30 PHILOSOPHICAL SOCIETY OF WASHINGTON. 
These are known as the Poggendorf corrections. They are due 
to the capacity of the tube from the zero to the ome-hundred degree 
mark, being different at different temperatures. 
In the following table are shown the results of the comparisons 
of a certain mercurial thermometer, Tounelot No. 4207, with the 
air-thermometer : 
Tounelot No. 4207. 
Correction . Correction to 
Scale reading. | as a normal ther- | reduce to the air Differences. 
mometer. thermometer. 
°c. ec, °c. oc. 
0.0 0.00 0.00 0.00 
5.6 + 0.05 + 0.01 + 0.04 
11.1 + 0.08 0.00 -- 0.08 
16.1 + 0.10 + 0.02 + 0.08 
22.2 + 0.15 + 0.05 + 0.10 
25.2 + 0.16 + 0.07 + 0.09 
30.1 +. 0.20 + 0.09 + 0.11 
35.4 + 0.22 + 0.10 + 0.12 
40.4 + 0.25 + 0.12 + 0.13 
45.1 + 0.25 + 0.16 + 0.09 
50.1 + 0.24 + 0.15 -+ 0.09 
52.7 + 0.24 + 0.16 + 0.08 
55.8 + 0.24 + 0.18 + 0.06 
If it be supposed that these differences are due to sensible terms 
in the expansion of glass and mercury dependent on the squares of 
the temperatures, an equation can be derived which will show that 
the maximum difference must be at 50°. But this is not so; the 
greatest difference is at about 40°. This agrees with what has 
been found by others. Rowland at Baltimore found the greatest 
difference at 40° to 45°; Mills in England found it at 35°, and 
Grunmach in Berlin at 30°. 
Forming a theory of the differences on the supposition that they 
depend on the third powers of the temperatures as well as the 
squares, equation (4) is obtained, which gives the relation between 
the thermometer reading, 7, and the true temperature, ¢. 
ee A, ) Sine B, 2 
— 4,144,100 baerrneny ta ane y 
1+ 4, 1 4%? 100 + Ys — Bs (100)° 
n— n—?, 
This is only approximate. The effects of the second and third 
(4) 
