The winner, John Hayes, of the first modern marathon in 1908 said that he thought drinking water during a marathon was a grave mistake, saying: "I merely bathed my face with florida water, and gargled my throat with brandy". Quote can be found here and is also in "The 1908 Olympic Games: Results for All Competitors in All Events, With Commentary" by Bill Mallon and Ian Buchanan.

Current research shows male elite runners today drink on average half a liter of water during a race. Raceorganizers will also set up aid stations throughout the race for contestants to get water. When did drinking water during a marathon become common practice?

  • 1
    @MarkC.Wallace I added a reference to the quote. I changed the question slightly as the importance of drinking water is debated. Commented Mar 4, 2019 at 14:57
  • 1
    It might prove a little hard to give a definite answer if you are only interested in drinking water as the advice to increase fluid intake in the mid 1970s also led to a boon in sales of sports drinks. Commented Mar 4, 2019 at 15:47
  • For more quaint/outdated/dangerous advice on how to run a marathon, check out Jon Bois's video Rat Poison and Brandy: The 1904 St. Louis Olympic Marathon. Commented Mar 5, 2019 at 3:54

3 Answers 3


Before 1970

Up until about 1970, it was thought to be harmful to drink water while running a marathon, and a marathon course might have no aid stations or only one. Marathon runners were actively discouraged from drinking water. For a description of this period, see this interview.


In 1969, a paper by Wyndham and Strydom ("The danger of an inadequate water intake during marathon running," S Afr Med J. 19;43(29):893) argued that marathon runners should be allowed to drink fluids in order to prevent heat stroke. From about 1972 to 1981, a South African researcher and runner named Tim Noakes started a campaign through publications such as Runner's World to get people to pay attention to the Wyndham paper and drink fluids during a marathon. The American College of Sports Medicine wrote guidelines saying that people should drink regularly while running.

This was also the period when so-called sports drinks were becoming popular. In 1969, Gatorade started to be marketed in the US, and it was designated the official sports drink of the NFL. Aided by these commercial interests, the scientifically unfounded idea began to be propagated that drinking before thirst was necessary in order to prevent heat stroke. Studies were carried out with football players in which it was claimed that drinking early versions of Gatorade led to increased performance, but the methodology was not capable of discerning whether this was a placebo effect.

A physiologist named David Costill had distance runners run on a treadmill without water, and then in separate trials drinking 1.2 liters per hour (which is a huge amount of water). He found that body temperatures were lower if they drank water. The meager evidence from studies like Costill's was overinterpreted, and in many case overzealous people began to simply invent scientifically unsupported guidelines for hydration. The US military adopted guidelines calling for soldiers to drink 64 ounces of water per hour.

During this same period, a cultural idea in the US of "getting back to the land" led to a craze for wilderness backpacking, and with this came further anxiety about water. This included a scare about contracting giardiasis from backcontry water sources in the US, which turns out to have been unfounded (Zell, "Epidemiology of wilderness-acquired diarrhea: implications for prevention and treatment," Wilderness and Environmental Medicine 3 (1992) 241).


During this period, scientists began to do more complete and careful studies, which showed that drinking large amounts of water was bad. It decreases performance, and can lead to a dangerous condition called hyponatremia. Studies with larger samples in real-world conditions showed that post-race body temperatures were not reduced by drinking more water (Noakes et al., "The danger of an inadequate water intake during prolonged exercise," European Journal of Applied Physiology and Occupational Physiology 57 (1988) 210).

Despite the scientific consensus that emerged by the end of the century, the irrational and extreme fear of dehydration began to grow in the popular consciousness, leading even sedentary people to believe that they were in danger. For example, there was a folk belief that one should drink at least eight 8-ounce glasses of water per day ("8x8") (Valtin, "'Drink at least eight glasses of water a day.' Really? Is there scientific evidence for '8x8'?," Am J Physiol Regul Integr Comp Physiol 283: R993-R1004, 2002).

  • 2
    Drinking large amounts of water may be bad, but what about smaller amounts?
    – JAB
    Commented Mar 4, 2019 at 19:05
  • 5
    @JAB: As far as I can tell, the present scientific consensus is that it's reasonable to listen to your body, and drink when you feel thirsty. But it's at most an issue of comfort and performance, not safety; this article emsworld.com/article/10324701/… says, "there is not a single case report or clinical trial that unambiguously links exercise-induced dehydration with specific life-threatening, exercise-related disorders," citing Noakes TD. Hyponatremia in distance athletes. The Physician and Sportsmedicine 28(9):71–76, Sept. 2000.
    – user2848
    Commented Mar 4, 2019 at 19:15
  • 2
    "There was a folk belief that one should drink at least eight 8-ounce glasses of water per day." In my experience this folk belief persists until the present day. Commented Mar 5, 2019 at 14:37
  • 3
    Part of the problem with the idea that you should drink vast amounts of water is that we also live with non-scientific recommendations to limit salt intake to the bare minimum needed to live. Hyponatremia is a result of upsetting your salinity i.e. your fluids are not saline enough.
    – JimmyJames
    Commented Mar 5, 2019 at 19:12
  • 2
    Long before dehydration becomes a problem for your biochemistry, it will become and exceedingly annoying sensation in your mind. This is called feeling thirsty. You will stop worrying about anything other than getting a drink. On the other hand, sweating during a race removes water and salt from your body. You can easily replace the water, but the salt is harder to come by. As has been noted, this can lead to hyponatremia. This is largely symptomless until it becomes critical. The first thing you might notice is cardiac arrythmia, followed quickly by arrest. Commented Mar 6, 2019 at 6:52

The constant availability of water to marathon runners probably developed in the 1970s.

According to an article in Outside magazine, drinking alcohol during competitive races was rather common in the late nineteenth and early twentieth centuries:

Fueling typically meant a shot of whiskey, brandy, or other alcohol. Spyridon Louis, winner of the marathon at the 1896 Olympics, sipped cognac with fewer than six miles remaining. The 1924 Paris Marathon featured a fluid station offering pours of wine to runners.

Citing the book Waterlogged, which is skeptical towards modern hydration practices, the Outside article claims that significant changes in these practices came only after the invention of Gatorade in 1965. "And even then, it wasn’t until the mid-1970s that thoughtful hydration became common practice."

Basically consistent with this, an article in the Encyclopedia of International Sports Studies describes a "landmark scientific study" which was published in 1969. The authors "concluded that marathon runners should aim to drink 250 ml of fluid every 15 min during exercise".

Wyndham, C.H., Strydom, N.B. "The danger of an inadequate water intake during marathon running." South African Medical Journal (1969) 43: 893-896

  • Is the answer to the question the last paragraph? The rest is just context?
    – MCW
    Commented Mar 4, 2019 at 15:19
  • Please don't reply in comments
    – MCW
    Commented Mar 4, 2019 at 16:58
  • 1
    Note that drinking alcohol was really much more common in pre-modern times and holdouts of the custom are everywhere. In the middle ages, water was simply not clean enough to be safe to drink in most places, and (lightly) alcoholic drinks were preferable as the alcohol killed most of the microbes.
    – Tom
    Commented Mar 6, 2019 at 13:54

Let's look beyond the marathon aspect in its over-specificity. The question asks for when was it recognised to drink water while exercising heavily would be advisable?

That means the scientific side has to be examined and not necessarily the practical application or even the commercial availability of specialised products, Brawndo-style.

This focus on water for performance began in Sweden during the 1930s, curiously as a by-product of analysing carbohydrate metabolism.

The development of nutritional beverages specifically geared towards improving athletic performance started with studies on carbohydrate and fat metabolism conducted in Sweden in the 1930s and continued into the late 1960s. The team of scientists led by Bjorn Ahlborg and Jonas Bergström studied the relationship among muscle glycogen storage, use, and resynthesis during prolonged exercise to exhaustion in a group of volunteers. The research by the Swedish team demonstrated a performance-enhancing role for carbohydrates during endurance exercise and showed that glycogen content and the long-term exercise capacity could be varied by instituting different diets after glycogen depletion.

Gustavo A. Galaz: "An Overview on the History of Sports Nutrition Beverages", Nutrition and Enhanced Sports Performance, Second Edition, p231, Elsevier, 2019.

The "exercise to exhaustion" is of course just a relative measure depnding on the tested individual and its fitness level.

This work continued

A slight sodium deficiency can impair athletic performance before any clinical signs of sodium lack are discernible. Therefore, during hot weather, adequate amounts of salt and water should be given to replace losses of these substances through the skin.
Theodore B. Van Itallie: "Nutrition and Athletic Performance", JAMA, November 17, 1956. (p 1126)

Key papers:

Molnar, G.W., Towbin, E.J., Gosselin, R.E., Brown, A.H. & Adolph, E.F.: "A comparative study of water, salt and heat exchanges of men in tropical and desert environments", American Journal of Hygiene 44, 411–433, 1946.

Adolph, A. & Associates: "Physiology of Man in the Desert". Wiley, New York, 1947.

Bass, D.E., Kleeman, C.R., Quinn, M., Henschel, A. & Hegnauer, A.H.: "Mechanisms of acclimatization to heat in man", Medicine 34, 323–380, 1955.

Buskirk, E.R., Iampietro, P.F. & Bass, D.E.: "Work performance after dehydration: effects of physical conditioning and heat acclimatization", Journal of Applied Physiology 12, 189–194, 1958.

Grande, F., Monagle, J.E., Buskirk, E.R. & Taylor, H.L.: "Body temperature responses to exercise in man on restricted food and water intake", Journal of Applied Physiology 14, 194–198, 1959

Senay, L.C. & Christensen, M.L.: "Cardiovascular and sweating responses to water ingestion during dehydration", Journal of Applied Physiology 20, 975– 979, 1965.

Moroff, S.V. & Bass, D.E.: "Effects of overhydration on man’s physiological responses to work in the heat", Journal of Applied Physiology 20, 267–270, 1965.

Strydom, N.B. & Holdsworth, D.L.: "The effects of different levels of water deficit on physiological responses during heat stress", Internationale Zeitschrift für Angewandte Physiologie 26, 95–102, 1968.

Cage, G., Wolfe, S., Thompson, R. & Gordon, R.: "Effects of water intake on composition of thermal sweat in normal human volunteers", Journal of Applied Physiology 29, 687–690, 1970.

American College of Sports Medicine: "Position statement on prevention of heat injuries during distance running", Medicine and Science in Sports and Exercise 7, vii–ix, American College of Sports Medicine, 1975.

It seems quite clear that it's been known for long in the medical literature that it's not really "water is essential for marathons", but that adequate hydration is really just one small angle to view a balanced scale.

Therefore, allow me a small link to an article that describes the current stae of trickle-down science, with a highlight on Exercise-associated hyponatraemia:

James M Winger et al.: "Beliefs about hydration and physiology drive drinking behaviours in runners", Br J Sports Med 2011; 45: 646–649. doi:10.1136/bjsm.2010.075275

And the belief angle is indeed crucial to understand why pre 1970s and especially in the very early years of the last cebtury the advice given out to runners looks very strange today to us.

Fundamental science was way advanced in its understandings compared to what applied and empirical sports science had to say about it.

The early advice was based on empiricism, but a purely observational one that was tainted by attitudes:


Contrary to popular theories and beliefs, distance runners predating 1970 were encouraged to drink as little as possible. Early studies of elite distance athletes showed the more successful athletes were the ones losing the most fluids of anyone in the race. These data concluded that a runner should not replenish the fluid lost during a race. To achieve even better results, it was suggested to have no fluid intake at all. In Tim Noakes book, Waterlogged, he talks about running a marathon in 1969 that had only water stop, at mile 20. As running became more popular, scientists conducted further research and came to some different conclusions...


Health risks of dehydration became more apparent in research after 1970, leading scientists to recommend that distance athletes drink during exercise. Frequent drinking was encouraged and promoted as the dangers of dehydration became even more evident as people suffered through injuries and even succumbed to death during races. Sports drink companies, such as Gatorade, were founded during this time period and began running advertisements in which professional athletes endorsed their product and talked about how hydration and sports drinks would lead to increased performance. (src)

No drinking….ever
Going back to the early days of marathon running, it was thought that the consumption of most fluids during long races like a marathon was not needed and even detrimental. Why? Because runners were studied and it was found that at the end of the race, the winners or top finishers lost the most body weight. The logic was that the best runners lost the most water weight, therefore losing fluids was necessary to maximize performance and hydration should not occur. The top runners were the most dehydrated, so dehydration is good! This line of thinking is used often, even to this day (i.e. The Kenyans do X, so X should be done…). This should be a cautionary tale to doing something just because the fastest guys do it. So early in the history of hydration we have a policy of no drinking. What happens next?

An Overreaction
With the rise of mass participation running, an increased awareness of illnesses associated with dehydration and the ability to measure hydration status very easily and quickly, we overreacted. The norm went from drinking nothing during exercise to trying to replace all of your fluid loss during exercise by drinking water or sports drinks. The common advice of measuring yourself before and after exercise to calculate hydration needs reached mantra status with coaches, nutritionist, trainers, and the common exerciser.

According to a nice summary by Mundel (BJSM-2011), one reason for this overreaction was the design of studies which measured the effect of drinking on tests at fixed intensities which essentially found how long you could go, and not how fast you can go over a fixed distance, which is what we do in the real world. As mentioned above, the other reason is that heat exhaustion and similar illness became more prevalent with the rise of mass participation. The thinking was simple, extreme dehydration caused some problems and helped contribute to heat exhaustions, therefore if we eliminate dehydration heat exhaustion and similar illnesses would be eliminated. The problem with this thinking is similar to the “no drinking” logic. Just because a lot of dehydration is bad, doesn’t mean we need to eliminate all of it. It’s only bad if it gets to a dangerous point outside of the norms. Until it gets to that point, which is hard to do unless you force yourself not to consume any fluids (which is what was occurring in the previous period), you are fine. You see this “all or none” thinking in a myriad of different places. Some obvious examples through history are: free radicals, carbohydrates, fat, lactate, etc. Just because a lot is bad, doesn’t mean a little is.

The history of Hydration : A lesson in the scientific method and the Hype cycle.

That resonates nicely with the mentioned above Toby Mündel; "To drink or not to drink? Explaining “contradictory findings” in fluid replacement and exercise performance: evidence from a more valid model for real-life competition", British Journal of Sports Medicine, Med 2011;45:2. http://dx.doi.org/10.1136/bjsm.2009.058594

Carl Heneghan: "Forty years of sports performance research and little insight gained", BMJ 2012; 345 doi: https://doi.org/10.1136/bmj.e4797 Cite this as: BMJ 2012;345:e4797

  • Re Ahlborg and Bergström, note that they were studying "exercise to exhaustion." This sort of runs counter to the first paragraph of your answer, where you seem to be saying that you want to consider less extreme endurance activities than marathons. My understanding of the (current) science is that for anything significantly below marathon distance, taking carbs is not necessary and doesn't enhance performance, because you aren't running out of glycogen.
    – user2848
    Commented Mar 4, 2019 at 19:20
  • @BenCrowell That should be read as 'relative exhaustion', which in untrained people happens much quicker than in trained for a marathon athletes. The point is that fundamental science was way advanced in its understandings compared to what applied and empirical sports science had to say about it. But you're right, that might need some clarification. Thinking mode activated, if meanwhile you have ideas, be my guest in edits ;) Commented Mar 4, 2019 at 20:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.