8 Mr, Grove^ on the Electrical Discharge^ [Jan. 28, 



which secondary current the brilliant effects of the Rnhmkorff coil are 

 due : but in addition to this current in the secondary wire, there is 

 also a secondary current in the primary wire, flowing in the same 

 direction ; the induction spark, at the moment following the disruption 

 of contact, completing the circuit of the primary, and thus allowing the 

 secondary current to pass. This secondary current in the primary wire 

 produces in its turn another secondary, or what may be termed a 

 tertiary, current in the secondary wire, in an opposite direction to the 

 secondary current. There are thus, almost synchronously, two currents 

 in opposite directions in the secondary wire ; these, by causing a 

 conflict or irregular action on the rarefied medium, would give rise to 

 waves or pulsations, and might well account for the stratified appear- 

 ance. The experimental evidence in favour of this view is as follows : 

 when a single break of battery contact is made by drawing a stout 

 copper wire over another wire, the striae do not invariably appear 

 in the rarefied medium through which the current of the secondary 

 wire passes. This would be accounted for on the above theory by 

 supposing that in some cases of disruption the induced spark passes 

 across immediately on disruption, and thus completes the circuit for 

 the secondary current in the primary wire ; while in other cases, either 

 from want of sufficient intensity, or from the mode or velocity with 

 which contact is broken, or from the oxidation of the points where 

 contact is broken, there is no induced spark by which the current can 

 pass : in the former case there would be a tertiary current in the 

 secondary wire, and therefore striae ; in the latter there would be 

 none. 



But the following experiment is more strongly in favour of the 

 theory. It is obvious that the secondary must be more powerful than 

 the tertiary current. Now supposing an obstacle or resistance placed 

 in the secondary circuit, which the secondary current can overcome 

 but the tertiary cannot, we ought by the theory to get no striae. If an 

 interruption be made in the secondary current in addition to that 

 formed by the rarefied medium, and this interruption be made of the 

 full extent which the spark will pass, there are, as a general rule, no 

 striae in the rarefied medium, while the same vacuum tube shows the 

 striae well if there be no such break or interruption. The experiment 

 was shown by a large vacuum cylinder (16 inches by 4) of Mr. 

 Gassiot, and his micrometer electrometer ; this tube showed numerous 

 broad and perfectly distinct bands when the points of the micrometer 

 were in contact ; but when they were separated to the fullest extent 

 that would allow sparks to pass, not the slightest symptom of bands 

 or striae were perceptible, the whole cylinder was filled with an uniform 

 lambent flame. With a spark from the prime conductor of the elec- 

 trical machine, the striae do not appear in tubes which show them well 

 with the Ruhmkorff coil ; occasionally, and in rare instances, striae 

 may be seen with sparks from the electrical machine, but, not as far 

 as Mr. Grove has observed, when the spark is unquestionably single. 

 All this is in favour of the theory given above j but without regarding 

 that as conclusive or as a proved rationale, it is clearly demonstrated 



