178 



ROYAL SOCIETY OF CANADA 



Table No. 2 presents Uie data expressed in terms of tlie ions; prob- 

 ably the most correct method of recording the results : 



II. Results of Analysis expressed as ions. 



March 12th 

 1907 



SiO/' 



SO/' 



Hj CO3' (by calculation) 



PO/" 



CL' 



Fe'" 



Al'" 



Mn" 



Ca" 



Mg" 



K' 



Na' 



Oxygen to form FcjOg AljOgjMngO^ 



p.p.m. 

 6.52 



4.43 



not det. 

 .50 

 .23 

 .20 



not det. 

 8.98 

 2.33 



not det. 



not det. 

 .19 



July I. 5th 

 19Ô7 



p.p.m. 

 7.06 



3.01 

 25.43 

 .58 

 .50 

 .49 

 .28 

 .62 



5.84 



1.67 

 .56 



1.59 

 .69 



Table 3 shows the combinations of the mineral constituents as they 

 possibly exist in the water. It is true that any such combination is 

 fopen to objection on the ground that it is necessarily of a more or less 

 hypothetical nature, but as it has been customary in the past to report 

 the results of analyses of mineral waters in such a form, it has been 

 thought well, more particularly for the purposes of comparison, to pre- 

 pare this table. As the March analysis is incomplete it is only possible 

 to report the results of the July sample according to this scheme. 



From these data it will be seen that the mineral matter of this 

 water is composed mainly of the carbonates of the alkaline earths witli 

 some silica and small quantities of the alkalies, iron, alumina, and 

 manganese. 



Comparing the results of the two analyses as given in Tables Nos. 

 1 and 2, a .Tiiarked difForcnce in the amounts of lime is to bo obsorvod. 

 Thus, in March the lime (CaO) was 12.57 p.p.m., while in July only 



