Heat-production Associated with Muscular Work. 



99 



Now I am aware of the fact that each of these additional sources of heat is 

 also an additional source of error, subtracting to some degree from the 

 theoretical precision of the instrument and diminishing its value as gauged 

 by the measurement of any standard source of heat. Tn the practical usage 

 of the instrument as applied to varying sources of heat they have a great 

 value, which, in my opinion, fully justifies their insertion. Of the value of 

 the lamps I have already spoken. The fan is used to keep the air inside the 

 calorimeter in a state of satisfactory admixture, so that samples withdrawn 

 from it in the general air-current are given a value in reference to the general 

 air within the calorimeter of a simple and easily determined kind, and may 

 thus be used to estimate changes in the value of this air. In these particular 

 experiments this point is only of importance in dealing with the storage of 

 aqueous vapour within the calorimeter. 



Air is made to enter the calorimeter through a tube guarded by thermo- 

 couples, a connected set being placed in the exit tube. Thus any difference 

 between the temperature of the entering and leaving air is known. Means 

 lor warming the entering air are provided, but I have not so far made any 

 arrangements for cooling it when necessary. The adjustment of these 

 temperatures has not then always been as good as might be, but they are also 

 observed by means of mercurial thermometers and any observed difference 

 allowed for. 



Air is sucked out of the calorimeter by a rotary pump at a measured rate, 

 varying in these experiments from 300 to 450 cubic feet per hour. In the tube 

 forming the path of the leaving air, and in the similar tube for the entering 

 air, a dilatation is provided in the form of an interpolated glass box. In 

 these glass boxes " wet and dry bulb " thermometers are placed, the stems 

 outside for observation, so that an estimate may be formed of the excess 

 aqueous vapour derived from the calorimeter. In my earlier experiments the 

 water-vapour was weighed after absorption by sulphuric acid, or rather a 

 fraction obtained from a section of the air-path was thus treated, but this 

 plan was abandoned after comparative trial, mainly as not lending itself so 

 well to the observation of five-minute periods of heat-production. Here 

 perhaps the interests of absolute precision have apparently suffered from the 

 desire to study the events of shorter periods of time. 



Much more important than this path of heat escape with the aqueous 

 vapour of the air current is the main path which issues along the stream of 

 water from the internal radiator system. This radiator system, originally 

 distributed along the line of junction of the walls and roof of the calorimeter, 

 I have extended so as to be co-extensive with the roof of the calorimeter, 

 beneath which it is suspended in an insulated fashion. The entering water is 



