ON MENTAL AND MUSCULAR FATIGUE. 175 



and have therefore had to pay nothing more than secondary attention 

 to the amount of water condensing on, or evaporating from, the radiator 

 in the calorimeter. "We have thus been set free from any necessity for 

 weighing the water that is condensed within the calorimeter. 



Measurements of the water vapour leaving the calorimeter were 

 obtained at first by direct weighing of absorption apparatus placed in its 

 path, but we have now substituted ' wet and dry ' bulb readings taken 

 in an accessory chamber of suitable dimensions, and apparently with 

 advantage. 



One of the greatest difficulties experienced at first was the great 

 delay in the external delivery of heat which was due to the great heat 

 capacity of the calorimeter. This has been satisfactorily eliminated by 

 observations which have given us its ' water equivalent ' and a means 

 of rendering all our figures free from this source of error. Using these 

 corrections we are enabled to express our experimental results in con- 

 tinuous curves showing variations in heat production of quite short 

 duration. 



That item in the construction of Atwater and Benedict's calorimeter 

 which appeared at first as likely to need an extraordinary amount of 

 experience in handling — namely, the equalisation of the temperature 

 of the metal box surrounding the calorimeter to the temperature of the 

 calorimeter in its several sections— we have found much more simple 

 than was anticipated, and have now proof that it is managed with 

 perfect success. 



The only occasions when difficulties arise in this matter are when 

 very great changes occur in the amount of heat produced within the 

 calorimeter. To meet these special difficulties we have placed within 

 the calorimeter an extra source of heat, a number of incandescent 

 lamps, using them frequently to balance such violent changes. We 

 have considerable evidence to support the statement that no new errors 

 of moment are introduced by this plan. It has the further advantage 

 that the interior of the calorimeter is lighted and its occupant always 

 visible. 



In our more recent experiments we have been enabled to take the 

 records of the occupants' temperature using the thermo-couple method 

 elaborated by Gamgee, and have thus now all the data required for an 

 estimation of the heat production of man as distinguished from his heat 

 elimination. 



In expressing the results of experiments it has been found essential 

 to take full account of the surface of each subject; a consideration of 

 the published results of others and of our own direct measurements 

 have led us to adopt a modified formula for the surface in terms of the 

 data of height and weight 



S=3H^W 



A large number of experiments have been carried out with a number 

 of subjects under conditions of rest, sleep, and work, and we hope in a 

 short time to publish a full account of these experiments and their 

 bearing upon the questions referred to this Committee. 



