Long-term changes of synaptic efficacy, in particular when they are use-dependent, are candidate mechanisms for the storage of information in the nervous system. In a variety of brain structures, including the neocortex and hippocampus, synapses are susceptible to long-term potentiation (LTP) and long-term depression (LTD). It has been hypothesized that the polarity of the synaptic gain change depends on the amplitude of the postsynaptic [Ca2+]i rise, the threshold for the induction of LTD being lower than that for the induction of LTP. To test this assumption, we characterized Ca2+ signals in layer II/III pyramidal cells of rat visual cortex slices, using the fluorescent Ca2+ indicator fura-2, during application of stimulation protocols that had been adjusted to reliably induce either LTP or LTD in cells not loaded with fura-2. At dendritic sites activated by the stimulated afferents the intracellular [Ca2+] concentration ([Ca2+]i) reached higher amplitudes and decayed more slowly with stimuli inducing LTP than with those inducing LTD. To directly analyse the functional significance of the observed difference in the Ca2+ signal amplitude, we examined whether a tetanization protocol suitable for the induction of LTP can be converted into a protocol inducing LTD by injecting the postsynaptic cells with Ca2+ chelators that reduce the concentration of effective free Ca2+. In the presence of fura-2 or BAPTA [bis(2-aminophenoxy) ethane-N,N,N',N'-tetraacetate], the stimulation protocol that would normally produce LTP induced either LTD or failed to induce synaptic modifications altogether. These results support the hypothesis that the amplitude of the postsynaptic rise in [Ca2+]i is a key factor in the determination of the polarity of synaptic gain change.