-21200 yards apart on a farm differed by a factor of seven. To attempt to estimate quantitatively the amount of iodine-131 in milk by measurement of external gamma readings incorporates not only the uncertainties just mentioned but also adds those due to possible (a) fractionation of the fission product debris (b) incorporation of varying amounts of induced activities in the fallout (c) wide variances of retention of the debris on the foliage (as a function of particle size distribution and other factors) and (d) other variables such as accurate instrument response, especially at relatively low exposure rates (where most studies have been performed) and extrapolation of external gamma readings by the time71.2 relationship. All of these leave one with an uneasy feeling of confidence in the conclusions. The most gross relationship might be inferred in com- paring different types of data such as external gamma levels and iodine-131 in milk but then only as an alert for possible additional monitoring. In fact, as stated in the report 18. on the study made following the March 13, 1964 event, " .. . the external beta plus gamma measurements were background throughout the study . . . utilizing such relationships in this instance would have led to the conclusion that there would have been no measurable I-131 milk levels found whereas our data indicate that levels could actually have reached values near 700 pc/1l had the study been started at an optimum time." liter.) (The highest measured value was 420 picocuries per DOE ARCHr Even less can be said for using concentrations of radioactivity in the air as the basis for a model to predict quantitatively the amount of