38 Sir Robert Kane on the Composition and Characters 



white, as such would require a temperature capable of mate- 

 rially altering the constitution of the ash, a fact of which I 

 have been long aware, and which has latterly fixed the atten- 

 tion of several chemists. The ash so prepared was carefully 

 dried in a stove, and then treated in the following manner: — 



Dilute muriatic acid having been poured over the quantity 

 of ash selected for analysis, the whole was heated in a water- 

 bath until it dried completely down ; water was then added, 

 and when the soluble materials had been completely taken up, 

 the whole was thrown upon a weighed filter and the liquor 

 separated ; there remained upon the filter such particles of 

 sand or soil as had been adherent to the plants, the unburned 

 charcoal of the ash, and the silica which had existed in the ash, 

 either free or in combination with alkaline or earthy bases. 



The weight of this insoluble residue having been properly 

 determined, it was boiled in a strong solution of caustic potash, 

 by which all the proper silica of the ash was taken up, and 

 the residue then remaining being weighed, gave the sand and 

 charcoal, the silica being thus determined by difference. 



The muriatic solution was then divided into three parts for 

 the determination — 



1. Of the alkaline constituents. 



2. Of the phosphoric acid, manganese, alumina, magnesia, 

 and lime. 



3. Of the sulphuric acid and oxide of iron. 



The first portion of solution was rendered slightly alkaline 

 by carbonate of ammonia, and then mixed with solution of 

 caustic barytes in excess, and allowed to stand for some hours. 

 By this means the sulphuric and phosphoric acid were perfectly 

 removed, as well as the earthy constituents, except a small 

 quantity of lime, which remained dissolved in a caustic state, 

 and which was then perfectly removed by the addition in ex- 

 cess of a mixture of caustic and carbonated ammonia. The 

 liquor, after filtration, was evaporated to dryness, and the 

 residue gently ignited, when the ammoniacal salts were per- 

 fectly expelled : there remained the alkalies of the ash as 

 chlorides. This residue was weighed, then dissolved in water, 

 and a solution of bichloride of platinum added. The liquor 

 and precipitate were then evaporated nearly to dryness, the 

 potash platinum salt washed by a mixture of alcohol and aether, 

 and the amount of platinum determined in the usual way. 

 The soda was ascertained by subtracting the weight of the 

 chloride of potassium from the weight of the mixed chlorides, 

 as given in the first instance. 



To the second portion of the liquor was added so much 

 ammonia as nearly neutralized it without producing any per- 



