of Induced Electric Sparks* 195 



poles of zinc, mercury, or carbon. Their brilliance was 

 augmented and their length was increased to above 20 millims. 

 when w was withdrawn so far from T that only a faint brush 

 discharge passed across from w to T, a distance of about 

 50 millims. ; and a still greater effect was produced when T 

 was directly connected with earth. This modification virtually 

 changed the arrangement to that indicated in fig. 4 *. 



6. When the spark obtained between Q and the earth-wire 

 C exceeds 4 millims. in length, it is capable of illuminating a 

 small vacuum-tube. The illumination differs, however, in a 

 remarkable manner from that produced by the ordinary direct 

 discharge of the induction-coil. Instead of the usual pheno- 

 mena of a small brush of light at the positive pole, and a 

 bright aureole or glow surrounding the negative pole, the 

 illumination is symmetrical, alike at each pole, having both 

 brush and glow at each, and is unaltered by reversing the 

 primary circuit of the coil. An occasional flash, however, 

 seemed to glow at one pole or at the other. 



On making single contacts by holding the break-piece of 

 the induction-coil in the fingers the effect was as follows : — 

 at each make-of-contact no illumination ; at each break-of- 

 contact an illumination of the same symmetrical nature, afford- 

 ing no evidence of polarity of discharge in one direction more 

 than in another. 



7. To ascertain whether the discharge at the two poles was 

 simultaneous or alternating, a rotating mirror was placed be- 

 side the vacuum-tube. The images were found to be, roughly 

 speaking, alternate, the discharges being alternately from 

 either pole. The irregularity of the alternation appeared to 

 be dependent partly upon the distribution of the resistances 

 in the circuit, partly upon the irregularities in the vibrations 

 of the break. The occasional flashes of light at one pole or 

 the other were explained by the occasional appearance in the 

 mirror of two contiguous images having the same position, 

 indicating the occurrence of two consecutive discharges in the 

 same direction. 



With rotating slits the same phenomenon of reciprocating 

 discharge was observed, though less advantageously. 



* Since the communication of these experiments to the Physical Society, 

 my attention has been called to an article in the Journal of the Franklin 

 Institute for June 1871, by Prof. E. J. Houston, of Philadelphia, where, 

 under the title of " A New Connexion for the Induction-coil," he de- 

 scribes a somewhat similar arrangement, in which an insulated conductor 

 is employed in connexion with the coil. With this combination Prof. 

 Houston observed some curious effects of induction in neighbouring 

 bodies ; but he seems to have thought them due to escape of electricity, 

 and apparently had no suspicion of the real nature of the sparks so 

 produced. 



02 



