*90 Researches upon the Heal developed 



When experiments are underlaken with a view to eluci- 

 date the phaenomena of nature, it is always more satisfac- 

 tory to avoid errors, or to compensate for them, than to rely 

 ujKjn calculations to appreciate their effects. 



As the law of the variation of the specific heat of water 

 at ditTerent temperatures is not known, and as we are but 

 imperfectly acquainted with the true measurement of the 

 mtervals of temperature which are marked by the divisions 

 of our thermometers; in order to prevent the effects of our 

 uncertainty on this head, as to the results of the inquiry in 

 question, I took care to make my experiments in a room 

 in which the temperature varied very little, and to confine 

 them to a very trifling elevation of the temperature of the 

 water in the receiver. It is true that I performed some ex- 

 periments in a room where the air was much colder, and 

 in which I filled the receiver with ice instead of water; 

 but these experiments were for a particular purpose, and 

 they are not arranged along with the others. Besides, they 

 never yielded results so constant or satisfactory as those of 

 the experiments made under other circun)s»ances. 



In order to give an idea of the confidence which may be 

 placed in the results of the experiments made with the new 

 apparatus which 1 have described, I shall subjoin the details 

 of an experiment made with the express view of discover- 

 ing the measurement of its perfection. 



Having filled two receivers properly attached to each 

 other, with water at the temperature of the air of the room, 

 that of .^5" Fahrenheit, I burned a taper under the mouth 

 of the principal receiver, so that all the products of the 

 combustion passed through the worm of the secondary re- 

 ceiver, after having passed through that of the principal. 

 Each of the receivers contained 2371 grammes of water. 



The following were the results of this experiment: 



It 



