Intracellular sodium activity (aiNa), intracellular pH (pHi) and membrane potential were directly and continuously measured in sheep cardiac Purkinje fibers using neutral carrier liquid membrane ion-selective microelectrodes. Changing the superfusing medium from normal Tyrode's solution to the cardioplegic solution "HTK" according to Bretschneider (6) a depolarization from -73.7 +/- 7.2 mV to -55.0 +/- 9.5 mV (n = 25), a decrease of aiNa from 9.1 +/- 1.9 mM to 4.0 +/- 1.4 mM (n = 25) and an intracellular acidification from pHi 7.18 +/- 0.06 to pHi 7.01 +/- 0.06 (n = 11, mean +/- S.D.) occurred at 35 degrees C. The decrease of intracellular sodium activity was not effected by replacement of K, Mg, or histidine by mannitol in the cardioplegic solution. Addition of 4 mM Ca somewhat enhanced aiNa decline. Inhibition of the sodium pump with the cardiac steroid dihydroouabain (10(-4) M) lowered the effect of "HTK" on intracellular sodium by approximately 35% (n = 5). Sodium decline was also sensitive to equilibration temperature, giving a Q10 of 1.54 for the initial decrease velocity (temperature range 20 to 35 degrees C), which is less than that found by other investigators for pure sodium pump activity. It is suggested that although the electrochemical sodium gradient remains inward throughout, sodium may leave myocardial cells on induction of Bretschneider's cardioplegia because of a reduction of inward fluxes by simultaneously increasing sodium pump activity, thus increasing Na efflux. Na/Ca exchange is assumed to be of minor importance and the Na/H exchange may be involved. With respect to the clinical application of the low Na and nominally Ca-free cardioplegic solution "HTK" lowering of intracellular sodium activity is interpreted as a factor minimizing the risk of a "calcium paradox" on reperfusion with Ca at serum levels, as well as a possible mechanism preventing early development of cellular edema.