HOT SPRINGS, ARKANSAS. 17 



burette was recorded, and to the clear solution was added one drop of 

 methylorange. A pure 3 7 ellow color resulted. The titration was con- 

 tinued with the HKS0 4 without refilling- the burette until thecbange toa 

 very slightly darker and reddish color was noted. The change was faint 

 and required practice to detect. The reading at this point was also 

 recorded. 



The first reading recorded gives the amount of alkali carbonates 

 present and must be multiplied by the factor 0.002970 for the result in 

 grams of C0 3 ions. 



For the number of grams of HC0 3 ions present the first recorded 

 reading is multiplied by two and the result subtracted from the second 

 reading, and this remainder is multiplied b} 7 the factor 0.003028. In 

 no case were carbonates found in any of the springs by the above 

 method, but only bicarbonates. 



Having now obtained the weight of HC0 3 ions in 1,000 cc of water, 

 we next calculate the weight of C0 2 given off when a like volume is 

 evaporated to dryness, and dividing this result by the weight of one 

 cc of carbon dioxide at 0° C. and 760 mm. pressure the number of cc 

 of carbon dioxide given off from the bicarbonates is the result. Sub- 

 tracting the number of cc of carbon dioxide given off from the 

 bicarbonates from the number of cc of carbon dioxide in excess of 

 that necessary to form normal carbonates, we have left the number of 

 cc existing in solution in a free state. 



Bicarbonic acid. — The amount of this substance present in the 

 spring was estimated during the process of determining the amount of 

 carbon dioxide given off from bicarbonates in the paragraph above. 



It will be noticed in several of the analyses of the different springs 

 that the amount of carbon dioxide (set free from bicarbonates on 

 evaporating to diyness) and calculated from the bicarbonic acid does 

 not agree with the amount of bicarbonic acid found in solution. This 

 is because the samples for determining the carbon dioxide and bicar- 

 bonic acid were taken at widely different periods, and the amount of 

 bicarbonic acid had evidently changed somewhat during the interven- 

 ing time. This is easily explained when we remember that many of 

 the springs are supplied from two or three different spring heads, 

 which doubtless vary from time to time both in their amount of flow 

 and in the amount of bicarbonic acid held in solution. 



For the determination of nitric and nitrous acid, free and albumi- 

 noid ammonia, and oxygen consuming capacity the methods as given 

 in Mason's Examination of Water were followed in all their principal 

 details. They are as follows: 



Nitric acid. — In determining nitric acid, 100 c. c. of the spring water 

 was treated with 2 drops of a saturated solution of sodium carbonate 

 and evaporated to dryness on the water bath. The residue was treated 

 with 2 c. c. of phenol sulphonic acid (made by mixing 118 c. c. of pure 

 sulphuric acid, 12 c. c. of water, and 21 grams of phenol), a littleVater 

 added, and then an excess of ammonia. The solution was transferred 

 to a 100 c. c. Nessler jar, the volume made up to 100 c. c. with distilled 

 water, and the depth of the } 7 ellow color compared with that produced 

 by treating different measured amounts of standard potassium nitrate 

 (containing 0.01 milligram of nitrogen as nitrate in each c. c.) in the 

 same manner. 



Nitrous acid. — For this determination 100 c. c. of the water was 

 placed in a 100 c. c. Nessler jar and treated with 1 drop of concen- 



S. Doc. 282 2 



