two or more tubes, which frequently
happens when we have not a sufficient number of bottles with two or
three necks, we must use a cork with two or three holes, Pl. IV. Fig. 8.

When the whole apparatus is thus solidly joined, so that no part can
play upon another, we begin to lute. The lute is softened by kneading
and rolling it between the fingers, with the assistance of heat, if
necessary. It is rolled into little cylindrical pieces, and applied to
the junctures, taking great care to make it apply close, and adhere
firmly, in every part; a second roll is applied over the first, so as to
pass it on each side, and so on till each juncture be sufficiently
covered; after this, the slips of bladder, or of linen, as above
directed, must be carefully applied over all. Though this operation may
appear extremely simple, yet it requires peculiar delicacy and
management; great care must be taken not to disturb one juncture whilst
luting another, and more especially when applying the fillets and
ligatures.

Before beginning any experiment, the closeness of the luting ought
always to be previously tried, either by slightly heating the retort A,
Pl. IV. Fig. 1, or by blowing in a little air by some of the
perpendicular tubes S s s s; the alteration of pressure causes a
change in the level of the liquid in these tubes. If the apparatus be
accurately luted, this alteration of level will be permanent; whereas,
if there be the smallest, opening in any of the junctures, the liquid
will very soon recover its former level. It must always be remembered,
that the whole success of experiments in modern chemistry depends upon
the exactness of this operation, which therefore requires the utmost
patience, and most attentive accuracy.

It would be of infinite service to enable chemists, especially those who
are engaged in pneumatic processes, to dispense with the use of lutes,
or at least to diminish the number necessary in complicated instruments.
I once thought of having my apparatus constructed so as to unite in all
its parts by fitting with emery, in the way of bottles with cristal
stoppers; but the execution of this plan was extremely difficult. I have
since thought it preferable to substitute columns of a few lines of
mercury in place of lutes, and have got an apparatus constructed upon
this principle, which appears capable of very convenient application in
a great number of circumstances.

It consists of a double necked bottle A, Pl.