492 Dr. Norman Campbell on the "Break " 



break when the current broken was about 002 amp., the 

 speed o£ the break being about 250 cm./sec. The "spark" 

 now looked like a spark, and did not take the form of the 

 " flash " which appeared when the current was broken in 

 the primary. Further, it could now be definitely established 

 that the limiting current, which could be broken without 

 the appearance of a spark, increased notably with the speed 

 of the break, and, as far as could be ascertained, it was 

 independent of the form of the terminals and of the pressure 

 between them. All these differences are precisely those 

 which are to be expected if the discharge is now a true 

 spark due to the potential across the gap exceeding the 

 normal spark potential. However, one discrepancy must 

 be noted. Though the limiting current increased with the 

 speed of separation, it did not increase as rapidly as was to 

 be expected. When the speed of separation was doubled, 

 the limiting current increased only some 50 per cent.; the 

 reason for this discrepancy has not been determined. 



Observations similar to those shown in fig. 1 were made of 

 the rise of the potential across the gap, both when the spark 

 did occur and when it did not. It was found that, when the 

 spark occurred, the initial part of the curve (corresponding 

 to a time of the order of 10 ~ 5 sec. and a separation of the 

 contacts of the order of *03 mm.) was precisely the same as 

 when the spark did not occur ; but that, at some point on 

 the curve, the readings became wholly irregular and no more 

 consistent measurements could be obtained. The point at 

 which this irregularity sets in is, of course, that at which the 

 spark occurs. 



The observations need not be described further. Those 

 already mentioned are sufficient to prove the only fact which 

 is of much importance in this connexion, that the obser- 

 vations when the spark occurs are wholly different from 

 those in the conditions investigated in the earlier experi- 

 ments and, accordingly, that in those conditions the discharge 

 which is seen to occur is not a true spark. It may be 

 asserted confidently that the conditions necessary for the 

 passage of the true spark are never likely to occur when the 

 primary of a magneto is broken, and that if arcing can be 

 avoided, the break will be perfect. 



(7) Note on the Resistance of Contacts. — In the course of 

 these experiments some observations have been made which 

 are of some interest and do not seem to have been described 

 previously. 



Suppose that we pass through a contact between two 



