1. Post-synaptic excitation produced in motoneurones of the isolated perfused frog spinal cord by different monosynaptic inputs and by ionophoretically applied glutamate was analysed with intracellular recording technique. 2. Ca2+-deficient, high Mg2+ (5--20 mM) media or addition of Mn2+ (2mM) or Co2+ (5 mM) reversibly abolished chemically mediated e.p.s.p.s derived from medullary reticular formation, ventral and lateral columns, but not the short-latency, rapidly rising e.p.s.p.s derived from dorsal roots or muscle nerves, suggesting electric coupling between some primary afferents and spinal motoneurones. This conclusion is consistent with the dynamic properties of dorsal root e.p.s.p.s, their small sensitivity to cooling, and with results of correction of intracellular records made for contribution of extracellular field potential. E.p.s.p.s evoked by ventral root stimulation were also insensitive to Ca2+-lack and presence of 5--10 mM-Mg2+. 3. As the post-synaptic membrane was made more negative the amplitude of electrotonic dorsal root e.p.s.p.s was increased, and it was decreased by depolarizing currents. No reversal of the early part of the electrotonic e.p.s.p. was observed, although the presence of the local response would account for the occasional reversal of its later phase seen with depolarization. 4. When hyperpolarizing and depolarizing currents were applied to motoneurones in which chemically mediated e.p.s.p.s of the reticular cells, the ventral and lateral columns, were evoked, the actual reversal of the early part of e.p.s.p. was not observed, and there was no correlation between the sensitivity of the e.p.s.p.s to injected currents and their time course. The positive values of the extrapolated reversal potentials and the effects of changes in ionic content of perfusing media suggest that synaptically released transmitter triggers off the Na permeability of the subsynaptic membrane. 5. The amplitude of depolarization produced by ionophoretically applied glutamate depends non-linearly on membrane potential and the curvature of this dependence differs from that seen with chemically mediated s.p.s.p.s. The asymptotic nature of this relationship is explicable by a dependence of the membrane conductance change upon the membrane voltage. 6. The results of conductance measurements during the glutamate induced depolarization, the values of apparent reversal potentials and their dependence on external Na+ and K+ and internal Cl- is explicable by the opening post-synaptic channel gates for Na+ and closing post-synaptic channel gates for K+. 7. Chemical and electrical transmission in the amphibian cord is discussed in relation to recent anatomical findings.