68 EXPERIMENT STATION RECORD. 



by alcohol and electric tests, can be measured with an average error of 0.9 per 

 cent, 0.6 per cent, and 1.6 per cent, respectively. In periods 1 hour long the 

 average error for heat measurement was 1.2 per cent, for carbon dioxid 1.6 per 

 cent, and for oxygen 3.2 per cent." 



Clinical calorimetry. — IV, Tlie determination of the basal metabolism of 

 normal men and the effect of food, F. C. Gephakt and E. F. DuBois {Arch. 

 Int. Med., 15 (1915), No. 5, pp. 835-S67). — Employing the apparatus described 

 in the previous article, measurements of the basal metabolism of 7 normal 

 men were made to supply control data for use in the intended study of metabo- 

 lism under pathological conditions. 



The average basal metabolism (at perfect rest, 14 to 18 hours after eating) 

 was 34.8 calories per hour per .square meter of body surface ; 34.7 calories has 

 been adopted as the average for normal men between the ages of 20 and 50 years. 



" All of the subjects studied in the bed calorimeter were within 11 per cent 

 of this average. . . . 



" The conclusion is drawn that among groups of men of varying weights 

 metabolism is proportional to surface area according to Rubner's law and is 

 not proportional to body weight. By using the surface area as a basis one 

 can refer all individuals to a single average normal figure, 34.7. If one uses 

 tlie body weight as a basis a different normal figure is required for each 

 weight. 



" The methods of direct and indirect calorimetry in disease agree in 2 and 

 3 hour periods, and in health may be found to agree in hourly periods. In 

 the total measurement of 4,577 calories in the experiments reported in this 

 paper the two methods have agreed witliin 0.17 per cent. In a total of 30 

 one-hour periods on one normal subject the two methods have agreed within 

 5 per cent in 21 individual hours and within 10 per cent in 27 of the periods. 



" The method of indirect calorimetry, using the oxygen consumption as a 

 basis, gives the best results in hourly periods. The method of direct calorimetry 

 in short periods is made difficult by imcertainty as to the correct specific 

 heat of the body and also by the fact that the different parts of the body do 

 not always change their temperatures at the same rate. . . . 



" The most satisfactory method of determining the effect of food in increasing 

 heat production in normal subjects and patients is to determine the basal 

 metabolism at frequent intervals, and on days shortly after a basal determina- 

 tion administer the food before the subject is sealed in the calorimeter. It has 

 been found that 200 gm. of dextrose or its equivalent in commercial glucose, 

 or a casein meal with 10.5 gm. of nitrogen, increase the heat production by 

 about 12 per cent over a period of 3 to 6 hours." 



Clinical calorimetry. — V, The measurement of the surface area of man, 

 D. and E. F. DuBois {Arch. Int. Med., 15 {1915), Xo. 5, pp. S6S-8S1, figs. 2).— 

 A method for measuring the area of the body surface is described in detail, and 

 the data of five different measurements reported. The total surface area as 

 determined by this method has been found to vary with that calculated from 

 a new formula by an average error of 1.7 per cent. The new formula is based 

 on the factors of length and average breadth, rather than on Avelght. 



The protein need of infants. — Being metabolism, studies of a two months' 

 old infant fed with varying proportions of cow's milk protein, B. R. IIooni.ER 

 (Amer. Jour. Diseases Children, 10 {1915), No. 3, pp. 153-171). — The subject 

 of these experiments, a robust, healthy boy, was under observation for 16 days. 

 The metabolism was determhiod for 37 periods of 1 hour each. A special 

 respiration chamber, previously described by Murlin (E. S. R., 32, p. 860), was 

 used during some of the periods, and during the remainder a small calorimeter 



