Nitrous oxide and halogenated anaesthetic agents are often administered concurrently during neuroanaesthesia. To assess the interactive effects of these agents on cerebral blood flow (CBF), we have measured CBF by 14C-iodoantipyrine autoradiography, in normocapnic rats receiving either 0.5 or 1.0 MAC of volatile agent alone (halothane or isoflurane) or 0.5 MAC of volatile agent+0.5 MAC nitrous oxide. CBF (global and regional) was significantly greater at 1.0 MAC compared with 0.5 MAC, regardless of how 1.0 MAC was achieved (agent alone or agent+nitrous oxide) (P < 0.05). The addition of 0.5 MAC nitrous oxide to 0.5 MAC halothane resulted in flows similar to those produced by 1.0 MAC halothane alone. In contrast, the addition of nitrous oxide to 0.5 MAC isoflurane resulated in flow values significantly greater than those measured during 1.0 MAC isoflurane alone. We conclude that the use of nitrous oxide, as opposed to an increased dose of volatile agent, has no advantage in respect of minimizing anaesthetic-induced increases in cerebral blood flow.