INTRODUCTION From 1951 through 1962, about 100 above-ground nuclear tests were conducted at the Nevada Test Site (NTS). Some of the fallout patterns included inhabited areas in Nevada, Utah, and surrounding states. Radionuclides in the fallout debris produced human exposures from external irradiation, inhalation, and ingestion of contaminated agricultural products. Epidemiological data, public concern and litigation prompted a need to estimate doses to the public from all exposure pathways. During the periods of heaviest fallout (1951-1958), few direct measurements of specific radionuclides in foods from the contaminated region were available. Foodchain transport simulation models were the only practical means of estimating internal doses to human tissues from ingestion pathways. The need to simulate discrete fallout events, seasonal changes in agricultural and dietary practices, and complex, time-dependent radioecological transport processes, required a dynamic, process-oriented model. Our goals dictated that the model provide realistic dose estimates with specification of uncertainties. Rigorous comparisons of model predictions against real, independent measurements were used as the primary means of establishing model credibility. This paper provides a brief overview of the methodology developed to estimate the transport of radionuclides through agricultural ecosystems to persons of various ages, lifestyles, and geographic locations. The methodology, embodied in the computer code PATHWAY, 1 ' was used to convert estimates of fallout deposition to time-dependent concentrations of radionuclides in food products, total intakes by people, and organ-specific doses from 21 radionuclides in fallout from 86 nuclear test events. A summary of model prediction uncertainties, an assessment of predictive accuracy, and a discussion of the relative importance of different exposure pathways is also provided. THE PATHWAY MODEL PATHWAY simulates the transport of various isotopes of I,