Nov. 22,1915 Improved Respiration Calorimeter 329 
As the slider moves back and forth on the bar which supports the slide 
wire, it carries a pen which draws a curve on ruled paper by which the 
movement of the contact point on the slide wire is expressed in tempera¬ 
ture. The total width of the paper scale, 25 cm., represents a difference 
of 2 degrees between the temperature of the water entering and that of 
tfie water leaving the heat absorbers, and corresponds exactly to the 
length of the slide wire by which the bridge is balanced for the inequality 
of resistance in the thermometer coils resulting from such a temperature 
difference. The temperature difference indicated by the position of the 
pen on the paper scale coincides with that to which the position of the 
contact point on the slide wire is equivalent. The paper scale is ruled 
with 100 lines, each representing 0.02 degree, and as the distance between 
the lines is 2.5 mm., the curve may easily be interpreted to 0.01 degree. 
The paper is moved forward at a very regular rate, approximately 7.5 cm. 
per hour, by the motor which moves the slider, the speed of the motor 
being regulated by a governor so that it is uniform, even with wide 
fluctuations in voltage of the current by which the motor is driven. 
Since the necessary changes in the position of the slider are made every 
7 seconds, the curve gives a practically continuous record of the tem¬ 
perature difference. 
The difference between the temperature of the water as it enters and 
that as it leaves the heat absorber may thus be easily read at any 
instant to 0.01 degree. The accuracy of the measurement of tempera¬ 
ture difference by the apparatus may be tested at any time, even during 
the course of an experiment, without interfering with the record, and 
such tests are made at frequent intervals. In the water channel in the 
center of each resistance-thermometer bulb is the end of a differential 
thermoelement of 0.125 mm. copper and constantan wires, having 11 
junctions in each end, inclosed in 4-mm. glass tubing, with thin wall. 
The element remains permanently in position, though it may be easily 
removed if necessary. The terminals of the element are joined by insu¬ 
lated 1-mm. copper wire to binding posts on the observer’s table, from 
which connection can be made with a potentiometer whenever a test is 
to be made. With this differential thermoelement, which has been cali¬ 
brated over a wide range of temperature at the United States Bureau of 
Standards, an electromotive force of over 4.5 microvolts results from a 
difference of 0.01 degree in the temperatures of the two ends. By means 
of the potentiometer and galvanometer with which it is employed, an 
electromotive force of half that magnitude is easily measured; conse¬ 
quently temperature differences may be measured by it to an accuracy 
at least as good as 0.01 degree. Measurements made with this apparatus 
therefore serve to indicate the accuracy of those with the recorder. 
The agreement of results obtained by the two methods of measuring the 
* increase in the temperature of the water flowing through the heat ab- 
