^6 NEW YORK STATE MUSEUM 



poorest 0.0002I per cent. Stated in this way, there is a very obvious 

 and important difference between the mineral waters and the sea 

 waters. 



But we may run comparisons in another way: Dr F. W. Clarke 

 summarizes upon page 95 of his invaluable work 264 analyses from 

 eighteen different localities, whose salinity varies from a maximum 

 of 5.854 per cent in the Red sea, to a minimum of 0.7215 per cent in 

 the Baltic, where, as the writer has personally observed, both people 

 and animals freely drink the water. When expressed as elements 

 and ions there is extraordinary uniformity in the dissolved salts of 

 the sea despite the widely varying salinity : 



MAXIMUM MINIMUM RANGE 



CI 55.96 54.62 1.34 



Br .19 .13 -06 



SO^ 8.01 7.47 .54 



CO3 46 .01 .45 



Na 31.21 30.20 1. 01 



K 2.10 .64 1.46 



Rb .04 .03 .01 



Ca 1.67 .89 .78 



Mg 3.896 3.36 .536 



Fe, SiO„ PO^, 1 



NH^, Nbg, Al^O,, J 08 .02 .06 



(Fe303) " J 



The last mentioned ingredients which are collectively grouped are 

 only determined in a few instances. The same is true of rubidium 

 (Rb). The greatest variability is shown by CO3 and the next by 

 potassium (K), but their amounts are relatively small. Calcium 

 (Ca) also varies widely, but it also is small in amount. The most 

 important ingredients, taken in order of abundance, chlorin (CI), 

 sodium (Na), the sulfuric acid ion (SO4), and magnesium (Mg), 

 change but slightly in this wide range of conditions. 



When we come to compare the analyses of the Saratoga and 

 Ballston waters with the above, certain allowances must be made. 

 The mineral waters are heavily charged with carbonic acid which 

 gives them greatly increased dissolving power upon limestone. We 

 would expect the calcium and perhaps also the magnesium to be 

 much in excess of the percentage in sea water. In making com- 

 parisons we would be justified in tentatively assigning the excess 

 of calcium and of carbon dioxid over the figures of sea water to 

 contributions provided in this way, and then recasting the remainder. 

 We must also deal with analyses that are based upon bona fide 

 natural waters. I therefore select a few chiefly from the series of 



