196 Mr. S. P. Thompson on some Phenomena 



8. One terminal of a long-coil Thomson's reflecting galva- 

 nometer, by Elliott Bros., was put to earth, the other being 

 connected with Q (fig. 3). When the distance w T was less 

 than 5 millims. no deflection was produced upon the needle 

 and mirror, a tremor being the only result. But when the 

 distance iv T was gradually increased, and the secondary dis- 

 charge had consequently to accumulate upon its conductors 

 to acquire sufficient density to overleap the resistance of the 

 increased thickness of air between w and T, the tremulous 

 motion of the galvanometer-needle became much more marked, 

 the spot of light wandering about on either side the zero of 

 the scale, never at rest, but showing no permanent deflection. 

 When the distance iv T was increased to nearly 50 millims. 

 the brisk flow of sparks gave place to the brush discharge, 

 varied only by the passage of an occasional spark. In this 

 case the wanderings of the spot of light on the scale were 

 still more exaggerated, the needle apparently being turned 

 little by little with an increasing deflection during some 

 twenty or thirty vibrations of the contact-breaker of the coil, 

 then receiving a sudden impetus in the opposite direction 

 as the secondary spark discharged across the interval w T. 

 This result was entirely confirmatory of the occasional flashes 

 previously observed in the vacuum-tube, with which, on further 

 examination, the more violent oscillations of the galvanometer- 

 needle were found to be simultaneous. The experiment also 

 accords with the previous observations of physicists with re- 

 spect to the oscillations of the induced currents of the coil. 

 It explains, too, why with an ordinary needle-galvanometer 

 no deflection had been obtained with the minute sparks yielded 

 by the earlier experiments. 



9. Having again diminished the distance iv T to 5 millims., 

 a wire was led from Q to the knob of a gold-leaf electroscope. 

 No divergence of the leaves took place ; but on drawing- 

 sparks by the hand or by an earth-wire from the electroscope 

 knob, from Q, or even from T', a divergence of the leaves 

 took place, rapidly followed by collapse at every spark so 

 drawn, the gold leaves being thrown into a state of violent 

 flutter. If the hand, or earth-wire, were suddenly removed 

 from contact, a residual charge usually remained upon the 

 leaves of the electroscope ; and this charge was variously posi- 

 tive or negative, changing sign apparently according to the 

 last oscillation of the secondary discharge. 



Experimentation with the electroscope could not be carried 

 on when w was removed more than 10 or 12 millims. from T, 

 as the momentary charges received by the gold leaves were 

 so violent as to destroy them. 



