286 Prof. J. J. Thomson. The Resistance of [Jan. 17, 



the case ; thin sheet metal and tin-foil when placed between the coils 

 were found to completely stop the sparks in C. I then coated a plate 

 of glass, which of itself had no eifect upon the sparks, with a film of 

 Dutch metal about yyVo f a centimetre in thickness, and found 

 that it completely stopped the sparks, and I have not been able to get 

 a film of metal thin enough to allow sufficient induction to pass 

 through to produce sparks in the secondary. 



This is in accordance with the results of our investigation on the 

 screening effect of conducting plates, for we saw by equation (7) that 

 when a screen of thickness h was interposed the electromotive force 

 is only 



when ff is the conductivity of the metal. 



Since the electromotive force in the plane of the screen, which is 

 taken as the plane of yz, is of the form 



2 cos by cos cz, 



v/(6 2 -f c 2 ) will be of the order 2?r/R where R is the radius of the primary 

 coil ; several coils were used whose radii varied from 13 to 23 c., so 

 that -v/(6 2 + c 2 ) will be of the order 1/2. The length of the coils 

 varied from 81 to 140 c., and the balls at the extremities from 1 to 

 2 c. in diameter, so that the length divided by the capacity is large, 

 and, therefore, by equation (18) the wave-length will be twice the 

 length of the coil, or for the largest coil about 3 metres ; thus p will 

 be about 2n- x 10 8 , and if we suppose the film is a-oV Q of a millimetre 

 thick, h will be 5 X 10~ 4 , we may take a to be 10 4 . A film of this 

 kind will, by the above formula, dimmish the induction about 800 

 times, and we should, therefore, not expect the electromotive force 

 acting on the secondary to be sufficient to produce sparks. 



A thick plate of ebonite was next placed between the coils bat did 

 not produce any appreciable diminution in the sparks in the secondary ; 

 thus ebonite, though opaque to vibrations as rapid as those of light, 

 still allows vibrations of which 10 8 take place in a second to pass 

 through without interruption. 



The effect of interposing a film of electrolyte was next tried. A large 

 square glass trough was placed between the coils B and C and carefully 

 levelled, the electrolyte was then poured in ; when only a very small 

 quantity of electrolyte was in the trough the sparks still passed, but 

 they got feebler and feebler as the quantity of electrolyte in the trough 

 increased, until finally, when the electrolyte was dilute sulphuric acid, 

 they ceased altogether when the depth of the electrolyte in the trough 

 amounted to 3 or 4 millimetres. The criterion adopted for the dis- 

 appearance of the sparks was to allow 60 sparks to pass into the 



