The quantitative importance of the adrenergic response of carp erythrocytes during severe oxygen restriction is not clear at present. Quantitative differences between in vivo and in vitro studies suggest that the response of carp erythrocytes may be dependent on the actual hypoxic condition. To our knowledge, a clear picture of the blood gas status, erythrocytic responses and catecholamines measured simultaneously in carp exposed to deep severe hypoxia or anoxia has not yet been reported. Therefore, we studied the physiological response of carp exposed to deep hypoxia at 0.3 kPa and subsequent recovery. Carp were fitted with an indwelling cannula in the dorsal aorta for repeated blood sampling and the blood was analysed for hematocrit, hemoglobin, mean cellular hemoglobin content, intra-and extracellular pH, pO2, pCO> total CO2 and catecholamines. Large fluctuations in arterial pO2 levels were observed in normoxic control carp, probably caused by the alternating breathing pattern of carp. Even at water pO2 levels of 0.3 kPa, arterial pO2 levels were maintained at about 0.2-0.3 kPa. Catecholamine levels were increased during deep hypoxia with noradrenaline as the predominant catecholamine. Hematological variables showed that the number of circulating erythrocytes was increased during hypoxia. The intracellular pH of carp red cells was maintained at pre-exposure values despite a considerable decrease of pHi. In this in vivo study, a marked decrease of the proton gradient across the red cell membrane (pHi-pHi), as high as 0.35 pH units, was observed, which is quantitatively similar to that usually observed in salmonids during hypoxia. It is suggested that the regulation of the carp erythrocytic pHi is probably caused to a major extent by deoxygenation of hemoglobin (Haldane effect) while adrenergic activa-