1. Previous evidence for the existence of Ca2(+)-activated K+ channels in lymphocytes comes from measurements using voltage-sensitive dyes and from tracer flux studies. We have now directly measured these channels in human tonsillar B lymphocytes and rat thymocytes in single-channel recordings from cell-attached and excised patches. 2. In cell-attached recordings, intracellular Ca2+ was raised by either ionomycin or replacement of external Ca2+ following incubation in Ca2(+)-free medium. Indo-1 measurements during the Ca2(+)-replacement technique showed that [Ca2+]i rose from approximately 90 to 260 nM. Both techniques activated two channels of approximately 25 and 8 pS (slope conductance at 0 mV applied, with 140 mM-K+ in the pipette). Over 90% of patches displayed this activity, indicating a high density of these channels in the membrane. 3. Both channels reversed near the K+ equilibrium potential with either KCl or potassium aspartate in the pipette, when the cells were bathed in normal or high-K+ saline. Therefore, these channels are selective for K+. 4. The larger channel was studied in more detail. It displayed inward rectification in symmetrical K+ solutions. The open-channel probability was weakly dependent on membrane potential. 5. Ca2(+)-dependent K+ channels were also recorded from excised, inside-out membrane patches. The threshold for activation was 200-300 nM [Ca2+i]. 6. Patch excision altered some characteristics of IK(Ca). Channels were activated in fewer than 50% of patches and the main conductance level was approximately 34 pS (at -80 mV). The duration of single-channel events was shorter than in cell-attached patches; kinetic analysis suggested that this was due to the loss of an open state in excised patches. 7. We conclude that B and T lymphocytes have K(+)-selective channels which are activated by internal [Ca2+] in the physiological range and which will influence the membrane potential during cell activation.