BETWEEN THE FREEZING AND BOILING-POINTS. 
229 
short time of each other, between which the temperature of the calorimeter never 
varied. 
In selecting the limits of flow to be used in all the measurements, the accuracy of 
the limits was tested for two other temperatures by recording observations for three 
different flows at 13° and at 60° C. In these cases the theory of the method was 
o-iven a severe test. 
o 
I have tabulated here the two sets, one for a mean temperature of 13° C. and the 
other for a temperature of 60° C. I have taken the values of 8 for each set from the 
variation curve. 
Calorimeter C. —Mean Temperature, 13°-79 C. March 9, 1899. 
8 = — -00208. J = 4-1913. 
Temperature of surrounding air, 19° C. 
dd. 
Q- 
(EC-4-2 Q dd)/d9. 
Difference from 
mean, -03940. 
8-5768 
•372746 
•03946 
4- -00006 
8-5803 
•372262 
•03947 
- -00007 
8-5586 
•459149 
•03923 
- -00017 
8-5683 
•458194 
•03930 
- -oooio 
8-5499 
•573318 
•03953 
+ -00013 
8-5616 
•571920 
•03940 
•ooooo 
Calorimeter C.—Mean Temperature, 59°"80 C. June 17, 1899. 
8 = — -00360. J = 4-1849. 
Temperature of surrounding air, 22° C. 
dd. 
Q. 
(EC - JQ dff)jd6. 
Difference from 
mean, -07254. 
8-3805 
•612400 
•07277 
+ -00023 
8•3835 
•611227 
•07263 
+ -00009 
8-3158 
•462971 
•07236 
- -00018 
8-3395 
•461364 
•07240 
- -00014 
8-3534 
•388491 
•07261 
+ -00007 
8-3674 
•387534 
•07242 
- -00012 
The variation from the mean value in both sets is less than 1 part in 10,000. It 
is a matter of interest to compare the balance points on the bridge-wire for the 
