The patch-clamp technique was used to investigate the effect of intracellular Mg2+ (Mgi2+) on the conductance of the large-conductance, Ca(2+)-activated K+ channel in cultured rat skeletal muscle. Measurements of single-channel current amplitudes indicated that Mgi2+ decreased the K+ currents in a concentration-dependent manner. Increasing Mgi2+ from 0 to 5, 10, 20, and 50 mM decreased channel currents by 34%, 44%, 56%, and 73%, respectively, at +50 mV. The magnitude of the Mgi2+ block increased with depolarization. For membrane potentials of -50, +50, and +90 mV, 20 mM Mgi2+ reduced the currents 22%, 56%, and 70%, respectively. Mgi2+ did not change the reversal potential, indicating that Mg2+ does not permeate the channel. The magnitude of the Mgi2+ block decreased as the concentration of K+ was increased. At a membrane potential of +50 mv, 20 mM Mgi2+ reduced the currents 71%, 56%, and 25% for Ki+ of 75, 150, and 500 mM. These effects of Mgi2+, voltage, and K+ were totally reversible. Although the Woodhull blocking model could approximate the voltage and concentration effects of the Mgi2+ block (Kd approximately 30 mM with 150 mM symmetrical K+; electrical distance approximately 0.22 from the inner surface), the Woodhull model could not account for the effects of K+. Double reciprocal plots of 1/single channel current vs. 1/[K+] in the presence and absence of Mgi2+, indicated that the Mgi2+ block is consistent with apparent competitive inhibition between Mgi2+ and Ki+. Cai2+, Nii2+, and Sri2+ were found to have concentration- and voltage-dependent blocking effects similar, but not identical, to those of Mgi2+. These observations suggest the blocking by Mgi2+ of the large-conductance, Ca(2+)-activated K+ channel is mainly nonspecific, competitive with K+, and at least partially electrostatic in nature.