northern atoll reef. The island chain is quite long and narrow and has only sparse vegetation. Several tests of nuclear devices were conducted in the near vicinity of this island complex, and thus the extremes in radiation levels typical of blast areas were found here. A survey using the portable meters was made along a route down the middle of the complex. These measurements ranged from 5 to 20 ur/hr on Aomen, 20 to 330 uxr/hr on Lomilik, 10 to 40 ur/hr on troij, and 3 to 7 »r/hr on the causeways. Except for the higher valves on Lomilik, these exposure yates were typical of weathered, sandy soils capable of only low retention of radioactive materials. The hot spots on Lomilik were depressed areas with clay-like soil. A soil sample was taken from the area of greatest activity. The soii analysis indicated that °°Co was respon- sible for more than 3/4 of the total exposure rate, with *25ep and lesser amounts of ~°*™Rh and **’¢cs contributing almost all the remainder. Traces of + © Rh, *°lRh, 14406, *55 mu, and ***am, and thus *°°Pu were also detected (see Figure 12). A 1964 sample from Iroij indicated that at that time ©°Co contributed about 75% of the total exposure rate, *22 sm about 8%, *87 a5 about 9%, and eaten the remainder, in substantial agreement with the i967 data for Lomilik. Many pieces of fairly radicactive metal scrap were found throughout this area. Althcsgh mest of the time this contamination was due to ~"Co several samples exhibited only unidentified 240 keV gamma-ray activity.