1. A gradual and prolonged decrease of the response, termed here ''depression,'' evoked by repeated activation with transmembrane current stimuli was analyzed in rat CA1 hippocampal pyramidal cells under single-electrode current clamp by the use of the in vitro slice technique. 2. Depression was induced by 2-s duration 0.3- to 0.7-nA current pulses presented as a sequence of 12 stimuli at 3- to 60-s intervals. Sinusoidal currents (0.5-1.0 nA) at 5-Hz or 200-ms pulses repeated at 0.3-0.5/s, which may be more natural stimulations, also induced depression. 3. Depression outlasted stimulation up to 170 s in all cells tested. The initial high rate spike burst changed little (<20%), whereas the lower rate adapted response decreased markedly (>40%). Thus neurons increased their rate of adaptation. The afterhyperpolarizations following pulse-evoked responses increased in duration and amplitude with depression. There were input resistance (R(in)) reductions at depolarized membrane potentials and during pulses. However, R(in) reductions were considerably smaller or altogether absent late during interpulse intervals. Subthreshold current stimuli were ineffective, indicating that spike activity was necessary to elicit depression. 4. Depression was 1) insensitive to the toxin omega-Agatoxin-IVA (omega-Aga-IVA; 0.5 mu M), which blocked synaptic transmission, revealing a key involvement of intrinsic properties and little if any synaptic participation; 2) insensitive to 4-aminopyrydine (2.00-4.00 mM), which greatly enhanced excitatory and inhibitory synaptic efficacy, again suggesting little synaptic involvement and a principal postsynaptic participation, and no participation of the K+-mediated currents I-A and I-D; 3) abolished by carbamalcholine (5.0-20.0 mu M) - an effect blocked by atropine (1.0-10.0 mu M) - and reduced by Ca2+-free solutions, and by intracellular injection of the Ca2+ chelator 1,2-bis(2-aminophenoxy) ethane-N,N,N',N'-tetraacetic acid (BAPTA), suggesting that Ca2+-dependent K+-mediated currents are key factors, with a less important participation of the K+-mediated I-M current. 5. We conclude that depression was due to activity-dependent modifications in intrinsic properties, with little if any synaptic participation. Depression may be functionally significant because it was induced by potentially natural stimulations. A model is proposed that accounts for the main traits of depression. In the model, depression was induced by a gradual decline of the speed at which Ca2+ was buffered intracellularly; an increase in the IK(CaS) activation rate constant also simulated depression.