The possibility of the basal metabolism being a physiological constant has been much discussed. From the earliest measurements of the metabolism of man and animals, investigators have attempted to find if uniformity existed in the metabolism of different human individuals and even different species of warm-blooded animals. In the experiments made with the respiration calorimeter at Wesleyan University direct measurements of the heat production of man were made for the first time and the complete uniformity of direct and indirect calorimetry was first established. In common with other investigators at that time, no attention was paid to the importance of controlling the minor muscular activity in the restricted confines of the respiration chamber. The data obtained in these measurements were carefully searched to find if evidence existed of uniformity in the metabolism of individuals. It was soon found that with individuals inside the respiration chamber, living substantially the same routine of life, the metab- olism was relatively constant with the same individual from period to period. This was made the subject of discussion by one of us in conjunction with Dr. T. M. Carpenter,1 but even at this time differences in individuals, and particularly in individuals of different weight, were strikingly emphasized.

1 Benedict. Journ. Biol. Chera.. 1015. 20, p. 263.

2 Harris and Benedict. Carnegie Inst. Wash. Pub. No. 279, 1919.

Recourse was had by earlier writers to the comparison of the metabolism on the basis of per kilogram of body-weight, on the theory that a large animal would give off more heat than a small animal, and heat production per kilogram would thus be a better index than heat production per individual. This, of course, is based upon the arbitrary assumption that each kilogram of weight has the same heat-producing power. Although for general purposes the heat production per kilogram of body-weight was found to be reasonably constant, many writers were of the opinion that the heat production per square meter of body surface was a much better index than the heat production per kilogram of body-weight. For such comparisons the body surface was computed by the now archaic method of Meeh,2 using the formula 12.312 3√(body-weight).2 Much of the evidence implied that the discrepancies between individuals were in large part eliminated when the calculations were based upon body-surface. Indeed, many writers considered that they were so completely eliminated as to establish a "law of surface area".

The so-called "law of surface area" has had a rather remarkable history. Warm-blooded animals have a temperature which is usually much higher than the environment. It was argued that heat was lost to the environment in proportion to the extent of the body-surface and that for equal surfaces the heat loss would be very nearly the same. In other words, since the heat produced inside the body very nearly compensates for the heat lost, thus resulting in a practically constant body temperature under all conditions of external environment, the natural inference was that the heat production was determined by the heat lost, and the heat lost was, in turn, determined by the number of square centimeters of surface exposed to the environmental temperature. The early promulgation of this idea by Rameaux1 and later by Berg-mann,2 unsubstantiated, it is true, by experimental evidence, was followed in 1883 by a series of remarkable experiments by Rubner,3 who altered the environmental temperature and studied the basal metabolism under these conditions. The law of surface area, as finally enunciated by Rubner, and almost simultaneously by Richet,4 was to the effect that the heat production of warm-blooded animals is essentially proportional to the surface. So attractive did this general thesis appear that E. Voit5 made the claim, based upon computations and fragmentary metabolism measurements on various animals, that this law held true whether the living organism was a horse or a hen. In other words, he computed that approximately 1,000 calories per square meter per 24 hours were given off by an animal. This figure became so fixed in the minds of physiologists as to be almost a fetish, and every effort has been made to utilize it for practical purposes, particularly in the comparison of pathological measurements with "a normal standard".

1 Benedict and Carpenter, Carnegie Inst. Wash. Pub. No. 126, 1910, p. 105. 2 Meeh, Zeitschr. f. Biol., 1879, 15, p. 425.

The desirability of having a standard figure for comparison with pathological cases admits of no argument. That such a standard figure actually exists is, however, open to serious argument, for it requires the assumption that there is a constant basal metabolism per unit of body-surface. Furthermore, the advocates of the law of surface area give little recognition to the fact that at least 25 per cent of the heat produced during conditions of repose is lost by the vaporization of water from the lungs and skin, warming of the expired air, etc.

One difficulty in interpreting metabolism data has been the lack of a sufficient number of individuals who have been studied under comparable conditions to provide values with the high degree of accuracy required for the deduction of such important factors as the relationship between the heat production and the body-weight or the heat production and the body-surface. Recently values obtained with over 150 men and women were brought together and charted.6 These measurements were made for the most part in the Nutrition Laboratory, and all were secured with the subjects in complete muscular repose and in the post-absorptive condition. The general picture presented by these values was far from indicating constancy. As would be expected, large men as a rule produced more heat than small men and large women more than small women. But when the calculations were made on the basis of per kilogram of body-weight, it was found that the heat production even on this basis varied for the men from 19.7 to 32.3 calories, with an average value of 25.5 calories. Nearly one-third of the observations fell outside of the extreme limits of plus or minus 10 per cent. Thus, on the basis of per kilogram of body-weight, there appeared to be no evidence with a sufficient degree of constancy to establish a "law".

1Rameaux. Bull. Acad, de med.. Paris, 1838-39, 3, p. 1094; Bull. Acad. roy. d. sc. de Brux., 1839. 6, p. 121; Mem. Couron. Acad. d. sc. de Belgique, Brux., 1855-58, 29, 64 pp.

2 Bergmann, Uber die Verhaltnisee der Warmedkonomie der Thiere su ihrer Grosse. Gottingen, 1848.

3Rubner. Zdtachr. f. Biol., 1883. 19. p. 535.

4 Richet, La chaleur animate, Paris, 1889. His earlier writings are here summarised.

5Voit. Zeitachr. f. Biol., 1901. 41. p. 120.

•Benedict. Emmea. Roth, and Smith, Journ. Biol. Chem., 1914, 18, p. 139; Benedict, ibid., 1915, 20. p. 263.

Closer analysis of certain of the figures showed that in a group of athletes practically no values were found in the lower range, and that the heat production for the group lay for the most part somewhat above rather than below the average value of 25.5 calories per kilogram of body-weight. The measurements obtained for the women indicated a lower metabolism per kilogram of body-weight than the metabolism of men of corresponding height and weight. Like comparisons on the basis of per square meter of body-surface showed similar lack of constancy. From these values, therefore, it would be perfectly legitimate to conclude that athletes as a class have a somewhat higher metabolism than non-athletic individuals of the same height and weight and that women have a somewhat lower metabolism than men. Thus, we have the first definite proof of differences in the metabolism of different classes of individuals.

This lack of constancy in metabolism is further confirmed by an examination of all our data obtained throughout many years of experimenting, new-born infants, young children, youths, and elderly individuals alike indicating very considerable variations from the so-called standard or normal figures. In recent years the attempt has been made to recognize these variations and to replace the single standard by a convenient scale of standard figures which should take into consideration age and sex. This is certainly a step in the right direction, but must be looked upon as a preliminary to the abolishment of the surface-area law. An extensive biometrical treatment of the basal metabolism data obtained in the Nutrition Laboratory is in press. From these data a series of tables has been derived which may be used for the prediction of the probable metabolism of men and women of varying weights, heights, and ages.

Method Of Presenting Data For Basal Metabolism

While there has been much quibbling over the method for presenting the data for basal metabolism, some writers stoutly maintaining that a basal metabolism determined in short periods should be expressed in values per half hour or per hour, the fundamental computation of basal metabolism must, in the last analysis, deal with the 24-hour period.

Furthermore, a great many people spend over one-half of their time either in bed or sitting quietly with minimum muscular activity. Since sitting involves an increase of but 10 per cent above the basal, a considerable proportion of the 24 hours of the day may be properly represented either by basal metabolism or by a slight percentage above it.

For the great needs of this Nation the fundamental basal metabolism per 24 hours is a factor of prime importance. For the physiologist the basal metabolism has even a greater refinement of definition. Theoretically, the basal metabolism is the minimum metabolism, but this is rarely observed in man, and the minimum metabolism compatible with life may be very much lower than the ordinary basal metabolism of a normal individual. However, it is commonly assumed that the basal metabolism, measured during periods of complete muscular repose, 12 hours after the last meal, and with the subject in deep sleep, represents the minimum, normal basal metabolism. The factors influencing this may be divided into two classes: (1) extraneous or superimposed factors, such as muscular work and the ingestion of food, and (2) inherent factors, such as sex, age, composition of the body (proportion of muscular tissue), condition of sleep versus awake, and disease. Many of these may pertain to the same individual at different periods of life. In studying groups it is important to note the influence, if any, of sex, age, state of being awake or asleep, and also the composition of the body, i. e., a study of individuals who are distinctly fat as compared with those who are distinctly thin. Finally, there are ever-increasing data with regard to the influence of disease upon metabolism, but the opportunity is infrequent for studying this with a single individual, for rarely can a subject be observed in health and again subsequent to acquiring a disease, such as diabetes and hyperthyroidism. One must therefore resort to a study of groups of normal individuals and compare the results with groups of individuals having the disease. But a study can be made with a single individual on the influence of sleep, the digestion of food, and muscular work. All three factors influence metabolism in an increasing degree. So, by insisting upon complete muscular repose and the absence of food, the two most prominent factors influencing basal metabolism are eliminated. There is still left sleep. Thus far the studies on metabolism in deep sleep are extremely few and are practically limited to observations with the respiration-chamber method.