SOT SPRINGS, ARKANSAS. 23 



ferred to a crucible, burned and blasted in the ordinary way, and 

 finally weighed as calcium oxide. The combined nitrates were evapo- 

 rated* to dryness in platinum and the major part of the ammonium 

 salts driven off by the aid of heat. The residue was dissolved in dilute 

 hvdrochloric acid and filtered. The filtrate was made slightly ammo- 

 niacal, enough sodium phosphate solution added, a drop at a time, to 

 precipitate all magnesium, and 10 cc. of concentrated ammonia finally 

 added, drop by drop. The beaker was covered and allowed to stand 

 over night, filtered, washed with dilute ammonia water, dried, blasted, 

 and weighed as magnesium p3 T rophosphate. 



Sulphuric acid, potassium, sodium, and lithium. — Another por- 

 tion of the above filtrate was precipitated while boiling with hot, dilute 

 barium chloride, and after standing filtered from the precipitated 

 barium sulphate, which was washed, dried, burned, and finally weighed 

 in the ordinary way. 



The filtrate was evaporated to diyness and taken up with water. 

 This solution was precipitated with a solution of barium hydrate and 

 filtered off from the insoluble magnesium hydrate. The magnesium 

 hydrate precipitate was well washed and the combined filtrate and 

 washings treated with ammonia, ammonium carbonate, and a little 

 ammonium oxalate to precipitate barium and calcium. This precipi- 

 tate was allowed to stand over night, filtered off and well washed. 

 The filtrate and washings were evaporated to diyness on the steam 

 bath, dried, and all of the ammonium salts driven off by gentle heat. 

 The residue was taken up with water, filtered through a small filter, 

 using as little wash water as possible, evaporated to a small volume, 

 and finally again precipitated with a drop of ammonia and two to 

 three drops of ammonium carbonate and oxalate. If any precipitate 

 appeared, which was not usually the case, it was filtered off and the 

 same process repeated. In any case, the solution was filtered from the 

 magnesium hydrate that had precipitated out on concentrating the 

 solution. The filtrate was then evaporated to diyness and all ammo- 

 nium salts driven off by heating in platinum to a little below redness. 

 The residue was taken up with a little water and filtered through a 

 small filter, again using as little wash water as possible, and again 

 heated in platinum to a point slightly below T red heat. By this time 

 all of the magnesia should have been removed. The residue was 

 then taken up with a little water, filtered into a weighed platinum dish, 

 treated with a few drops of hydrochloric acid, and evaporated to dry- 

 ness. This residue was thoroughlv dried, heated to a little below 

 redness, cooled in a desiccator, and finally weighed as the combined 

 chlorides of potassium, sodium, and lithium. 



The determination of lithium was then made according to the method 

 of Gooch a — i. e., the combined chlorides were dissolved in water and 

 transferred to a small beaker,- where they were again evaporated nearty 

 to diyness. About 30 cc. of anyl alcohol was added and the contents 

 of the beaker boiled until the temperature had risen to approximately 

 the boiling point of the anyl alcohol, showing that all of the water had 

 been driven off. The liquid was cooled slight!} 7 , and a drop of hydro- 

 chloric acid was. added to reconvert small amounts of lithium hydrate 

 to lithium chloride. The boiling was then continued to again drive off 

 all water, until finally the liquid had reached a volume of about 15 cc. 



a American Chemical Journal, vol. 9, p. 33. 



