2o8 Continuation of the Experiments and Ohfervations 
Fourthly, If phlogifton could be decompofed by heat, and 
then leave a refiduum of phlogifticated air, amounting to one- 
third of its bulk, the diminution a riling from its inflammation 
with common or dephlogifticated air could never be fo great as 
it is found to be by repeated experiments ; for when inflam- 
mable and common air are flred in the proportion of eleven of 
the latter to four of the former, a bulk equal to the whole of the 
inflammable air, and to one-fifth of the common air, difap- 
pears, according to Mr. volta*, and the diminution is about 
two-fifths of the whole, or more exadtly out of fifteen mea- 
fures, only 8,8 remain ; but if the inflammable air were decom- 
pofed, and one-third of it, being phlogifticated air, fhould 
remain, then not quite one-fifth of the whole would vanifh, 
and the refiduum fhould be 10,54 meafures. This evidently 
proves, that pure inflammable air is never decompofed (unlefs 
the lofs of its fire be called a decompofition) ; but in the a£l of 
inflammation is totally transferred upon the pure part of relpira- 
ble air to which it unites. Fifthly, To obtain ftill a clearer 
infight into this matter, I intreated Mr. cavallo, who is very 
expert in the management of the blow-pipe, as well as in pneu- 
matic experiments, to repeat this experiment in my laboratory. 
We accordingly filled a tube 10,5 inches long, and one- fourth 
or an inch in diameter, with inflammable air from iron re- 
ceived over mercury, and having made the tube red-hot 
tnroughout and black, and foftened it fo far as to endanger the 
elcape of the air, we opened it 011 mercury. The air was di- 
minifhed only one-tenth, and inflamed with an explofion as 
loud as an equal quantity of the fame inflammable air that had 
not been heated. 
■' Roz. April 1779,4). 2 95 * 
The 
