28 Dr. Hare’s improved Eudiometers. 
der these circumstances, one of the calorimotors is surrounded 
with the acid contained in the jug, and an explosion almost 
invariably succeeds. Before effecting the explosion, the num- 
ber of the degrees of the sliding-rod which are out ofits tube 
should be noted; and it must afterwards be forced into the 
tube, in order to compensate the consequent condensation of 
the gases as nearly as it can be anticipated. A communication 
with the gauge must then be opened gradually. If the water is 
disturbed from its level, the equilibrium must be restored by 
duly moving the rod. Then deducting the degrees of the 
sliding-rod remaining out of the tube from those which it 
indicated before the explosion, the remainder is the deficit 
caused by it; one-third of which is the quantity of oxygen gas 
in the included air. Or, the residual air being expelled by 
the rod, and the quantity thus ascertained deducted from the 
amount included before the explosion, the difference will be 
the quantity condensed. 
It may be proper to mention, that as other metals are al- 
most universally acted upon by mercury, the cocks, sockets, 
screws, and sliding-rods of the mercurial eudiometers are 
made of cast steel. The tubes containing the rods are of 
iron. 
Since the drawings (figs. 1 and 4) were made, verniers have 
been attached to the screws through which thé sliding-rods 
pass; so that the measurements are made to one-tenth of a 
degree. 
I have alluded to the water-gauge without explaining its 
construction. It consists of three tubes. A small tube of 
varnished copper (which is fastened into the only perforation 
which communicates with the cock, and of course with the 
glass recipient) passes up in the axis of a glass tube (T, fig. 4), 
open at top, cemented into a socket (S, fig. 4), which screws 
on to the cock. A smaller glass tube is placed in the inter- 
stice between the external glass tube and the copper tube in 
its axis. This intermediate glass tube is open at its lower ter- 
mination, but at the upper one is closed or opened at pleasure 
by a screw. The interstices between the three tubes are par- 
tially supplied with water, as represented in the drawing 
(W, fig. 4). When the passage between the gauge and the 
recipient is open, if the pressure on the included air be more 
or less than that of the atmosphere, the water will rise in one 
of the gauge-tubes, and sink in the other. Other liquids may 
be substituted for water, in the gauge, when desirable. 
In addition to the principal collar of leathers, and screws 
for rendering that collar compact, there is in the mercurial 
eudiometers a small hollow cylinder (a piece of a gun-barrel), 
with 
