In another series of experiments they subtracted from the expenditure of energy during work, the amount spent when the subject, instead of lying on a couch, sat on the ergometer and allowed the pedals to be turned under his feet. Using this method of estimation they were able by careful adjustment of speed and load to realize with a professional bicycle rider an efficiency of 33 per cent or as much as Zuntz and his associates had estimated from the walking experiments.

Only under the most favorable circumstances and with subjects fully accustomed to the kind of work being performed will the actual mechanical effect produced amount to as much as one fourth to one third of the extra energy expended during work over that during rest, i.e. to an efficiency of 25 to 33 per cent. Not only do most occupations involve kinds of work which in their nature must be done with less efficiency than walking (or riding a stationary ergometer) but the usual hours of labor are longer than those in which the maximum mechanical efficiency is attained. The efficiency may begin to decline before any sensation of fatigue is felt.

Thus Leo Zuntz found, when he rode his bicycle for four successive hours at an average rate of 15 to 17 kilometers (about 9 miles) per hour, that he experienced no feeling of fatigue, but his determinations showed that the expenditure of energy necessary to produce a given effect had increased about 9,13,10, and 23 per cent at the end of 1, 2,3, and 4 hours respectively. This is because if the same kind of work be performed for a series of hours, auxiliary muscles are gradually brought increasingly into action, partly for the performance of the work itself, partly for the fixation of the bodily framework (maintenance of posture). These auxiliary muscles work less economically than those which are used first and most naturally. For much the same reasons there is a lower efficiency in the case of work which is from the first of too fatiguing a nature because of being either excessive or unsuitably distributed. When Leo Zuntz increased his speed 2.4 times, he found his metabolism increased 4.3 times, implying a considerable loss of efficiency. Under the conditions of Benedict and Cathcart's experiments also, the efficiency was usually decreased upon increasing the speed; on the other hand a moderately heavy load was more economical than a light one.

From the data determined by Atwater and Benedict, Lusk, Becker, and their respective collaborators, it is now possible to estimate the approximate average expenditure of energy per hour under a considerable number of conditions of muscular activity. For convenience of comparison and application the original data have been reduced to a common basis of a man of 70 kilograms (154 pounds), then averaged and the average approximated to the nearest "round" number, with the results shown in the accompanying table.

Energy Expenditure Of Average-Sized Man (70 Kilograms) Per Hour Under Different Conditions Of Activity. (Approximate Averages Only)

By the use of these estimates the probable food requirement for a person of 70 kilograms (154 pounds) may be calculated very simply, as, for instance, in the following example:

Tigerstedt, in his Textbook of Physiology, gives estimates of food requirements for different degrees of activity as indicated by means of typical occupations, which may be useful in checking results calculated as above.

* Going to and from work, for example.

According to Tigerstedt:

2000-2400 Calories per day suffice for a shoemaker. 2400-2700 Calories per day suffice for a weaver. 2700-3200 Calories per day suffice for a carpenter or mason. 3200-4100 Calories per day suffice for a farm laborer. 4100-5000 Calories per day suffice for an excavator. Over 5000 Calories per day are required by a lumberman.

Lusk gives the following summary of energy requirements of women at work at typical occupations as investigated by Becker and Hamalainen in Finland:

A seamstress sewing with a needle required 1800 Calories.

Two seamstresses, using a sewing machine, required 1900 and 2100 Calories, respectively.

Two bookbinders required 1900 and 2100 Calories.

Two household servants, employed in such occupations as cleaning windows and floors, scouring knives, forks, and spoons, scouring copper and iron pots, required 2300 to 2900 Calories.

Two washerwomen, the same servants as the last named, required 2600 and 3400 Calories in the fulfillment of their daily work.

Benedict and Cathcart find that when muscular work is severe there is a rise in the respiratory quotient, the rise being greater the more severe the work. In such cases the respiratory quotient is found to fall during the rest period following the work, and usually to a lower figure than that observed before the work was begun. They interpret this to mean that hard muscular work draws upon the stored carbohydrate of the body in slightly greater proportion than upon the stored fat. That the work is performed at the expense of both carbohydrate and fat is shown by Benedict and Cathcart's data as well as by those of many previous experiments. Apparently it is only severe muscular activity which has any appreciable influence upon the relative proportions of fat and carbohydrate burned. In the experiments cited on page 181, for example, the respiratory quotient was not changed by walking either on a horizontal surface or up an inclined plane. It should also be noted that Benedict and Cathcart found the same mechanical efficiency in work whether preceded by a carbohydrate-rich or a carbohydrate-poor diet.