One primary concern about nonionizing radiofrequency radiation RFR has been accumulation of subtle injury over a long period of time, resulting in a delayed expression of harmful effect. In the case of RFR, a general consensus has been achieved among standard-setters that, for acute exposures, the threshold for ill effects lies above a specific absorption rate SAR of 4 Wkg. Since hazard injury has been verified only above this threshold, one might conclude that the current approach of applying a safety factor of 10 or more to this level would adequately protect the population at risk. Unfortunately this is not the case. Lack of data derived from true long-term experiments or from epidemiological studies of defined exposures to a human population has prevented a consensus from being reached for these kinds of exposures. In 1978, the U.S. Air Force, with the University of Washington, embarked on the most ambitious long-term low-level RFR study ever attempted. After 2 years of facility and equipment design, exposure device construction, protocol development and refinement, and pilot operation, a definitive study was initiated in September 1980. After 25 months of continuous exposure, the test was completed in September 1982. Data analysis and tissue examination were completed 1 year later, and the results are being published in a series of nine technical reports, of which the ninth is an overall summary and interpretation.