Human activities are increasing both the frequency of hypoxic episodes and the mean temperature of aquatic ecosystems, but few studies have considered the possibility that acclimation to one of these stressors could improve the ability to cope with the other stressor. Here, we used Atlantic killifish, Fundulus heteroclitus, to test this hypothesis. Hypoxia tolerance was measured as time to loss of equilibrium in hypoxia (LOE hyp ) at 0.4 kPa oxygen. Time to LOE hyp declined from 73.3±6.9 min at 15°C to 2.6±3.8 min at 23°C, and at 30°C no fish could withstand this level of hypoxia. Prior acclimation to warm temperatures significantly increased time to LOE hyp . Hypoxia tolerance of the southern subspecies of killifish, F. heteroclitus heteroclitus, was greater than that of the northern subspecies, F. heteroclitus macrolepidotus, measured both as critical oxygen tension (P crit ) and as time to LOE hyp . Warm acclimation offset the negative effects of temperature on time to LOE hyp to a similar extent in the two subspecies. Warm acclimation increased total lamellar surface area of the gill in both subspecies as a result of regression of an interlamellar cell mass (ILCM). However, differences in total lamellar surface area could not explain differences in time to LOE hyp between the subspecies, suggesting that the lower time to LOE hyp of northern fish is related to their higher routine metabolic rate. These data suggest that thermal plasticity in gill morphology can improve the capacity of this species to tolerate hypoxia, and shows how existing plasticity may help organisms to cope with the complex interacting stressors that they will encounter with increasing frequency as our climate changes.