, a : Dr. John Garric on the 
exact limitation of the heat set free from air by compression. a4 
He states that to ignite sulphur, it is necessary to reduce‘atmos- 
pherie air to one- -eighteenth of its volume, and to set fire to sigigh, 
to one-thirtieth.* 
Philosophers, in their experiments on air, have been: in the 
practice of dividing it into certain proportions—as the half, the 
fourth, the eighth of the volume experimented upon—and of: am 
considering each division as affording an oned measure of heat 
N 
as the result of condensation. Now, the number of such divis- 
jons in reducing air to one-eighteenth of its ns eh is a little 
more than four, and to one-thirtieth, nearly five; while the tem- 
peratures at which the above named substances burn, may be 
stated respectively, at nearly 500° F. and 600° F. Here we 
have data, which if correct, would at once solve the question in- 
volved. By dividing these numbers by each other we shall have 
reduction of it to half its previous volume. If these data could 
be relied upon, the subject would need no further investigation ; 
but so different have been the conclusions from analogous experi- 
ments in the hands of the other philosophers, that new and more 
satisfactory ones are demande or while, as we have seen, 
M. Colladon says it requires a ‘condensation of eign air 
equal to one-thirtieth of its volume to ignite tinder, M. L 
sac, who made similar experiments on a larger Siti ics 
that a reduction to one-fifth is sufficient. 
Nor have other experiments and processes of reasoning, for 
deducing a precise solution of this apparently simple question, 
though conducted by the most eminent chemists and mathema- 
ticians of the past and present ages, been more uniform or more 
show a similar variance, of from one to five and upwards, a in their 
deductions. 
und, in experimenting with the common air-pump © 
alton fo of 
the lebaedony, that if the density of air be suddenly doubled by 
compression, its temperature rises 5U° F. This experiment is 
among the earliest we find on the subject, and comes, if we can. 
regard the 50° LE". as computed at the standard of the specific heat 
of water, nearer the truth, for the first reduction of air from its 
ordinary temperature and pressure, to half its volume, than any 
subsequently recorde 
According to some "experiments of Leslie, condueted upon a 
somewhat different plan, 2t would appar that atmospheric air 
Journal of Science [I], vol. xx, cPd 
fort it will be ot ec in the coutse of this co phe AO how this very great dif- 
in apparent t, may arise from ccidhiarats ively small errors of the ex- 
~ 
