80 MINERAL WATERS OF THE UNITED STATES. 



water when the flow of the spring is very much increased, but do remember on 

 several occasions having to bottle when the flow was 1 to 1 gallons per minute to 

 supply the demand. We expect the sample you obtained was water taken at that 

 time. 



OSCEOLA WATER (No. 1011 1. & W.). 



While there is some variation between the advertised and Bureau 

 of Chemistry analyses, the difference could easily be accounted for by 

 natural changes in the water from time to time. 



GOLINDO LlTHIA WATER (No. 1012 I. & W.). 



The same may be said for this spring as was said for the Osceola. 

 To this may be added that the spring hardly contains enough lithium 

 to be called a lithia water. 



CHERRYDALE WATER (No. 1013 I. & W.). 



The authors were unable to obtain the advertised analysis of this 

 spring. 



DEEP ROCK WATER (No. 1020 I. & W.). 



The authors were unable to obtain the advertised analysis of this 

 water, but a comparison of their analysis with the analysis in Crook's 

 shows an astounding variation, which can hardly be explained on the 

 basis of natural changes in the composition of the water of the spring. 

 This variation is particularly noticeable in the determination of silica 

 and potassium. A sample of the water obtained directly from the 

 spring showed the same small amounts of the last two mentioned 

 elements. 



CHAMPION WATER, SARATOGA, N. Y. (No. 1021 I. & W.). 



The variation between the Bureau of Chemistry analysis and the 

 advertised analysis is in this case so large as to make it appear that 

 the Champion water has grown much weaker in its mineral content 

 since the time the latter examination was made. 



ALLOUEZ MAGNESIA WATER (No. 1022 I. & W.). 



The authors were unable to obtain the published analysis of this 

 spring, but the analysis in Crook's differs materially from the Bureau 

 of Chemistry examination. 



MOUNT CLEMENS WATER (No. 1023 I. & W.). 



In this case again the authors were unable to obtain the published 

 analysis. The analysis in Crook's differs very materially from the 

 Bureau of Chemistry analysis. In a water so highly mineralized as 

 this one, however a large amount of variation is to be expected from 

 time to time. 



