water a canvas bag fixed to the extremity of a
rubber tube, it is possible to produce in the interior a tumultuous
disengagement of gas without any bubbles passing through.

[Illustration: FIG. 2.--ARRANGEMENT FOR THE STUDY OF CAPILLARY
REACTION IN POROUS VESSELS.]

From an electrical point of view, partitions are of very unequal
quality. Various partitions having been placed between electrodes
spaced three centimeters apart, currents were obtained which indicated
that, with the best of porcelains, the rendering of the apparatus is
diminished by one-half. Asbestos cloth introduces but an insignificant
resistance.

To this inconvenience of porous vessels is added their fragility,
their high price, and the impossibility of obtaining them of the
dimensions that large apparatus would call for. The selection of
asbestos cloth is therefore clearly indicated; but, as it does not
entirely separate the gases, except at a pressure that does not exceed
a few centimeters of water, it was always necessary to bring back the
variation of the level to these narrow limits by a special
arrangement. We cannot, in fact, expect that the entire piping shall
be always in such conditions that no difference in pressure can occur.
The levels are brought back to equality within the effective limits by
interposing between the voltameter and the piping an apparatus called
a compensator, which consists of two vessels that communicate in the
interior part through a large tube. The gases enter each vessel
through a pipe that debouches beneath the level of the water. If a
momentary stoppage occurs in one of the conduits, the water changes
level in the compensator, but the pressure remains constant at the
orifice of the tubes. The compensator is, as may be seen, nothing more
than a double Mariotte flask. When it is desired to obtain pure gases,
there is introduced into the compensator a solution of tartaric acid,
which retains the traces of alkalies carried along by the current of
gas. The alkaline solution, moreover, destroys the ozone at the moment
of its formation.

It will be seen that laboratory studies have furnished all the
elements of a problem which is now capable of entering the domain of
practice. The cheapness of the raw materials permits of constructing
apparatus whose dimensions will no longer be limited except by reasons
of another nature. The electrodes may be placed in proximity at will,
owing to the use of the porous partition. I