(875) 
however, sufficiently accurate for the purpose of checking the experi- 
mental results. The data employed in the calculation are 
Aye = 1.0006 
a BX 10° C><10° 
0°.C.. —0:57440 -) 2.9594 
HOC. 1040405} > -+-3,0178 
99°.4C. + 0.25075 + 3.5669 
If the value of C, = 0.2389 cal. is used, and if expansion always 
takes place against atmospheric pressure, equation (3) gives for the 
cooling effect in air at a temperature of 0° C. the relation 
T,—T, = 0.273 (p, — 1) — 0.000208 (p,? — 1). 
This is in good agreement with the JouLE-KELvIN experiments which 
were carried out with pressures up to 6 atm. and satisfied the relation 
? ry 273 
Ti hi Pipi) ort 2 
§ 3. A diagram of the first apparatus used is shown in fig. 1, 
and it can be seen that it did not differ much from that described 
by OrszewskKr *). Two spirals of copper tubing, s, and s,, joined in 
series served to bring the gas to the required temperature; s,, in 
front of which was coupled a manometer, was 5 m. long and of 
3.5 mm. bore; s,, 7 m. long and of 1.25 mm. bore’), ended in 
a reduction valve , which opened into a small silvered vacuum- 
vessel, g. This vacuum glass was enclosed in a german-silver box, 
B,, having free communication with the outside through a wide 
tube, B,, and was protected from radiation from above by felt 
faced with nickel-paper. Instead of a resistance thermometer as used 
by OrszewskKi, a thermoeletnent served to indicate the difference in 
temperature between the compressed and expanded gas. One junction, 
Th,, was in the liquid bath surrounding the spirals, and the 
second, 7h,, the wires of which were insulated by short glass 
tubes, 6, and 6,, soldered into the top of the box, was in the 
expansion chamber, g. By this means a direct measurement of the 
heat-effect was possible. 
1) K. Orszewski: Phil. Mag. |6). 3. p. 535. (1902). 
Ann. Phys. 7. p. 818. (1902). 
*) This cross-section was taken so small on account of the low temperature of 
the liquid hydrogen in which the apparatus filled with helium was lo be placed, 
and because at that time pure helium was still very difficult to obtain. 
59% 
