Vol.:(0123456789)1 3 European Journal of Applied Physiology (2024) 124:1–145 https://doi.org/10.1007/s00421-023-05276-3 INVITED REVIEW A century of exercise physiology: concepts that ignited the study of human thermoregulation. Part 3: Heat and cold tolerance during exercise Sean R. Notley1,5  · Duncan Mitchell2,3  · Nigel A. S. Taylor4 Received: 26 January 2023 / Accepted: 4 July 2023 / Published online: 5 October 2023 © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2023 Abstract In this third installment of our four-part historical series, we evaluate contributions that shaped our understanding of heat and cold stress during occupational and athletic pursuits. Our first topic concerns how we tolerate, and sometimes fail to tolerate, exercise-heat stress. By 1900, physical activity with clothing- and climate-induced evaporative impediments led to an extraordinarily high incidence of heat stroke within the military. Fortunately, deep-body temperatures > 40 °C were not always fatal. Thirty years later, water immersion and patient treatments mimicking sweat evaporation were found to be effec- tive, with the adage of cool first, transport later being adopted. We gradually acquired an understanding of thermoeffector function during heat storage, and learned about challenges to other regulatory mechanisms. In our second topic, we explore cold tolerance and intolerance. By the 1930s, hypothermia was known to reduce cutaneous circulation, particularly at the extremities, conserving body heat. Cold-induced vasodilatation hindered heat conservation, but it was protective. Increased metabolic heat production followed, driven by shivering and non-shivering thermogenesis, even during exercise and work. Physical endurance and shivering could both be compromised by hypoglycaemia. Later, treatments for hypothermia and cold injuries were refined, and the thermal after-drop was explained. In our final topic, we critique the numerous indices developed in attempts to numerically rate hot and cold stresses. The criteria for an effective thermal stress index were established by the 1930s. However, few indices satisfied those requirements, either then or now, and the surviving indices, including the unvalidated Wet-Bulb Globe-Thermometer index, do not fully predict thermal strain. Keywords Body temperature · Cold injuries · Exercise · Heat exchange · Heat illness · Hyperthermia · Hypothermia · Non- shivering thermogenesis · Sweating · Shivering thermogenesis · Vasomotor Abbreviations BAT Brown adipose tissue EHS Exertional heat stroke Emax Maximal attainable evaporation Ereq Required evaporation HSI Heat stress index IREQ Required clothing insulation IREQmin Minimal required clothing insulation IREQmax Maximal required clothing insulation IREQneutral Clothing insulation required for thermal comfort ISO International Standards Organization PHS Predicted Heat Strain UTCI Universal thermal climate index WBGT Wet-bulb globe thermometer Communicated by Michael I. Lindinger. * Nigel A. S. Taylor nigelastaylor@gmail.com 1 Defence Science and Technology Group, Department of Defence, Melbourne, Australia 2 Brain Function Research Group, School of Physiology, University of the Witwatersrand, Johannesburg, South Africa 3 School of Human Sciences, University of Western Australia, Crawley, Australia 4 Research Institute of Human Ecology, College of Human Ecology, Seoul National University, Seoul, Republic of Korea 5 School of Human Kinetics, University of Ottawa, Ottawa, Canada http://crossmark.crossref.org/dialog/?doi=10.1007/s00421-023-05276-3&domain=pdf http://orcid.org/0000-0002-5065-5000 http://orcid.org/0000-0001-8989-4773 http://orcid.org/0000-0002-3655-5249 2 European Journal of Applied Physiology (2024) 124:1–145 1 3 Introduction This review is Part 3 of four historical manuscripts that focus on the thermal physiology of most relevance to recreational and occupational athletes. Our use of those terms is deliber- ately inclusive and relates to adults of varying abilities who are exposed to thermally stressful conditions during work (e.g., emergency workers and military personnel) and exer- cise. In Part 1 (Notley et al. 2023a), we described the scien- tific steps that revealed the principles of thermodynamics which govern thermal energy exchanges. Those exchanges must obey physics. We also summarised the research that led to our contemporary understanding of body temperature regulation, by controlling physiological (thermoeffector) responses that produce, conserve or eliminate heat (Notley et al. 2023a [Figs. 20 and 21]). Armed with that informa- tion, we reviewed and critiqued the historical development of physiological measurements that were based upon those principles, and that withstood scientific scrutiny to become valid and reliable tools for measuring body temperatures and our thermoeffector responses (Notley et al. 2023b). We will now explore the physiological strain experi- enced by, as well as the external stresses imposed upon, the men and women who rest, work and compete in ther- mally challenging environments. We again focus on three epochs, as defined by the research of two Nobel laureates, August Krogh (1874–1949; Denmark) and Archibald V. Hill (1886–1977; England), from the century of exercise physi- ology: before 1900, the years 1900–1930 (the Krogh–Hill epoch) and the modern epoch (beyond 1930). We do not doubt that Krogh and Hill would feel comfortable being described as physiologists with passionate interests in exer- cise and body temperatures. In Part 1 of this series, we introduced the Serbian physiol- ogist Ivan Djaja (1884–1957 [Jean Giaja]), who established the first Institute of Physiology in the Balkans during the Krogh-Hill epoch (Andjus et al. 2011). We return again to Djaja, but now to one of his publications at the start of the modern epoch (Giaja 1938), and his schematic representa- tion of the zones of temperature regulation (Notley et al. 2023a [Fig. 21]). In this third part of our historical series, we will initially travel beyond the zone of thermoneutral- ity (“Température neutralité thermique”; thermoneutrality), through the “Zone de l’hyperthermie” (hyperthermia) and into the region of potentially lethal hyperthermia (“Tempé- rature de la mort par le chaud”). We will then leave the thermoneutral zone for a second time, and pass through the zone of hypothermia (“Zone de l’hypothermie”) to arrive at potentially lethal states of hypothermia (“Température de la mort par le froid”). Since recreational and occupational athletes, by definition, may spend hours beyond, and some- times well beyond, their comfort zones, they may experience levels of physiological strain that can have serious patho- physiological consequences. Whilst this review series was written to facilitate their independent reading, each of its four parts builds upon, and is linked to, its predecessor. Across the series, we have gone deep into the archival resources of over 25 countries, across five continents, with the aim of providing a significant col- lection of primary-source references and recommended read- ings (> 3800 across the series) that represent the foundations upon which our understanding of human thermal physiology during exercise sits. That work was undertaken by exercise and thermal physiologists, as well as by ergonomists, engi- neers, chemists and physicians, and we have woven their work into the geopolitical and technological climates of their times. It is our aim that those resources, along with the stories that lay behind that research, might be part of the antidote for scientific insularity. It is further hoped that readers might appreciate that, as sub-disciplines become progressively narrower in their foci, they often become less able to tackle complex physiological questions. That reality was emphasised by Nathan Zuntz (1847–1920) during the Krogh–Hill epoch (Gunga 2009). Beyond the summer stroll—accommodating, tolerating and suffering exercise in the heat Our journey beyond the summer stroll could be said to have begun in France, in the late eighteenth century, when the polymath, Antoine-Laurent Lavoisier (1743–1794), and his collaborator, Armand J.F. Séguin (1767–1835), discovered that oxygen consumption, and therefore metabolic heat production, increased during physical activity (Séguin and Lavoisier 1789; Candas and Libert 2022). Well before the construction of the First Law of Thermodynamics (Joule 1850), they observed that heat production was accompa- nied by increased evaporative cooling (Candas and Libert 2022), although it was not until 1845 that the retention of that metabolic heat was associated with an increase in deep- body (core) temperature. A Cornish physician, John Davy (1790–1868, England; Davy 1845), made that connection, although the elevations of deep-body temperature that he observed were too small to have any pathophysiological consequences, and he would not have any concept of the potentially fatal aftermath of very high body temperatures, though lethal heat stroke would be reported by other British doctors within 15 years (Gordon 1860). This summer journey will reveal the potentially fatal consequences of excessively elevated body temperatures, but winter strolls can also be perilous (“Beyond shiver- ing—accommodating, tolerating and suffering exercise in the cold”). The association of exercise with thermal strain will lead us to physiologists, physicians and engineers who 3European Journal of Applied Physiology (2024) 124:1–145 1 3 have explored human thermoregulation. Before setting off, we provide road maps (Figs. 1 and 2) with sign posts to high- light the main protagonists and their critical observations, discoveries and hypotheses. Since physiologists commenced their journeys on the heels of physicians, who were concerned with the patho- physiological consequences of excessive heat storage, then our journey also commences with those physicians, and their recommended treatments for hyperthermia that varied from dousing with water (Gordon 1860) to opium administration (Wrench 1868). Those physicians may have been the first to treat, but they were far from the first to encounter, heat ill- nesses, which are perhaps among the oldest known diseases (Osler 1893; Wakefield and Hall 1927). Heat illness in epoch one: treatments with water and opium Heat illness: killed by forced marches and military uniforms? There are ancient allusions to heat illnesses, with possibly the earliest report appearing in the Book of Judith (King James Bible; Chapter 8): “And her husband was Manasses, who died in the time of the barley harvest; for he was stand- ing over them that bound sheaves in the field; and the heat came upon his head, and he died in Bethulia” (Wakefield and Hall 1927 [P. 94]). Some important words were missing from that quote, which originally read: “heat came upon his head, and he fell on his bed, and he died” (Horowitz 2022 [P. 541]). Those missing words told us that deaths from heat illness are not necessarily immediate, but can occur several days later. Somewhat later, we discover that heat illness compro- mised the military campaign of the Roman Governor of Egypt, Gaius Aelius Gallus (26–24 BCE): “for the desert, the sun, and the water (which had some peculiar nature) all caused his men great distress, so that the larger part of the army perished. The malady proved to be unlike any of the common complaints but attacked the head and caused it to become parched, killing forthwith most of those who were attacked, but in the case of those who survived this stage it descended to the legs, skipping all the intervening parts of the body, and caused dire injury to them” (Jarcho 1967 [P. 767]). Had Gallus also provided the first record of rhabdomyolysis? Although it had been known from the time of Hippocrates that cold-water affusions might have helped those victims (Hippocrates 1817), cold water is not easily accessible to desert sojourners, even though it may well have been just below their feet. Heat casualties also afflicted the crusaders in the Middle Ages, who were unacclimatised, heavily armoured and exer- cising in Middle-Eastern weather. There were catastrophes further east too: “In Peking, in July 1743, 11,000 persons are said to have perished on the streets from the effects of heat” (Wakefield and Hall 1927 [P. 94]). At the end of the eighteenth century, that perfect storm also compromised Napoleon’s forces in Egypt (Steinman 1987). Bonaparte’s Chief Surgeon, and the originator of the flying ambulance and medical triage (Wood 2008), Dominique-Jean Larrey (1766–1842, France), described heat disorders that varied from ophthalmia to testicular atrophy (Commission des Sci- ences et Arts d’Egypte 1809), although his thermal contri- butions are best known for increasing our understanding of cold injuries and their treatment (“Beyond shivering: the descent into hypothermia”). It was another Frenchman, Gabriel Andral (1797–1876), who was perhaps the first per- son to formally associate high ambient temperatures with heat stroke (Andral 1843 [Pp. 113–116]). In the context of our current interests, heat illness in well-prepared work- ers and athletes, it is often the highly motivated who are at greatest risk (Sonna 2001). Manasses perished from the heat, and the sun caused the Roman soldiers much distress, but it was when our journey entered tropical climates that we encounter the fatal conse- quences of elevated body temperatures that were not neces- sarily caused by the sun. It was the British military physi- cians, during the colonisation of India, who observed and described those scenarios (e.g., Martin 1837). They lacked the support of a framework for understanding thermoregula- tory function, but by observing their successes and failures, we can see how knowledge accumulated, and we might also appreciate why some, such as Edward Wrench (1833–1912; England), might recommend strange treatments: “the sheet- anchor in such cases, if you can get hold of them before coma or convulsions have come on, is opium. I tried it over and over again and found it answer[s] in a most marvelous man- ner” (Wrench 1868 [P. 179–180]). In the following year, another English physician prescribed “large doses of quinine by the mouth, or hypodermically” (Waller 1869 [P. 132]). At that time, physicians could not distinguish between malaria and heat stroke. Fortunately, quinine is efficacious in patients with malaria and harmless in those with heat stroke. In the ninteenth century, the term applied to a suite of heat illnesses was heat apoplexy, and was coined by the military surgeon Charles Gordon (1820–1899, England and India; Gordon 1860). The French term was “coup de soleil”, and other terms include insolation, apoplexy of the hot winds, erythismus tropicus, heat stroke, coup de chaleur, Hitzschlag, sunstroke, Sonnenstich, heat exhaustion, heat prostration, solis ictus, thermic fever and thermoplegia (Gor- don 1860; Pembrey 1914; Casa et al. 2010a). In the nin- teenth century, no distinction was made between potentially lethal sunburn, which we now know to be a toxic condition caused by photochemical rays of the sun and unrelated to body temperature, and heat stroke, in which hyperthermia 4 European Journal of Applied Physiology (2024) 124:1–145 1 3 1860 1870 1880 1890 1900 1910 1920 1930 Sucquet (1862) Goltz et al. (1875) Luchsinger (1877) Ott (1879) Langley (1891) Williams and Arnold (1899) Blake and Scannell (1903) Ward (1910) Bardswell and Chapman (1911) Barbour (1912) Epoch one (<1900) Epoch two (1900-1930) Barbour (1913) Barcroft and Marshall (1923) MacKeith et al. (1923) Adolph (1924) Sundstroem (1927) First description of cutaneous arteriovenous anastomoses Sciatic nerve stimulation induced eccrine sweating Cholinergic agonists could stimulate profuse sweating Stimulation of the medulla oblongata could activate sweating Described the sympathetic chain and ganglia First records of sweat rates in marathon runners First records of deep-body temperature in marathon runners Cutaneous blood flow is elevated in the heat Optimal evaporation requires warm, dry air and continual airflow Body temperature elevation was proportional to exercise intensity Blood volume increases during exercise in the heat Cardiac output increased to match the elevation in cutaneous blood flow That redistribution could be counterproductive if protracted Cutaneous blood flow could be ~50 times greater in the heat Cardiac output is redistributed during exercise in the heat 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 Lewis and Pickering (1931) Lewis and Pickering (1932) Dale and Feldberg (1934) McSwiney (1934) List and Peet (1938) Asmussen and Bøje (1945) Spealman (1945) Barcroft and Edholm (1945) Adolph (1947) Rothstein et al. (1947) Bazett (1948) Haimovici (1948) Wyndham et al. (1953) Roddie and Shepherd (1956) Roddie and Shepherd (1957) Grande et al. (1959) Zotterman (1959) Landis (1976) Epoch three (>1930) Described active vasodilatation of cutaneous vessels Acetylcholine was the neurotransmitter for eccrine sweat glands Psychogenic stress can activate sweating Warming up can improve physical performance Hand blood flow can increase twenty-fold from cold to hot state Exercise limits can be associated with hypotension Deep-body temperature increases during exercise with progressive hypohydration First description of voluntary hypohydration Blood temperature varies substantially between deep-body and peripheral tissues Eccrine sweat glands respond to noradrenaline Body temperature during exercise is not independent of external heat Non-acral skin also has neurally mediated, active vasodilation Two types of sympathetic nerves control blood flow of non Deep-body temperatures increased linearly with hypohydration Human cutaneous blood vessels are under sympathetic control Thermoreceptors are distributed throughout the body Sweat is hypo-osmotic relative to blood Phenotypic conversion of sweat glands to cholinergic activation A B Fig. 1 Historical timelines: three epochs of discovery related to human heat stress during exercise. A Epochs one and two. The photo- graph of Isaac Ott was obtained and used under the Wikimedia Com- mons agreement (Public Domain). Source: https:// en. wikip edia. org/ wiki/ Isaac_ Ott#/ media/ File: Isaac_ Ott. jpg Accessed: December 30th, 2022. B Discoveries during the modern epoch https://en.wikipedia.org/wiki/Isaac_Ott#/media/File:Isaac_Ott.jpg https://en.wikipedia.org/wiki/Isaac_Ott#/media/File:Isaac_Ott.jpg 5European Journal of Applied Physiology (2024) 124:1–145 1 3 is the cardinal event (Mitchell and Laburn 1985). Thus, heat stroke and sunstroke were often used interchangeably (Lev- ick 1859), with that lack of differentiation is still evident within the contemporary glossary of thermal physiology (IUPS 2001). Gordon reported heat apoplexy in 0.5% of his regiment, with an appalling mortality rate of 88% (Gordon 1860). Waller (1869) reported a mortality rate of 51%. Gordon associated the condition in some regiments with exercise in the heat: “soldiers were most liable to the disease when, after a debauch of spirits over night, they were forced to march armed and accoutred during the heat of the succeed- ing day” (Gordon 1860 [P.989]). In less active regiments, he observed that “… it is among those who, from compul- sion or inclination, are much within doors during the intense heat of the day that it principally occurs” (Gordon 1860 [P. 986]). In 12 years in India, he saw no cases of heat apoplexy in women; their work was conducted inside. He identified the meteorological conditions with the highest risk to be oppressive days with cloudy skies and little wind. Soldiers were at higher risk when forced to march along narrow roads in dense brushwood than in open country, and he correctly concluded that “diminished evaporation from the skin in a most hot condition of atmosphere thus explains why we then suffer most (in health), and why hot climates are most unwholesome” (Gordon 1860 [P. 990]). However, Gordon incorrectly concluded that diminished evaporation imposed a threat because it failed to carry away toxins, and we revisit those ninteenth century European perspectives (e.g., Lind 1771; Balfour 1923) when discussing heat adaptation in the last of these historical communications. Nevertheless, Gordon described a method for the successful treatment of a heat-stroke patient, which has withstood the test of time: “two mussucks [skin bags] of water were immediately thrown over him” (Gordon 1860 [P. 993]). George Johnson (England; Johnson 1868), agreed with Gordon’s risk assessments, noting that muscular exertion and fatigue, hot clothing, alcohol and close over-crowded rooms, always with very high air temperatures, were associ- ated with heat illnesses. His explanation was that hot blood relaxed the pulmonary arteries, flooding the capillaries and causing pulmonary congestion. He proposed that “hot air and hot blood are the cause of this form of apnoea, so cold air and cold blood are the chief means of cure” (Johnson 1868 [P. 103]). His recommended treatment, which has also stood the test of time, was to lay the patient in the coolest place, with free-air current, wetting the entire body with cold water (without necessarily removing the clothes), giv- ing cold water by mouth or by injection, and ceasing only when the skin was cool and moist. William Strange (1816–1893, England) agreed about the circulatory problem, as did Wrench (1868), but not necessar- ily with its site being the lungs (Strange 1868). Strange also pointed out the vast inter-individual differences in symp- toms, possibly due to confusing sunburn and heat stroke, and the difficulty of distinguishing heat illness from the fever of tropical infections (Fayrer 1879, 1882). The British Surgeon- General, Joseph Fayrer (1824–1907), did not consider lethal sunburn and heat stroke as different conditions, but regarded them as the second and third stages of sunstroke, with the first stage being heat exhaustion and syncope (Fayrer 1879). Syncope was due to cardiovascular insufficiency (uncom- pensable hypotension) resulting from over-exertion in the heat and typically seen in furnace stokers on steam ships in the tropics. Other casualties were labourers, factory workers and farm workers. The second stage resulted from “exposure to the direct action of the sun’s rays when the atmospheric temperature is also high, and especially when unusual exer- tion is made” (Fayrer 1879 [P. 800]). Many thought that the sun’s rays directly affected the brain and spinal cord, and that misconception led to the wearing of pith helmets to cover the head and neck (Fig. 3A), which became a feature of many caricatures of the British Army; “Mad dogs and Englishmen go out in the midday sun”. The third stage of Fayrer’s sunstroke could occur with- out exposure to the sun (so-called heat apoplexy): “the most serious cases are those that come on under cover by night as well as by day, and apart from the direct solar rays … Heat alone, especially when the atmosphere is loaded with mois- ture so as to prevent evaporation from the person, is the real cause of the disease” (Fayrer 1879 [P. 301–302]). He later had doubts about heat alone being the cause, because the heat was accompanied by many other atmospheric changes (Fayrer 1884). He said that body (presumably sublingual) temperature could rise to 43 °C, the skin was hot and some- times dry, and delirium, coma and convulsions were com- mon, as were fatalities, with hundreds of deaths per year (Fayrer 1882). In India, those fatalities occurred predomi- nantly amongst recently arrived Europeans, but if the heat was severe enough, even the local residents would succumb (Fayrer, 1884). He reported that, at post mortem, “the blood is dark and grumous [clotted], often imperfectly coagulated, and effused in patches of ecchymosis [bruising], rendering the body rapidly livid” (Fayrer 1879 [P. 303]). We might associate his description with the coagulopathy of heat stroke. He recognised army clothing as an agent of heat ill- ness (Fig. 3B): casualties “were brought to me, and laid out in rows, perfectly unconscious, in their red coats and black leather stocks (they wore them, in those days, even in action under a tropical sun)” (Fayrer 1879 [P. 304]). In agreement with Johnson (1868), and as we would agree today, he noted that “heat being the primary cause of the disease, the object is to reduce temperature as speedily as possible and before tissue changes have been caused” (Fayrer 1879 [P. 305]). Heat waves were another phenomenon known to those whose summer stroll took them eastwards. G. Douglas 6 European Journal of Applied Physiology (2024) 124:1–145 1 3 A B 1800 1810 1830 1840 1850 1860 1870 1880 1890 1900 Larrey (1809) Martin (1837) Andral (1843) Gordon (1860) Gordon (1868) Johnson (1868) Fayrer (1879) Fayrer (1880) Fayrer (1881) Lagrange (1889) Withington (1895) Withington (1896) Phillips (1897) Weaver (1897) Sambon (1898) Weaver (1898) First clinical descriptions of heat disorders First description of heat illness in the tropics Associated high ambient temperatures with heat stroke Linked heat illness with reduced evaporative cooling Treat heat illness by dousing with water Linked exercise and heavy clothing with heat illnesses Treat heat illness by dousing with cold water and air movement Confirmed the links between heat illness, exercise and heavy clothing Linked heat exhaustion with syncope (cardiovascular insufficiency) Heat illness was due to reduced evaporative cooling Soldier deaths from heat illness are due to exercise, not the sun Described symptoms of endotoxaemia and hyponatraemia Survivable deep-body temperatures of 44.3°-46.1°C Body temperature did not necessarily increase during heat exhaustion Heat illnesses were linked to absolute (not relative) humidity Recommended cold bathing of casualties of heat illness Reported rebound hyperthermia following initial cooling Deep-body temperatures of 46°C were not always fatal Epoch one (<1900) 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 Mayer (1900) Pembrey (1902) Pembrey (1903) Pembrey (1914) Simpson (1914) Melville (1916) Pattison (1920) Willcox (1920) Levine et al. (1924) Mavrogordato and Pirow (1927) Ferris et al. (1938) Barcroft and Edholm (1945) Buskirk and Beetham (1960) Pugh et al. (1967) Hart (1982) Epoch three (>1930) Sweat removes more water than solutes, and can lead to hypertonic hypovolaemia Heat stroke was linked to heavy military clothing Recommended wetting and fanning heat-stroke casualties First psychrometric chart for environments during exercise Recommended drinking requirements for soldiers Recommended rubbing ice on the skin of heat-illness casualties Exertional heat stroke does not necessarily impair sweating Fatigue in marathon runners associated with hypoglycaemia Recommended heat adapting workers to reduce heat illness Ice-water immersion to treat heat-illness casualties Exercise limits can be associated with hypotension Post-race, body-mass deficits of 6% in marathon runners Post-race, rectal temperature >41°C in a marathon runner Advised against water-restriction practices Classical heat-stroke patient survived with a deep-body temperature of 47.0°C Epoch two (1900-1930) 7European Journal of Applied Physiology (2024) 124:1–145 1 3 Hunter (England) recorded the consequences of the heat wave of June 1886 (Egypt; minimum nightly 31 °C, maxi- mum daily 50 °C [dry bulb]). Of the 2469 men in the army garrison, 48 developed heat stroke and 25 died; most were under the influence of alcohol (Hunter 1887). He railed against the bleeding of casualties, which accelerated death in heat-stroke casualties. Some of his intoxicated casualties may have preferred that outcome to his treatment: immedi- ately strip the patient, lie him in the coolest place possible, splash cold water on the head and spine, and give a large cold-water enema (Hunter 1887). Theo Hyslop (1863–1933, England) reported 55 cases of insanity following sunstroke, 23 of whom were Europe- ans in India (Hyslop 1890), and he believed sunstroke and heat stroke to have different causes, but the same symptoms. Hyslop suggested sunstroke was due to the rays of the sun, which affected “the vasomotor centre in the medulla oblon- gata, especially by striking on the unguarded occiput and neck” (Hyslop 1890 [P. 495]), whilst heat stroke was caused by hot air and the “heat of the body produced by exercise which is not attended by perspiration” (Hyslop 1890 [P. 494]). However, at the end of the ninteenth century, Louis (Luigi) W. Sambon (1867–1931, England) caused an upset and damaged his career, by contesting that sunstroke had anything to do with the action of the sun or heat on the body (Sambon 1898a). He also challenged the view that Euro- peans could not tolerate or adapt to hot-tropical climates (Sambon 1898b). Sambon was born in England, trained in London and Naples (Italy), began his medical career in Rome and then worked in England as a tropical medicine specialist (Hold- royd 1931). He classified heat illnesses on the basis of deep- body temperature changes. If the body temperature did not rise, the condition was classified as heat exhaustion, which was common in soldiers forced to march in close order with heavy clothing and equipment. He considered that term to be a misnomer since, if body temperature did not rise, the condition could not have been caused by heat; we return to this concept in “Warmed to the work”. For conditions in which deep-body temperature did rise, he resurrected the ancient name of siriasis. The “ancients called it siriasis because of its prevalence during the hottest season, when Sirius, the dog star, rises and sets with the sun” (Sambon 1898a [P. 748]). Siriasis was also characterised by coma, pulmonary congestion, and high mortality, and dark blood and venous engorgement at post mortem. It did not always occur on the hottest days, but high water vapour content imposed a serious risk. Sambon concluded that siriasis had the characteristics of an infectious disease prevalent in India, but absent in Europe, which he incorrectly attributed to a soil microorganism that flourished when the air was wet. Sambon’s paper (1898a) solicited a flurry of rejoinders. For example, O’Grady (1898; Royal Navy, England) reported a fatal case of heat stroke in a ship’s stoker in European waters. McCartie (1899a, b) dismissed arguments about the hottest days, pointing out that heat stress also depended on wind speed and humidity. He also identified long forced marches and the tight army uniforms as causal agents of heat illness, noting that “every workman in the world strips to the waist except the soldier, who is trussed and swathed up as if his life depended on it” (McCartie 1899a [P. 191]). Across the Atlantic, Theodore Deecke (1866–1882, U.S.A.) reviewed cases of fatal heat stroke (Deecke 1883), and reported deep-body temperatures of up to 44 °C, with regular observations of dark blood post mortem. Much higher deep-body temperatures (46 °C), which were not always fatal, were reported by Weaver (1897), who reviewed 2038 U.S. cases of fatal sunstroke. He diagnosed death to be due to the absorption of toxic products through a stomach wall, which had been damaged due to the loss of sodium chloride. Those toxins needed to be made innocuous, and “Nature attempts to do it by burning it up with a high fever” (Weaver 1897 [P. 621]). Weaver mistakenly attrib- uted the reduced sodium chloride concentration to sweat- ing, which actually induces an hypertonic hypohydration, but he correctly identified drinking too much water as a potential pathophysiological problem. We now know that to be hyponatraemia (“How should the fluid lost in sweat be replaced?”), but did he also describe endotoxaemia (“Have early ideas about exertional heat stroke been supported?”)? William Phillips (1863–1935, U.S.A.) reviewed those same fatalities, finding that 72% had occurred in low-lying western cities, and he gave readers a lesson in the first prin- ciples and the psychrometry of heat stroke: “the action of heat is much influenced by the hygrometric condition of the atmosphere. A dry, hot air, it is claimed, is better tolerated than a moist one at a lower temperature, because it favors perspiration, and thereby keeps the body cool, while damp air diminishes evaporation and the refrigerating processes of the body” (Phillips 1897 [P. 225]). He correctly identified that the risk factor associated with damp air was absolute humidity (proportional to water vapour pressure), not rela- tive humidity. Charles Withington (1862–1917, U.S.A.) analysed 100 cases of heat prostration (1882–1894; Withington 1895), which included 28 fatalities, 20 of whom died unconscious, with 11 of those surviving > 24 h. All casualties presenting with rectal temperatures > 43.3 °C died, whilst all those with Fig. 2 Historical timelines: three epochs of discovery related to heat illness during exercise. A Epoch one. The portrait of Dominique- Jean Larrey was painted by Larrey’s sister-in-law, Marie-Guillemine Benoist (1804). The work was obtained and used under the Wikime- dia Commons agreement (Public Domain). Source: https:// commo ns. wikim edia. org/ wiki/ File: Portr ait_ of_ Baron_ Larrey_ by_ Marie- Guill emine_ Benoi st. jpg Accessed: December 30th, 2022. B Discoveries during the Krogh-Hill and modern epochs ◂ https://commons.wikimedia.org/wiki/File:Portrait_of_Baron_Larrey_by_Marie-Guillemine_Benoist.jpg https://commons.wikimedia.org/wiki/File:Portrait_of_Baron_Larrey_by_Marie-Guillemine_Benoist.jpg https://commons.wikimedia.org/wiki/File:Portrait_of_Baron_Larrey_by_Marie-Guillemine_Benoist.jpg 8 European Journal of Applied Physiology (2024) 124:1–145 1 3 temperatures < 39.4 °C survived. Some casualties reported a cessation of sweating: “in one such case a longshoreman while “tending hatch” on deck of a vessel at 3 P. M., had been sweating profusely. This suddenly stopped and he “felt cold”, then lost consciousness” (Withington 1895 [P. 513]). Treatment consisted of cold and ice-cold water bath- ing, sometimes with ice rubbing. Withington reported the consequences of over-cooling: “sometimes the chilling of the surface drives the blood in upon the viscera, and collapse occurs” (Withington 1895 [P. 513]). He noticed, but did not attempt to explain, rebound hyperthermia, sometimes back to the temperature at admission, which could occur follow- ing bouts of cooling. By the end of the ninteenth century, cold water and ice were established firmly as treatments for exertional heat stroke (Costrini 1990). Insights and blind spots about thermoregulation during exercise We introduced T. Clifford Allbutt (1836–1925; England), the inventor of the first practical, clinical thermometer, in our previous communication (Notley et al. 2023b). He was also interested in the physiology of thermoregulation during healthy exercise, and, when contemplating what determined human body temperature, he noted that “the effects of hard and prolonged exercise upon the body had not been tested, and this seemed a serious omission” (Allbutt 1872 [P. 106]). Allbutt, along with Séguin and Lavoisier (1789), would have been dumbfounded by the ideas of Austin Flint (1812–1886, U.S.A.), who was soon to become President of the American Medical Association. Flint rejected the idea that physiology had to conform to the laws of physics: “our physiological facts, if definite and well established, should be treated as facts not to be distorted into arguments in favour of laws which we have enacted rather than discovered” (Flint 1877 [P. 95]). One of those laws was the First Law of Thermo- dynamics: “If a man take a certain amount of food, and should do no work except that required to maintain circula- tion respiration, and assimilation, there is no evidence that the force “locked up” in the food is evolved in the form of heat” (Flint 1877 [P. 94]). Flint also denied that exer- cise could elicit hyperthermia; “… we have all been wrong often” (Hill 1965 [P. 167]). The views of Flint (1877) were not shared by Fernand Lagrange (1845–1909, France), whom we would recognise as an exercise physiologist. He said that “if to the action of high temperature there is added that of muscular work, the organism has not to fight against the heat of the sur- rounding medium, it has also to defend itself against the increased heat developed in its own organs” (Lagrange 1889 [P. 151]). He had no doubt that “excessive temperature is a consequence of excessive vital combustion” (P. 152) or that “a man who succumbs during a forced march, under the hot sun, is not killed by the sun, but by the forced march” (Lagrange 1889 [P. 151]). That exercise caused hyperthermia was the unqualified conclusion of two British thermal physiologists, who pro- vided further insights about thermoregulation during exer- cise. The first was William Hale White (1857–1949 [hyphen- ated after 1900]; Milton 2022). He developed a gravimetric method for measuring local sweat rates (Hale White 1897), which was possibly the forerunner of the absorbent-patch (pouch) method (Notley et al. 2023b). He had the physi- ological insight to say that if the skin temperature, deep- body temperature and sweat rate all increase within a fixed environment, then metabolic heat production must also have Fig. 3 A An Indian shola-style pith helmet, worn to protect the head and back of the neck from direct solar radiation. Creative Commons Attribution 2.0. B A British officer in the uniform of the 74th High- land Regiment. The tight woollen clothing covering almost all of the body impeded evaporation and did not allow air circulation near the skin. John Opie, Wikimedia Commons and in the Public Domain. C The cotton uniform and felt hat worn by Robert Baden Powell, which replaced the redcoat, woollen uniform and pith helmet. Painting by John Edward Chapman: Baden-Powell in the Uniform of the South African Constabulary, Standing by His Charger. The Scout Associa- tion: Creative Commons CC0 1.0 Universal Public Domain Dedica- tion 9European Journal of Applied Physiology (2024) 124:1–145 1 3 increased. The second physiologist was Marcus S. Pembrey (1868–1934; Douglas 1935; Milton 2022). He and his stu- dent showed that “muscular work causes a marked rise in the internal temperature” (Pembrey and Nicol 1898 [P. 392]). By the end of the ninteenth century many, but certainly not all, would have agreed with Pembrey and Nicol (1898) that a hallmark of thermoregulation during exercise was an elevated deep-body temperature. That elevation was exac- erbated by suppression of evaporative cooling, either by ambient conditions (high water-vapour pressure, low wind speed) or by clothing, and it could lead to fatal heat stroke. An important cautionary message emerged concerning the ventilatory artefacts accompanying oral thermometry dur- ing exercise, although it took some time for that message to be realised across the Atlantic. In perhaps the first deep- body (sublingual) temperature measurements taken before and after running a marathon (Boston Marathon), Harold Williams and Horace Arnold (1899, U.S.A.) arrived at the incorrect conclusion that deep-body temperature actually decreased over the course of the race (air temperature 7 °C). They also measured body-mass changes, with 10 of the 17 runners losing an average of 1.6 kg. Whilst that paper offers little to modern physiologists, it sets a marker close to the commencement of applied research in exercise physiology (Maron and Horvath 1978). Both sublingual and rectal temperatures were reported for the next three Boston Marathons (1900–1902; Blake and Larrabee 1903), from which the authors concluded that “mouth temperature is not a reliable factor” (P. 197). Five runners had post-race, rectal temperatures of 40 °C in 1900 (2.5 kg body-mass loss), whilst the mean temperatures for all runners studied in 1901 were remarkably low (38.4 °C; 1.8 kg mass loss). Such low temperatures imply a slow pace; the authors were not to know how high deep-body tempera- tures would be in the modern epoch, nor how much sweat could be lost. Indeed, water restriction was often considered a sign of fortitude, and the Oxford University rowing crew of 1860 were restricted to two pints a day during training (Noakes 1995), although other fluids appear not to have been restricted. During marathon racing, “contestants may and probably did drink certain amounts of brandy while running” (Blake and Scannell 1903 [P. 198]). If for no reason other than it compromises thermoregulation and increases the risk of exertional heat stroke (Yeo 2004; Yoda et al. 2005), that practice would be unthinkable today, yet competitors often consumed alcohol in early marathons, and wine was availa- ble along the route of the 1924 Paris Olympic Games (Peiser and Reilly 2004). In a curious twist to this story, Levine et al. (1924) would soon describe fatigue-related hypoglycaemia in runners at the end of the Boston marathon (“Fuelling the fires: the competition for blood glucose”;) a state that has an adverse impact on thermoregulation in the cold and which is tightly linked to alcohol consumption. In spite of their impressive insights at that time, the nin- teenth-century physiologists who recognised that exercise elevated deep-body temperatures, also considered hyper- thermia to reflect a failure of thermoregulation: “the resist- ing power of the heat regulator fails or becomes exhausted and the man’s temperature rises to pyrexia or hyperpyrexia” (McCartie 1899b [P. 436]). It was only well into the twenti- eth century that the connection was made between elevated body temperatures, thermoeffector activation and heat bal- ance during exercise. The contemporary term, “regulatory failure”, does not imply regulatory collapse, but a reduced ability to defend a stable body temperature, regardless of its level. Epoch two: exercise hyperthermia, friend, foe or fiction? Heat stroke in the Krogh‑Hill epoch: a military perspective In the Krogh-Hill epoch, research concerning thermoregula- tion during exercise in the heat continued to be dominated by an emphasis on heat illness, and particularly when the British military arena of interest shifted from India to Meso- potamia and the Middle East. According to Alexander (Eng- land), “heat waves (locally known as “date ripeners”) occur, when the temperature keeps up for about a fortnight at 124 to 128 °F. (51–53 °C), and it is during these distressing periods that most of the heatstroke cases occur” (Alexander 1922 [P. 358]). Unfortunately, many protagonists of applied research at that time showed only a scant acquaintance with physiology. In a different class, however, were the contributions of physiologists such as Marcus Pembrey and Joseph Barcroft. In an observation that will soon ring an iden- tical bell for cold injuries (“Beyond shivering—accom- modating, tolerating and suffering exercise in the cold”), Pembrey observed that “in many an army heat-stroke has been more fatal than the bullets of the enemy” (Pembrey 1902 [P. 261]). Pembrey dismissed the pronouncement that the blame for heat stroke lay with the chosen lifestyle of the soldiers. He had no doubt that exercise, even without ambient heat, could elevate deep-body temperature “as high as 101 F (38.3 C)” (Pembrey 1902 [P. 263]), but he did not consider that elevation as pathological. Indeed, he thought it was beneficial to skeletal muscle activation. Joseph Barcroft (1872–1947, England), and his student showed that elevated temperatures shifted the oxygen hae- moglobin dissociation curve to the right, facilitating the offloading of oxygen within exercising muscles (Barcroft and King 1909); some degree of body-temperature eleva- tion was indeed advantageous. 10 European Journal of Applied Physiology (2024) 124:1–145 1 3 Pembrey (1902) doubted that the exercise-induced ele- vation of body temperature would attain heat-stroke levels during exercise, even on a hot day, if the air was dry. What caused heat stroke was exercise in the heat when the water- vapour content (absolute humidity) was high, which inhib- ited evaporative cooling. Farm labourers seldom developed heat stroke even if they worked all day in the sun, so he laid blame squarely on the army: “the soldier on the march and under the order and care of his officers is clothed in open defiance of common sense and physiological principles” (Pembrey 1902 [P. 265]). He recognised that coping with exercise in the heat required increased cardiac output and coordinated thermoeffector activation. Whilst his expertise far exceeded that of army doctors, neither he nor his con- temporaries understood thermoregulatory function properly. Nonetheless, Pembrey (1902) dismissed the contemporary belief that drinking cold water during hyperthermia could be fatal; soldiers died from what made them drink, but not because they drank. His recommended treatment was wet- ting heat-stroke casualties with water and creating an airflow over them: “under no other treatment do so many patients recover” (Pembrey 1902 [P. 269]). He also recommended: “light clothing, light loads, open order, a proper supply of water, and training on hot as well as on cold days” (Pembrey 1902 [P. 269]). Pembrey credited the British Army with the develop- ment of methods for cooling heat-stroke casualties (Pembrey 1914), but he used the words of the military surgeon, James R. Martin (1796–1874, England), to chastise the military for its regulations: “Parades, formalities, the majestic Eng- lish march, “Regulations” and appearances, must here be utterly and at once discarded; for it is a question of life and death. The open, disorderly-looking order of march, however slovenly it may seem to the lieutenant-colonel, must here be used, the close order being nothing short of stifling and sickening the men “by Regulation”” (Martin 1861 [P. 409]). Pembrey knew that sweating was under neural control, but he thought it was influenced by the temperature and com- position of the blood. He would find allies among a small group of thermal physiologists, but not the majority. In his view, the brain temperature could be lowered selectively by external cooling: “the application of ice to the head and neck seems to give prompt results owing probably to the fact that the high temperature especially affects the brain” (Pembrey 1914 [P. 634]). Pembrey also advised against the practice of restricting drinking and was clearly contradicting the contemporary dogma, which advocated water restriction. For example, James Sullivan, an organiser of the 1904 Summer Olympics in St. Louis (U.S.A.), pronounced: “don’t get in the habit of eating or drinking in the Marathon race; some prominent runners do, but it is not beneficial” (Sullivan 1909; cited by Noakes 1995 [P. 124]). Sullivan also campaigned against women participating in the Olympic Games. Leonard Rogers (1908, England) evaluated the hypoth- esis that heat stroke had a microbial origin (Sambon 1898a, b). He studied 425 fatal cases, and correlated those data with changes in the weather (Bengal Meteorological Office, India), finding that he could explain the heat-stroke inci- dence “on purely physiological grounds, without the assis- tance of a hypothetical microbe” (Rogers 1908 [P. 30]), and asserted “there is a most intimate relationship between heat- waves, the degree of moisture in the air … and the preva- lence of heat-stroke; and all the facts are readily explain- able on the hypothesis that the hyperpyrexia is produced by a failure of the cooling mechanism of the body during exposure to great heat, especially if accompanied by much moisture in the air and of prolonged duration” (Rogers 1908 [P. 32]). Rogers was not alone (Buchanan 1900; Duncan 1908), although Milner attributed the cessation of sweat- ing, and the resultant heat stroke, to a “paralysis of the heat-regulating centre, due to intoxication by the parasite of malignant tertian malaria” (Milner 1918 [P. 639]). He claimed that quinine reduced mortality substantially. Inhibi- tion of the heat-loss effectors is indeed one consequence of pyrogens, but Leonard Hill (1866–1952, England; Douglas 1953) pointed out that many heat-stroke cases occurred in malaria-free troops who recently had arrived in Mesopota- mia from England (Hill 1920). In a debate that continued within the modern era, Pem- brey (1914) believed that dry skin was a hallmark of heat stroke and that it allowed one to distinguish heat stroke from heat exhaustion. Similarly, Hearne (1919, 1920) claimed that a cessation of sweating occurred 1–48 h before other symp- toms of heat stroke, and urged that the temperature of anyone with dry skin be monitored. He contended that he had not seen a heat-stroke casualty with a deep-body temperature of 43.3 °C, or higher, who still was sweating. Hill (1920) pointed out that it was not the sympathetic nervous system that failed in heat stroke, but the sweating system. The view that sweating always failed in heat stroke was challenged by Willcox (1920a; London), who noted that: “suppression of sweating did not, in my experience, always precede heat hyperpyrexia, and though it is undoubtedly an important predisposing cause it cannot be regarded as the primary one” (Willcox 1920a [P. 648]). British Army physicians working in the Middle East dis- covered differences in the prevalence of heat illnesses when compared to India. For example, Willcox (1920b) recorded that, for equally high ambient temperatures (July–August, 1917), there were 3156 heat-stroke cases with 458 deaths (15%), whilst in India, there were just 601 cases, with 62 deaths (10%), and it was not just soldiers who suffered. In 1916, the co-inventor of the Horsley-Clarke stereotactic apparatus that enabled many advances in neural and thermal 11European Journal of Applied Physiology (2024) 124:1–145 1 3 physiology (Notley et al. 2023a), Victor Horsley (Semon, 1916), was a volunteer surgeon in Mesopotamia. That role required walking long distances daily, with shade tempera- tures often exceeding 43.3 °C: he became ill, was admitted to hospital, lost consciousness and died (MacNalty 1957). Whatever the cause of his and other heat-stroke deaths, it was not the direct effect of the sun on the brain and spinal cord (Alexander 1922); “the Eastern sun is not, as so many people think, a strange avenging deity with totally different powers from those met with in England” (Finny 1918 [P. 369]). Army physicians knew that recovery from heat stroke required prompt cooling (Turner 1917), and reducing the period of unconsciousness seemed to be associated with a better chance of recovery (Rogers 1908), but how can that best be achieved? Some used evaporative cooling, with water thrown over the casualty (dousing) and increased air move- ment (ventilation). One physician used a water-filled trench to successfully treat 40 heat-stroke cases (McIntosh 1920). The Persian Oil Company built a cooled, anti-heat stroke ward (Rennie 1930). Others insisted that rubbing the body surface with ice was the best option (Withington 1895; Pat- tison 1920; Hehir 1922; Wakefield and Hall 1927). Not only did ice rubbing result in subsequent rebound hyperthermia (Gauss and Meyer 1917; Hehir 1922), it was irrational on physical grounds. The latent heat of fusion of water, the discovery of which Tunbridge (1971) attributed Joseph Black in 1761 (Notley et al. 2023a), is ~ 334 J.g–1, but its vaporisation would dissipate seven times more heat (2426 J.g–1). Nonetheless, ice rubbing was not as irrational as another popular cooling method; ice-water enemata (Beamish 1907; Howard 1920). According to Hill (1920), 70 g of water evaporated from the skin took away as much heat as a 1-kg, ice-water enema. Ice-water enemata also prevented measuring rectal temperatures and sometimes had pathological consequences. In one patient who died, “post-mortem the rectum was found to be partly gangrenous, the pathology being probably identical with the gangrene following frostbite, and due to prolonged exposure of the mucous membrane to a very low temperature” (Hull and Reed 1907 [P. 341]). Ice water was not the only noxious enema. Some used turpentine (Duncan 1903) and others brandy (Henderson 1902), but the physiology underlying those noxious treatments remains a mystery. The British army was also engaged in the South Afri- can War (the Anglo-Boer War; 1899–1902); to the Boers, it was the second war of independence. Heat illnesses again occurred, although cases were remarkably rare. Hospital admissions were 1625 (1899–1902), or 2.7 per 1000. Only 15 deaths were recorded; case mortality < 0.1% (Simpson 1909). Protagonists of the malaria hypothesis might have attributed that low mortality to an absence of the parasite over the combat zone. Thermal physiologists might prefer the interpretation of Simpson (England): “the service dress in South Africa could hardly have been improved on; it was eminently suited to the climate. At the beginning of the war there was a tendency to too close fitting, and the helmet was generally worn, but these two faults were eliminated very early; the felt hat in particular was far better suited to the climate than the helmet” (Simpson 1909 [P. 260]; Fig. 3C). If 15 British soldiers died of heat illness in that war, and heat-related deaths in India and the Middle East ran into the thousands, why did just four heat-stroke deaths in South African gold mines (1925–1927; Village Deep mine, Johannesburg) generate so much attention (Mavrogordato and Pirow 1927)? In the first instance, heat illnesses reduced productivity, but the real significance was that those deaths changed the way that occupational heat illnesses were reduced and prevented, using regimented artificial heat- adaptation (acclimation) procedures. Indeed, at the end of the Krogh-Hill epoch, there was a paradigm shift, and almost the entire underground workforce of South Africa’s gold mining industry was heat acclimated under the supervi- sion of thermal physiologists. That programme was hailed by Jim Hardy (U.S.A.) as the greatest achievement in applied physiology (Mitchell and Laburn 2022). Mavrogordato and Pirow were not physiologists, but they understood heat transfer, although they thought that all body temperature elevations were pathological. Those four casualties had been engaged in hard physical work under- ground, and it became apparent that the wet-bulb tempera- tures were > 30.0 °C, with air movement < 0.2 m.s–1. Most importantly, they had been working in their first or second shifts, after periods away. Mavrogordato and Pirow recom- mended that air movement at hot sites underground should be increased and that men, new or returning to work, be eased into work in the heat by first working at cooler sites: “it is neither kind nor wise to set a raw boy to learn lash- ing [rock shovelling] at 86F. wet-bulb” (Mavrogordato and Pirow 1927 [P. 110]). At the end of the Krogh-Hill epoch, there was a body of knowledge about the circumstances in which heat stroke occurred, and about how it can be prevented and treated. The causal agents were a combination of hard physical work and the wearing of clothing in conditions that prevented the evaporation of sweat at rates sufficient to dissipate meta- bolic heat and that acquired from the environment. It did not escape Charles Martin (England) that “The coolie works with his nice brown body exposed and covered with sweat, and is jolly, whereas the white man distressfully labours in a hyperthermic condition, straining his heart to work a refrigerating plant which he has rendered inefficient because his sense of dignity forbids him to expose his skin” (Martin 1930 [P. 677]). For sweat to evaporate efficiently, the boundary layer of warm, moist air around the body has to be removed, which 12 European Journal of Applied Physiology (2024) 124:1–145 1 3 requires air movement. In that era, all elevations of deep- body temperature were considered to result from a failure of the thermoregulatory centre, though some considered limited elevations to be beneficial to exercise. Fatalities resulted from thermal damage to body tissues. Although recovery was once thought to be impossible from rectal tem- perature > 43.3 °C, survivable, deep-body temperatures of 46 °C were reported (Weaver 1897; Tigerstedt 1906). Heat- stroke risk could be reduced by wearing lighter clothing, by frequent drinking and by easing people into working in the heat. The most successful cooling treatment mimicked the evaporation of sweat, whilst aggressive surface cooling often resulted in rebound hyperthermia (e.g., Withington 1895). Further contributions to heat illness of occupational relevance Like their colleagues in the military, civilian researchers of the Krogh-Hill epoch knew that it was not just the dry-bulb temperature that determined the physiological impact of the climate. Robert Ward (1867–1931; U.S. climatologist; Rohli and Bierly 2011), reminded us that “the hotter the air, the greater its capacity for water vapour; the drier the air, the more water can still be evaporated into it; the more wind, the greater the opportunity for evaporation into the fresh supply of air which is constantly brought to the body” (Ward 1904 [P. 131–132]). Contrary to the views of Haldane (1905), not even the wet-bulb temperature indicated how readily the air would accept heat. Simpson (1914) concluded that the risk of heat stroke was greatest when the absolute air humid- ity was 25–30 g.m–3. Whilst absolute humidity was not a concept familiar to either physicians or physiologists of that time, he showed how it was related to dry-bulb temperature and relative humidity using a psychrometric chart (Simpson 1914). He had also identified hypohydration as a greater challenge to athletes and workers than was a hindrance to evaporation when exercising in hot-dry environments. But other occupational environments could be just as challenging. Leonard Hill had observed that sailors (stok- ers) worked with wet-bulb temperatures never lower than 26.7 °C, and sometimes reaching 36.7 °C, and presuma- bly with little air movement (Hill 1912). Watkins (1917) noted heat hazards in U.S. factories after they changed from water power to steam power. Like Finny (1918), Watkins had understood the role of the boundary layer in evapora- tive cooling: “in hot working zones, if the air be still, even though it be dry, the body becomes quickly surrounded by an air envelope, saturated with body moisture, which, acting like a blanket, prevents the cooling of the body by evaporation” (Watkins 1917 [P. 2119]). Horace M. Vernon (1870–1951, England; Milton 2022) identified problems in cotton-weaving sheds, which were humidified to prevent the cotton breaking, so wet-bulb temperatures were only 1.1 °C below the dry bulb (Vernon, 1921). He also found that the work efficiency of miners increased with wind speed (Ver- non 1928). Of course, evaporation can occur in saturated air, as long as the vapour pressure gradient is favourable (Mitchell et al. 2018). Pembrey was convinced that deep-body temperatures were elevated during exercise (up to 38.9 °C), and he believed “there is no evidence to show that the rise in the internal tem- perature is injurious; evidence is gradually forthcoming to show that it may be beneficial” (Pembrey 1904 [P. 476]). The benefit was being pre-warmed and ready to commence work (Cook and Pembrey 1913). Those elevations were believed to be independent of the external work rate, but the rate of heat production was dependent upon that work rate (Lippmann 1913; Melville 1910, 1916). We will meet Charles Melville again (“Water loss, thirst and drinking: “a pint at the half-way halt””), and it is presumed here that he meant that the tem- perature elevation was independent of both the environment and the exercise intensity, and was regulated at that level by controlling heat loss at a rate required for that work intensity. So elevated body temperatures during exercise not only were normal, rather than pathological, but they were maintained physiologically. A substantial further advance during exercise came from an experiment at a tuberculosis hospital, where the physi- cians measured their own rectal temperatures whilst walk- ing on a level road. Those temperatures increased during exercise and then reached a plateau (Bardswell and Chap- man 1911; England). Those plateaux were intensity (walking speed) dependent (Fig. 4), and they were so consistent that they could reproduce any rectal temperature up to 39.6 °C in themselves “simply by prescribing to ourselves varying degrees of muscular effort” (Bardswell and Chapman 1911 [Pp. 1107–1108]). Those observations were made 27 years ahead of Marius Nielsen (1903–2000, Denmark; Nielsen 1938; Nielsen Johannsen 2022), who also reported that deep-body temperature was determined by exercise intensity. Indeed, most researchers, including ourselves (Notley et al. 2023a), usually attribute that discovery to Nielsen (1938). Accordingly, an elevation of deep-body temperature not only occurred during exercise, but it was proportional to the exercise intensity, although Bardswell and Chapman (1911) made no connection between that elevation and the excita- tion of heat-loss mechanisms. The data in Fig. 4 are not from exercise in the heat. Those data were subsequently presented by Young et al. (1920) and Nielsen (1938). The former were researchers from the Australian Institute of Tropical Medicine (Townsville; Tay- lor et al. 2022), who measured the deep-body (rectal) tem- peratures during outdoor walks at the hottest time of the day and year; wet-bulb temperature 27.2 °C. They compared those data with values obtained for subjects resting in an iron chamber in the sun, in which water was boiled to achieve 13European Journal of Applied Physiology (2024) 124:1–145 1 3 a wet-bulb temperature of 38.9 °C. Body temperatures did not stabilise in the chamber, but they did during exercise outdoors, following an elevation of 1.1–1.7 °C over the first half of walks lasting 2.0–3.5 h, after which temperatures rose only a few tenths further. What did researchers of the Krogh-Hill era know about the thermoeffectors? They knew that cutaneous blood flow was elevated in the heat (Barbour 1912; Stewart, 1913a, b, c; Sundstroem 1927), but were uncertain about the ther- moregulatory vasomotor responses to exercise in the heat. They could have measured cutaneous blood flow, at least in the arms and feet, because plethysmographic and calori- metric methods were well established (Notley et al. 2023b). However, those investigating exercise tended not to measure cutaneous blood flow, but inferred changes from other vari- ables, like skin temperature and heart rate. That approach was confounded somewhat by the heterogenous nature of skin temperatures (Benedict and Slack 1911; LeFèvre 1911; Wer- ner and Reents 1980), though it is less so in the heat. Henry C. Bazett (1885–1950, U.S.A.; Blatteis and Sch- neider 2022) predicted that the skin temperature of work- ing limbs would increase during exercise because he had assumed that the blood passed through the working mus- cles en route to the skin (Bazett 1927). He expected general vasodilatation of the cutaneous vasculature during exercise, so he would have expected skin temperature to rise else- where too, though not as much. One of his contemporar- ies was Francis G. Benedict (1870–1957, U.S.A.; Blatteis and Schneider 2022), who measured 13 different skin tem- peratures on a man cycling in the air at 20 °C (Benedict 1925). Benedict indeed found that the skin temperature over working muscles increased, but skin temperatures elsewhere fell. Regardless of the environmental temperature, research- ers of the Krogh-Hill epoch knew that metabolic heat was carried by the blood to the body’s surface (Melville 1910). The primary means for increasing cutaneous blood flow was via dilatation of peripheral arterioles and venules (Adolph 1924; Winslow 1926). Without cardiovascular compensa- tion, that diversion of blood to the periphery was thought to compromise blood flow elsewhere; “when in a hot and humid atmosphere the blood vessels of the skin are dilated and overcharged with blood, the brain and spinal cord among other organs are rendered correspondingly anemic” (Lee 1912 [P. 866]). One form of compensation would be an increase in cardiac output, and, based on experiments car- ried out on themselves, Barcroft and Marshall (1923) found that, during body warming, the cardiac output increased by an amount equal to the elevation in cutaneous blood flow. Sundstroem (1927) suggested cutaneous blood flow could be ~50 times greater in the heat, which Rowell et al. (1969a) confirmed experimentally. Another compensatory adjustment involves a redistribu- tion of the available cardiac output (MacKeith et al. 1923). Indeed, splanchnic blood flow, which supplies the gastroin- testinal tract, liver, spleen, pancreas and kidneys, is dramati- cally reduced when skin blood flow is significantly elevated, and when humans exercise in the heat (Rowell 1973, 1986; Rowell et al. 1968, 1969a, 1970; Perko et al. 1998). That redistribution can, if protracted, result in localised under- perfusion, and Adolph (1924) thought that such increases in cutaneous blood flow could be counterproductive. Other compensatory adjustments relate to intravascular changes, such as an increased blood volume (Barbour 1912; Hamilton et al. 1924), which accompanies an elevation in plasma pro- tein content (Bazett 1927). In a now-classical paper, August Krogh showed that some of those compensations, which included changes in the ventilatory and heart rates, were initiated very soon after exercise started (Krogh and Lind- hard 1913), and presumably before any substantial change in deep-body temperature. They had demonstrated feedforward thermoregulatory control (Notley et al. 2023a). Feedforward control was neither widely known, nor was it fully understood by researchers of that epoch, and some thought that an accelerated heart rate might be attributable to increased blood temperature (Mansfeld 1910). That possibil- ity was evaluated by Martin et al. (1914) who dismissed the hypothesis. On the other hand, Adolph (1924) found a better association between heart rate and superficial temperature, but he thought that the control mechanism was that “the heart gradually compensates for the lack of venous return of the blood by beating faster” (Adolph 1924 [P. 584]). We now know that the low-pressure baroreceptors will drive the heart rate to regulate central venous pressure (Rowell 1993). Fig. 4 Exercise-induced and intensity-dependent changes in rectal temperature for men walking outdoors on a level road (N = 3), includ- ing the authors (Bardswell and Chapman 1911). Over walking speeds from 3.2 to 9.7  km.h–1, rectal temperature averages were a linear function of walking speed, and have been plotted, with 95% confi- dence limits, from tabulated data 14 European Journal of Applied Physiology (2024) 124:1–145 1 3 At least part of the confusion about the origin of the increased heart rate resulted from uncertainty about how the thermoeffectors were controlled. Indeed, by 1912, it had not only been established that a thermoregulatory centre existed, but it was located within the hypothalamus (Not- ley et al. 2023a). Unfortunately, many of those investigating human thermoregulation during exercise in the heat were either unaware of, or not familiar with, earlier non-human research, although Barbour (1912) was an exception. In Bazett’s (1927) review, Isaac Ott (1847–1916, U.S.A.; Blat- teis and Schneider 2022), who was responsible for pioneer- ing work in the U.S.A. on that regulatory centre, was not cited, although Bazett acknowledged the existence of such a centre. To illustrate the state of the art at that time, we provide the following examples, with a more detailed discussion contained within Notley et al. (2023a). Isaac Ott had pro- posed that a hypothalamic regulatory centre was responsible for the production of sweat in cats, whilst Leonard Rowntree (1883–1959, U.S.A.) proposed that spinal centres excited the sweat glands of humans. Barbour (1912) recognised that warming and cooling a site at the base of the brain influ- enced vasomotion (rabbits), but still asked whether sweating was “primarily set into action through the temperature-sense nerve endings and the central nervous system?” (Barbour 1912 [P. 303]). Edward F. Adolph (1895–1986, U.S.A.) stated “that the dilatation reflex does not concern any part of the central nervous system has been shown by various authors who obtained the response of sweating and of capil- lary dilatation after central connections were cut” (Adolph 1924 [P. 581]), adding “that the three primary responses to high temperatures are initiated by temperature conditions in the skin, and not by those in the central organs” (Adolph 1924 [P. 584]). Mavrogordato and Pirow (1927), as well as most contemporary thermal physiologists, hold the view that a regulatory centre within the brain controls the blood ves- sels, respiratory muscles and sweat glands. Water loss, thirst and drinking: “a pint at the half‑way halt” Whilst Marcus Pembrey condemned the British practice of restricting water for soldiers, the German approach during the First World War was diametrically opposed: “on hot or long marches, mounted or cycle orderlies shall be sent ahead to warn the villagers to turn out and line the streets with tubs and buckets of water so that the men may get some as they pass through” (Melville 1916 [P. 69]). Though it would have been clear to physiologists and physicians of our first epoch that exercise in the heat led to progressive dehydration (hypohydration), they did not know its conse- quences for osmoregulation or thermoregulation. Nor did they know which fluid, how much or on what time sched- ule fluid should be administered to correct water deficits, or whether one could simply rely upon thirst to maintain water balance. Herein, we have adopted the recommenda- tion of Michael Sawka (1992, U.S.A.) that “hypohydration” is preferable entomologically, for referring to reductions in body-fluid content, whilst “dehydration” should refer to the process involved in the progression towards hypohydration. In the past, we have not consistently applied that distinction. For the South African gold miners, Mavrogordato and Pirow (1927) recommended replacing lost fluids with 0.4% sodium chloride solution, which has a salinity nearly half that of blood. In Britain, there was controversy about the pathophysiology of hypohydration and rehydration. For example, Rayner Thrower (1928) identified salt and water loss as a cause of heat cramps, while John S. Haldane (1860–1936, England) was adamant that the cause was acute poisoning by water (Haldane 1928), which could be prevented by eating “red herrings or highly salted bacon” (Hancock et al. 1929, [P. 55]; “… we have all been wrong often” (Hill 1965 [P. 167]). Charles Melville (1863–1943, Scotland) was the British expert on the physiology of the route and forced marches during the Krogh-Hill epoch (Melville 1910, 1916); his German counterpart was Nathan Zuntz (Zuntz and Schum- burg 1901). Melville attempted to analyse fluid losses and replacement during marching, and, for a typical soldier (male 68 kg), whom he presumed would be 67% water (45 kg; total body water for lean males is ~ 650 mL.kg–1; Maw et al. 1996, Australia), he predicted that a fluid loss of 4.5 kg (10% of body water) could be tolerated without the risk of death. Almost a century later, the World Health Organisation (2005) classified that level of hypohydration as “Severe dehydration” (P. 7). Melville (1916) also believed that a 3.4-kg (5%) loss should neither reduce work efficiency nor induce intolerable suffering in well-trained soldiers, while a loss of only 1.1 kg (1.6%) would have similar effects in poorly trained individuals (Melville 1916). The possibility that hypohydration might compromise physical performance is taken up again in “Does hypohydration compromise exer- cise performance?”. Based on the estimated energy cost for a loaded sol- dier during a forced march, Melville estimated that it would require sweat to evaporate on the skin at a rate of 118 mL.km−1 to dissipate that metabolic heat. So even a poorly trained man should be able to march 11 km (50% of the daily marching distance) without a significant loss of work efficiency (body-mass loss: 1.2 kg). Accordingly, Melville recommended drinking a little less than “a pint at the half-way halt” (11 km; Melville 1916 [P. 67]), which would also suffice for the next 11 km. For longer marches, he insisted that soldiers drank every hour and on command. Presumably, he ignored both the presence or absence of thirst: “no man should be allowed to use his water-bottle without orders, any more than he is allowed to fire a round 15European Journal of Applied Physiology (2024) 124:1–145 1 3 of ammunition without orders” (Melville 1916 [P. 70]). He also advised against drinking larger amounts, because excess water “merely passes through the kidneys, without doing the body much permanent good” (Melville 1916 [P. 70]). Melville did not address water loss in the heat, but Edward Adolph did (Adolph 1921). He established the Rochester Desert Unit (U.S.A.; Blatteis and Schneider 2022) from which arose numerous contributions to our understand- ing of exercise in the heat (Adolph 1947a). Since sweat is hypotonic with respect to blood (Hunt 1912), heavy sweat- ing leaves the remaining body fluids hypertonic, so Adolph expected renal chloride loss to increase, but he actually found that it decreased (Adolph 1921). In an experiment in which one of Haldane’s students “refrained from bathing for 1 week” (he truly was English; Hancock et al. 1929 [P. 45]), and then sweated in a chamber with air saturated at 33.3 °C, sweat chloride concentration increased with sweat rate, which they attributed to sweat- gland fatigue. We now know that electrolyte reabsorption within the sweat duct is inversely related to flow (Dill et al. 1938; Kuno 1956; Sato and Dobson 1970). Hancock et al. (1929) also found that the chloride concentration in urine decreased, as would be expected from reductions in splanch- nic blood flow (MacKeith et al. 1923; Rowell 1973, 1974; Rowell et al. 1968). Adolph made two other announcements that are relevant to an ongoing debate. Firstly, he said that hypohydration induced by sweating had no consequences for body tem- perature: “In dehydration I have not found a poorer regula- tion of temperature than in plethora; the limits to the body’s use of water for regulating temperature were not reached” (Adolph 1921 [P. 126]). He conceded that he may not have dehydrated his participants enough to affect body tempera- ture, and subsequently changed his view when he, and others from his group, found evidence linking an elevation in rectal temperature to hypohydration (Adolph 1947e; Brown 1947c; Rothstein and Towbin 1947). Secondly, he said that hypohy- dration did not reduce sweat rates in standardised conditions. As we shall see, it may or may not. Like most others of the time, Adolph worked on the assumption that the water in sweat came from the blood, at least initially. Of course, body water, as well as its location within the body-fluid compart- ments, urine and sweat, is an entirely passive molecule that obeys changes in osmotic pressure (Starling 1896). Contrary to Adolph’s second statement, William Marriott (U.S.A.) wrote: “When the blood and tissues become concentrated by water loss the amount of water available for evaporation is diminished and ultimately becomes less than that required for removal of the heat of metabolism. Fever then occurs.” (Marriott 1923 [P. 286–287]). He considered evaporative cooling to be compromised by reduced body water, but since the total body water of a lean, 70-kg individual would be ~ 45 L (Maw et al. 1996), we have a large, accessible water volume. Nevertheless, body-water loss and the accompanying increase in plasma (serum) osmolality result in the sensation of thirst. Walter B. Cannon (1871–1945, U.S.A) confirmed that thirst was not a sensation of local origin: “the thirsty man does not complain of these general conditions. He is tormented by a parched and burning throat, and any expla- nation of the physiological mechanism for maintaining the water content of the body must take into account this promi- nent fact” (Cannon 1918 [P. 293]). He attributed the parched sensation to the paucity of saliva, which was confirmed by Rowntree (1922), in his translation of a statement by Schiff and Levier (1867, Germany): “the feeling of dryness in the mouth, although it accompanies thirst, has only the value of a secondary phenomenon, and bears no deeper relation to the general sensation than the heaviness of the eyes bears to the general sensation of sleepiness” (Rowntree 1922 [P. 126]). Whether thirst could be trusted to maintain water balance, or whether that maintenance required coercion, of which Melville was convinced, was not debated in the Krogh-Hill epoch, but it certainly has been in the modern epoch. With the exception of a few planned experiments, atten- tion to thermoregulation during exercise in the heat had largely been devoted to men at work, sometimes in indus- try, but mostly in the military. In their analysis, Mavrogor- dato and Pirow remarked that the deep-body temperatures observed in miners were the same as those “found in athletes during violent exercise on the surface” (Mavrogordato and Pirow 1927 [P. 111]). They seem to have possessed informa- tion that was not generally available, presumably because research on thermoregulation during exercise was almost devoid of experiments conducted during recreational exer- cise. After a brief period of intense interest in thermoregula- tion during the Boston Marathon, interest waned (Maron and Horvath 1978). However, there was one significant paper in which the poor physical condition of runners at the end of that marathon was attributed to hypoglycaemia, rather than to heat strain (Levine et al. 1924), and we shall return to that paper in “Fuelling the fires: the competition for blood glucose”. Epoch three and a touch of politics for exercising in the heat Let us advance our journey whilst looking back at the obser- vations and interpretations that have been supported over time. We will linger at several mileposts along the way at which our understanding of thermoregulation during exer- cise in the heat was advanced substantially, but readers are also directed to Rowell (1974, 1993), Gisolfi and Wenger (1984), Montain et al. (1994), Hales et al. (1996), Sawka et  al. (1996, 2011), Taylor et  al. (2008c), Crandall and 16 European Journal of Applied Physiology (2024) 124:1–145 1 3 González-Alonso (2010), González-Alonso (2012), Johnson et al. (2014), Nybo et al. (2014), Taylor (2019) and Cramer et al. (2022). Because some topics have no counterpart in the Krogh-Hill epoch, we will not address some prominent features of the contemporary landscape, such as thermoregu- lation during exercise in the aged (Hellon and Lind 1956; Dill and Consolazio 1962; Kenney 1997; Inoue et al. 1999b; Meade et al. 2019) and in children (Rowland 2008; Notley et al. 2020a, b), thermoregulation during anaerobic exercise (Cheuvront et al. 2006; Zhao et al. 2013), sex (gender) dif- ferences (Kenney and Anderson 1988; Gagnon and Kenny 2012; Notley et al. 2017) and differences due to ancestry (Samueloff 1987; Taylor 2006a, 2014; Lambert et al. 2008; Muia et al. 2020). Another view of this journey can be found in Schneider and Moseley (2014). Body temperature elevation during exercise in the heat is physiological and essential The very clear message from some of the most accomplished researchers across our three epochs (e.g., Marcus Pembrey, John Haldane, Erik Christensen [1904–1996, Denmark] and Marius Nielsen), was that exercise could elevate body temperatures independently of the ambient heat load. That message was not accepted universally at the time. Those who opposed the view did so not because they had contrary data, but on theoretical grounds and a partial understand- ing of thermoregulation. If body temperature is regulated, then surely it would be regulated during exercise too? Thus, body-temperature deviations would be opposed and over- come by heat-loss thermoeffectors, and then restored or maintained at its basal level. That view is still held today by some (e.g., Ramsay and Woods 2014). Researchers of the Krogh-Hill epoch had no concept of either control theory or negative-feedback control. They did not know what the stimulus was for the excitation of sweat- ing or cutaneous vasodilatation, and some even denied that the nervous system was involved. Though some had the nec- essary data, nowhere in the papers cited thus far does one find the development of a clear linkage between deep-body temperature and a heat-loss mechanism. That was provided by Cyril H. Wyndham (Fig. 5; Wyndham 1967) from South Africa (1916–1987; Anonymous 1987; Wolstenholme 1987; Mitchell and Laburn 2022). Figure 5 represents an important milepost along our jour- ney, and it showed that feedback from the deep-body ther- moreceptors alone seemed to drive sweating. As we estab- lished within our first communication (Notley et al. 2023a), that interpretation is no longer considered valid, although it could be derived from those data because, in all trials, the participants had approximately the same skin temperature (~ 35.0 °C; Notley et al. 2023a [Figure 13]), so feedback from the peripheral thermoreceptors would have remained stable. Today, we know that feedback comes from thermore- ceptors distributed throughout the body (Werner 1980, 2010; Jessen 1996; Werner et al. 2008). But the principle stands. Elevations of body temperature during exercise are essen- tial, because those changes excite the thermoreceptors that provide the feedback needed to recruit, and then maintain, the activity of our heat-loss mechanisms. Under the con- stant load of an elevated metabolic-heat production during exercise, sustained body-heat balance is possible only if the appropriate temperature deviation is also sustained. Without that feedback, our thermoeffectors are turned off, and heat loss becomes wholly dependent upon passive exchanges. In addition to demonstrating that relationship (Fig. 5), it is clear that heat-adapted individuals were sweating more profusely at the same deep-body temperature, and we return to that topic in Part 4 of this historical series. Bardswell and Chapman (1911) drew our attention to the fact that, after some time at each walking speed, rectal temperatures would stabilise at levels determined by walking speed (Fig.  4). That intensity dependence was made famous by August Krogh’s assistant, Marius Nielsen (1938; cycle ergometry; Notley et al. 2023a [Fig- ure 23]). With Erling Asmussen, Nielsen later reported that, at the same external work rate (cycling), both rectal and gastrointestinal temperatures were lower, when work was performed using the arms, than with the legs (Asmus- sen and Nielsen, 1947). That contrasted with evidence reported by Erik Christensen (1931; Notley et al. 2023b), from the same institute. Bodil Nielsen (1968), also from Fig. 5 Sweat rates for exercising men in saturated conditions (box stepping) plotted as a function of deep-body (rectal) temperatures. Data points are four-hour averages, with sweat rates grouped in class intervals (0.3  °F) of rectal temperature, obtained from 13 heat adapted and 353 non-adapted miners of African ancestry. Data were collected across 45 different combinations of wet-bulb temperature, wind speed and work rate, with 465 data pairs for the adapted miners and 262 data pairs for those who were not heat adapted. Drawn using data extracted from Wyndham (1967 [Table 2]) 17European Journal of Applied Physiology (2024) 124:1–145 1 3 that institute, replicated the observation of Asmussen and Nielsen (1947), but now using oesophageal temperature. However, the interpretations of those more recent studies were domesticated by Sawka et al. (1984c). He also meas- ured both rectal and oesophageal temperatures and found that the deep-body thermal responses were not exercise- mode specific, but work-rate dependent. Curiously, none of those papers cited Christensen (1931). Sawka et al. (1984c) attributed those different observations to under- powered experimental designs, not achieving physiologi- cal steady states, difficulty using the arm-crank ergometer, or some combination of those factors. One can imagine some interesting conversations in the family home since Bodil Nielsen is Asmussen’s daughter. As noted by Notley et al. (2023a), Marius Nielsen is also credited with the claim that the deep-body (rectal) tempera- tures reached during steady-state exercise were not influ- enced significantly by the dry-bulb temperature. However, dry-bulb temperatures were varied only between 8 and 29 °C (Nielsen 1938), and in a companion investigation, only three temperatures were used: 5, 20 and 30 °C (Nielsen and Nielsen 1962). From those investigations, arose the belief that deep-body temperature would be independent of air temperatures from 5 to 30 °C (Nielsen Johannsen 2022). But is the deep-body temperature during exercise in the heat truly independent of the ambient thermal load? Credit for revealing the fallacy of that independence is generally given to Alexander Lind (1925–1990, U.S.A.; Blatteis and Schneider 2022). Lind was then working in England, and by extending the ambient conditions above those imposed by the Danish duo, he established a region of ambient conditions, which he called the prescriptive zone, in which the rectal temperature of just three individuals seemed to be independent of the ambient heat load. Above that zone, deep-body (rectal) temperature rose above the level expected for the metabolic-heat production; a very important observa- tion, but was it original? We must also consider the possibil- ity that those data may not have reflected true thermal steady states, since those trials lasted just 60 min, and we know that rectal temperatures can take longer to stabilise (Notley et al. 2023b). One of the earliest observations that demon- strated that reality was provided by Withington (1895), who recorded deep-body temperatures prior to, and for some time after, death due to heat stroke: “… the thermometer contin- ued to show per rectum, at five-minute intervals for fifteen minutes after death, the same temperature of 107° [41.7 °C]” (P. 514). To address both the originality and the methodological (steady-state) limitation, we must travel 13,000 km to Johan- nesburg (South Africa), and set our next milepost. A decade before Lind (1963), Cyril Wyndham and his team reported the same phenomenon, but now in miners working in the heat (Fig. 6; Wyndham et al. 1953 [4 h, N = 13]). At lower ambient heat loads, deep-body temperatures stabilised, after about 60 min, at levels that depended upon the work rate, and were not significantly influenced by the ambient heat load. At higher heat loads, those temperatures either stabi- lised at a higher level, or did not stabilise at all, but became dependent upon the ambient heat load. Wyndham’s paper was not published in an easily acces- sible journal. However, we are obliged to seek out and famil- iarise ourselves with all possible resources. Notwithstanding, the principal conclusion from both groups remains unchal- lenged. Deep-body temperature is stable and elevated during exercise for physiological, not pathological reasons. The level of that temperature is a function of metabolic heat produc- tion, and is dependent upon the physics of heat transfer and the constraints of negative-feedback control. Curiously, Lind and Wyndham knew each other quite well, and they even worked together (U.K. Climatic and Working Efficiency Unit; Milton 2022), yet, whilst Lind cited Wyndham’s paper, he confused the author sequence, misspelt one author’s name and underplayed its outcome. That led to tension between Lind and Wyndham (Mitchell and Laburn 2022). Opportunities to conduct experiments similar to those led by Wyndham were only possible because they were funded by a mining company. The company itself was perhaps not so interested in basic science, but in how the acquired knowledge might be applied to reduce work-related heat illness and to simultaneously increase productivity. Clinically relevant deep‑body and skin temperatures Although Wunderlich reported that the “highest tem- perature yet met in a living man, noted by a trustworthy observer, amounted to 112.55 °F (= 44.76 °C)” (Wunder- lich 1871 [P. 2]), we will soon describe still higher tem- peratures from both the Krogh-Hill (Tigerstedt 1906; 46 °C [possibly malaria]) and modern epochs (Hart et al. 1982; 47.0 °C [classical heat stroke]). Wunderlich (1871) classi- fied deep-body temperatures from 38.5° to 39.0 °C as states of moderate fever, > 39.5 °C in the morning as high fever and those > 42.0 °C as probably being fatal. Contemporary clinical classifications provide us with broad classifica- tions of deep-body temperatures for use with the general population, which might then be used to determine treat- ment strategies: >  40  °C (profound clinical hyperther- mia), > 39.5 °C (profound hyperthermia), 38.5°–39.5 °C (moderate hyperthermia), 37.2°–38.5 °C (mild hyperther- mia) and 36.5°–37.0 °C (normothermia; Taylor et al. 2008c). From an occupational perspective, deep-body temperatures of 38.0°–38.5 °C are widely accepted as an upper limit for worker health and safety (Jacklitsch et al. 2016), although such thresholds appear to be directed at preventing not only heat stroke, but also less-severe heat illnesses (e.g., syncope, heat exhaustion) in nearly all workers (Malchaire 18 European Journal of Applied Physiology (2024) 124:1–145 1 3 et al. 2000). To achieve the same objective, some have indi- cated that threshold may be even lower (e.g., 37.5 °C; Sawka et al. 2001a), which would be inappropriately restrictive. Furthermore, we would expect most competitive marathon runners to have deep-body temperatures > 40 °C at the end of their races, but without sequelae (“Exertional heat stroke in sport”). This raises a larger issue as to whether deep-body temperature should even be the variable of primary inter- est when setting limits to mitigate injury risks in occupa- tional athletes, although we are currently without a suitable Fig. 6 Deep-body (rectal) temperatures of 13 heat-adapted miners across four hours of exercise (box stepping in an underground cli- matic chamber) at each of three work rates, across five different dew- point temperatures (saturated air) and with a wind speed of 0.77 m. s−1. Redrawn from Wyndham et al. (1953) 19European Journal of Applied Physiology (2024) 124:1–145 1 3 alternative. For clinically relevant skin temperatures, we have the following general classifications: > 50 °C (second- degree), > 45 °C (tissue damage), 41°–43 °C (burning pain), 39°–41 °C (pain) and 33°–39 °C (warm to uncomfortably hot; Taylor et al. 2008c). Practice improves thermoregulation during exercise in the heat When Marcus Pembrey recommended “training on hot as well as on cold days” (Pembrey 1902 [P. 269]), as a way of improving the resistance of soldiers to heat stroke, he did so by intuition. He had no knowledge of the physiology of ther- mal adaptation. Does physiological adaptation to exercise in the heat require training in the heat, or can it be achieved by passive heat exposure? Does training on cold days have any benefit for our capacity to exercise in the heat without harm? These and other questions concerning natural and artificially induced adaptations to both heat and cold are addressed in Part 4 of this review series. Warmed to the work Our sub-title is taken from the catchphrase that Marcus Pem- brey (Cook and Pembrey 1913) used to encapsulate his view that the elevation of deep-body temperature during exercise was beneficial physiologically. It would have been evident that an upper limit to that elevation should exist, because excessive elevations caused heat illness. Pembrey believed that, between those limits, either the capacity for exercise or the efficiency of exercise would be enhanced by increasing (warming) body temperatures. Archibald Hill received a Nobel Prize in 1922 for his dis- coveries that contracting muscle converted chemical energy into work and heat. At the most fundamental level, we must ask how skeletal muscle activation might be affected by muscle temperature. Whilst we know more about the rela- tionship between muscle function and muscle temperature within poikilotherms (Bennett 1985; James 2013; James and Tallis 2019), it seems that the temperature dependence of in situ muscle activation in homeotherms, and especially the static properties of muscle, is quite weak (Racinais et al. 2019a). We do know that muscle tension tends to be maximal at resting muscle temperatures and declines a little at higher temperatures (Rall and Woledge 1990). Circulating water at 44 °C around the forearm appeared to make no difference to the isometric force developed by the forearm muscles (Mal- lette et al. 2019), although changes in muscle temperature were not actually measured. Rates of muscle activation and relaxation, and consequently the development of maximal power output, are more dependent on muscle temperature and are accelerated at higher temperatures (Bennett 1984). The temperature of resting muscle usually is a few degrees below rectal or oesophageal temperature (Booth et al. 2004), and resistance to fatigue is better at that temperature than at higher temperatures (Malak 2020). So Pembrey may have been disappointed if he had known how little being “warmed to the work” affected the contractile machinery of skeletal muscle. Of course, there is far more to musculoskeletal perfor- mance during exercise in the heat than how the muscle fibres respond to being warm. The turnover of oxygen and nutri- ents could be influenced by temperature, as could energy delivery and the removal of metabolic wastes. Muscular activity requires activation of the motor nerves, which also could be influenced by temperature. Coordinated exercise also requires optimal cognitive function, particularly during teams sports, which could be affected by body temperature (Schmit et al. 2017; van den Heuvel et al. 2017). Accord- ing to Cook and Pembrey (1913), the nett outcome of those temperature dependencies would be improved performance. That view was shared by Asmussen and Bøje (1945), who showed that the performance of a 100-m sprint (cycling) and a 1500-m run were improved when the body was warmed, either by prior exercise or by diathermy. They associated that improvement with elevated muscle temperature, which approached 39 °C (Asmussen and Bøje 1945). If muscle temperatures had exceeded 39 °C, they might have come to the opposite conclusion. In the same Danish laboratory, Bengt Saltin (1935–2014; originally from Sweden; Nielsen Johannsen 2022) and col- leagues asked subjects to cycle to exhaustion at work rates of 90%, 100% and 115% of peak aerobic power, but in different air temperatures, with and without pre-heating (Saltin et al. 1972). The higher the external heat load, the sooner the sub- jects stopped work, and always at leg-muscle temperatures of 39°–40 °C, irrespective of workload, work time and ambi- ent temperature. D. Bruce Dill (1891–1988, U.S.A.; Blatteis and Schneider 2022) had previously encountered premature exhaustion during exercise in the heat, and attributed it to the heart reaching its pumping capacity (Dill et al. 1931). If that limit resulted in blood pressure regulation being compro- mised, then participants might well have terminated exercise due to uncompensable hypotension (cardiovascular insuf- ficiency; Barcroft and Edholm, 1945; Bass 1963; Krediet et al. 2004), possibly accompanied by pre-syncopal episodes (Wilson et al. 2006) and even incapacitation (Noakes 2008a; Epstein and Yanovich 2019). In the workplace, miners shov- elling rock under thermally stressful conditions reduced their performance as ambient conditions became more stressful (Strydom et al. 1963). So being “warmed to the work” was not necessarily beneficial. On the contrary, being too warm can compromise performance, and it is well established that marathon race times increase, particularly within sub- elite runners, with greater external thermal loads (Ely et al. 2007a, b). 20 European Journal of Applied Physiology (2024) 124:1–145 1 3 Over time, a pattern emerged concerning exercise perfor- mance and body warming, which Marcus Pembrey would have supported. Firstly, warming up can improve exer- cise performance (Asmussen and Bøje 1945; McGowan et al. 2015), although the benefits are often hard to dem- onstrate, and they may involve injury prevention (Maughan and Shirreffs 2004). When warming up is beneficial, those changes are mainly related to increased body temperature, but they may be quite small; a 1 °C increase in exercising mus- cle temperature enhanced subsequent performance by 2–5% (McGowan et al. 2015). However, at the level of elite sport, such marginal benefits may be crucial, especially in explosive events. In a cool environment, cyclists achieved higher peak power in an ergometer sprint test if their leg muscles were prevented from cooling between their warm-up and the actual event (Faulkner et al. 2013). That benefit has to be traded off in hot weather, especially in humid conditions, because elevated body temperatures can compromise performance (McGowan et al. 2015). In hot weather, warm-up practices should be tailored so as not to raise body temperatures mark- edly (Maughan and Shirreffs 2004). One potential way of preparing for endurance exercise without inducing