and, if done with due attention, it works! - it did for me on several occasions.

'The way we were' - before the mile - Loughborough vs ULAC vs Cambridge University 1963

Georges Cmela with camera - John Brotherhood ( Bro ) seated in middle - Gus Schumacher ( USA ) seated right

John Farrington ( now Aus. ) in England 'top' with Dan Tunstall-Pedoe ( often London Marathon senior MD ) on his right

Human Fuel and Water Stores and Endurance  :  How they may be Increased


In his article, " Physiological Factors in Marathon Running " ( R.R.C. Newsletter , August  1971 ), Dr. Griffith Pugh named three systems that may limit per­formance in long-distance races.

They are :

1.  Body temperature;         2. Water and fuel stores;         3. Circulation.

Of these, " water and fuel stores " are undoubtedly the most important in races of marathon and longer distance. In his paragraph on Practical Suggestions he states the " Saltin Diet" as a means of increasing endurance. Most runners have heard of this regiment but many do not know exactly what is involved. The purpose of this article is to outline the rationale behind the regimen and to des­cribe it in detail.


The Energy Cost of Running

" Energy expenditure " can either be expressed as power, that is the amount of energy or work expended per time, or it can describe the total amount of energy or work expended to perform a task. The energy cost of running speed is expressed as a power output in units of oxygen ( or Calories ) per body weight per time, so that the energy cost of running varies with speed and body weight. The total energy ( usually expressed in Calories ) required to cover a unit of horizontal distance varies with body weight, but is independent of running speed ( that is to say the faster the running speed the less total time spent at a higher energy out­put ). For middle and long-distance running, speed is limited by the maximum rate at which oxygen can be supplied to the muscles: this is known as the maximum oxygen intake. Racing speeds for 3 miles or 5,000 metres correspond to maximum oxygen intake. Speeds are higher for shorter distances since they are run at 100% maximum oxygen intake plus the oxygen debt. The average speed for an athlete's best 3 mile or 5,000 metre performance is a useful measure of his physiological ability and will be referred to in this article as maximum running speed. For reasons not yet understood, athletes are not able to maintain their maximum speed for much further than about 5,000m. As the distance increases so the proportion of the maximum speed decreases. These relative running speeds are as follows: 3-5 miles, 100%; 10 miles, 95% 15-20 miles,90%; marathon; 80-90%; 40 miles, 70-80%; 50 miles, 60-70%. The relative running speed is a more important concept than the absolute speed, since it is this that determines the body's physiological responses to the demands of prolonged exercise.

Body Fuel

The body obtains its energy from two types of fuel: carbohydrate in the form of glucose, and fat. The brain can use only glucose supplied by the blood either from glycogen stores in the liver or directly absorbed from food in the intestine. The muscles, however, can burn both fat and glucose. Fat is largely stored in remote depots under the skin and has to be transported by the blood to the muscle cells. Some fat is present in the muscle cells and can be used. The body's carbohydrate is stored in the liver and the muscles as insoluble granules of glycogen. Glycogen is a composite molecule made up of many units of glucose associated with water and potassium; it provides an immediate supply of glucose within the cell. An adequate level of blood glucose is essential to the proper function of the brain and nervous system. The blood glucose supplies only a small proportion of muscle fuel, but prolonged exercise can deplete both liver and blood carbohydrate, depriving the brain of essential fuel with disastrous effect on performance. So that, though the body has an unlimited store of energy in the form of fat, it has limited stores of carbohydrate ( glycogen ), essential to optimal performance.

Muscle Fuels

It has been known for many years that working muscles use both fat and carbohydrate as fuel. By examining the ratio of carbon dioxide expired to the oxygen absorbed by the body - known as the respiratory quotient - it is possible to determine the proportions of carbohydrate arid fat being used At light exercise levels most of the energy is derived from fat; however, in severe exercise exceed­ing 80% of the maximum oxygen intake, energy is provided almost entirely by carbohydrate. Recently it has become possible to measure directly the glycogen in minute portions of muscle removed from active men. This technique was devised by Bergstrom and Hultmann, and they and Saltin have studied the effects of diet and exercise on muscle glycogen. They have demonstrated that the rate of use of glycogen in the muscles increases with work rate. At a constant work rate the total amount of work that can be done to exhaustion is dependent on the amount of glycogen in the muscles at the start, and at exhaustion the muscles are virtually empty of glycogen. Furthermore, the nearer the work rate is to the maximum the less total work can be done for an initial glycogen level. When the muscle glycogen is greatly depleted, work can be continued, but only at a consider­ably reduced rate. Endurance, then, is directly determined  by the amount of glycogen present in the muscles, and the nearer to maximum that the athlete runs the shorter the distance that he can run on the available glycogen. This is almost certainly borne out by long-distance runners' performances over races by varying length. Races of 10-15 miles can be run at 90% without drop in speed, but many athletes find that they are not able to maintain level pace after about 20 miles at 90% in marathons; in 40-mile track races runners lose speed after about 27 miles at 70% and about 30 miles in 50-mile races; in the 100-mile track race the runners lost speed after running 35 miles at speeds corresponding to 70% of their maximum. After this failure runners are not able to maintain speeds exceeding 55-60%, which is about 7 m.p.h. Though there is not at present any direct evidence that this failure to maintain speed is due to muscle glycogen exhaustion, it seems most likely. Assuming that these drops in running speed are due to glycogen exhaustion, it is possible to estimate the glycogen used and the requirement for any distance. For the average athlete the energy cost of run­ning one mile is about 100 Calories. The proportion of carbohydrate to fat used increases with relative running speed. A man running a marathon at 90% of his maximum oxygen intake is burning virtually pure carbohydrate, so that for each mile he uses 100 Calories of carbohydrate. One gram of glycogen provides 4 Calories, so that 1 mile consumes 25 gm of glycogen, and he requires 25 x 26 = 650 gm of glycogen to maintain 90% for the full marathon. Many marathon runners find that they are not able to maintain speed after 20 miles, so that it appears that their glycogen store is 20 x 25 = 500 gm. Similarly it is estimated that the 100 mile runners started the race at about 70% of their maximum; at this relative speed about 60% of the energy is supplied by carbohydrate, so that 60 Calories per mile come from glycogen. Their running speed dropped irretrievably after 35 miles or the probable expenditure of 15 x 35 = 525 gm of glycogen. It appears, therefore, that long-distance runners eating a normal mixed diet can store about 500 gm of glycogen. Athletes who have used the Saltin procedure have been able to sustain relative running speeds of 80% for at least 40 miles.


Water Stores

An athlete running at 90% of his maximum speed may lose 4lb. of sweat an hour; weight losses of 9lb. are not uncommon in marathons on hot days. When an athlete suffers water losses of this magnitude he is not able to maintain his speed. A large amount of water is stored in association with glycogen, about 2.5 gm to every gram of glycogen; a further 0.5 gm of  " water of combustion" per gram is also produced by metabolism, so that glycogen provides three times its weight in water.

It is not difficult, then, to appreciate that great advantages may be gained if the long-distance runner were able to increase his muscle glycogen stores.


Factors relating to the Size of the Glycogen Stores

It has been shown that exercise reduces muscle glycogen. It is important to know how long it takes to restore the muscle glycogen to normal levels. The nature of the food eaten has a most important influence on the restoration and the magnitude of the glycogen stores. Hultmann and Bergstrom have studied the effect of diet on muscle glycogen content. They have found that the normal resting level of glycogen for people eating a mixed diet is 1.5-2.0%. With no exercise, but eating a low carbohydrate diet, or starvation, muscle glycogen falls steadily over 4 to 5 days from 1.5% to 0.9%; when a high carbohydrate diet ensues, the level returns to 1.5%. If the muscle glycogen is depleted by pro­longed exercise, the levels remain low with a low carbohydrate diet or starvation over a number of days. But if the exercise is immediately followed by a high carbohydrate diet, glycogen rises within 48 hours to levels exceeding the initial. Their most significant finding was the combined effect of exhausting exercise followed by a low carbohydrate diet to maintain low glycogen levels for 2-3 days, followed by a high carbohydrate diet. This procedure resulted in a rapid increase of glycogen far exceeding the initial levels and in the order of 4-4.5%. It appears that during the period of glycogen depletion the muscles build up a substance that enables them to restore very much more glycogen than normally. They have also demonstrated that the liver glycogen responds in the same way. Endurance at all work levels was greatly increased with these high glycogen levels. Saltin demonstrated that cross-country skiers could reduce their time for a 30km ( 16 mile ) race from 143 to 135 minutes using this procedure.


Methods of increasing Muscle Glycogen

It appears that an athlete eating a normal mixed diet with adequate carbo­hydrate can race up to 15 miles without running out of glycogen. For races of this distance it is sufficient for the athlete to reduce his running to zero or minimal and to eat plenty of carbohydrate 24 hours before the race. Similarly a high carbohydrate intake with normal training 3 days before will probably enable the athlete to maintain level pace up to 20 miles. For distances further than this the full exhaustion, low carbohydrate, high carbohydrate regimen must be used.


A Regimen to increase Muscle Glycogen Stores

This procedure lasts eight days, starting on Day 1 with a glycogen depleting run, and ending on Day 8 with the race. The depleting run is followed by 3 days normal training and a low carbohydrate diet; this is followed by 3 days of reduced exercise and a high carbohydrate diet. Suggested running speeds for 25 mile glycogen depleting run, assuming marathon to be run at 90% maximum :

Maximum                 Marathon                                                             25 miles

running speed       Speed 90%                                                                 75%

m.p.h.                                            m.p.h.                                                                                                 m.p.h.

13.2                                                     12                                                                                                     10

12.7                                               11.5                                                                                   9.5

12.1                                                     11                                                                                                     9.1

11.5                                                 10.5                                                                                          8.7

11                                                         10                                                                                                     8.3

If higher speeds are used, the mileage may be reduced. Splitting the run into sections of 75% and racing speed will also reduce distance required.



Normal training should continue as far as possible during the low carbo­hydrate phase. During the 3 days high carbohydrate phase prior to the race, the amount of exercise should be greatly reduced both in miles and speed, and to an absolute minimum 24 hours before the race.



The low carbohydrate diet ( sometimes known as " high protein and fat diet" ) must starve the individual of carbohydrate. The permitted foods are listed below  :

Meat, bacon and offal, fish, eggs, cheese, butter, cream, vegetable oils, margarine, mushrooms, asparagus; in unlimited quantities. Moderate portions of the following fruit and vegetables may be eaten in addition: French and runner beans, cabbage, cauliflower, broccoli, spinach, celery, marrow—all boiled; endive, lettuce, mustard and cress, watercress, tomatoes as salad; rhubarb and goose­berries, saccharine - sweetened only; lemons and grapefruit, olives. No sugar, sweet or starchy food or drink should be taken, and only small quantities of milk in beverages. This diet should not restrict the total amount of food normally eaten - merely reduce the carbohydrate intake to very low amounts.

The high carbohydrate diet: the aim here is to eat as much carbohydrate as possible without greatly exceeding usual food intake. Food rich in starch and sugar should predominate at every meal, and! the quantities of protein and fat eaten be reduced. Spices and meat extracts can be used for flavouring. Though the diet need not be protein free, it must be remembered that meat takes up " stomach" space that may otherwise take carbohydrate. When preparing for races of marathon length there is probably no need to take extra carbohydrate other than at normal meal times, but for longer races extra sugar, cakes, etc., between meals may be beneficial.

Plenty should be taken to drink throughout the procedure and especially during the high carbohydrate phase, when electrolyte replacers could be taken with benefit.


Pre-Race Meal

Should be eaten at the time before the race that the athlete has found best by experience ( 3-4 hours ). It should not be too heavy and should have a pre­ponderance of carbohydrate rather than protein or fat; plenty should  be drunk at this meal. The athlete might also with advantage continue to drink sweetened drinks until within an hour of the race, especially on a hot day, to ensure that he is fully hydrated and that he has a good blood sugar at the start of the race.


After the Race

Diet and training return to normal.


Practical Considerations

Weight changes: It has already been stated that glycogen has 2.5 times its own weight of water associated with it, so that there will be considerable changes in weight during the course of the regimen. All athletes should weigh themselves in a standard fashion as precisely as possible on bathroom scales, nude and with their bladder empty on rising in the morning, at which time they should normally be fully hydrated. They will find their weight remarkably constant. After the glycogen-depleting run and during the low carbohydrate phase their weight will be 3 – 4lb. lower than normal. When they go on to the high carbohydrate diet their weight will increase -very rapidly for 2 days 6 – 7lb. and then less rapidly as glycogen and water is taken up in the muscles. Much of this increase will be in the legs, and they may feel rather heavy and stiff. Marathon runners should aim to keep their weight increase to about 3lb. over normal: when they have attained this they should reduce the amount of carbohydrate they are eating. Ultra-long distance runners should gain as much weight as possible in 3 days.

General effect of glycogen depletion: Just as the muscle glycogen is depleted by exercise and low carbohydrate diet, so also will be the liver stores and the blood glucose. This may result in a variety of relatively unpleasant symptoms. The athlete may feel tired and weak and his training performance will almost certainly deteriorate. He may also suffer headaches and spells of dizziness or faintness, and find concentration difficult. Individuals who work with machinery or drive motor vehicles should be particularly aware of these possibilities; it may not be possible for them to do this regimen and continue their normal activities.

These symptoms are due to low blood sugar, and may be quickly overcome by taking sugar or carbohydrate food. Any individual who uses this regimen should carry sugar with him during the low carbohydrate phase and take it if he  has a severe attack of these symptoms. If he finds that his training performance is reduced too much, he may find it an advantage to have a sweet drink or small amount of carbohydrate before his run.


Practical Experience

There is very little published information on athletes' practical experience of this regimen. From the information in this article athletes should be able to work out routines that best suit themselves. Certainly all endurance athletes should experiment with this procedure. It is not known how frequently the regimen may be used with benefit.


Some runners would undoubtedly be interested in reading the original scientific publications, and a short list of the most important papers is given below.



Review Article :

Bergstrom, J., Hultman, E. (1972). Nutrition for maximal sports performance. Journal of the American Medical Association: Vol. 221 No. 9. 999-1006.

Scientific Publications :

Hultman, E., Bergstrom, J. (1967). Muscle glycogen synthesis in relation to diet studied in normal subjects. Acta. Med. Scand. 182, 109-117.

Ahlberg, B., Bergstrom, J. et al (1967). Human muscle glycogen content and capacity for prolonged exercise after different diets. Forsvarsmedicin (Stockholm) 3, 85-99.

Bergstrom, J., Hultman, E. (1969). Muscle glycogen synthesis after exercise: an enhancing factor localised to the muscle cells in man. Nature. 210, 309-310.

Karlsson, J., Saltin, B. (1971). Diet, muscle glycogen and endurance per­formance. J. Appl. Physiol. 31, (2) 203-206.

Saltin, B., Karlsson, J. (1971). Muscle glycogen utilization during work of different intensities. In: Muscle Metabolism during exercise. New York, Plenium 1971, Vol. 11, 289-300.


Back to the runCW Home page