHILIP, 0C68), and ong other cistern was reported as out. Each cistern opening was about 0.65 m* and was fed by galvanized metal sheeting used for catching rainwater (Sh57). A little rain was reported on the afternoon of March 1, 1954 (Sh57). It was assumed that the additional cistern catchment area did not contribute water or activity to the cistern. Results of the analysis for gross bega activity concentration in ciste n water ranged from 1.8 x, 10 2.0 x 10° Bq 2 to, (0.005 to 0.054 Ci ml™*) with a mean of 5.0 x 10° Bq kL (0.027 yzCi ml!) at 50 hours post detonation (0C68). The fallout from Castle series coral surface bursts including BRAVO was barely soluble in water (Ka66). Estimates of solubility were based on BRAVO fallout which was collected with mixtures of rain and sea spray, 20-50% of the iodine activity was found in the liquid phase (Ka66). The servicemen at Rongerik Atoll examined the terrestrial fallout under a microscope and reported that the sand like granules were not soluble in water on the microscope slide (Sh57). Therefore, most BRAVO activity probably remained with fallout granules at the bottom of a cistern. Mean air concentration estimates of the activity of selected nuclides were based on the deposition rates of fallout granules and total activity deposited. The air activity concentration at a point in time was assumed 1) directly proportional to the fraction of total activity deposited per minute, 2) directly proportional to the total activity on the ground at the end of fallout (decay corrected back to that point in time) and 3) inversly proportional to fall velocity of granules. The cumulated air activity concentrations for Rongelap Island which were derived by us from Bikini Ash activity results were about three times less than the cumulated air activity concentrations given by Peterson (Pel). i ; , t : “i 28