DISCUSSION OF ANALYSES. 99 



drawn. It is, of course, probable that some of this difference is due 

 to natural variation in the water from time to time, but it is hardly 

 possible that such a difference as appears is entirely due to such 

 periodic fluctuations. 



A rather curious case is presented in the Chief water, the adver- 

 tised analysis of which, made some years ago, s*hows only about half as 

 much mineral matter as the analysis made recently by the Bureau of 

 Chemistry. 



To summarize the results in a few words, it will be noted that the 

 following points have been brought out by the study of Saratoga 

 waters: 



(1) The waters are in nearly all cases markedly weaker in mineral 

 content than they were about thirty-five years ago. 



(2) There is a great variation in the total mineral content of indi- 

 vidual springs from time to time. 



(3) The rarer elements, such as lithium and bromin, seem to vary to 

 a greater extent than the other elements present. 



INDEX TO ANALYSES OF WATERS ARRANGED ACCORDING TO 



CLASS. 



Alkaline: Page. 



Bicarbonated 



Sodic 



Augusta White lithia water 67 



Geyser water 88 



Jeff ress lithia water 40 



Manitou \vater 70 



Powhatan water 43 



Thompson's bromin and arsenic water 73 



Magnesic 



Osceola water 52 



Calcic 



Allouez water 57 



Augusta White lithia water 67 



Bear lithia water .- 68 



Crockett arsenic lithia water 64 



Golindo lithia water 53 



Great Bear water 61 



Jeff re.ss lithia water 40 



Londonderry lithia water 62 



Manitou water 70 



Mardela water 48 



Massanetta water 34 



Missisquoi water 



Osceola water 52 



Otterburn lithia water 42 



Poland water 32 



Powhatan water 



Rubino Healing Springs water 71 



Sublett lithia water 51 



Vitan water 4& 



Ferruginous 



Mardela water . . 48 



