Washed human erythrocytes incubated with glucose and S8 and purged with N2 produced H2S at a nearly constant rate of 170 mumol (L cells)-1 min-1, which continued for several hours. In sealed vials up to 25 mM HS- accumulated. Glucose caused the fastest H2S production, although either lactate or glycerol could support slower rates. When glucose was added without S8, anoxic H2S production nonetheless occurred at about 1.5% of the maximum rate, after 24 hr totaling 0.5 mmol H2S (L cells)-1, suggesting the presence of endogenous reducible sulfur. Anaerobic conditions were not required, since oxygenated cells produced H2S from S8 at 80% of the anoxic rate. Using cell lysates, production of H2S occurred after addition of either glutathione, NADH, or NADPH. The observations suggest possible physiological roles for H2S as an electron carrier, and are consistent with an evolutionary relationship between eukaryotic cytoplasm and sulfur-reducing Archaea.