VOL. LVI.J ■ PHILOSOPHICAL TRANSACTIONS. 305 



small space was left between the glass-tube and the pearl-ashes in the cylinder. 

 When the zinc was dissolved, the whole apparatus was weighed again, and was 

 found to have lost II4 grains in weight* ; which loss is principally owing to the 

 weight of the inflammable air discharged. But it must be observed, that before 

 the ettervescence, that part of the bottle and cylinder, which was not occupied 

 by other more solid matter, was filled with common air ; whereas, after the effer- 

 vescence, it was filled with inflammable air ; so that, on that account alone, 

 supposing no more inflammable air to be discharged than what was sufficient to 

 fill that space, the weight of the apparatus would have been diminished by the 

 difference of the weight of that quantity of common air and inflammable air. 

 The whole empty space in the bottle and cylinder was about 98O grain measures; 

 there is no need of exactness ; and the difference of the weight of that quantity 

 of common and inflammable air, is about one grain : therefore the true weight 

 of the inflammable air discharged, is lOf grains. The quantity of zinc used 

 was 254 grains, and consequently the weight of the air discharged is Vt or Tx^f 

 the weight of the zinc. 



It was before said, that one grain of zinc yielded 356 grain measures of air ; 

 therefore 254 grains of zinc yield 90427 grain measures of air ; which we have 

 just found to weigh lOf grains; therefore inflammable air is about 8410 times 

 lighter than water, or lO-^ times lighter than common air. The quantity of 

 moisture condensed in the pearl-ashes was found to be about 1^ grains. By an- 

 other experiment, tried exactly in the same way, the density of inflammable air 

 came out 8300 times less than that of water. 



The specific gravity of the air, produced by dissolving zinc in spirit of salt, 

 was tried exactly in the same manner ; 244 grains of zinc being dissolved in spirit 

 of salt diluted with about 4 times its weight of water, the loss in effervescence 

 was 10^ grains, the empty space in the bottle and cylinder was 9 14 grain mea- 

 sures; whence the weight of the inflammable air was 9^ grains, and conse- 

 quently its density was 89 10 times less than that of water. By another experi- 

 ment, its specific gravity came out 9030 times lighter than water. 



A like experiment was tried with iron : 250^- grains of iron being dissolved in 

 oil of vitriol diluted with 4 times its weight of water, the loss in effervescence 

 was 13 grains, the empty space 1420 grain measures. Therefore the weight of 

 the inflammable air was 1 If grains, i.e. about Vt of the weight of the iron, and 

 its densitywas 8973 times less than that of water. The moisture condensed was 

 1+ grains. 

 A like experiment was tried with tin : 607 grains of tinfoil being dissolved in 



* As the quantity of lute used was but small, and as this kind of lute does not lose a g^at deal 

 of its weight by being kept in a moderately dry room, no sensible error could arise from the drying 

 of the lute during the experiment. — Orig. 



VOL. XII. R R 



