THE MINERAL SPRINGS OF SARATOGA 21 



The analysis of a water is conducted by taking a suitable and 

 carefully measured sample from the spring and evaporating it to 

 dryness so as to secure the dissolved matter. The composition of 

 the evaporated residue is then determined by the chemist. It must 

 be remembered that the residue from a water such as is afforded 

 by the Saratoga springs is a complex mixture containing a number 

 of bases and acids, each of which is determined by itself. Thus 

 in the Saratoga waters there have been customarily reported ten 

 elements which are present in acids, namely, chlorin, bromin, 

 iodin, fluorin, carbon, sulfur, phosphorus, boron, nitrogen and 

 silicon. Each of these makes an acid of its own. Some ten ele- 

 ments furnish the bases, namely, sodium, potassium, lithium, mag- 

 nesium, calcium, barium, strontium, iron, aluminum and manganese. 

 With these must be placed the rarer ammonium and still rarer 

 rubidium. When the chemists have determined these acids and bases 

 they endeavor to combine the results as they believe them to exist in 

 the original w^ater. They do so on assumptions of the relative 

 chemical affinities of the elements for one another and divide the 

 acids among the bases so as to use up the supply of each. Probably 

 the results represent with a fair degree of accuracy the combinations 

 as found in the springs, but as subsequent discussion will show, 

 there inevitably arise minor differences among chemists. 



In later years the theory of ions has gained great support. 

 According to this conception, the dissolved compounds exist in dilute 

 solutions in the form of incomplete molecules, the same being held 

 in this condition by electromotive force. The incomplete molecules 

 are called ions. Thus, whereas in an older analysis, we would find 

 reported NaCl, NaXO,, (or NaHCOJ. XaBr, Xal, ^^,FO„ KG, 

 K.SO,, MgCOg (or Mg(HC03),), CaCO, (or Ca(HCO,)3)^ CaF,, 

 FeCO, (or Fe(HC03)3), BaC63 (or Ba(HCOJ,), LiXO, (or 

 Li(HCO,), SrCO, (or Sr(HCO,)2), A1,0,, and SiO, ; in one in the 

 last ten years we find, CI, HCO,. Br, I, PO„ SO,, Na, K, Mg, Ca. 

 Fe, Ba, Li, Sr, Al and SiO.,. As a separate statement the latter 

 authority might then combine them into salts but whereas as the 

 earlier chemists sometimes calculated carbonates, such as XaXO.., 

 MgCO-j, CaCO;,, etc., the later analysis gives them all as bicarbon- 

 ates, respectively NaHCO.^ Mg(HC03)„ Ca(IICC\),, etc. Each 

 chemist determined boron — the former combined it with sodium a^ 

 a biborate ; the latter as a metaborate. 



There is no very serious objection to cither of these ccntrses as 

 above outlined. Fashions change among chemists as among other 

 people, but the method of reporting by ions probably involves fewer 



