102 Mr. S. P. Thompson on some Phenomena 



magnet-core b\ an earth-wire or by any conducting substance 

 bold in the hand and presented to the core *. These sparks 

 are best seen when the electromagnet-coils, interrupter, and 

 battery-cells are completely insulated from the ground. The 

 substitution of a bar of any metallic substance, or even carbon 

 or wood coated with tinfoil, in place of the iron core of the 

 electromagnet, does not prevent the production of the spark. 

 With bars of zinc or cadmium, or, better still, of either of these 

 metals amalgamated on the surface, the sparks are brighter. 



The minuteness of the sparks makes observation of them a 

 matter of some difficulty ; between two points of graphite 

 pencils in a dark box they are more readily seen. They are 

 conducted by all metallic bodies, and by the body; and even 

 pass into insulated conductors of large surface, apparently re- 

 quiring no completion of the circuit. This circumstance, 

 however, as well as the supposed inability of the sparks to pro- 

 duce any indications upon the galvanometers and electroscopes 

 of ordinary use, may be explained by a further knowledge 

 of their nature. It was this apparent lack of polarity in 

 the nature of the discharge that led to the adoption of the 

 hypothesis of a new form of energy; and the assumption 

 seemed further to be borne out by an apparent absence of 

 physiological effect. 



Two forms of the apparatus at first employed to produce 

 the sparks in question are given in figs. 1 & 2. In the first 

 case, where the circuit was rapidly interrupted by an auto- 

 matic vibrator, the sparks were drawn from the electromagnet- 

 core or from the vibrating armature. In the second they 

 were drawn from a bar of metal placed above, but not in con- 

 tact with, an electromagnet in a circuit completed by a depressor 

 key. 



3. Careful repetition of the experiments failed to elicit 

 evidence of polarity in the spark, or to procure response to 

 the ordinary electrical tests. And the wish to obtain further 

 information upon the nature of this peculiar discharge by 

 studying its effects in illuminating the rarefied gases of 

 vacuum-tubes, rendered it necessary to seek a means of more 

 plentiful production, and to abandon the methods of the 

 original investigators already described; for the sparks so 



* As an interrupter almost any form will answer ; a Morse key is good 

 for slow interruptions by hand ; or an automatic vibrator, similar to the 

 "break" of an induction-coil may be employed. Similar sparks have 

 doubtless been often noticed by those who have used any form of appa- 

 ratus involving rapid interruption of the current, and as often attributed 

 to defective insulation. The writer of this paper observed the same thing 

 some years ago in a trembler of particularly good construction, and in 

 vain rewound the coils in hope of eliminating the supposed leak. 



