38 Kansas Academy of Science. 



hydroxide. This is illustrated by some laboratory experiments 

 made on the Lawrence (Kan.) waters: 



Table I. 



Part3 per million. 

 Fe^Oa+AljOj Manganese. 



No. 25 (a) Taken from well 14.8 2.62 



(6) After standing 48 hours 5.2 1 . 04 



No. 26 (a) City water, after treatment be- 

 fore settling 14 4 2.51 



(6) After standing 48 hours 3.2 .93 



Sample No. 25 was taken directly from the large well from which 

 a great portion of the city supply is drawn. 



Sample No. 26 was taken from the outlet to the settling basin 

 after the water had been treated with a small amount of lime and 

 aerated. In both cases the waters were allowed to stand forty- 

 eight hours and the clear solution above the precipitate pipeted off. 

 The results above are the average of several analyses. 



However, the ratio of the iron to the manganese removed from 

 the water as delivered to the consumer is different, there being a 

 larger per cent of manganese removed than iron. Results of an- 

 alyses made on seven different days on the tap water showed an 

 average of 1.16 parts per million of manganese and 10 parts per 

 million of iron and aluminum oxides. The tap water thus con- 

 tained about three times as much iron and aluminum oxides as the 

 water which had settled in the laboratory. This discrepancy was 

 due, no doubt, to a larger amount of the ferric hydroxide remain- 

 ing in suspension, the first deposit carrying down the precipitated 

 manganese, or the unoxidized organic matter still in the water has 

 a tendency to keep the iron in solution while in contact with the 

 iron pipes of the system, there being no solution of manganese go- 

 ing on while in the pipes. 



The difficulty of determining manganese by long gravimetric 

 methods, or by volumetric methods of doubtful accuracy in cases 

 of small amounts of manganese, was obviated by making use of the 

 bismuthate methods as adapted to waters by Mr. Robert Spurr 

 Weston,^ with a slight modification on account of the high chlorine 

 in the Kansas waters. The method used in this work is as follows: 



Evaporate with 5 cc. of concentrated sulphuric acid to strong 

 fumes, 250 cc. of the water of unknown manganese content. Take 

 up with 50 cc. of nitric acid (sp. gr. 1.135), and when the solution 

 is cold add about 0.5 gram of sodium bismuthate. Heat until the 

 pink color disappears, with or without the precipitation of manga- 



3. Jr. Amer. Chem. Soc, vol. XXIX. p. 1073; Blair, Chetn. Anal. Iron and Steel, p. 135: Chem. 

 News. vol. LXXIV. pp. 247-302. vol. LXXXV. p. 59: Trans. Chem. Soc. 1895. p. 268; Ding:. Poly. 

 Jr.. pp. 224-269; Brearley & Ibbotson. That Anal, of Street Wks. Materials. 



