In clinical obstetrics, magnesium sulfate (MgSO(4)) use is widespread, but effects on brain development are unknown. Many agents that depress neuronal excitability increase developmental neuroapoptosis. In this study, we used dissociated cultures of rodent hippocampus to examine the effects of Mg(++) on excitability and survival. Mg(++)-induced caspase-3-associated cell loss at clinically relevant concentrations. Whole-cell patch-clamp techniques measured Mg(++) effects on action potential threshold, action potential peak amplitude, spike number and changes in resting membrane potential. Mg(++) depolarized action potential threshold, presumably from surface charge screening effects on voltage-gated sodium channels. Mg(++) also decreased the number of action potentials in response to fixed current injection without affecting action potential peak amplitude. Surprisingly, Mg(++) also depolarized neuronal resting potential in a concentration-dependent manner with a +5.2 mV shift at 10 mM. Voltage ramps suggested that Mg(++) blocked a potassium conductance contributing to the resting potential. In spite of this depolarizing effect of Mg(++), the net inhibitory effect of Mg(++) nearly completely silenced neuronal network activity measured with multielectrode array recordings. We conclude that although Mg(++) has complex effects on cellular excitability, the overall inhibitory influence of Mg(++) decreases neuronal survival. Taken together with recent in vivo evidence, our results suggest that caution may be warranted in the use of Mg(++) in clinical obstetrics and neonatology.