225 



To eliminate the effect of damping due to eddy currents two sugges- 

 tions have been made. (1) to insert in the galvanometer circuit a special 

 key so arranged as to brealv the galvanometer circuit a brief instant after 

 the charge has passed, thus securing always the open circuit conditions of 

 curve D ; (2) to insert a special key in the galvanometer circuit so ar- 

 ranged that the galvanometer will always be closed through a circuit of 

 constant resistance. Both of these methods are satisfactory, but only with 

 perfectly operating keys, which condition is not easy to secure. 



By far the safest and most convenient procedure is then to calihrait 

 the gnlvanoir.etcr for the precise eomlitions under which it is to be used. 

 This may readily be carried out by permanently including in the galvano- 

 meter circuit the secondary coil of a standard nuitual inductance, and by 

 simply reversing a known current in the primary circuit the constant can 

 be accurately determined from the resulting throw. 



For a standard of nuitual inductance it has long been customary tn 

 rely on the long solenoid with a short coaxial solenoid fur a secondary 

 coil. Unless these are well made, with exceptional care and by experienced 

 hands, they are by no means standard. The writer has measured the mu- 

 tual inductance of a large number of such coils from different makers, and 

 the subjoined table will show the discordances between me:isui-ed and cal- 

 culated values for a few of them. 



The calculated values were all secured from the approximate formula 

 based on 4 ^r ni as the value of the field at the center of a solenoid, while 

 the measured values were obtained by IMaxwell's method, by comparison 



[15—29034] 



