The therapeutic potential of a variety of bioreductive agents, including misonidazole, RSU-1069, NFVO, mitomycin C, porfiromycin, and SR-4233 was evaluated using Chinese hamster V79 multicell spheroids in vitro. Fluorescence-activated cell sorting techniques were used to selectively recover cells from various depths within the spheroids to measure the differential cytotoxicity in the cells near the hypoxic core of the spheroid relative to the well oxygenated peripheral cells. At the high cell density found in spheroids (as in tissues in vivo) the differential toxicity observed was typically much less than expected, based on data from single cell systems. In some cases, this was due to lack of sufficient hypoxia in the spheroids; in other cases, drug treatment itself produced reoxygenation through metabolic or toxic effects during treatment. An unexpected observation of considerable concern was rapid bioreduction of the more active agents; this sometimes occurred at rates that exceeded drug delivery, resulting in considerably less efficacy when large hypoxic fractions were present (e.g. mitomycin C, NFVO, and SR-4233). This suggests that induction of hypoxia prior to bioreductive agent therapy may not be the most productive approach. Though none of the agents showed "ideal" properties, porfiromycin was judged to give the best combination of differential toxicity, longevity in situ, and ability to reach the entire hypoxic cell subpopulation.