The passive Ca efflux from human red cell ghosts was studied in media of differing ion compositions and compared to the ATP-dependent Ca efflux. Cells were loaded with(45)Ca during reversible hemolysis, and the loss of radioactivity into the non-radioactive incubation medium was measured, usually for 3 hr at 37°C. Analysis of the efflux curves revealed that(45)Ca efflux followed the kinetics of a simple two-compartment system. In the concentration range between 0 and 1MM Ca in the external solution ([Ca(++)] o ), the rate constant of passive Ca efflux (k min(-1), fraction of(45)Ca lost per minute into the medium) increased from 0.00732 to 0.0150 min(-1). There was no further increase at higher [Ca(++)] o . The relation between the rate constant of Ca efflux and [Ca(++)] o is thus characterized by saturation kinetics. The passive transfer system for Ca could also be activated by Sr. The alkali metal ions Na, K and Li did not seem to have any significant influence on passive Ca transfer. The passive Ca efflux was slightly inhibited by Mg and strongly inhibited by Pb. Under most experimental conditions, a fraction of 15 to 50% of the intracellular Ca seemed to be "inexchangeable". The inexchangeable fraction decreased with increasing [Ca(++)] o and increased with increasing [Ca(++)] i . It was not influenced by alkali metal ions, CN or Pb, but it could be completely removed from the cells by the addition of 0.1MM Mersalyl to the incubation medium or by hemolysis with addition of a detergent. The active ATP-dependent Ca transport differed characteristically from passive transfer; the rate constant decreased with increasing [Ca(++)] o , and the inexchangeable Ca fraction increased with increasing [Ca(++)] o . The experimental results suggest that there exists a carrier-mediated Ca-Ca exchange diffusion in the erythrocyte membrane and that only a fraction of the ghost cell population participates in the Ca exchange diffusion.