Electrodiagnostic assessments of the peripheral nervous system often require limb wan-ning. The traditional warming method, the hot water bath, has several disadvantages. We present an alternative: hot water blankets containing circulating water of a constant temperature. In addition to having practical advantages, these blankets allow for accurate and continuous monitoring of limb temperature and nerve conduction velocity (NCV) during warming. We therefore aimed to determine (1) the temperature dependence of the NCV of the median motor and sensory nerve, deep peroneal nerve, and sural nerve during warming with blankets, and (2) the optimal warming strategy for clinical purposes. We warmed the limbs of 10 subjects for 90 minutes with blankets set at a temperature of 41 degrees C. During warming, the NCV of the three nerves and distal motor latency of the two motor nerves were measured every 2 1/2 minutes. After 25 minutes of warming from an initial temperature of 28 degrees C, the NCV and distal motor latency of all nerves were within the normal range. However, after median nerve warming until a predefined temperature of 34 degrees C (i.e., for 17 minutes on average), two of nine subjects showed abnormal NCVs. We therefore conclude that for limb wanning with hot water blankets, a fixed 25-minute wanning period is optimal.