472 Notices respecting New Boohs. 



In the Third part of the work, Dr. Goldstein investigates the 

 effect upon the deflection thus shown to take place, of varying the 

 nature of the gas in the tube, or of the metal of which the elec- 

 trodes consist, the density of the gas, the intensity of the charge 

 communicated to the kathodes, the source of the electricity, and so 

 ou. It is found that neither the nature of the gas, nor its density, 

 nor the nature of the metal employed for the electrodes, has any 

 influence upon the phenomena observed ; and also that the inten- 

 sity of the charge is of no consequence provided that the two 

 kathodes are equally charged, as they must be if placed in good 

 metallic communication with each other. If, however, they are 

 connected by a bad conductor, such as a moistened thread, the dark 

 surfaces produced are unequal in size, the largest being that nearest 

 the kathode which is most strongly charged. The repulsion exerted 

 by a kathode is thus increased by increasing its electric charge ; 

 and, on the other hand, the stiffness, so to speak, of the rays emitted 

 by a kathode is increased by increasing its electric charge. 



In the Fourth part of the work, the different theories of the elec- 

 tric discharge are examined, in order to see how far they can explain 

 the new phenomena. Three kinds of repulsion are known — that 

 due to mechanical action (such as takes place in the impact of pon- 

 derable masses), electrostatic repulsion, and electrodynamic repul- 

 sion. 



In adopting the mechanical hypothesis, we must suppose that 

 molecules of gas charged with electricity fly off from each kathode, 

 that they come into collision, and that the two motions combine to 

 a resultant motion whose direction will be that of the deflected ray. 

 To decide this question, a cylindrical tube was taken provided 

 with a kathode at one end, and a second kathode and diaphragm 

 some distance from the first in the side of the tube, from which, 

 when excited alone, a sharply-defined pencil of rays falls on the 

 opposite side of the tube. It is found that the rays from the ter- 

 minal kathode are deflected on passing the lateral kathode, but that 

 the rays from this second are not affected by the supposed impact 

 of the molecules forming the first bundle of rays. 



It is further clearly shown that the phenomena are not con- 

 sistent with the known laws of the mutual action of currents ; so 

 that the electrodynamic hypothesis must be rejected. 



The third hypothesis is that the action is an electrostatic one, be- 

 tween the electricity with which the one kathode is charged and 

 the like electricity residing in the molecules of the stream from the 

 other kathode. 



According to what may be termed the " convective theory," the 

 electric discharge is a sort of convective process, consisting in the 

 motion of molecules charged with electricity, which bring about the 

 transference of electricity from the kathode in the direction of the 

 stream of rays. This convective theory is the one which has 

 attracted most attention. The particular form of the theory ex- 

 pounded by Crookes assumes that the gas-molecules charged with 

 electricity are driven off from the kathode as a sort of wind which 

 drives the uncharged molecules before it ; so that, surrounding the 



