322 
Journal of Agricultural Research 
Vol. V, No. 8 
when the coil is not in series, the reading is from 2 to 12 degrees. In 
each degree there are 10 subdivisions which are about 4 mm. apart, so 
that it is possible to change the setting of the indicator by as little as 
0.05 degree, and the galvanometer connected with the indicator is sen¬ 
sitive to a change of this magnitude in the balance of the bridge. The 
resistance thermometer which is in one arm of the bridge is a coil of 
platinum wire, of approximately 25.5 ohms' resistance at 20° C., the 
resistance of which changes 0.1 ohm with a change of 1 degree in its 
temperature. The wire is wound in a flat, narrow coil and inclosed in a 
very thin silver case, resembling that of the heating coil, and similarly 
mounted in the upper end of the water channel. The resistance wire 
very rapidly acquires the temperature of the water flowing in the milli¬ 
meter space surrounding the case, and changes in resistance instantly 
follow very small changes in the temperature. The thermometer is a 
short distance from the bridge, as shown in Plate XXXVI, figure 2, 
and connected with it by leads that are compensated so that the effect 
of the resistance of the leads and of change in their resistance due to 
change in temperature is eliminated. 
From the final heater the water flows into a bottle of about 1-liter 
capacity, nearly full of broken pumice, and then into the heat absorber. 
Measuring the Temperature Increase in Heat Absorber 
The water that has passed through the heat absorber will have increased 
in temperature according to its rate of flow and the rate of production 
of heat in the chamber. The accuracy with which the increase in tem¬ 
perature is determined is of fundamental importance in the measurement 
of heat generated. 
MEASUREMENT BY MERCURY THERMOMETERS 
The difference between the temperature of the ingoing and that of 
the outgoing water was formerly determined by reading two mercury 
thermometers installed in the water circuit, with the bulb of one in the 
water just entering the chamber and the bulb of the other in the water 
just leaving it. The thermometers were as sensitive as it was practicable 
to employ and were very accurately calibrated. Each had a range of 
about 12 degrees, with graduations of 0.02 degree, the one in the ingoing 
water reading from o° to 12 0 and the one in the outgoing from 8° 
to 20 0 , and by judging the position of the mercury between the 
graduations the temperature was estimated to 0.01 degree. The observer 
read the thermometers and recorded the temperatures every two or four 
minutes, which, in addition to the other duties at the observer’s table, 
was rather tedious and trying. Both thermometers were supposed to 
be read simultaneously, but as this was impracticable for one observer 
the two thermometers were read as quickly as possible, and then the 
