VOL. LVI.j PHILOSOPHICAL TRANSACTIONS. 311 



fastening a glass tube to the wooden cap of the bladder, and luting that to the 

 mouth of the bottle containing the effervescing mixture, in such manner as to 

 be air-tight. The bladder was kept on till it was quite full of fixed air : being 

 then taken off and weighed, it was found to lose 34 grains, by forcing out the 

 air. The bladder was previously found to hold 100 ounce measures. Whence 

 if the outward air, at the time when this experiment was tried, be supposed to 

 have been 800 times lighter than water, fixed air is 511 times lighter than water, 

 or 1-iVo- times heavier than common air. The heat of the air during the trial of 

 this experiment was 45°. By another experiment of the same kind, made when 

 the thermometer was at 65% fixed air seemed to be about 563 times lighter than 

 water. 



Exper. g. — Fixed air has no power of keeping fire alive, as common air has ; 

 but, on the contrary, that property of common air is very much diminished by 

 the mixture of a small quantity of fixed air ; as appears thus : A small wax 

 candle burnt 80* in a receiver, which held J go ounce measures, when filled with 

 common air only. The same candle burnt bV in the same receiver, when filled 

 with a mixture of one part of fixed air to ig of common air, i. e. when the fixed 

 air was -jV of the whole mixture. When the fixed air was -^ of the whole mix- 

 ture, the candle burnt 23'. When the fixed air was -iV of the whole, it burnt 



I V. When the fixed air was -^ or — oi the whole mixture, the candle went 



out immediately. 



Hence it should seem, that when the air contains near i its bulk of fixed air, 

 it is unfit for small candles to bum in. Perhaps indeed, if I had used a larger 

 candle and a larger receiver, it might have burnt in a mixture containing a larger 

 proportion of fixed air than this ; as I believe that large flaming bodies will burn 

 in a fouler air than small ones. But this is sufficient to show, that the power 

 which common air has of keeping fire alive, is very much diminished by a small 

 mixture of fixed air. This experiment was tried, by setting the candle in a large 

 cistern of water, in such manner that the flame was raised but a little way above 

 the surface; the receiver being inverted full of water into the same cistern. The 

 proper quantity of fixed air was then let up, and the remaining space filled with 

 common air, by raising the receiver gradually out of water ; after which, it was 

 immediately whelmed gently over the burning candle. 

 Experiments on the Quatdity of Fixed ^ir, contained in Alkaline Substances. 

 Exper. 10. — The quantity of fixed air contained in marble, was found by 

 dissolving some marble in spirit of salt, and finding the loss of weight, which it 

 suffered in effervescence, in the same manner as I found the weight of the 

 inflammable air discharged from metals by solution in acids, except that the 

 cylinder was filled with shreds of filtering paper, instead of dry pearl ashes ; for 



