On a Standard Tension- Galvanometer. 27 



must be introduced into the bulb either by alternately sucking 

 and blowing through a piece of india-rubber tubing connected 

 with the end of i, or by disconnecting the india-rubber tube 

 leading from the reservoir, emptying the cistern cd, and 

 cautiously inclining the distiller until small bubbles of air 

 enter at the end of the tube a b and rise into the bulb. The 

 mercury then sinks into the cistern, and may be withdrawn 

 through the tube/. 



The Sprengel-tube i should be carefully cleaned and dried 

 before putting the apparatus together. 



The first time of using the distiller, it is interesting to 

 observe that, as the tube a b gradually becomes heated, the 

 surface air-film detaches itself from the glass and rises into 

 the bulb. 



The quantity of mercury distilled by an apparatus of the 

 size and form here described is about 2 lb. an hour: very 

 little gas is used, as the flames should be less than a quarter 

 of an inch high and never allowed to come in contact with the 

 glass bulb. For commercial purposes, an iron mercury-bottle 

 and iron gas-pipe might advantageously replace the glass bulb 

 and tube. 



It seems to be possible that the latter modification of the 

 method may be applicable to the distillation of some other 

 metals, such as zinc, magnesium, cadmium, arsenic, &c. 



III. On a Standard Tension- Galvanometer. 

 By E. H. M. Bosanquet. 



To the Editors of the Philosophical Magazine and Journal. 

 Gentlemen, 

 npENSION-galvanometers employ the small current flowing 

 JL through a high resistance under a given difference of 

 potential as a measure of that difference of potential. 



In all cases, so far as I am aware, such instruments have 

 hitherto been standardized by comparison with normal voltaic 

 cells of some description, whose E.M.F. in volts is supposed 

 to be known ; such as Daniell's cell, or the cell of Latimer 

 Clark. 



This proceeding is liable to uncertainties, which may be 

 called gross when compared with other electrical measures. 



I propose to describe a form of tangent-galvanometer 

 for the measurement of tensions independently of any such 

 process of calibration. The accuracy depends only on the 

 usual measures of dimensions and the knowledge of the hori- 

 zontal component of the earth's magnetism, as in the case of 

 the ordinary tangent-galvanometer, besides the measurement 



