68 J. Bliaduri — Transformation of [No. 2, 



trace of *' ic " iron reduced by a small crystal of sodium sulphite and the 

 solution freely boiled to expel every trace of sulphur dioxide. After 

 the precipitation of iron as ferrous hydrate, by pure sodic hydrate, the 

 alkaline chlorate was added. The whole was then boiled for a few 

 minutes. When the reduction of tlie chlorate was complete, hydro- 

 chloric acid was added to redissolve the precipitate. Reduction in 

 alkaline solution was found imperative. Many experiments were lost 

 owing to the evolution of distinct amounts of chlorine in the attempt 

 to effect the reduction in presence of free sulphuric and hydrochloric 

 acids. In all cases, however, the boiling of solutions and subsequent 

 cooling and titration were effected in an atmosphere of carbon dioxide. 



Objections have been raised by Fresenius that titrations by per- 

 manganate are not admissible in presence of free hydrochloric acid, or 

 chloride and sulphuric acid. They have, however, been disputed by 

 Dr. Fleischer. The liberation of free chlorine can be entirely prevented, 

 according to Zimmermann, by the presence of a sufficient quantity of 

 manganous sulphate, so that the determinations are rendered more exact. 

 My own observations have led me to believe that the results are perfectly 

 concordant when the titrations are made in cold dilute solutions and in 

 presence of not too large an excess of free hydrochloric acid. When excess 

 is suspected, it is to be partially neutralised with pure sodic carbonate, 

 but care must always be taken to maintain the solution acid, otherwise 

 the volume of permanganate required will be appreciably higlier. To 

 establish the above points the following experiments were performed : — 



A. Without hydrochloric acid (that present being in stannous 

 chloride). 



(i) 10-230 gms. of Sn Cl2 = 51-20 c.c. Permanganate 

 (ii) 5-897 „ „ =29-40 „ 



(iii) 5-259 „ „ =26 25 „ * 



.'. 10 gms. Sn 012=49*97 c.c. Permanganate. 



B. With 1 c.c. strong hydrochloric acid for 5 c.c. stannous chloride 

 solution present. 



(i) 5-39 gms. of Sn 012 = 26*90 c.c. Permanganate 

 (ii) 7-5 „ „ =37-45 „ 



(iii) 8-32 „ „ =41-50 „ 



.*. 10 gms. Sn Ol2 = 49'9l c.c. Permanganate. 

 It is evident from the results given above, that the difference 

 between two series of experiments (006 in 50) is very slight, and can 

 be accounted for as errors of experiments. 



The following analysis of ferrous ammonium sulphate shows strik- 

 ing concordance of the results : — 



(i) 11054 gms. of the double salt in presence of 2 c.c. concen- 



