INTRODUCTION 9 



conduction and radiation, so that the temperature of the copper disk re- 

 mains essentially constant. 



This form of electric brake has proved particularly advantageous, in- 

 asmuch as it is at ordinary temperatures absolutely constant; for com- 

 parison experiments, in which the same individual uses the apparatus 

 under differing conditions of diet, or in which different subjects use the 

 same apparatus, the results are especially satisfactory, since a definitely 

 known amount of muscular activity for all subjects is obtained. Never- 

 theless it is necessary at times to know not only the relative but also the 

 absolute energy value of the work done upon the apparatus, and hence 

 it was considered desirable to calibrate the instrument in order to find 

 in absolute units the heat output per revolution of the pedals. 



METHOD OF CALIBRATION. 



The bicycle ergometer may be calibrated in a number of ways. Either 

 a cradle dynanometer can be used or the rear wheel of the bicycle can be 

 driven by means of a belt or sprocket chain, or, better still, by a gear 

 or direct-friction drive; but these methods involve so many and so large 

 corrections that the final value might be seriously in error. A method 

 for calibrating this type of instrument suggested by the late Prof. W. 0. 

 Atwater has been in use for some time and has given admirable results. 

 Professor Atwater's method involved placing the ergometer, exactly as 

 it was to be used for an experiment, inside of a respiration calorimeter, 

 and driving the apparatus from the outside by means of a shaft, magnet- 

 izing the field as desired. The number of revolutions of the pedal could 

 then be counted and the heat directly given off by the instrument meas- 

 ured as in an ordinary calorimeter experiment. 1 



This method of calibration was extensively employed on the first 

 apparatus and it has already been described. 2 The calorimeter then 

 used was the original respiration calorimeter in the chemical laboratory 

 of Wesleyan University in Middletown, Connecticut. This calorimeter 

 was a universal apparatus, inasmuch as with it experiments could be made 

 with men not only at rest but also undergoing severe muscular work. 

 The apparatus t was consequently of considerable size. While the 

 earlier check-tests showed an agreement among themselves that was highly 

 satisfactory, certain apparent abnormalities in the curve of calibration 

 have been adversely criticized by European observers in private communi- 

 cations. The earlier calorimetric calibrations of the ergometer indicated, 

 for example, that within the limits of speed used, namely, from 55 to 85 

 revolutions per minute, the heat per revolution was approximately constant, 

 with the same degree of magnetization in the field. From elementary 



1 It is interesting to recall in this connection that Violle (Comptes rendus, 1870, 70, 

 p. 1283) many years ago made determinations of the mechanical equivalent of heat by 

 expending a known amount of energy in rotating a disk of metal between the poles of an 

 electro-magnet and then quickly plunging the disk into a water calorimeter. 



2 Benedict and Carpenter, loc. cit., p. 14. 



