ueeze out the air. The ebonite should not be perfectly warm
and dry, for it is then sure to be electrified, and this will give
trouble. It must not be wet, because then it will conduct, and the
sealing-wax will produce no result. If it has been used as the support
for the rings for some of the previous experiments, it will have been
sufficiently splashed by the bursting of bubbles to be in the best
condition. It must, however, be well wiped occasionally.

A stick of sealing-wax should be held in readiness under the arm, in a
fold or two of _dry_ flannel or fur. If the wax is very strongly
electrified, it is apt to be far too powerful, and to cause the bubbles,
when it is presented to them, to destroy each other. A feeble
electrification is sufficient; then the instant it is exposed the
bubbles coalesce. The wax may be brought so near one bubble in which
another one is resting, that it pulls them to one side, but the inner
one is screened from electrical action by the outer one. It is important
not to bring the wax very near, as in that case the bubble will be
pulled so far as to touch it, and so be broken. The wetting of the wax
will make further electrification very uncertain. In showing the
difference between an inner and an outer bubble, the same remarks with
regard to undue pressure, electrification, or loss of time apply. I have
generally found that it is advisable in this experiment not to drain the
drops from both the bubbles, as their weight seems to steady them; the
external bubble may be drained, and if it is not too large, the process
of electrically joining the outer bubbles, without injury to the inner
one, may be repeated many times. I once caused eight or nine single
bubbles to unite with the outer one of a pair in succession before it
became too unwieldy for more accessions to be possible.

* * * * *

It would be going outside my subject to say anything about the
management of lanterns. I may, however, state that while the experiments
with the small bubbles are best projected with a lens upon the screen,
the larger bubbles described in the last lecture can only be projected
by their shadows. For this purpose the condensing lens is removed, and
the bare light alone made use of. An electric arc is far preferable to a
lime-light, both because the shadows are sharper, and because the
colours are so much more brilliant. No oil lamp would answer, even if
the light were sufficie