Historical Review of Experiments on Heat. 189 



This habit of many years' standing, by force of which 

 I seize with the greatest eagerness, and endeavour to 

 investigate, each and every phenomenon related even 

 in the slightest manner to heat and its operations which 

 comes to my knowledge, has suggested to me almost all 

 the experiments that I have performed with reference to 

 this subject. 



In the year 1778 I was engaged in investigating the 

 force of gunpowder and the velocity of bullets dis- 

 charged from fire-arms. For this purpose I discharged 

 many times a musket-barrel which was loaded in vari- 

 ous ways, and which rested on two iron rods, perfectly 

 free (that is, without any stock), in a horizontal posi- 

 tion, about four feet from the ground.* This gave me 

 occasion to make a very striking observation. 



Since these experiments were intended principally 

 to determine, from the recoil of the barrel, the veloci- 

 ties with which the bullets were discharged, it was 

 first necessary to ascertain how much the weight of 

 the powder which caused the discharge of the bullets 

 had to do with this recoil. In order to solve this 

 problem, I made several successive experiments, 

 some with a charge of powder without any bullet, 

 and some with two, three, or even four bullets, one 

 upon another. 



According to my usual practice, I seized the piece 

 with my left hand immediately after each discharge, in 

 order to hold it firmly until I had wiped it out with 

 some tow fastened to the rammer. I was therefore 

 not a little astonished to notice, on this occasion, that 



* A detailed description of these investigations may be found in the seventy-first 

 volume of the Philosophical Transactions, and in the first volume of my Philosophical 

 Papers, which was published at London, in the year 1802, by Cadell and Davics. Sec 

 also Vol. I. p. I. ' 



