When single-barrelled electrodes (5-60 Momega) were advanced under manual control from the mucosal side of the epithelium the mucosal membrane was on average indented by about 40 micron before the microelectrode pentrated the cell. Since this dimpling was comparable with the total depth of the cell, which recovered its original shape within 0.5 sec, the steady intracellular potential was recorded only about 14 micronm from the basal (serosal) membrane. Fast recording of the associated change in potential revealed an abrupt drop to -26 mV at a mean rate of 84 V/sec, followed by a further slow drop to a steady value of about -50 mV at a mean rate of 0.28 V/sec. The initial level of -26 mV may be regarded as the potential difference across the mucosal membrane. This conclusion was confirmed by mounting the microelectrode on a piezoelectric probe, which delivered 3 micron jabs in less than 0.5 msec. With this device in operation to prevent dimpling, the mean potential difference across the mucosal membrane was recorded as -29 mV. In all cases the potential across the basal membrane was recorded as -52 mV. Manual advance of the microelectrode tip within the cytoplasm yielded an intracellular potential gradient of 0.6 mV/micron. The same potential profile and membrane potentials were demonstrated on penetrating the epithelium from the serosal side, and measurements with multibarrelled electrodes whose tips were staggered in depth gave roughly the same internal potential gradient. The resistivity of the cytoplasm was determined by a triple-barrelled microelectrode, and varied from 10 times that of Necturus saline at the mucosal end of the cell to 4 times in the middle and 6 times at the serosal end.