20 HOT SPRINGS, ARKANSAS. 



tion for iodine and bromine as did a known sample of water containing 

 0.2 milligram of both iodine and bromine, as iodides and bromides, to 

 the liter. 



An attempt was made to determine iodine and bromine quantita- 

 tively in spring- No. 15 by evaporating- down a large volume of 

 water, but the attempt failed because both these elements were pres- 

 ent in such minute traces. The method used was the same as that 

 described by Gooeh and Whitfield a and is as follows: The iodides and 

 bromides were extracted with 90 per cent alcohol in the same manner 

 as described above. 



The alcohol extract was evaporated to dryness, acidulated with dilute 

 sulphuric acid, mixed with a ferric sulphate solution, and distilled 

 from a retort which was joined to a condenser sealed by a U-tube filled 

 with water and carbon bisulphide. If a very small amount of iodine 

 had been present it would have colored the carbon bisulphide and 

 could have been titrated with sodium thiosulphate, but not enough 

 was present. 



After the distillation had been continued long enough to be sure 

 that all iodine had been volatilized, crystals of potassium permanga- 

 nate were added and the distillation continued the same as before, 

 except that the U-tube acting as a seal was now filled with water and 

 chloroform. The contents of the tube were treated with sodium 

 hydroxide and zinc in a breaker and the chloride and bromide solution 

 so formed acidified with nitric acid and precipitated with silver nitrate. 

 The precipitate was dried and weighed. It was then dissolved in potas- 

 sium cyanide and the silver precipitated b} r electrolysis. 1 ' In this way 

 data on the weight of the combined silver chloride and bromide and 

 the weight of the silver in same was determined. From this the weight 

 of the bromine could be calculated, which in this case was nothing. 



Arsenic acid. — An aliquot portion of the above filtrate was acidified 

 with hydrochloric acid, the solution heated to 70° C, and a current 

 of hydrogen sulphide passed through for several hours. In case either 

 arsenic, copper, or lead were present they would be precipitated. No 

 precipitation took place in any of the springs. 



Boric acid. — A test for boric acid was made in the following manner: 

 A part of the above filtrate was evaporated to dryness, treated with 

 a cubic centimeter or two of water and slightly acidified with hydro- 

 chloric acid. About 25 or 30 c. c. of absolute alcohol was added, the 

 solution boiled and filtered. This was repeated. The filtrate was 

 made slightly alkaline with sodium hydrate and evaporated to dry- 

 ness. A very little water was added, the solution slightly acidified 

 with hydrochloric acid, and a strip of tumeric paper placed in the 

 liquid. The whole was evaporated to dryness on "the steam bath, and 

 the heating continued until the tumeric paper had become entirely dry. 

 In case boric acid were present the tumeric paper took on a cherry- 

 red color. 



A quantitative determination of boric acid was made in the case of 

 two springs to serve as an example of all the other springs. 



The method used was the same as that described by Gooch, c except 



that a slightly different form of apparatus was used. 



— , 



: ' Bulletin 47 of U. S. Geological Survey. 

 ''American Chemical Journal, vol. 8, p. 421. 

 "American Chemical Journal, vol. 9. p. 23. 



