262 Prof. J. A. Fleming on a Method of determining 



In the case I am considering the Plante cells evidently act 

 like leaky leyden jars coupled in series. If C is the apparent 



capacity of one cell, — would be the capacity of n cells. 



An examination of the photographs of the oscillations pro- 

 duced by 2500 cells showed an apparent capacity of about 

 1000 electrostatic units. Five thousand cells gave an apparent 

 capacity of about 500 units. The small apparent capacity C 

 results from the leaking of the condenser due to the conduction 

 through the electrolyte. Since the discharge from an accumu- 

 lator of a large number of cells is in general oscillatory, I am 

 led to the belief that the discharge from any primary battery 

 is also oscillatory for all cases in which we have to deal with 

 capacity and self-induction. It is evident that a galvanometer 

 or telephone in circuit with a Geissler tube cannot detect the 

 oscillatory discharge, since it is of high period. Moreover, 

 when a Geissler tube is lighted by a large battery with no 

 resistance save that of the tube and the battery in the circuit, 

 and the light is examined in a revolving mirror by the eye, 

 no oscillation or intermittence of light can be perceived on 

 account of the flaming discharge through the rarefied gas. 



The oscillatory discharge may be said to be the common 

 occurrence in nature in the case of electrical discharges, and 

 the one-direction discharge the uncommon. This has been 

 expressed by the remark that electricity takes the path of 

 least resistance, this common belief must be modified, however, 

 under certain conditions of resonance ; but in general nature 

 avoids a unidh-ectional discharge, 



Jefferson Physical Laboratory, 



Harvard University, 



Cambridge, Mass., U.S. 



XXXV. A Method of determining Magnetic Hysteresis loss in 

 Straight Iron Strips. By J. A. Fleming, M.A., D.Sc, 

 F.R.S., Professor of Electrical Engineering in University 

 Collfqe, London *. 



[Plate V.] 



THE exact determination of the magnetic hysteresis loss 

 in sheet iron as used for alternating current transformer 

 construction has become of commercial importance of late 

 years. Owing to the obvious necessity for reducing the core 

 loss in transformers to the lowest attainable limit, it is now the 

 custom to carefully test the iron to be used and to express 

 * Communicate^ by the Physical Society: read June 11th, 1897. 



