such small magnitude that decay is greater than build-up. Another check on the validity of the analysis using the experi- menvel model was a comparison of the particle size distribution as measured from samples collected on the atolls and the size distribution that would be expected from consideration of the trajectories of the particles. Table 6.7 tabulates the measured particle size distribution found in samples from the atolls as taken from the data presented in Chapter 5. - TABLE 6.7 - Shot 1, Measured Particle Size Smallest Particle Station (u) Bikini Ailinginae Rongelap Village Rongelap North End Rongelap, Kabelle Utirik . Largest Particle (») 25 16 >1000 172 16 518 10 16 126 394 6 134 Geometric Mean (4) 112 60 70 45 The calculated trajectories showed particles from 2000 to 100 p arrived as primary fallout within the Bikini Lagoon. This fact agrees _ very well with the measured size distribution shown in Table 6.7. ConSideratiun of the cloud diameter and stem diameter, in the experimental model, on the arrival points of the particle trajectories indicates particles from 150 to 75 # diameter would arrive at the north end of © Rongelap with the linit of the 250 p# particles falling approximately 10 nautical miles north of Rongelap Atoll. The steep gradient of particle size distritution in a north-south line is also clearly indi-. cated from the model study which agrees well with the size distribution found at Ailinginae some 15 nautical miles south of north Rongelap. Also the calculated size limits the particles arriving at a distance or 300 neatical miles to a maximm diameter of 75 » as compared to a measured geometric mean size of 45 The only discrepancy of eay magnitude between observed data and those calculated from the experimental model is that no fallout arrived at Utirik based on the model analysis. It must be realized that at this distance the model analysis is weakest because the wind data used were extrapolated as being constant fron 0 + 6 hr to O + 20 hr, the latter being the time of arrival of fallout at a distance of approxiMately 300 nautical miles. This extrapolation was necessary because no analysis. Even better corralation of measured to calculated particle size would be obtained if a larger cloud diameter were used in the experimental modei. For this analysis the value used of 66 nautical miies was conservatively chosen; Project $.1 cloud dimension data indicate 8) , ee ee ee eaen ee eS \ vy wind data for periods beyond O + 6 hr was available at the time of this