1. Whole-cell patch-clamp recordings of excitatory postsynaptic currents (EPSCs) were made from guinea pig hippocampal CA1 pyramidal cells. The sensitivity of paired pulse facilitation (PPF) and EPSC variance to changes in synaptic transmission was investigated and the results were compared with the changes in these parameters evoked by long-term potentiation (LTP). 2. Presynaptic manipulations, such as activation of presynaptic gamma-aminobutyric acid-B receptors by baclofen, blockade of presynaptic adenosine receptors by theophylline, blockade of presynaptic potassium channels by cesium, and increasing the Ca(2+)-Mg2+ ratio in the external recording solution, each reliably changed PPF in a fashion reciprocal to the change in the EPSC amplitude. However, recruitment of additional synaptic release sites by increasing stimulus strength and antagonism of non-N-methyl-D-aspartate (NMDA) glutamate receptors by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) failed to alter PPF. 3. Presynaptic manipulations including increased stimulus strength gave the predicted changes in the value of mean 2/variance (M2/sigma 2). Moreover, postsynaptic manipulations that altered EPSC amplitude, including blockade of non-NMDA receptors by CNQX, or changing the holding potential of the postsynaptic cell, gave little change in M2/sigma 2, as would be predicted for manipulations resulting in a uniform postsynaptic change. 4. LTP caused no change in PPF, whereas the presynaptic manipulations, which caused a similar amount of potentiation to that induced by LTP, significantly decreased PPF. On the other hand, LTP did increase M2/sigma 2, although the increase was less than that predicted for a purely presynaptic mechanism.(ABSTRACT TRUNCATED AT 250 WORDS)