a rapid succession of Electrical Discharges. 327 



vanced along the surface of the wood towards each other, and in 

 the course of a minute or two united near the middle, forming 

 a long crooked line of fire between the points. This effect took 

 place most rapidly when the points were placed on the longitu- 

 dinal section of the wood of the same year's growth ; for if placed 

 indiscriminately on different portions of the wood, the sparks 

 would advance from each point a certain distance, following the 

 course of the grain, and then suddenly jump over into the next 

 layer, and so on at irregular intervals until they both reached 

 the same longitudinal layer, when they would rapidly advance 

 towards each other, and the combustion would instantly increase 

 throughout the whole of the crooked circuit thus produced. 



On connecting with the wires of the universal discharger the 

 ends of a slip of deal about as thick as a goose-quill, and 5 or 6 

 inches long, slightly moistened with nitric acid, it instantly takes 

 fire at the points of contact, and the sparks pass along its surface, 

 rapidly kindling it throughout its whole length, when it assumes 

 the appearance of a bar of tire in vivid combustion and is speedily 

 consumed. Sometimes the spark, instead of passing over its sur- 

 face and kindling it, will rise from one portion and leap to the 

 middle or some other spot, where it instantly burns through the 

 wood and generally breaks it. 



This remarkable action is not confined to inflammable imper- 

 fect conductors, and would appear to arise either from the heat 

 elicited from the air in the passage of the spark, or from an im- 

 mediate heating power of the spark itself; for even thin glass 

 will fuse and become vividly ignited when sparks are made to 

 pass over its surface. 



The interposition of the Leyden jar, as is well known, pro- 

 duces a marked change in the character of the spark : it increases 

 its volume, light and sound, but diminishes its length ; and the 

 larger the jar or battery, the more is the length of tlie spark di- 

 minished, the excitation of the coil being the same. The ther- 

 mal effects also undergo a curious change, and the induced cur- 

 rent appears to be endowed with a new character which it did 

 not before possess, viz. that of heating an intermediate portion 

 of the metallic circuit when of small dimensions (Phil. Mag. 

 vol. xii. p. 380. par. 7). Smce I first noticed these results, I 

 have repeated them with more powerful machines, and the fol- 

 lowing experiment exhibits the effect in a striking manner. 



I connect with the terminals of the induction coil, by means 

 of two long and flexible conductors, a discharging electrometer, 

 set 80 as to give sparks of 0-3 or 0*4 of an inch in length. In 

 immediate connexion with the discharger, and forming an inter- 

 mediate portion of the circuit, I place a piece of fine platina wire 

 about 4 or 5 inches in length. I then connect one of the con- 



