302 Mr. A. Gray on the Number of Electrostatic 



rately than if the cell had been applied directly to the absolute 

 electrometer, in absolute electrostatic units a difference of 

 potential equal, or nearly equal, to the E.M.F. of a DanielPs cell, 

 by a quadrant-electrometer. Thus the quadrant-electrometer, 

 a much more convenient instrument than the absolute electro- 

 meter for rapid experimenting, and for the purpose to which it 

 was to be applied, could be used in the remainder of the inves- 

 tigation, which, as appears clearly from Mr. Shida's paper, 

 supplemented by his note in the June number, consisted in 

 the measurement, both electrostatically and electromagneti- 

 cally, of one and the same electromotive force. 



For the electromagnetic measurement, the cell was placed 

 in circuit with a resistance-coil and a tangent-galvanometer, 

 and the current measured as described by Mr. Shida. But Mr. 

 Shida has since stated that, while the current was thus flowing 

 through the galvanometer, he measured the difference of poten- 

 tials at the two ends of the external resistance by the quad- 

 rant-electrometer, and calculated from this, by a simple appli- 

 cation of Ohm's law, the E.M.F. of the cell while generating 

 a current. There could therefore be no question here of pola- 

 rization-error ; for the method is plainly equivalent to deter- 

 mining simultaneously, in both kinds of units, the difference 

 of potentials between the two ends of the external resistance. 

 These measurements, Mr. Shida has told me, he has since care- 

 fully repeated with different resistances in circuit, obtaining 

 results which one and all differed only very slightly from the 

 E.M.F. of the cell with its poles insulated. Hence, even if 

 the assumption deprecated by Dr. Wright had been made, no 

 appreciable error in the result would have been introduced. 

 Of course I do not overlook the fact that the determination 

 of the resistance of the cell by the ordinary method of measu- 

 ring the difference of potentials between the poles of the cell 

 when insulated, and again when connected by a known resist- 

 ance, will be in error if there is a diminution of E.M.F. due 

 to polarization in the latter case ; but even a considerable per- 

 centage of error from this cause will not, in the calculation by 

 Ohm's law of the E.M.F. of the cell when generating a cur- 

 rent through a large resistance (such as that used by Mr. 

 Shida), mask the diminution, if sensible, of E.M.F. produced 

 by the current through that resistance. 



These results of Mr. Shida' s have been confirmed by expe- 

 riments which I have myself made to determine the diminu- 

 tion, if any, of E.M.F. caused by polarization in a Thomson's 

 gravity Daniell* similar to that used by Mr. Shida. By 



* This cell consists of a copper plate about 720 square centirn. in area 

 and a zinc grating of about the same area separated by a diaphragm of 



