To determine the relationship between respiratory muscle strength and changes in normal vital capacity (VG), graded levels of muscle weakness were produced in six healthy supine male subjects by four successive doses ofd-tubocrarine (0.05 mg/kg each). The maximal effect ofd-tubocurarine abolished hand-grip strength and ability to sustain head lift for 5 s, but VC was decreased to only 66 ± 3 per cent of control. At each level of weakness decreases in VC were significantly less than were decreases in respiratory muscle strength (RMS) monitored by maximum static inspiratory and expiratory pressures. The first dose decreased RMS to 86 ± 3 per cent of control, but VC was unchanged. Following the second dose, VC (97 ± 1 per cent of control) was minimally affected compared with RMS (71 ± 4 per cent). The VC after the third dose was still 85 ± 3 per cent of control, while RMS had decreased to 58 ± 2 per cent. Following the final dose ofd-tubocurarine (cumulative total 0.20 mg/kg), RMS was 39 ± 2 per cent of control, compared with VC, 66 ± 3 per cent of control. The relationship between VC and RMS was curvilinear and conformed to predictions based on the static mechanical characteristics of the normal respiratory system. These findings demonstrate that while VC is relatively spared during partial curarization, this sparing of VC does not indicate a similar extent of preserved RMS. Rather, it reflects the relative ease with which weakened respiratory muscles are able to drive the normal respiratory system in the supine subject. The same weakened muscles may be unable to generate sufficient force to handle mechanical challenges such as coughing and vomiting.