HIGH FREQUENCY AND HIGH POTENTIAL CURRENTS. 279

When a new bulb was put on, the mercury was always raisedabove stopcock c h which was closed, so as to always keep themercury and both the reservoirs in line condition, and the mer-cury was never withdrawn from iq except when the pump hadreached the highest degree of exhaustion. It is necessary to ob-serve this rule if it is desired to use the apparatus to advantage.

By means of this arrangement I was able to proceed veryquickly, and when the apparatus was in perfect order it was pos-sible to reach the phosphorescent stage in a small bulb in lessthan fifteen minutes, which is certainly very quick work for asmall laboratory arrangement requiring all in all about 100 poundsof mercury. With ordinary small bulbs the ratio of the capacityof the pump, receiver, and connections, and that of reservoir kwas about 1 to 20, and the degrees of exhaustion reached werenecessarily very high, though I am unable to make a precise andreliable statement how far the exhaustion was carried.

What impresses the investigator most in the course of theseexperiences is the behavior of gases when subjected to great rap-idly alternating electrostatic stresses. But he must remain indoubt as to whether the effects observed are due wholly to themolecules, or atoms, of the gas which chemical analysis disclosesto us, or whether there enters into play another medium of agaseous nature, comprising atoms, or molecules, immersed in afluid pervading the space. Such a medium surely must exist,and I am convinced that, for instance, even if air were absent,the surface and neighborhood of a body in space would be heatedby rapidly alternating the potential of the body; but no suchheating of the surface or neighborhood could occur if all freeatoms were removed and only a homogeneous, incompressible, andelastic fluid—such as ether is supposed to be—would remain, forthen there would be no impacts, no collisions. In such a case,as far as the body itself is concerned, only frictional losses in theinside could occur.

It is a striking fact that the discharge through a gas is es-tablished with ever-increasing freedom as the frequency of theimpulses is augmented. It behaves in this respect quite contrarilyto a metallic conductor. In the latter the impedance entersprominently into play as the frequency is increased, but the gasacts much as a series of condensers would; the facility withwhich the discharge passes through, seems to depend on the rateof change of potential. If it acts so, then in a vacuum tube even