The design of small mammal studies should receive careful considerations at the nuclear sites. The designs presently in use at the safety-shot areas yield data that are difficult to relate to soil and vegetation concentrations. These difficulties arise primarily from the mobile nature of the animals and the lack of detailed knowledge relative to the levels of contamination to which individuals are exposed. Explaining the variability observed among animals captured at the same trap location is, therefore, a difficult task. The procedures used to prepare soil before being sent out to the laboratories for analysis need careful review before sampling begins at the nuclear sites. The practice of sending ball-milled samples to some laboratories and sieved (after ball-milling) samples to other laboratories (the case for TTR and Area 11 soil samples) has created problems of the comparability of soil results between laboratories. Hence, this practice should not be continued for the nuclear sites. All aliquots of soil leaving the soils preparatory lab should be prepared in the same way, regardless of which analytical laboratory is to receive the aliquot for analysis. Furthermore, the recommended procedure is to ball-mill but not to sieve the soil samples. This recommendation is based on data from the safety-shot sites that suggest concentrations in sieved soil aliquots are greater than those in unsieved (but still ball-milled) aliquots. This effect would presumably also be present for soils at nuclear sites. EXPERIMENTAL CLEAN-UP PLOTS Plans are being made to establish small experimental clean-up plots on nuclear and/or safety-shot sites for purposes of choosing among several clean-up procedures on the basis of cost, effectiveness, and posttreatment stabilization of the areas. The design and layout of these plots will require particular attention to the specification of detailed objectives, variables to be measured, kinds of field and laboratory instrumentation to be used, control (if possible) of cross contamination of the treatment plot from the contaminated sites, and the selection of the sampling and experimental design. 110 The placement