144 Lord Rayleigh and Mrs. Sidgwick. [June 10, 



the attraction to the current, the error is halved by the extraction of 

 the square root. 



The currents actually employed were about ^ ampere. Much more 

 powerful currents could not be passed for the necessary time through 

 the suspended coil without risk of undue heating. Had it been 

 desirable to use stronger currents, it would, of course, have been 

 possible to do so by the use of thicker wire. With given grooves to 

 be filled up, the ratio of the electromagnetic attraction to the heat 

 developed is independent of the gauge of the wire ; and the only 

 further modification required would be the multiplication of the fine 

 copper wires by which the flexible connexions are made. 



Thirteen determinations of the ratio of the square root of the 

 double attraction of the coils to the rate of silver deposit gave 

 numbers ranging from 2413' 7 to 2415'5, mean 24l4'45 ; whence, 

 after introduction of the constant of the apparatus, the value of the 

 electrochemical equivalent in C.G.S. measure is found to be 



0111794. 



Tn terms of practical units, we have as the quantity of silver 

 in grams deposited per ampere per hour 



4-0246. 



With use of this number, it is now easy to determine by the deposit 

 of silver currents up to l ampere. For currents of greater power it 

 is necessary either to increase the size of the (3-inch) platinum basins 

 serving as voltameters, or to dispose several such in multiple arc. For 

 ordinary practical purposes, many of the precautions which we 

 thought it necessary to adopt may be dispensed with. The deposits, 

 after a few rinsings with distilled water, may be left to soak for an 

 hour, and then, after a further rinsing, dried off over a spirit lamp. 

 In an hour's time, the basin may be weighed correctly to a few tenths 

 of a milligram. With regard to the materials, it is sufficient to use 

 for the anodes a sheet of ordinary fine silver (such as is sold at 5*. 

 per ounce), and a 15 or 30 per cent, solution of nitrate. It is hoped 

 thafc this method may come into general use for the verification of 

 current-measuring instruments, whose indications depend upon the 

 constancy of springs, or of steel magnets. Silver presents so many 

 advantages as compared with copper, that its greater cost should not 

 stand in the way of its adoption, more especially as there need be no 

 great waste of material. 



In view of the importance of obtaining a convenient standard of 

 electromotive force, a prolonged examination has been made of a 

 number of Clark cells. Of these two patterns have been used, the 

 first constructed according to the directions of Clark himself and 

 of Alder Wright, with some simplifications ; the second, called for 



