[ '5^ ] 



of nitre; the confequence however is not juft, for undoubtedly 

 five parts nitre v\^ould confume one of charcoal, but it does not 

 thence follow that they would not confume ftill more. On the 

 other hand he found that ten parts nitre were necejfary for the 

 confumption of one part of plombago, whence it follows that one 

 part of plombago decompofes ten of nitre, otherwifc nine parts 

 nitre would fuffice to confume it, and the tenth would have been 

 unneceflary, as it ads only as it is decompofed. Now this propor- 

 tion approaches very nearly to my refult, namely, i of charcoal to 

 9,6 of nitre. 



Hence, and fince Kilkenny coal in the preceding experiments 

 fnewed no fign of its containing any thing bituminous, I take it 

 for granted that it confifts almoft entirely of pure carbon ; and 

 fmce 50 grains of it alkalize 480 grains of pure ignited nitre, that 

 in all the fubfequent experiments on other fpecies of coals or bi- 

 tumens free from fulphur and iron, the dccompofition of this 

 ftandard quantity of nitre will indicate in the quantity of coal 

 ncceffary for that dccompofition, the prefence of 50 parts of mere 

 carbon. 



Before I proceed to the recital of other experiments I muft 

 mention another circumftance that occurs in making them, which 

 is, that after the inflammation ceafes a hilling noife is perceived 

 for a long time, and is increafed on adding frefli quantities of 

 coal, even when the nitre is feemingly decompofed ; this feemed 



to 



