472 



SCIENCE. 



[N. S. Vol. XXII. No. 663. 



The diagram enables one to find such prod- 

 ucts as those involved in equation (4), and so 

 a graphic solution of this equation as it stands 

 can be carried out if desired. 



Since the radiating lines cut the lines 

 a; = ± 1 and y = 1 in scales of tangents 

 and cotangents, such products as cot h tan c, 

 if the factors are not too great, can be ob- 

 tained by first finding cot h in the upper 

 margin of the diagram and then going dov?n- 

 ward (keeping at a distance cot h from the 

 central line) until meeting the radiating line 

 numbered c. The distance thence to the ini- 

 tial or base line is cot h tan c. The cosine 

 scale of the diagram enables one to find the 

 angle whose cosine is equivalent to cot & tan c. 

 This is the angle J. of a spherical triangle 

 right-angled at B. E, A. Harris. 



ANALYSIS OF THE MISSISSIPPI RIVER. 



A SHORT time ago, in conversation vpith Dr. 

 E. W. Hilgard, of the University of Cali- 

 fornia, I learned, to my great astonishment, 

 that he had been unable to find in any publi- 

 cation a recent and complete analysis of the 

 Mississippi River. Deeming this a serious 

 oversight on the part of chemists at large, a 

 sample was secured for me through the kind- 

 ness of Mr. J. L. Porter, chemist for the New 

 Orleans City Sewerage and Water Board, and 

 analyzed by me with the greatest of care. The 

 methods employed in the mineral analysis were 

 very similar to those recommended by Pro- 

 fessor Bailey, of the Kansas Geological Sur- 

 vey, while the nitrogen determinations were 

 patterned after those made by the Massachu- 

 setts State Board of Health. 

 /" The sample was taken by J. L. Porter, 

 chemist of the New Orleans City Sewerage 

 and Water Board about noon of May 23, 1905. 

 Location of the sample was opposite Nine 

 Mile point • just above Carrollton, in mid- 

 stream, and about six feet below the surface. 

 Temperature of the water at the time was 

 23° C; Turbidity was about twice the average 

 for the year. Oxygen was about one hundred 

 per cent, of saturation and the free carbonic 

 acid about three parts per hundred thousand. 



The results of the analysis are as follows: 



Results of Analysis Expressed in Parts per 



100,000. .-;, 



Total solids (unfiltered) 106.9 



Total solids (filtered) 16.75 



Loss on ignition (unfiltered) .... 7.4 



Loss on ignition (filtered) 2.75 



Si 0.35 



Al 0.009 



Mn 0.012 



Ca 2.95 



Mg 0.68 



Fe 0.008 



K 0.23 



Na 1.00 



SO4 2.87 



PO4 0.04 



CO3 0.00 



HCO3 : 11.04 



CI 1.61 



Nitrogen as free ammonia 0.016 



Nitrogen as albumenoid ammonia. 0.014 



Nitrogen as nitrites 0.000 



Nitrogen as nitrates 0.023 



Oxygen consumed (unfiltered) . . . 1.42 



Oxygen consumed (filtered) 0.33 



Hardness 10.92 



Turbidity . Heavy. 



Sediment Large. 



Odor (cold) Practically none. 



Results of Analysis Calculated as Oxides. 



SiO, 0.74 



AloO:, 0.017 



FeA 0.011 



MujOi 0.016 



CaO 4.12 



MgO 1.13 



K2O 0.28 



Na^O . . : 1.35 



SO3 2.39 



C0„ 7.96 



The silica was rather higher than I ex- 

 pected, being about the same as that found in 

 the Hot Springs of Arkansas by Mr. Haywood. 

 Still, it is not a quarter of that occurring in 

 many of our "western streams. The ratio of 

 lime to hiagnesia is about normal, as is the 

 ratio of Na^O to K„0, but the amount of 

 bicarbonate seems unusually large, indicating 

 a large percentage of drainage from the arid 

 lands to the northwest. Sulphates form a 

 rather large per cent, of the total solids, but 

 this also is to be expected when we consider 



