Hypothermia is a clinically effective therapeutic strategy to prevent neuronal injury during ischaemia. We hypothesise that glucopenic injury to white matter is also temperature sensitive. However as well as the injury mechanism, glycogen in the tissue acts in a neuroprotective manner during glucopenic episodes. In this study we have sought to differentiate between the temperature-dependent effects of glycogen and the temperature dependence of the injury mechanism using electrophysiological techniques to record the stimulus evoked compound action potential (CAP) from acutely isolated mouse optic nerve (MON). MONs were superfused with artificial cerebrospinal fluid (aCSF) in a recording chamber, maintained at 37°C unless otherwise stated, and constantly aerated with 95 % O2 / 5 % CO2. Data are presented as mean ± S.D. (n = 4) for all experiments. In MONs exposed to either 1 or 2 hours of glucopenia followed by a 1 hour recovery period in 10 mM glucose aCSF, the CAP area fell after 15.7 ± 2.9 minutes and was reduced to 46 ± 19 % or 17 ± 7 % of baseline, respectively, indicating a time-dependent component of the injury process. To investigate the temperature dependence of axonal injury, MONs were maintained at varying temperatures during a 1-hour glucopenic episode, followed by 1-hour recovery at 37°C in 10 mM glucose aCSF. Both the latency to CAP failure and the degree of recovery increased with decreasing temperatures (i.e. 42°C, 37°C, 32°C or 27°C); the latency to failure was 13.0 ± 4.3, 15.7 ± 2.9, 17.7 ± 3.8 or 19.9 ± 4.5 minutes, while the degree of recovery was 15.6 ± 5.3, 34.7 ± 3.8, 50.3 ± 9.9 and 73.8 ± 5.9 %, respectively. We have previously shown that the latency to failure during glucopenia is determined by the glycogen content of the tissue (Brown et al, 2003), thus we sought to determine the effect of temperature on the neuroprotective effects of glycogen. MONs were incubated for 2 hours in 2 mM glucose aCSF, a condition that we have previously demonstrated depleted glycogen to its nadir (Brown et al, 2003). Subsequent exposure to glucopenia at the range of temperature described above altered the latencies to CAP failure and extent of recovery. The latency to CAP failure at 42°C, 37°C, 32°C or 27°C was 5.3 ± 3.2, 8.0 ± 2.6, 9.4 ± 1.9 or 11.3 ± 1.0 minutes, respectively, and the degree of recovery was 0 ± 0, 13.2 ± 6.0, 37.8 ± 13.4 or 61 ± 8.9 %, respectively. A comparison of the two sets of data reveal that at all temperature tested the presence of glycogen increased both the latency to CAP failure and the recovery of the CAP after the glucopenic insult. These findings show that the neuroprotective effects of glycogen in the mouse optic nerve are temperature sensitive, and that temperature range tested the presence of glycogen improved recovery rates after a period of glucopenia. Experiments conform with Society ethical requirements.