objectives of the survey had to be interpreted very carefully, bearing in tind that this survey was a precleanup effort and not the cleanup operation, Estimates would necessarily have to be crude, some very crude and not exact in any way. Enough information would have to be generated to make possible policy decisions and budgetary estimates for cleaning up the Atoll, The conclusions which could be drawn would depend on the number of samples collected and analyzed, as well as the judgement of the designer as to what information he would really expect to be required to obtain, It was a question of defining approximately the conditions on an island or a large fraction thereof, with a realistic cost/sample number estimate versus defining a specific volume of soil on a specific island and having an astronomically expensive and unmanageable quantity of samples. A compromise had to be made resulting in the sampling program requirements indicated in the following pages, Stratification The Atoll was stratified into groups of islands, individual islands, or specific areas of islands, depending on what was known or suspected about the area and what was required to be known, Although other radio- nuclides were present, the soils effort was based on 239pu concentrations. Previous experience with this long half-life isotope indicated it was of major importance in cleanup considerations. Using the results from pre~ vious surveys, the ranking of the islands according to fallout insult, knowledge of particular problems, and considerable insight, the Atoll was divided into four main phases based on assumed contamination conditions. Additional stratification was accomplished, depending on: the infor- mation required, the known or assumed contamination levels, the number of samples required to give the information, and assumed soil concentration