90 M. Achille Cazin on Internal Work in Gases. 



§ III. Course of an experiment. 



The stopcock C (fig. 2) is open, and the pressure of the gas 

 contained in the reservoirs differs but little from the atmospheric 

 pressure ; for an instant the valve R is opened, and the levels in 

 the manometer Q are allowed to settle themselves. The stop- 

 cock C and the valve are closed. By means of the pump some 

 gas is caused to pass from the reservoir B to A ; and when the 

 temperature is stationary, the pressure p } is measured in the 

 manometer K, and the pressure p^ in the manometer M. The 

 stopcocks r, r n are closed, so that the gas contained in the mano- 

 meters may be separated from that which the reservoirs contain, 

 and not disturb the expansion; without this precaution the 

 effects observed during the expansion would depend on the move- 

 ment of the gas in the tubes L N O, IPO, and it would be im- 

 possible to obtain an accurate result. 



After these preparations, the strip of paper is put in motion; 

 the stopcock C is opened by an assistant, who writes down the 

 various positions of the exterior level of the liquid contained in 

 the manometer Q as they are read aloud on the scale of this 

 manometer, closing at the same moment the circuit of the elec- 

 tromagnet. The highest and lowest points which the level has 

 reached are particularly noted. 



In general this level does not return quite to its initial position, 

 because there remains some gas in the pump and the manometers; 

 but the difference is not very great, and the manometer Q is 

 sufficiently true, as has been already stated. When several con- 

 secutive experiments are made, the final level is always at the 

 same point as in the first experiment, provided there has been 

 neither leakage in the apparatus nor change in the exterior tem- 

 perature and pressure. The essential condition of a successful 

 experiment consists in the initial level being near the final level. 



In order to complete the observations, the external tempera- 

 ture and the atmospheric pressure are measured. From the 

 position of the final level, we know the pressure p\ which esta- 

 blishes itself in the reservoirs when every motion has ceased and 

 the reservoirs have returned to the ordinary temperature. Thus 

 we know that the gas has undergone an expansion under an ex- 

 cess of pressure p l — p% without external work, that it has finally 

 attained the pressure p', and that before attaining this pressure 

 it has had a variable pressure p, which must be deduced from the 

 observed variations of the level in the manometer Q. 



It is in order to find this variable pressure that the curve de- 

 fined in § I. was constructed, of which the general form is 

 represented in fig. 1. The abscissa? are calculated by the traces 

 of the pencil on the strip of paper, and the ordinates by means 



