Length of the Break-Spark in an Inductive Circuit. 653 



evasement disparait graduellement quand la temperature 

 s'eleve ; pour les gaz comme Thydrogene, Fair et l'azote, la 

 variation avec la temperature est tres peu sensible. 



" Si on compare la decroissance de ces coefficients entre les 

 memes limites de pression a differentes temperatures on ne 

 trouve aucune variation assez nettement indiquee pour en 

 preciser le sens, d'autant que precisement pour les groupes 

 de series comportant des limites de temperature assez etendues 

 les limites de pression sont restreintes a 1000 atmospheres ; 

 il est done difficile de prevoir si le coefficient angulaire sous 

 uue pression suffisante, ou a une temperature assez elevee 

 tend vers une valeur limite ; dans tous les cas, les plus petites 

 valeurs de ces coefficients sont des limites sup^rieures des 

 plus petits volumes possibles ; il pourrait etre interessant de 

 comparer " 



If I have insisted on the preceding facts it is because several 

 physicists in addition to Mr. Rose-lnnes have brought them 

 forward recently without taking into account my last memoir 

 on the subject, in consequence of which I might appear to 

 have introduced into science results altogether different from 

 those that I have, on the contrary, taken care to describe at 

 length, with details, and on which I have insisted many times 

 since the publication of my memoir of 1893. (Cf. on this 

 subject mv Report to the Congres international de Physique 

 held in Paris in 1900.) 

 Paris, 23 August, 1901. 



LXII. Note on the Length of the Break-Spark in an Inductive 

 Circuit. By R. Beattik, B.Sc * 



RECENTLY 1 had occasion to make a few measurements 

 of the length of the spark which appears on suddenly 

 interrupting an inductive, current-carrying, electric circuit, 

 my object being to gain some definite information regarding the 

 influence of the material of tin 1 break-poles on the spark-length; 

 and as I have been unable to find any reference to experiments 

 of a similar kind done elsewhere, it may not be out of place 

 to describe here, in brief, the main results arrived at. 



Jn order to obtain sparks of easily measurable length the 

 self-inductance of the circuit was made large — about 1 henry 

 — by the introduction of an electromagnet which, as it 

 happened, had a core of somewhat small cross-sectional area. 

 For the particular purpose in view this was rather advantageous 

 than otherwise, since it led to the rate of decay of current at 



* Communicated bv the Author. 



