The excitotoxic theory of stroke, which implicated N-methyl-d-aspartate (NMDA) receptors as mediators of excessive Ca2+ entry and neuronal death, generated a great deal of enthusiasm for the prospect of using NMDA receptor antagonists to prevent the associated brain injury. Unfortunately, these receptor antagonists failed to provide effective treatments for human stroke. In part, the failure is likely a consequence of having to administer these drugs within a very short therapeutic window after stroke and to the intolerable psychomimetic side effects associated with their use. However, new possibilities for therapeutic intervention are revealing themselves as our understanding of excitotoxicity evolves. We now recognize that ischemia and Ca2+ toxicity in central neurons can be attributed to a variety of mechanisms recruited downstream of NMDA receptor activation. These include the activation of Ca2+-permeable transient receptor potential channels of the melastatin family. The more-delayed activation of these channels offers the tantalizing possibility that drugs targeting selected members of this family may possess a wider therapeutic window for preventing the debilitating consequences after stroke onset.