ON THE SILVER VOLTAMETER. 585 



had been retained, and that this was the cause of the abnormal character of the 



ilr|i"sit. Tin 1 Iliakrix \\n\\ anli-d ;i v.-,-,,::,! v ;l lli|i]r \\!il<-!i MM <TV>-talli-i-il t'p-Ill a 



neutral liquid, and normal results (W,) were obtained with this. 



C. Standard Method of Preparing Silver Nitrate Solutions for Electrolysis. 



It is now possible to state the conditions that 'should be complied with in preparing 

 silver nitrate solutions for use in the voltameter. 



(1) If commercial " pure " silver nitrate is used, a part of it should be purified by 

 recrystallising twice from water and the deposit compared with that from the 

 original sample. If the values agree, this can be used without purification ; if not, the 

 \\hole of the sample should be twice recrystallised. For rough work in which an 

 error less than O'l per cent, may be neglected the commercial nitrate may be used 

 directly without testing or purifying. 



(2) If recovered silver nitrate is used it should be freed from acid by evaporating to 

 dryness and heating in the oven at 140 C. until the blue colour of the copper nitrate 

 (if present) is destroyed, then dissolved, filtered and recrystallised until the mother- 

 liquor drained from the crystals is colourless, then once again recrystallised before 

 being used for electrolysis. 



(3) In crystallising the nitrate it is desirable to effect the dissolution of the crystals 

 by heating on a water-bath, rather than over a bare flame, so as to avoid all risk of 

 overheating the solution, and to dissolve in a conical flask of Jena glass rather than in 

 a beaker, so as to reduce the risk of contamination by exposure to the air. 



(4) For filtering the hot solution we prefer to use a Hirsch porcelain funnel, the 

 perforated plate of which is covered by two discs of filter paper. The funnel is 

 attached to a filter pump and warmed by pouring lx>iling distilled water through it ; 

 there is then but little risk that the hot solution will crystallise in the filter. The 

 filtered solution is allowed to crystallise in the pump-flask, so as to avoid unnecessary 

 exposure to the air. If the flask is cautiously cooled and shaken it is generally 

 possible to secure the separation of the nitrate in small crystals ; these can subse- 

 quently be drained much more effectively than the larger crystals, which separate 

 when the solution is allowed to cool slowly and without disturbance. As an alterna- 

 tive the filtered solution may be left to deposit large crystals from which the mother- 

 liquor can be poured off, but in this case the separation of the mother-liquor is much 

 less complete and a larger number of crystallisations is required. Towards the 

 end of the crystallisation the flask may be cooled in ice, so as to reduce the amount 

 of material left behind in the mother-liquors.* A porcelain filter funnel may be used 

 without filter paper for collecting and draining the crystals ; when these are well 

 pressed down in the funnel, most of the mother-liquor can be removed by means of a 



* 100 grammes of water dissolve 115 grammes AgNOj at 0, 160 grammes at 10*, and 215 grammes 

 at 20 C. 



VOL. CCVII. A. 4 F 



