1. Whole-cell and sharp electrode recordings from adult rat dentate gyrus GCs were performed in the 400-microns-thick hippocampal slice preparation maintained at 34 +/- 1 degrees C. Intrinsic membrane properties of granule cells (GCs) were evaluated with the use of a switching current-clamp amplifier. 2. With the whole-cell technique, the average resting membrane potential (RMP) of GCs was -85 mV when a potassium gluconate electrode solution was used versus -74 mV measured with potassium acetate-filled sharp microelectrodes. The membrane voltage response to injected current was linear over two membrane potential ranges, greater than 10 mV hyperpolarized from RMP and between 10 mV more negative than RMP and -62 mV. The average input resistances (RN) calculated over these ranges were 107 and 228 M omega in the whole-cell recordings versus 37 and 54 M omega in the sharp electrode recordings. There was no correlation between RMP and RN with either recording technique. The membrane time constant (tau m) determined at the RMP was 26.9 ms for whole-cell recordings and 13.9 ms for sharp electrode recordings. 3. There was no evidence of time-dependent changes in RMP, RN, and tau m in whole-cell recordings, although the slow inward rectification seen at hyperpolarized potentials decreased over 30-60 min. Addition of calcium buffers to the whole-cell recording solution did not result in a significant change in the average RMP, the average RN, or the average tau m. 4. Action potential threshold was comparable in whole-cell (-49 mV) and sharp electrode (-52 mV) recordings, but action potential amplitude was larger in whole-cell (126 mV) than in sharp electrode (106 mV) recordings. Spike frequency adaptation was present in the whole-cell recordings and could be abolished by addition of calcium buffers to the electrode solution. 5. We estimated rho, the ratio of dendritic to somatic conductance, to be 5.1 for the whole-cell records and 2.1 for sharp electrode recordings. The electrotonic length of the equivalent cylinder representing the cell processes was estimated to be 0.49 from the whole-cell data and 0.79 from the sharp electrode recordings. This implies that at rest there is only a 10% decrement in steady-state membrane voltage along the length of the dendrite due to shunting across the membrane resistance; small synaptic events occurring in the distal dendritic tree will therefore have a more substantial influence on the soma than previous analyses suggested.(ABSTRACT TRUNCATED AT 400 WORDS)