FOODS HUMAN NUTRITION. 367 



Calorimetric observations on man, J. S. Macdonald {Jour. Physiol., JfJ^ 

 (11)12), No. 1-2, pp. IV, V). — A brief account is given of a calorimeter modeled 

 on the Atwater-Rosa-Benedict apparatus devised at Middletown, Conn., having 

 the same general dimensions, and protected from surface loss or gain of heat 

 in the same manner. The calorimeter is not equipped to deal with the respira- 

 tory exchange of gases and is therefore not a respiration calorimeter. 



" Departures from the original model have been made, notably in an ex- 

 tension of the internal radiator system and the insertion of an additional 

 internal source of heat. The latter arrangement is used as a compensating 

 system w^hereby the total heat measurement is maintained as far as possible 

 a constant. The heat output of the subject is found by subtracting from this 

 the momentary .values for the heat output of the compensating system. 



" This change in method is due to experimental observation of a definite re- 

 lation between the temperature of the calorimeter (Tc) and the mean tem- 

 perature of the internal system (Tr) such that the heat measured by reference 

 to the temperature and quantity of the water stream = k (Tc — Tr). 



"The value of "k" in this calorimeter was such as to render difficult all 

 attempts to follow abrupt variations in the subject's heat output. Thus when 

 a period of work interrupted a period of rest it became evident that Tc — Ti- 

 must have a new value of perhaps 8° C, entailing a sudden change in the tem- 

 perature of the water entering the radiator system of 10° C. Because of the 

 slow rate, and the lack of precision, with which such alterations in temperature 

 could be obtained the preceding method was adopted." 



The data obtained every 5 minutes throughout an experiment are of three 

 kinds: "(1) Heat stored in, or lost from, the calorimeter walls as inferred 

 from observed variations in temperature; (2) heat removed by water travers- 

 ing an internal radiator system; (3) heat stored in a change of state in water, 

 and measured by reference to the excess aqueous vapor leaving the calorimeter 

 in the air stream." 



Experiments have been performed with a number of subjects under alternat- 

 ing conditions of work and rest and in two instances with a subject asleep. 



According to the author, the most interesting features of the results ob- 

 tained, briefly stated, are as follows: 



" The performance of work entails an increased heat output which gradu- 

 ally rises toward a maximum reached at the end of about 1^ hours. This 

 maximum is then maintained until the cessation of work. 



" In order to discover the degree to which this initial portion of the heat 

 output curve is rendered gradual by a storage of heat in the subject's body 

 as evidenced by a coincident rise in body temperature, the latter has in the 

 most recent of these experiments been measured . . . [by a thermocouple 

 method]. We have not, so far, found sufficient variations in temperature to 

 explain the whole effect, though their nature is always such as to explain some 

 fraction of this. It would appear, then, that the ' efficiency ' diminishes as the 

 temperature of the body increases." 



A new calorimeter for small warm-blooded animals, A. V. Hill (Jour. 

 Physiol., 1/4 {1912), No. 1-2, pp. I, II). — A small calorimeter for warm-blooded 

 animals is described, the essential details being: "(1) A large cylindrical 

 Dewar flask of depth 40 cm. and diameter 10 cm. ; (2) a coil of fine copper 

 tubing inside the flask, through which water runs at a uniform rate to take 

 away the heat formed; (3) a thermocouple with its ends in (a) the water 

 going into the flask, and (b) the water coming out of the flask; (4) a gal- 

 vanometer in another part of the laboratory registering the current through 

 the thermocouple; (5) a Mariotte bottle supplying water at a uniform rate; 



