102 Mr. J. S. Macdonald. Studies in the 



The staff at work upon the calorimeter in most of my earlier experiments 

 consisted of two men ; later, and in the whole of the experiments to which 

 reference is made here, of three men. I must mention in the first place my 

 laboratory assistant, Mr. A. Wallis, who has been a most valuable aid in the 

 construction of the calorimeter and in the design and construction of 

 accessory apparatus. In these experiments he has been responsible 

 throughout for the balance of the calorimeter surface, and has kept records 

 of the currents due to the various sets of thermocouples of very considerable 

 importance to a discussion of the data. In the second place the observer, 

 assisted by a note-taker, who recorded the observations as they were made 

 and immediately dictated. At the present time this staff is increased by 

 Dr. Duffield, who is now responsible for the gas-analysis side of the 

 investigation. If I may forestall one of his measurements here, it is to 

 record the fact that the air space of the calorimeter is 176 cubic feet. To 

 obtain this measurement carbonic acid gas was delivered into the calorimeter, 

 the enclosed atmosphere then thoroughly mixed by the fan, and its percentage 

 content of CO2 obtained from a sample. The rotary pump was then started, 

 and the total amount of the gas ascertained as it was gradually withdrawn 

 with the air current. The figure agrees very closely with that obtained by 

 measurements of the average length, width, and depth of the calorimeter, 

 but is of value because of uncertainty due to irregularities in the walls and 

 because of a subtraction made necessary by the presence of the radiator 

 system, etc. 



A most important part of the method has been the securing of financial 

 assistance, and in this connection I owe much to the British Association. 



Section B. — Results. 



The accompanying figure may serve to illustrate my need for considering 

 the results of these experiments under two headings : (1) the events of the 

 first hour of cycling ; (2) the events of the second hour. Fig. A represents 

 an average of data obtained in 24 separate experiments upon E. J. Briscoe 

 at various known revolution rates of the cycle, and with various known 

 values of the electrical brake, covering the whole range of work -performance 

 upon the cycle which he could maintain uniformly for the required length 

 of time. The lowermost curve in the figure is that of the heat-production ; 

 the middle curve gives his surface temperature ; the uppermost is the rectal 

 temperature. Fig. B is the average curve of such of these experiments — 

 14 in number — as are below the general average of the full series. Big. C 

 is the average curve of the 10 experiments which lie above the general 

 average of heat-production. 



