365 
of simple alimentary substances, ^c. 
tube, P, (fig. 1.) is of green or bottle glass, moderately stout, 
and about -j of an inch internal diameter. It is fixed between 
the horizontal parts of the vertical cocks, MM, so as to be 
perfectly air-tight; and when required, the whole, or any 
part of it, may be heated by means of the lamp, Q, at the 
pleasure of the operator. 
When the apparatus is to be employed, both the syphon 
gasometers, I L, IL, are to be filled with quicksilver, and a 
small green glass retort, containing the requisite quantity of 
chlorate of potash, (and which had been previously heated so 
as to completely expel the common air, and to fill it with 
oxygen gas) is to be attached to one of the cocks, as repre¬ 
sented in fig. 4, by means of a caoutchouc tube. Heat is then 
to be applied, and any quantity of oxygen gas that may be 
required, introduced into the tube, I. After the whole has 
acquired the temperature of the atmosphere, the exact quan¬ 
tity of the gas is to be accurately noticed, as well as the state 
of the barometer and thermometer at the same time. The 
tube, P, containing the substance to be analysed, is then to be 
firmly fixed between the cocks M M,* and subjected to heat, 
during which the oxygen gas is to be transferred from one 
syphon to another, through the red hot tube, with any velo¬ 
city that may be required, and which may be regulated by 
means o^one of the stop-cocks of the funnels, R R, and the 
stop-cock, S, of the opposite syphon. 
Such is a general view of the apparatus, and the principles 
» I have tried various modes of connecting the tube so as to ensure its being air¬ 
tight. Caoutchouc answers very well; but the best substance I have hitherto 
employed are slips of thin moistened hogs’ bladder, tied on very tightly with fine 
dry twine. The twine is then to be moistened also, and the whole kept in this 
state till the end of the experiment. 
