194 Mr. S. P. Thomson on some Phenomena 



was apparently that the effect was one of surface-charge. 

 This surmise might be directly tested by increasing the size 

 of the metallic core and the number of surrounding turns of 

 wire. 



4. The arrangement next to be described*, and shown in 

 diagram at fig. 3, was therefore directly suggested by the 

 preceding experiments. It remained practically unchanged 

 in the subsequent investigation. A large Leyden jar was 

 taken and its inner and outer surfaces joined by a metallic 

 wire, constituting it simply a conductor of large surface. 

 Around it were wrapped sheets of vulcanized rubber to a 

 thickness of 25 millims. About this again were coiled about 

 30 metres of stoutly insulated copper wiref. As before, one 

 end of this wire was led to the terminal T 7 (fig. 3) — the other 

 end, w, being placed near the terminal T, and so adjusted that 

 the distance w T might be increased or diminished at will. 



5. When the coil was set in action, sparks could be drawn 

 by an earth-wire, C, from the knob of the accumulator, Q. 

 When the discharging-distance of the coil w T was less than 

 5 millims., the sparks drawn from Q were minute, not ex- 

 ceeding 1 millim. in length, destitute of perceptible physio- 

 logical action, producing no deflection when led through an 

 ordinary needle-galvanometer, and, as before, were incapable 

 of imparting any charge to a proof-plane. 



When w was removed to a distance of 10 millims. from T 

 the sparks at Q were larger and brighter ; when the distance 

 was increased to 20, 25, and 30 millims. successively, the 

 sparks drawn at Q were much more vivid and exceeded 

 10 millims. in length. These sparks, though producing com- 

 paratively little effect in contracting the muscles of the arm, 

 were excessively painful and penetrating to the nerves of the 

 skin. Their light was brightest when they passed between 



* It will be plain, from the subsequent details, that this arrangement 

 plays the part of an accumulator or condenser, the interior surface of 

 which is represented by the Leyden jar, the exterior being represented by 

 the wire coiled outside. A Leyden jar is not essential; any cylindrical 

 conductor is as good if of large surface ; and a sheet of tinfoil laid upon 

 the exterior of the insulating material is more effective than 50 metres of 

 covered wire wound round it. The precise form of the accumulator is 

 immaterial; but thorough insulation must be secured. The above ar- 

 rangement is preferred because its parts are analogues of those of the 

 experiments that suggested it. The knob of the jar is moreover con- 

 venient for drawing sparks from. 



t To avoid confusion, in fig. 3 the condenser arrangement is given in 

 diagram, D standing for the insulating folds of rubber, P for one con- 

 densing surface (the external wire coils), Q for the other condensing sur- 

 face (the Leyden jar, which acts in this combination as if made of one 

 piece of metal). 



