THE 



AMERICAN JOURNAL OF SCIENCE 



[FOURTH SERIES.] 



Aet. XII. — On Rowland? s new Method for measuring 

 Electric Absorption, and Losses of Energy due to Hysteresis 

 and Foucault Currents, and on the Detection of short 

 Circuits in Coils ; by Louis M. Potts. 



The following investigation has had as its object the testing 

 of methods devised by Professor Rowland* for the measure- 

 ment, in the first place, of electric absorption ; further, of the 

 energy losses due to hysteresis and Foucault currents ; and, 

 finally, for the detection of short circuits in coils. 



I. Electric Absorption. 

 Historical. — It has long been known that a Leyden jar, 

 which has been charged and then discharged, will show another 

 charge after standing a short time. If this is discharged, after 

 a short time the jar will show another charge ; this may be 

 repeated indefinitely. These " after-charges " are known as 

 residual charges and are due to the phenomenon now called 

 electric absorption. Faradayf made some experiments on this 

 phenomenon in Leyden jars, and seems to have attributed it to 

 a conduction of the charge into the interior of the dielectric, 

 and after discharge creeping back again to the coatings and 

 manifesting itself as the residual charge. Kohlrausch^: was the 

 first to make any elaborate investigation of the subject. He 

 charged the condenser and then measured the potential at cer- 

 tain intervals with an electrometer. In this way he obtained 

 the relation between the potential and time. He advanced the 

 idea that the phenomenon was due to an electric polarity of 



*See this Journal for July, 1899, pp. 35-57. 



f Faraday, Experimental Researches. 



% Pogg. Annalen, vol. xci, pp. 59-82, 179-214. 



Am. Jour. Sci. — Fourth Series, Vol. X, No. 56.— August, 1900. 



7 



