Humans, like other tropical animals, can tolerate high levels of heat stress (albeit for short periods of time) provided that (1) they avoid contact with extremely hot surfaces, (2) they can produce and evaporate sufficient sweat, and (3) the cardiovascular system that supports homeostasis under such conditions is not compromised. Indeed, during episodic severe heat waves that result in increased death rates especially among the elderly, cardiovascular disease is the primary or contributing cause of death in > 80% of examined cases (Smenza et al. 1996). Crandall et al. (2008) point out in this issue of The Journal of Physiology that ‘whole-body heating has the potential to cause significant cardiovascular stress that … may be second only to [that] associated with exercise’. Because of the high compliance of the cutaneous circulation and the lack of muscle pump function during severe and prolonged passive heating, the strain on the left ventricle – which provides all of the energy to essentially double cardiac output under such conditions – may be even greater than that seen during exercise. The response of the cardiovascular system to passive heat stress is an elegant example of homeostatic regulation and integration, worthy of being taught in every physiology class. In the classic experimental model of passive whole-body heating, skin temperature is elevated and clamped using water-perfused suits and core temperature rises as heat is transferred internally by convection and conduction. Cutaneous blood flow increases remarkably from basal values of 200 ml min−1 to upwards of 8 l min−1, yet arterial pressure is well maintained by a doubling of cardiac output (due to an increase in heart rate and maintenance of, or a slight elevation in, stroke volume despite falling central venous pressure) coupled with a redistribution of flow from renal and splanchnic vascular beds (Rowell et al. 1969). Elderly subjects, even those with no apparent cardiac dysfunction, respond to passive heating with an attenuated rise in skin blood flow coupled with both a lower cardiac output response and a diminished ability to redistribute flow from splanchnic and renal circulations, relying on a greater portion of their chronotropic reserve to compensate for a reduced inotropic response (Minson et al. 1998). A compromised left ventricle, e.g. due to heart disease or advanced ageing, may not be capable of meeting these demands in light of the requisite increase in cardiac work (i.e. the area of the flow–volume loop). While flow characteristics of the integrated cardiovascular response to heating are well understood, some controversy has existed about capacitance characteristics of the response. Muller (1905) first hypothesized a reduction in central blood volume (CBV) during passive heating, yet Rowell et al. (1969) noted a slight increase in CBV which he later described as a ‘pre-ventricular sump’ phenomenon. Crandall et al. (2008) approached this question using γ camera imaging of 99mTc-labelled autologous red blood cells. Their results demonstrated that thoracic blood volume, an index of CBV, was indeed decreased during supine passive heating. Further, blood volume was reduced in splanchnic circulations. These two findings support the hypothesized shift of blood volume from the central circulation to the skin to sustain heat dissipation. A second important finding from Crandall and colleagues (2008) is the 13% increase in left ventricular ejection fraction (EF) that occurs in concert with the elevation in heart rate in young heated subjects. The fact that stroke volume does not fall (and may increase slightly) despite falling central venous pressure presents a strong argument for increased cardiac contractility. While the authors concede that EF is an imperfect index of contractility, the link between the two is particularly robust under conditions of supine heating. In combination with the data of Rowell and Minson, the preponderance of cardiovascular morbidity and mortality (e.g. Smenza et al. 1996) during heat waves – the most closely related natural occurrence of sustained passive heating – is predictable. It is important to note that the subjects in this meritorious investigation were young, healthy and medication free. Given the preponderance of heat-related problems in the elderly and the otherwise cardiovascularly compromised, it remains uncertain whether their attenuated flow responses (Minson et al. 1998) are balanced by disparate volume characteristics. However, the present contribution of Crandall et al. lends additional support to the hypothesis that a large number of heat-related deaths among the elderly during heat waves may be attributable to increased strain on an already compromised left ventricle.