340 EXPERIMENTS ON THE RELATION 



dexterity, we transfer the gas into the sealed leg of the syphon. 

 When we conceive enough to have been passed up, we remove 

 the finger, and next bring the mercury to a level in both legs, 

 either by the addition of a few drops, or by the displacement 

 of a portion, by thrusting down into it a small cylinder of 

 wood. We now ascertain, by careful inspection, the volume 

 of included gas. Applying the fore-finger again to the orifice, 

 so as also to touch the end of the platina wire, we then ap- 

 proach the pendent ball or button to the electrical machine, 

 and transmit the spark. Even when the included gas is consi- 

 derable in quantity, and of a strongly explosive power, we feel 

 at the instant nothing but a slight push or pressure on the tip 

 of the finger. After explosion, when condensation of volume 

 ensues, the finger will feel pressed down to the orifice by the 

 superincumbent atmosphere. On gradually sliding the finger 

 to one side, and admitting the air, the mercurial column in the 

 sealed leg will rise more or less above that in the other. We 

 then pour in tliis liquid metal, till the equilibrium be again re- 

 stored, when we read off as before, wi;hout any reduction, the 

 true resulting volume of gas. 



', As we ought always to leave two inches or more of air be- 

 tween the finger and the mercury, this atmospheric column 

 serves as a perfect recoil spring, enabling us to explode very 

 large quantities without any inconvenience or danger. The 

 manipulation is also, after a little practice, as easy as that of 

 the single tube. But a peculiar advantage of this detachable 

 instrument, is to enable us to keep our pneumatic troughs and 

 electrical machine at any distance which convenience may re- 

 quire ; even in different chambers, which, in the case of wet 

 weather, or a damp apartment, may be found necessary to en- 

 sure electrical excitation. In the immediate vicinity of the 

 water pneumatic cistern, we know how often the electric spark 



refuses 



