38 JOURNAL OF THE 



as precipitate formed ; [addiug the reagents, drop by drop with con- 

 stant stirring, is supposed to prevent the mechanical incorporation 

 by the precipitates of any of the potash salt, and for same reasons 

 unnecessary excesses of the precipitants are to be avoided,] filtered ; 

 evaporated the filtrate to dryness in platinum dish, drove off am- 

 monia salts and took up in hot water; filtered through a small filter 

 into a porcelain dish, added 10 c. c. of solution of chloride of pla- 

 tinum (PtCl4) and evaporated so low that the mass solidified on 

 cooling; took up in 85 per cent, alcohol, filtered through a small 

 filter and thoroughly dried, filter and precipitate; weighed them in 

 a weighing bottle; extracted KoPtClfi from filter with hot water, 

 thoroughly dried the filter again and weighed in weighing bottle. 

 Difference is KoPtCl,; yielded by 0.1 gram. K0SO4. 



b. Heated to boiling; added BaClg, drop by drop with constant 

 stirring, in slight excess; therj, without filtering, precipitated the 

 excess of barium by (NH^^-gCOg, drop by drop with constant stir- 

 ring, after having made the solution alkaline with 1 c. c. NH^HO 

 as in a; filtered ; the rest of the manipulation is identical with a. 

 [The only difference between a and 6 is one more filtration in a. In 

 6 the barium sulphate and barium carbonate precipitates were 

 thrown down together; in a, separately, a filtration intervening. 

 The double filtration in a is not necessary but it extends the manip- 

 ulation.] 



c. Was treated Jw6'i like a fertilizer, by the so-called " Washing- 

 ton method," proposed by the committee who sent out the samples. 



The remaining five samples were treated in the same way. In 

 case of the kainit (No. 2) 10 grams uere dissolved up to 1,000 c. c. 

 and 25 c. c. taken for the analysis, equal to 0.25 grams substance; 

 in the cases of 3, 4, 5 and 6, 10 grams were dissolved up to 1,000 c. 

 c. and 100 c. c. taken for the analysis, equal to 1.0 gram of sub- 

 stance. 



Perhaps a synopsis of the method had best be given : 

 Heated to boiling; added BaCL, drop by drop with constant stir- 

 ring, in slight excess; then added (without filtering) Ba(HO)o, drop 

 by drop with constant stirring, to al&aline reaction; filtered; to fil- 

 trate added 1 c. c. strong NH4HO, and then, drop by drop with con- 

 stant stirring, a saturated solution of (NH4)2C03 as long as a pre- 

 cipitate-formed. [Just here, instead of addina: ammonium oxalate 

 (or oxalic acid) immediately, as the method calls for, I filtered from 

 the precipitate with (NH4)oC03 and added the ammonium oxalate 

 (0.75 gram) to the filtrate. The result was that a slight precipitate 

 appeared in every case, except that of the chemically pure K2SO4, 



