as in the
preceding test, but the tap water was replaced by a saturated solution
of salt water. Eleven amperes with a potential difference of 7 volts
between the electrodes, which were 73/4 inches apart, were passed
through the solution for three hours, and the temperature at the end
of the run was 122 deg. F., and was rising very slowly.

Although the current per square inch is much greater, the watts
absorbed per cubic inch is much less in this case than when water was
used. With the water carrying 2 amperes the watts absorbed would be
over 10 per cubic inch, while for the saturated solution of salt when
carrying 11 amperes it would be only about 0.4 watt.

In test No. 3 use was made of a long, wooden rectangular trough (42
inches by 61/2 inches by 8 inches) with vertical, sheet iron electrodes.
The cross section of the liquid, which was a 10 per cent. solution of
salt in water, was 44 square inches, and with 10 amperes passing
through the solution for 13/4 hours the temperature rose to 95 deg. F., and
was rising slowly at the end of the run.

The plates were 413/4 inches apart, and at the end of the run the
voltmeter across the terminals read 20. This gives a current density
of nearly 1/4 ampere per square inch and 0.11 watt per cubic inch. These
values are too low to be considered maximum values, for this cross
section of a 10 per cent. salt solution would probably carry 13 to 15
amperes safely.

It appears that as the amount of salt in the solution is increased
from zero to saturation, the maximum current carrying capacity is
increased, but the watts absorbed per cubic inch are less.

A very small addition of salt to tap water makes the solution a much
better conductor than the water, and reduces greatly the safe maximum
watts absorbed. In using glass vessels, such as Daniell jars, there is
danger of cracking the jar if the temperature rises much above 165 deg. to
175 deg. F.

In test No. 4 an ordinary whisky barrel, filled up with tap water, was
used. Two horizontal circular iron plates (3/16 inch thick) were used
for electrodes. The diameter of the inside of the barrel was
approximately 19-1/2 inches. With the two plates 26-3/8 inches apart a
difference of potential of 486 volts gave a current of 2.6 amperes.
With the plates 7/8 inch apart, 228 volts gave 35.5 amperes at the end
of one hour, when all the water in the barrel was very hot (175 deg. F.),
and there was quite a good deal of gas given off. The