Measurements of the K activity (aiK) in ferret ventricle were made using either single‐ or double‐barrelled K‐sensitive micro‐electrodes. aiK was also estimated from the Nernst equation by measuring the membrane potential when changing the external K concentration. Microelectrodes filled with the Corning K sensor gave unreliable results most probably due to interference from some substance(s) in the heart cells. Reproduceable measurements were obtained using a valinomycin cocktail as the K sensor. The mean value ± S.D. for aiK from four experiments which met strict criteria for calibration, electrode penetration and drift was 104 mmol/l ± 9 mmol/l. If all nine experiments were taken the value was 101 mmol/l ± 8 mmol/l. Estimation from the Nernst equation gave values of aiK that were on average 20 mmol/l higher than the measured values with the valinomycin cocktail. It is recommended that a valinomycin cocktail be used to measure aiK.

Cell pairs isolated from adult rat liver were used to study the electrical properties of gap junctions. Each cell of a cell pair was connected to a suction pipette so as to enable whole cell tight-seal recording. A double voltage-clamp approach was adopted to control the voltage gradient across the gap junction and measure the transjunctional current. The current-voltage relationship of the gap junctional membrane was linear over the voltage range tested (+/- 50 m V). Under control conditions, the resistance of the gap junction, rj, was 15 M omega (n = 27; range, 4.6 to 45.8 M omega), corresponding to a conductance gj of 67 nS. rj was insensitive to the nonjunctional membrane potential, Vm (voltage range,-90 m V to + 40 m V). There was no indication of a time-dependent gating of rj (time range, 20 ms to 10 s). Dialysis with 1 mM CaCl2 produced irreversible electrical uncoupling without affecting the linearity of the relationship Vj/Ij.