degree. Such thermometers will be about 17 inches long.

It is very satisfactory to have _two_ instruments and a good outfit of
thermometers and heat-carriers, in order to take duplicate observations
for mutual verification and detection of errors.


HEAT CARRIERS.

For these platinum is greatly to be preferred to any other known
substance. Its rather high cost is the only objection to its use. Its
heat capacity is low, by weight, but its specific gravity is great, and
sufficient capacity can be obtained in moderate bulk, while its high
conductivity tends to shorten the duration of each experiment or
observation. A convenient outfit for each instrument consists of three
balls, hammered to a spherical form, one 1.1385 inches diameter,
weighing 4,200 grains=0.6 pound avoirdupois; one 0.9945 inch diameter,
weighing 2,800 grains=0.4 pound; and one 0.7894 inch diameter, weighing
1,400 grains=0.2 pound.

These can be obtained at 1-2/3 cents per grain, and will cost,
respectively, $70.00, $46.67, and $23.33, and collectively, $140.00.
At the assumed specific heat of Pt=0.0333+, the heat capacity of the
respective balls will be 1/100, 1/150, and 1/300 of 2 pounds of cold
water, and the two smaller balls used together will be equal to the
larger one. Corrections for varying specific heat of platinum may
be conveniently made by the tables given in a previous article.[1]
Corrections for varying specific heat of water are less important, but
may be made by the following table: