[ 76 ] 



IX. Intelligence and Miscellaneous Articles. 



NOTE BY DR. C. R. ALDER WRIGHT ON MR. SHIDA's PAPER " ON 

 THE NUMBER OF ELECTROSTATIC UNITS IN THE ELECTROMAG- 

 NETIC UNIT," AND ON HIS RECENT NOTE THEREON. 

 rpHE note by Mr. R. Shida, in the June Number of the Philoso- 

 -*- phical Magazine (p. 473), calls for some remark, in view of the 

 great interest and importance attaching to the exact valuation of 

 the different electrical magnitudes and their correlations. Mr. 

 Shida stated, in his original paper*, that he measured the E.M.F. 

 of a gravity Daniell cell electromagnetically by determining the 

 current generated by it in a circuit the total resistance of which 

 (inclusive of the internal resistance of the cell) was measured, the 

 product of the current into the resistance being the required 

 E.M.F. ; also that he compared, by means of a quadrant electro- 

 meter, the E.M.F. of the gravity-cell with that of a 30-cell Daniell 

 battery, which had been just measured by means of an absolute 

 electrometer — this electrostatic valuation being made just before, 

 and also just after, the electromagnetic valuation. [It is noticeable, 

 in passing, that, through an obvious slip in copying his results, Mr. 

 Shida gives in each of the papers the electrostatic E.M.F. of the 

 30-cell battery as 0*904187 C.Gr.S. units, and that of the gravity- 



0*904187 

 cell as ofl.ooo =0-034381 C.Gr.S. units, instead of only one tenth 



of these quantities in each case respectively ; otherwise the final 

 value of v deduced would be only one tenth of the quantity found.] 

 In his recent note, Mr. Shida says that he omitted to state in the 

 original paper that the electrostatic valuation depended upon by 

 him was made " while the current ivas actually flowing through the 

 tangent-galvanometer" as it was during the electromagnetic valua- 

 tion ; and he consequently infers that he eliminated the possible 

 source of error incidentally suggested by me (whilst discussing the 

 subject of "polarization" of so-called constant batteries) as exist- 

 ing in his experiments, due to the E.M.F. of the cell during the 

 electromagnetic valuation whilst a current was passing not being 

 identical with that subsisting during the electrostatic valuation 

 made (as would be naturally supposed on reading his paper) when 

 no current was passing. 



It is noteworthy, however, that in striving thus to avoid the 

 Scylla of possible error through battery-polarization, Mr. Shida has 

 fallen into the Chary bdis of a yet more serious error, due to the 

 circumstance that the potential difference registered by a quadrant- 

 electrometer to which a cell is applied is not the E.M.F. of that cell 

 if it generate a current, but is something less. That this is so, is mani- 

 fest from the following considerations. Let the actual E.M.F. of 

 the cell be E 1? and the E.M.F. obtained by the quadrant-electro- 

 meter be E 2 . Let C be the current generated, E x the internal 

 resistance of the cell, and R 2 the external resistance of the rest of 



E 

 the circuit. Then C= ^ — r 1 ^-. The total work which this current 



* Phil. Mag. Dec. 1880, p. 431 j also Brit. Assoc. Keports, 1880, p. 497. 



