26 



MINERAL WATERS OF THE UNITED STATES. 



II. Two grams of sodium chlorid, 4 mg of potassium bromid, and 

 0.8 mg of potassium iodid were dissolved in a liter of water. The 

 same procedure was followed as in experiment I. Four mg of potas- 

 sium bromid and 0.6 mg of potassium iodid were recovered. 



Ill* Two artificial waters were then made up having the following 

 composition: 



Composition of 3,000-gram samples of waters artificially made up for experiment. 



These waters were evaporated with sodium carbonate, the residues 

 extracted with water, filtered, and made up to definite volumes. Such 

 aliquot portions of the filtrates were taken for analysis as would rep- 

 resent one-third of the original sample. From these samples, contain- 

 ing in No. I 10.4 parts per million of potassium bromid and 0.3 part 

 per million potassium iodid and in No. II 3 parts per million of potas- 

 sium bromid and 0.1 part per million of potassium iodid, the following 

 amounts of the two constituents were recovered: In No. 1, 11 parts per 

 million of potassium bromid and 0.25 part per million of potassium 

 iodid, and in No. II, 3.5 parts per million of potassium bromid and 0.06 

 part per million of potassium iodid. 



IV. Working on Blue Lick water the following results (expressed 

 in parts per million) were obtained when the samples were run in 

 triplicate and the amounts of water used for the determination varied: 

 Potassium iodid, 0.17, 0.16, and 0.15; potassium bromid, 33.3, 33.4, 

 and 36. 



V. Working on Saratoga Champion the following duplicate results 

 were obtained: Potassium iodid 0.2 and 0.25 parts per million, and 

 potassium bromid 20 and 20 parts per million. 



While the authors do not claim any great scientific exactness for the 

 above method, they do believe that it gives fairly accurate results within 

 its limits. It certainly permits the determination of much smaller 

 amounts of iodin and bromin than can be made by any of the methods 

 in common use, and thus avoids the tedious evaporation of large quan- 

 tities of water and the consequent difficulties which arise from the 

 presence of large quantities of salts in solution. Fortunate ly most 

 mineral waters containing iodin and bromin fall within the limits of 

 this method. In only four of the numerous samples examined was it 

 necessarv to resort to the gravimetric method of determination. 



