Fig. 5. The effects of heat shock and anoxia on the thermosensitivity of action potentials recorded from locust forewing motoneurons. (A) Thermosensitivity of action potential latency in control (filled circles) (N=6), heat-shock (stippled circles) (N=9) and anoxia (open circles) (N=7) neurons. Latency in heat-shock neurons is significantly shorter than latency in control and anoxic neurons (two-way ANOVA with Tukey multiple pairwise comparisons, F=6.74, d.f.=2, P<0.05). No value for anoxia neurons at 45°C is shown because action potentials remaining at this temperature occurred spontaneously. (B) Thermosensitivity of action potential time to peak in control (filled circles) (N=6), heat-shock (stippled circles) (N=9) and anoxia (open circles) neurons (N=8). Time to peak in heat-shock neurons is significantly different from that of both control and anoxia neurons (two-way ANOVA with Tukey multiple pairwise comparisons, F=7.02, d.f.=2, P<0.05). (C) Thermosensitivity of action potential duration in control (filled circles) (N=6), heat-shock (stippled circles) (N=9) and anoxia (open circles) (N=9) neurons. Action potentials in anoxia neurons have significantly longer durations than those of both control and heat-shock neurons (two-way ANOVA with Tukey multiple pairwise comparisons, F=5.99, d.f.=2, P<0.05). (D) Thermosensitivity of action potential amplitude in control (filled circles) (N=6), heat-shock (stippled circles) (N=9) and anoxia (open circles) (N=6) neurons. Amplitude in heat-shock and anoxia neurons is significantly smaller than that in control neurons (two-way ANOVA with Tukey multiple pairwise comparisons, F=9.15, d.f.=2, P<0.05). Statistical tests include data only from temperatures between 22 and 35°C. At higher temperatures, the majority of control action potentials failed and, consequently, there were not enough data points for statistical comparisons at temperatures above 35°C. Values are means ± S.E.M.