36 THE SALTON SEA. 



During the first ten days in February 1907, when the lake had reached its maximum 

 volxune, a set of samples collected from different points on the lake and at different depths 

 were analyzed for their total solids and chlorine content ; and, as the samples were collected 

 from points extending over the entire range of the lake, the results obtained represent 

 the variation in concentration existing in the lake at the time of its greatest volume. 



These results show that with one exception no very great variation was found in any 

 of the samples analyzed. Sample No. 9, which was less salty than the others, was taken 

 within a few miles of the mouths of the New and Alamo Rivers, through which only a 

 short time before flowed the fresh water of the Colorado River. (Table 5.) 



The greatest concentration was found near the bottom of the lake at a depth of 60 

 feet. This would indicate that leaching of salts from the bed of the lake was still taking 

 place. On account of this result, and from the fact that the concentration found for the 

 lake in general was not as uniform as it was expected to become ultimately, no further 

 samples were collected until June 3 of the same year. On this date a sample was collected 

 for complete analysis at a point 4 miles from shore in a southwestern direction from Mecca 

 Landing. Each year following samples were collected in the same way and in approximately 

 the same location. 



METHODS OF ANALYSIS. 



The methods of analysis were selected according to what seemed best suited for the 

 concentrations of the constituents to be determined, but only those methods were used 

 which have been shown by extensive use to give accurate results. All determinations were 

 carried out in quadruple and blank determinations were made of the reagents used in order 

 to be able to judge accurately as to the increase in concentration which was taking place 

 from year to year. With certain constituents this was easy to determine accurately, but 

 with others which occurred in smaller proportions the experimental error, even with the 

 greatest care, must necessarily be considerable. 



SILICA. 



For the determination of silica and of the following constituents which were made in 

 the same sample, 2 Uters of the water were evaporated to dryness with hydrochloric acid 

 in a platinum dish and then dried for 2 hours at 120° C. As the concentration of the 

 water increased a correspondingly smaller volume was taken. The residue was taken up 

 in hydrochloric-acid solution, filtered, and washed. The filtrate was evaporated as before, 

 and the process repeated three times. The insoluble portions were combined, dried, and 

 ignited in a platinum crucible. The residue was then treated with a few drops of dilute 

 sulphiu*ic-acid solution and ignited to constant weight. The silica was driven off with 

 hydrofluoric acid, a few drops of sulphmic acid again added, and the ignition to constant 

 weight repeated. From the loss in weight noted the silica present was determined. Any 

 residue remaining after the final ignition was brought into solution and combined with the 

 filtrate from which the silica had been separated. 



IRON AND ALUMINIUM. 



The filtrate from the silica determination was heated almost to boiling and hydrogen 

 siilphide passed in for an hoxu". The slight precipitate which formed was found to consist 

 principally of platinum taken up from the platinum dish in which the water had been 

 evaporated. It was consequently rejected and tests for the heavy metals made in a solu- 

 tion prepared by evaporating water in a porcelain dish as described under "copper." 



The filtrate from which the precipitate was rejected was heated to drive off hydrogen 

 sulphide and then oxidized with bromine water. To the hot solution was added a slight ex- 

 cess of ammonia, and on filtering and igniting the combined oxides of iron, aluminium, and 

 phosphorus were obtained. The phosphate radicle was determined in a separate sample. 



