The presence and significance of Na+-induced Ca2+ release from rat liver mitochondria was investigated by the arsenazo technique. Under the experimental conditions used, the mitochondria, as expected, avidly extracted Ca2+ from the medium. However, when the uptake pathway was blocked with ruthenium red, only a small rate of 'basal' release of Ca2+ was seen (0.3 nmol Ca2+ X min-1 X mg-1), in marked contrast to earlier reports on a rapid loss of sequestered Ca2+ from rat liver mitochondria. The addition of Na+ in 'cytosolic' levels (20 mM) led to an increase in the release rate by about 1 nmol Ca2+ X min-1 X mg-1. This effect was specific for Na+. The significance of this Na+-induced Ca2+ release, in relation to the Ca2+ uptake mechanism, was investigated (in the absence of uptake inhibitors) by following the change in the extramitochondrial Ca2+ steady-state level (set point) induced by Na+. A five-fold increase in this level, from less than 0.2 microM to more than 1 microM, was induced by less than 20 mM Na+. The presence of K+ increased the sensitivity of the Ca2+ homeostat to Na+. The effect of Na+ on the extramitochondrial level was equally well observed in an K+/organic-anion buffer as in a sucrose buffer. Liver mitochondria incubated under these circumstances actively counteracted a Ca2+ or EGTA challenge by taking up or releasing Ca2+, so that the initial level, as well as the Na+-controlled level, was regained. It was concluded that liver mitochondria should be considered Na+-sensitive, that the capacity of the Na+-induced efflux pathway was of sufficient magnitude to enable it to influence the extramitochondrial Ca2+ level biochemically and probably also physiologically, and that the mitochondria have the potential to act as active, Na+-dependent regulators of extramitochondrial ('cytosolic') Ca2+. It is suggested that changes of cytosolic Na+ could be a mediator between certain hormonal signals (notably alpha 1-adrenergic) and changes in this extramitochondrial ('cytosolic') Ca2+ steady state level.