Intelligence and Miscellaneous Articles, 471 



The following is the principle of the method of observation. The 

 gas is at the outset under the same pressure in the two reservoirs, 

 which are then connected with a sulphuric-acid manometer, and the 

 level noted. A valve is closed, which cuts off communication between 

 the reservoirs and the manometer. By means of a pump, part of the 

 gas is driven from one reservoir into the other, A large stopcock 

 between the reservoirs being opened; sudden expansion takes place, 

 and the motion of the stopcock opens in turn the valve of the mano- 

 meter. The apparatus indicates the time when the expansion com- 

 mences, and at which the valve has been opened; and the level in the 

 manometer at known intervals is then observed. 



For this purpose a voltaic circuit is used containing an electro- 

 magnet the armature of which is provided with a pencil ; against 

 this pencil a band of paper is uncoiled with known velocity. The 

 stopcock in turning closes the circuit as soon as the expansion 

 commences, and then opens it at the same time as the valve ; 

 the circuit, finally, is closed for a moment each time the manometer is 

 read off. From the marks left on the paper the necessary times are 

 easily known. 



Let H be the difference of level observed at different times, H' the 

 final difference observed when the gas has resumed its original tem- 

 perature ; a curve is traced having for ordinates the values H 7 and H, 

 and for abscissae the times calculated from the moment the expan- 

 sion commenced. 



By gradually increasing the time which elapses between the com- 

 mencement of the expansion and the opening of the valve, leaving 

 other circumstances unaltered, a series of curves is obtained which 

 succeed each other regularly, and a comparison of which gives the 

 curve of real pressure which the gas possesses at various intervals. 

 By this device we avoid being compelled to attempt jthe impossibility 

 of having a manometer which shall instantaneously indicate the va- 

 rying pressure of the gas. 



This method is analogous to that which I used in 1862, in an in- 

 vestigation on the expansion and compression of gases. Ordinarily 

 changes of bodies are observed by means of an instrument in which 

 a moveable part indicates almost instantaneously the condition of the 

 body. When the change is very rapid, the moveable part is always 

 behindhand, and the displacements observed cannot directly lead to 

 a knowledge of the change in question. But if the degree of retar- 

 dation be varied, and if the law of the displacement be represented 

 in each experiment by a curve, the law of the change may be deduced 

 from a great number of such curves. In this manner phenomena 

 so fugitive as the movements of a gaseous mass become accessible to 

 observation. I have used a reservoir of 9 litres containing gas under 

 about 4 atmospheres, and a reservoir of 34 or of 60 litres containing 

 highly rarefied gas. The stopcock which separated the two reser- 

 voirs had a conduit of 4 centims. diameter. 



The general result of the experiments is as follows : — As soon as 

 the two reservoirs are connected, the pressure becomes equal in a 



212 



