VOL. LXXX.] PHILOSOPHICAL TRANSACTIONS. 721 



tion, it occupied the space of 1 inch and -^. Being agitated with lime-water, 

 very nearly -^ was absorbed. A portion of the insoluble residue was ex|X)sed to 

 a lighted taper, and burned with a faint blue flame.* 



The dephlogisticated air used in this experiment had been previously agitated 

 with water, to free it entirely from fixed air. It was the purest dephlogisticated 

 air he had ever seen : for when 1 measure of it was mixed with 1 measure and 

 ^^ of nitrous air, the residue occupied the space of only Vir of a measure. From 

 the foregoing trial it was evident, that 14- parts of pure air were insufficient to 

 saturate one of the animal air that had not been previously agitated with water. 

 The experiment was therefore repeated as follows. Two parts of pure air being 

 mixed with 1 of animal air, occupied ^-V of an inch. The mixture being fired 

 by the electrical shock, the residue stood at a little less than -^. When this 

 residue was agitated with lime-water, it was almost wholly absorbed. By a sub- 

 sequent trial it was found, that nearly half the animal air used in this experi- 

 ment was soluble in water. Hence it appears, that the quantity of pure air re- 

 quired to saturate the insoluble part of the animal air, is somewhat less than 

 that required to saturate the compound fluid which had not been previously agi- 

 tated with water. But the latter fluid has been shown to consist almost en- 

 tirely of heavy inflammable, animal, hepatic, and fixed air ; and as the last of 

 these is already saturated with pure air, it is manifest that the above-mentioned 

 difference must depend on the animal hepatic air. Whence it follows, that the 

 latter contains a large portion of the inflammable principle. From the quantity 

 of fixed air produced in the last of the preceding experiments, there is also the 

 utmost reason to believe, that the basis of heavy inflammable forms one of the 

 constituent parts of animal hepatic air. When equal parts of pure and animal 

 air were burned together, a considerable increase of bulk almost invariably took 

 place ; and when the proportion of the animal was to that of pure air as 2 1 to 

 15, the bulk of the mixture was increased one half. The air that remained 

 after the combustion in the last-mentioned experiments was inflammable : for a 

 portion of it being introduced into a small phial, and exposed to a lighted 

 candle, it first exploded, and then burned with a blue lambent flame. 



Being desirous of learning the cause of the increase of bulk in the foregoing 

 experiments, the following trials were made. Three measures of animal were 

 mixed with 2 of pure air, and several strong electrical shocks were made to pass 

 through the mixture ; but it would not take fire. Half a measure of pure air 



• When this experiment was first made, the residue did not appear to be inflammable. It had 

 been tried by applying an inflamed slip of paper to the mouth of a phial which was filled with it ; 

 but, upon repeating the experiment, when the phial containing the residuary air was carried into a 

 dark room, and an ignited wax taper was applied to its mouth, an evident inflammation took place. 

 — Orig. 



VOL. XVI. -* Z 



