Disruptions of glutamate- and Ca²⁺-dependent signal cascades are critical steps underlying “excitotoxic” cell damage in different forms of neurodegenerative disease, including Alzheimer disease, Parkinson disease, Huntington disease and amyotrophic lateral sclerosis (ALS). Recent investigations have provided increasing evidence that a highly interactive network of proteins and organelles precisely controls the spatio-temporal profile of [Ca²⁺]i and — accordingly — the activation pattern of “downstream” excitotoxic cascades. This network displays a substantial cell-to-cell variability, which might partially explain the profound heterogeneity in neuronal damage induced by a given insult. The cellular basis of this “selective neuronal vulnerability” has been studied in great detail in mouse models of ALS, which are characterized by a highly selective degeneration of motoneurons in the brain stem and spinal cord. Comparative studies between vulnerable and resistant cells demonstrated the critical importance of Ca²⁺ buffering, diffusion and uptake for neuronal vulnerability. As more detailed information becomes available, it might become feasible to develop novel neuroprotective strategies based on a targeted “stabilisation” of Ca²⁺ homeostasis in selectively vulnerable systems. The future potential of such strategies will be discussed.