more tissues rather than on counts of the entire organism, as was done for certain treatments of the fish data. Attempts to compare species by areas on the basis of the ratios of activity of their tissues were thwarted by a lack of some samples and by the presence of many samples with only background counts, 1.e., net sample counts of zero. Also, the method of ranking was considered but was believed to be inadequate because of the great effect of surface contamination upon the average of a limited number of sample counts (see p.18). Results. Appendix Table 1 gives individual sample values. Table 5, page 32, shows average amounts of radioactivity in the main invertebrate groups at the collecting localities. Blanks indicate no specimens were found. These values bear out the inverse relationship of radioactivity to distance from the test sites for operations previous to Ivy, which extended from Runit Island to Engebi Island. Within this range the only significantly low counts came from a small collection made by navy personnel on Piiraai Island. However, it is probable that because of this island's position relative to the prevailing winds, waves, and current it neither initially received nor retained large amounts of radioactivity, in spite of its intermediate position vetween two shot islands. Igurin and Japtan Islands were almost equally low, and Rigili higher. Because of their marked influence upon the averages, the high-counting samples included in Table 5 and Appendix Table are listed separately in Table 6. - 31 -