Long-term potentiation was studied in vivo in the rat barrel cortex. It was found that LTP lasting several hours could be induced in layer II/III by tetanic stimuli applied in layer IV. The probability of inducing LTP at a given site was high (86%) provided that the electrodes were not displaced too far horizontally. LTP was not observed if the stimulating electrode was located on the far side of the neighbouring barrel-column from the recording electrode. The strongest LTP was induced by stimulating layer IV septal locations or the edge of the barrel and recording in the near half of the neighbouring barrel. However, examples were found of LTP from layer IV to II/III within the same barrel, within the same septum and from barrel to adjacent septum. The probability of inducing LTP on a particular occasion was greatly increased by iontophoresis of bicuculline at the recording site during the tetanus (from 20 to 55% judged by a change in peak amplitude). The average increase in the peak amplitude was 29 +/- 3.2% for protocol 1 (urethane anesthesia, monopolar stimulation) and 23 +/- 7% for protocol 2 (barbiturate anesthesia, bipolar stimulation). The probability of inducing LTP was greater if the first tetanus was accompanied by BMI application (67%) than for any subsequent attempts (39%). These results suggest it should be possible to study the effect of LTP on sensory processing in defined positions within the barrel field.