Unipolar Induction, 187 



For the case where solenoid and condenser are set in 

 rotation at the same speed and in the same direction, the 

 mean value o£ the ratio 0'242 is 5 per cent, larger than the 

 theoretical value of 0*231. But immediately after these 

 observations a similar group of four sets taken at half speed 

 gave a mean ratio of 0*228 (range '211 to "263), which 

 is 1*4 per cent, too small. Then a group was taken at full 

 speed but by a different method : in quick succession the 

 observer threw on the current, insulated the electrometer, 

 read the spot, and reversed the current ; after 20 seconds 

 another reading was taken and the current was turned off, 

 and after another 20 seconds a final reading was taken. This 

 gives two deflexions in opposite directions, a third of wiiose 

 numerical sum should be the electromagnetic effect. The 

 values thus obtained ranged from *195 to *271, and averaged 

 0*238 ; including three readings rejected because the first 

 reading was not the greater, the average becomes 0*241. 

 Proportionality of the effect to current strength had been 

 tested previously ; a group of four sets using 13 amp. gave 

 a mean ratio (not comparable with the previous values) of 

 '25* while a group taken under the same conditions, but using 

 25 amp., gave *27. 



With the condenser alone rotating final observations were 

 not obtained because the solenoid became short-circuited, and 

 it was not thought worth while to rewind it. The case is, 

 however, deducible from the other two by simple super- 

 position ; and partial confirmation is furnished by a group 

 of observations which were taken under about the same 

 conditions, and gave a mean electrometer deflexion of 

 14 mm. 



The results thus remain somewhat rough, but they seem 

 sufficient to warrant the conclusion with fair certainty that 

 the condenser becomes charged by its own rotation in the 

 manner required by Lorentz's theory, independently of the 

 rotation of the solenoid. 



§ 7. Fate of the Moving-Line Theory. 



The view has been put forward by Poincare *, Abraham f, 

 and Barnett J that even observations on open circuits cannot 

 disprove the moving-line theory; and in reply to criticism 

 by the author, Barnett maintained this view with some hent§. 



* H. Poincare, Eel. Elect, xxiii. p. 41 (1900). 

 t Abraham & Fopp], Theo. der Elektr. p. 420. 

 t S. J. Barnett, Phys. Rev. Nov. 1912, p. 323. 

 § S. J. Barnett, Pins. Rev. Oct. 1913, p. 323. 



