116 Professors Ayrton and Perry on 



the lower part of the scale, this iron core is not saturated, and 

 differences in the values of the readings of as much -as 30 

 to 40 per cent, are obtained, depending on whether it is an 

 increasing or a decreasing current that is being measured with 

 such an electromagnetic instrument. 



XL The fourth error, arising from the volt standards em- 

 ployed by various people differing several per cent, from 

 one another, has been hitherto a very serious matter. The 

 standards employed have usually been — 



1. The E.M.F. of a Darnell's cell. 



2. The E.M.F. of a Latimer Clark's cell. 



3. A standard depending on the electrochemical equivalent 

 of silver and the value of the legal ohm. 



4. The indication of a voltmeter graduated by some well- 

 known maker, but probably graduated in a way quite unknown 

 to the user. 



Standard 1 is vague unless the details of the method 

 employed in setting up the Daniell's cell are more carefully 

 specified than heretofore ; for example, Carpart, in some 

 recent tests of the E.M.F of a Daniell's cell, made out of a 

 U-tube, with a contraction in the lower part, found it to vary 

 from 1-111 legal volts when the zinc-sulphate solution had 

 25 per cent, of ZnS0 4 , to 1*142 legal volts when the solution 

 had only 3 per cent, of ZnS0 4 . His method of testing is 

 based on the electrochemical equivalent of silver and the 

 legal ohm, and as a proof of its accuracy he finds that the 

 E.M.F. of a Latimer Clark's cell is 1*434 volts. If, however, 

 we compare his results for a Daniell's cell with 1*072 volts 

 as found by Sir W. Thomson when the zinc-sulphate solution 

 had a specific gravity of 1*2, we see how Daniell's cells differ 

 among one another*. 



The E.M.F. of a Latimer Clark's cell has also some vague- 

 ness attached to it, unless the temperature-coefficient of varia- 

 tion of E.M.F. of the particular Latimer Clark's cell is known 

 and taken account of. Generally the users of these cells are 

 contented to say that the E.M.F. of a Latimer Clark's cell, 

 as determined by Lord Rayleigh, is 1*435 true volts, or 

 1*438 legal volts, and they neglect the temperature variation 

 altogether, which amounts to about 1 per cent, diminution 

 per 12° C. elevation of temperature with certain Latimer 

 Clark's cells. 



* Since the reading of this paper this subject has been very fully 

 treated by Dr. Fleming, in a paper " On the Use of Daniell's Cell as a 

 Standard of Electromotive Force/' read before the Physical Society on 

 June 27, 1885 [Phil. Mag. [5] vol. xx. p. 126 (August 1885)]. . 



