42 ON THE LOSS OF HEAT 



perature of the interior room, from the absence of proper 

 means, in some of the apparatus experimented upon, to regu- 

 late the combustion ; but from very few trials with each, 

 it was found less difficult than had been anticipated, and 

 that the difficulty could be entirely avoided by making the 

 quantity of fuel administered to the fire, the regulator of the 

 rapidity and extent of combustion necessary to be produced, 

 which was effected by using the wood in small pieces. 



The results have been thrown into a tabular form, and ex- 

 hibit, as before stated, the comparison of each apparatus with 

 No. 9, in which it is assumed that no loss of heat is sustained, 

 this assumption being necessary, for the purpose of determining 

 the comparative loss of heat sustained by each apparatus, which 

 is the object of the experiments. 



The manner of obtaining the results in time, having been 

 stated, it is evident, that, as the same quantity of fuel was con- 

 sumed in every experiment, consequently the same quantity of 

 heat must have been generated. In all the experiments, (ex- 

 cept the standard experiment No. 9,) we find part of the heat 

 escaped by the pipe or flue of the grate and fire-place into the 

 chimney, and was lost, and proportional to this loss must have 

 been the quantity of the fuel required to be consumed in a 

 given time, to maintain the temperature of the room, and, con- 

 sequently, the duration of each experiment was proportionally 

 affected thereby. The numbers, therefore, which express the 

 duration of each experiment, are proportional to the heat 

 saved, and assuming the positive quantity of heat generated as 

 100, this being the result of apparatus No. 9, if the time occu- 

 pied in any other experiment is deducted from 100, the re- 

 mainder gives the positive loss sustained in every hundred parts 

 of heat generated by using this apparatus, and by which we de- 

 termine that in using No. 1, as only 10 parts in every hundred 

 parts of heat generated are saved, consequently we lose 90 per 

 cent of heat. 



As the important difference which exists in the quantity of 

 fuel required to be consumed in different apparatus to produce 

 the same effect, might not in all instances be obvious by a cur- 



