in this manner by as much as a thousand fold. Thus, for example, one ‘gram of plankton could contain a thousand times as much radioactivity as a gram of sea water adjacent to it. The radioactivity from these plankton which form a portion of fish diet tends to concentrate in the liver of the fish, and, if sufficiently high leveis of contamination are encountered, could have a marked effect upon the ecology of an ocean ares. However, the low level from world-wide contamination has not been identified to date in undersea life at great distances from the weapon test sites. The fish in close-in fall-out ereas have been shown to absorb radioactivity to varying degrees, one mechanism being the method of contamination just discussed. The slow world-wide fall-out of radioactive fission products involves all land and seaareas. Rain, as previously mentioned, scavenges small particles in the atmosphere quite efficiently. Thus, on a world- wide basis, arable land should exhibit higher concentrations of fission product radioactivity than desert lands. The New York Operations office has studied this problem as have the Lamont Laboratories of Columbia University and the Institute of Nuclear Studies of the University of Chicago. These studies have been reported under the project code name SUNSHINE and in general are classified SECRET. The data confirm the hypothesis that areas of high rainfall have considerably more artificial radioactivity in the soil than areas of low rainfall. The SUNSHINE data are considered reliable, but ultimate quantitative interpretations will have to await long-term animal experiments. The biological discussion of the SUNSHINE data will be included in the section on biological effects. Suffice it to say here that rainwater, ground water, soil, foods, and animal and human bones have been subjected to radiochemical analysis. All indications tend to substantiate the fact that there is a low level of world-wide contamination from the amount of fissionable material that has been fissioned in nuclear weapons to date. To return to the question of the land surface involved, the RAND 60