VOL. X.] PHILOSOPHICAL TRANSACTIONS. 273 



After this, spirit of wine was made to boil in vacuo in the same manner as 

 the water, and it boiled much sooner; and it made the mercury rise about an 

 inch in the gauge. After taken from the fire, it continued boiling; and even 

 sinking the receiver into cold water, it boiled much more strongly. 



Having caused some gunpowder to be burnt with a burning glass ; we were 

 much surprised to see that it burnt grain by grain, none of the grains kindled 

 fire in those which touched. Another time when the sun had less force, we 

 could not at all kindle the powder, but made it only boil and emit store of 

 smoke. We put a gauge in the same recipient, by means whereof we observed 

 that all that smoke produced no air ; for the quicksilver did not rise in the tube. 

 This smoke falling on the paste-board, on which the powder was put, appeared 

 yellow of the colour of brimstone. Then we took out the powder that re- 

 mained, being like a black mass, and having put it upon burning coals, it 

 burned like salt-petre ; so that the sulphur was almost all exhaled. 



Another time having put the weight of 12 or 15 grains of powder in a glass 

 shaped like a cupping-glass, capable of holding 14 ounces of water, and hav- 

 ing set fire to it, the powder boiled and smoked as usual. Afterwards seeing 

 that the grains began to crack very near one after another, we took away the 

 burning concave, for fear all should be kindled together ; but it was already too 

 late, for the corns continued to crack longer than a second of time, and at last 

 all kindled, though there was then nothing left to heat them but the fire which 

 they had kept within themselves. The receiver was lifted up above a foot high 

 without breaking. 



Another time putting the weight of 1 8 grains of powder with a gauge into 

 a receiver, holding seven pounds of water, and firing the powder, it made the 

 quicksilver rise to the height of an inch and a half in the gauge; and I am 

 very well assured that all that air was not come from without; for that part of 

 the receiver to which the cover is applied had always been under water. 



From what has been related it may be concluded, that there is a fifth part of 

 air in gunpowder, supposing, as other experiments show, that air is about 

 1000 times lighter than water. For, in this experiment, the mercury rose to 

 the 18th part of the height where the air commonly sustains it; and conse- 

 quently the weight of 18 grains of powder yielded air enough to fill the 18th 

 part of a receiver that contains seven pounds of water. Now this 1 8th part 

 contains 49 drachms of water: wherefore the air that takes up an equal space 

 being 1000 times lighter weighs -toVo- of 49 drachms, which is more than three 

 grains and a half. It follows therefore that the weight of 1 8 grains of powder, 

 which were employed in the experiment, contained more than three and a half 

 of air, which is about the fifth part of 18 grains. 



VOL. II. N N 



