- 51 - general level of ra dioactivity at Rongelap Island, but does illust rate the . ’problem introduced by sample variation when too few samp les are taken. ane other soil sample from Rongelap Island cont ained 1.7 uc/kg of top 3 7 inches of soil and agrees favorabl with tha y t expected from a consideration of the values obtained in previous collections when radioactivity decline is <7 taken into consideration. The rate of decline of radioactive contamination in the combined soil “:~ gamples from Kabelle and Labaredj Islands between the dates, March 26, 1054 and October 23, 1955 is best represented by a straight line on log-log coordinates with a slope of approximately -1.6. The decay curve for a Labaredj Island soil sample extending through the same period of time 5 is expressed by the formula r=t i. 31 . In both curves March 1, 1954 is the date of origin. Enibuk Island (Ailinginae Atoll) soil contained 1.2 yuc/kg of top inch of - soil or an average of 0.61 uc/kg of top 3 inches of soil on October 23, 1955. Thus the soil at Enibuk was about 1/3 as radioactive as was soil from _Rongelap Island, a relationship which was reflected in the radioactivity - / levels ofland plants from the same islands. ‘The sandat lowtideline at - . Enibuk also contained about 1/3 as much radioactive contamination as did a » Uke Sample from Rongelap. At both Enibuk and Rongelap the radioactivity “level. in the low tide line sand was about 1/18 that found in the island soil (top 3 inches). | The levels of radioactivity in the top 6 inches of sand profiles taken on the lagoon bottom off Kabelle Island varied from 1/3 to 1/1 that for the UNIVER Ape ge cc. mA ; Usre we