gas is the sum of the portions evolved at the two electrodes, and the
instrument is better adapted than either of the former as a measurer of the
quantity of voltaic electricity transmitted in ordinary cases. It consists
of a straight tube (fig. 64.) closed at the upper extremity, and graduated,
through the sides of which pass platina wires (being fused into the glass),
which are connected with two plates within. The tube is fitted by grinding
into one mouth of a double-necked bottle. If the latter be one-half or
two-thirds full of the dilute sulphuric acid (706.), it will, upon
inclination of the whole, flow into the tube and fill it. When an electric
current is passed through the instrument, the gases evolved against the
plates collect in the upper portion of the tube, and are not subject to the
recombining power of the platina.

710. Another form of the instrument is given at fig. 65.

711. A fifth form is delineated (fig. 66.). This I have found exceedingly
useful in experiments continued in succession for days together, and where
large quantities of indicating gas were to be collected. It is fixed on a
weighted foot, and has the form of a small retort containing the two
electrodes: the neck is narrow, and sufficiently long to deliver gas
issuing from it into a jar placed in a small pneumatic trough. The
electrode chamber, sealed hermetically at the part held in the stand, is
five inches in length, and 0.6 of an inch in diameter; the neck about nine
inches in length, and 0.4 of an inch in diameter internally. The figure
will fully indicate the construction.

712. It can hardly be requisite to remark, that in the arrangement of any
of these forms of apparatus, they, and the wires connecting them with the
substance, which is collaterally subjected to the action of the same
electric current, should be so far insulated as to ensure a certainty that
all the electricity which passes through the one shall also be transmitted
through the other.

* * * * *

713. Next to the precaution of collecting the gases, if mingled, out of
contact with the platinum, was the necessity of testing the law of a
_definite electrolytic_ action, upon water at least, under all varieties of
condition; that, with a conviction of its certainty, might also be obtained
a knowledge of those interfering circumstances which would require to be
practically guarded against.

714. The first point investigated was the