US infantry soldiers
© Eliot Elisofon/Time and Life Pictures/GettyUS infantry soldiers trudge across a hill in the desert in the El Guettar Valley, Tunisia, early 1943
On 17 August 1942, seven men set up camp near the isolated desert crossroads of Freda, California. Soon, a squad of soldiers arrived for what seemed a particularly brutal form of training. The troops worked and hiked up to 35 kilometres beneath the merciless sun. Then they did it all over again, periodically submitting to tests to see how their bodies were coping. The goal was to help win the second world war by learning how far a man could be pushed in the baking North African desert. Today the results remain vital for marathon runners, trekkers, and others who exert themselves in scorching heat.

California's lower Colorado desert is a forbidding place, particularly in the summer. It's a barren landscape of rocks and rattlesnakes, where little grows but creosote bushes and cactus. Midday temperatures top 43 °C and searing winds and afternoon sun combine to suck moisture from the body.

This is not the place for a midday march, but that is exactly what Edward Adolph had in mind when he took a group of soldiers and researchers there in the summer of 1942. Adolph, a physiologist at the University of Rochester in New York state, wanted to find out how people could live and work efficiently in the desert and how to get the best out of them.

He wasn't the first to ponder the effects of hot, dry conditions on the human body. The image of the traveller lost in the desert, crawling towards a shimmering mirage, is probably as old as desert travel itself. But earlier researchers always focused on survival. "They never looked at performance," says Timothy Noakes, an exercise physiologist at the University of Cape Town, South Africa, and veteran of some of the world's toughest ultramarathons.

Adolph was the first to test the presumptions most people still have about what to do if forced to make any sort of effort in extreme heat. Most, he discovered, were myths. Stripping to T-shirt and shorts, for instance, is not the best way to cope with dehydrating conditions. Long sleeves and long trousers may feel hotter, but they'll slow the loss of water. Nor is there any point in rationing water when supplies are low. Putting off drinking it merely makes you unhappier sooner. "It is better," wrote Adolph, "to have the water inside you than to carry it."

The most important of Adolph's findings was the simplest: drinking during exercise improves performance. Today, we take this for granted, but generations of coaches and distance runners were taught that drinking during exercise was for wimps. It would only make you thirstier, some claimed. Others said it could even trigger a heart attack. "Don't get into the habit of drinking and eating in a marathon race," advised the author of Marathon Running in 1909. "Some prominent runners do, but it is not beneficial."

Adolph tested the old assumptions by splitting his soldiers into two groups. Both marched through the desert for up to 8 hours during the time of year when the average afternoon high was 42 °C. The soldiers in one group were allowed to drink as much water as they wanted and the others weren't allowed any. The results were clear: the drinkers outperformed the non-drinkers, but the men in both groups gave up once they had sweated off 7 to 10 per cent of their body weight.

To Adolph, this made perfect sense. On days when the temperature is hotter than the average person's skin temperature - about 33 °C - the only way for the body to cool itself is by the evaporation of sweat, and he could calculate how much moisture that required. A brisk walk could easily require three-quarters of a litre or more of evaporative cooling each hour. Over an 8-hour stint, that's a lot of sweat.

Adolph's research was driven by the North Africa campaign, which ended in 1943. But he returned to the desert each summer during the war years and supplemented his experiments with tests in his heated lab. His findings stayed secret until 1947, when he was allowed to publish his pioneering Physiology of Man in the Desert. It went almost entirely unnoticed. In the late 1960s, marathon runners were still advised not to drink during races and until 1977, runners in international competitions were banned from taking water in the first 11 kilometres and after that were allowed water only every 5 kilometres.

Then there was a complete reversal of opinion. A trickle of studies began to warn of the dangers of running a marathon without enough water and suddenly runners were being told they must drink during a race - and if they didn't feel like it they should force themselves or risk heatstroke.

In 1978, Alberto Salazar, one of America's great distance runners, ran a 7.1-mile race on a 29 °C day. At mile six, he was in second place. "The last thing I remember," he said later, "was watching Bill Rodgers pull away from me. It was dreamlike. There was Bill, floating away, and I wasn't able to summon the energy to go after him. In the next mile, I faded from second to tenth, but I have no memory of being passed by anyone."

Salazar almost died. At the finish, his body temperature was 42 °C and he was saved only because a quick-thinking medical crew promptly dumped him into a tub of iced water. Everyone "knew" what Salazar had done wrong: he hadn't drunk enough before or during the race, became dehydrated and nearly killed himself. Even Salazar agreed. "Dehydration is insidious," he would later say.

At first glance, Adolph's findings seem to bear this out. His notes about his dehydrated soldiers are a litany of woe. "Only desire, to stop and rest," he wrote of one man, after 13.4 waterless kilometres in 40 °C heat. "Unsociable attitude," he wrote of another, who managed 29.8 kilometres at 34 °C. "Began to lag and finally stopped."

Runners and coaches in the 1970s and 1980s presumed that collapsing athletes like Salazar were simply extreme cases of the same thing. Dehydration and heat collapse were virtually synonymous in many minds. "Drink early and often," athletes were told - and not just when thirsty.

Yet, as Noakes points out, none of Adolph's dehydrated soldiers suffered heatstroke. "They just got very angry and stopped walking." What's more, they recovered rapidly when allowed to rest and drink. "Able to walk almost immediately after taking water," Adolph wrote in one case. In another: "exhaustion relieved by water". Salazar's brush with death wasn't the result of drinking too little: he had simply tried to run a world-class race on a very hot day. Under those conditions heat is the enemy, not dehydration.

Adolph had recognised this but thought it too obvious to warrant more than a few lines in his book. He had conducted most of his tests on marches - not because he wasn't interested in the effects of running in the heat but because when he made his soldiers run, even at a slow jog their body temperature soared by 2.5 °C in 30 minutes. "There is no doubt that men are limited in the physical work they can do in the desert," he wrote.

The advocates of drinking-early-and-often also overlooked Adolph's finding that even soldiers who were allowed to drink what they wanted still tended to dehydrate, and only made up their deficits at mealtimes. Adolph dismissed this as a "peculiarity of dehydration", but Noakes believes he had stumbled on a quirk of human evolution.

Humans, Noakes points out, are "delayed drinkers". He suspects that's an adaptation to hunting, which for early humans took the form of chasing game long distances under the African sun. There are good reasons for not stopping to drink during a hunt, not least the prospect of the quarry escaping. There's also the fact that we are not built like camels and other animals that can drink deeply and quickly. That makes us better runners - and running hunters - but means we cannot drink as fast as we can sweat, so we put off our thirst until it's convenient to drink, says Noakes.

Adolph never used the word evolution in his book but he would have understood Noakes's point. "Instead of blaming either performance or failure in the desert upon personality," he wrote, "it is better to know which working characteristics of the human body are basic to whatever undertaking men wish to accomplish."