330 
• Journal of Agricultural Research 
Vol. V, No. 8 
sorber is shown in Table I, which summarizes data obtained in an alcohol 
check test (see p. 342) of the calorimeter made in January, 1915, which 
continued for two consecutive periods of three hours each. 
Table I .—Comparison of data for heat measurement obtained by use of temperature 
difference recorder and of thermoelement with potentiometer 
Time. 
Water flow. 
Temperature difference. 
Heat computed from 
measurement. 
By recorder. 
By potentio¬ 
meter. 
By recorder. 
By potentio¬ 
meter. 
Kgm. 
Degrees. 
Degrees. 
Calories. 
Calories. 
i hour. 
20. 53 
3 * 99 
4. OI 
8l. 9 
82. 3 
Do. 
21. 20 
3 - 97 
3 - 96 
84. 2 
84. O 
Do. 
23.00 
3 - 73 
3 - 73 
86. 0 
86.0 
Total. 
252. 1 
252.3 
1 hour.. 
23. 00 
3-68 
3-68 
84. 6. 
84.6 
Do. 
23. 21 
3 - 57 
3 - 54 
82. 9 
82. 2 
Do. 
22. 72 
3 - 67 
3-67 
83-4 
83-4 
Total... 
250.9 
250. 2 
In order that the recording device may continue to measure tempera¬ 
ture differences with the accuracy required, not only must the bridge be 
sensitive to a change as small as 0.002 per cent in the resistance of the 
thermometer coils, but also the resistances of the various parts of the 
bridge circuit other than the thermometers must not change as much as 
0.003 per cent. Provision is made for testing the component parts of 
the bridge by the substitution of duplicate parts, which are mounted 
with the ratio coils of the bridge in a check box, and tests of this character 
are made at frequent intervals. After the apparatus had been in use for 
a short time a very slight change in one of the ratio coils was detected 
and corrected. Since that time the bridge has remained remarkably 
constant. It is possible also to test with the check box and recorder 
whether the thermometer coils remain alike in resistance at the same 
temperature. Provision is made in the check box for correcting slight 
inequalities in them by a variable shunt across a coil of small resistance 
in series with one of the thermometers. 
PREVENTING TRANSFERENCE OF HEAT THROUGH THE WAEES OF THE 
CHAMBER 
In order that the quantity of heat produced in the chamber may be 
accurately measured, either there must be no increase or decrease in it t 
due to the passage of heat through the walls of the chamber, or if heat 
is thus added or subtracted, the quantity must be determined and allow¬ 
ance made for it. This calorimeter is constructed and operated in 
accordance with the method employed in the original calorimeter of 
