WATERS from ICELAND. 103 



These phenomena appear to me to have proceeded from the 

 very weak and flow action of the acid and alkali on one ano- 

 ther, in confequence of the exceffively diluted date in which 

 they were mixed together, the alkali at the fame time not being 

 pure, but combined with the filiceous earth, a fubftance for 

 which it has a confiderable attraction. I therefore fuppofed 

 that when I added the fmall dofes of diluted acid, the acid par- 

 ticles remained for fome time difperfed through the liquor, 

 without joining the alkali, and the water contained, at the fame 

 time, a filicated alkali, if I may fo call it, and an unfaturated 

 acid ; but the colour of litmus being much more difpofed to 

 be affected and changed by acids than alkalis, it became red, 

 and retained this colour as long as any particles of the acid re- 

 mained unfaturated. Thefe, however, after fome time, being 

 all attracted and faturated by the alkali, the colour was again 

 changed by the remaining unfaturated alkali. 



It may perhaps be fufpected, that a fmall quantity of fixed 

 air, detached from the alkali, might be the caufe of this tem- 

 porary red colour, and that the colour returned again to blue, 

 when the fixed air evaporated from the water : And I know 

 that a very fmall quantity of fixed air, contained in water, is 

 fufficient to change the colour of litmus, and that a confiderable 

 time is required for its evaporation from the water, fo that the 

 litmus may recover its natural tint ', but it is equally true, that 

 the fixed air never requires fo long a time for its evaporation as 

 feveral weeks, and that it has not the power to redden litmus, 

 when an alkali is prefent, except when the quantity of the al- 

 kali is exceedingly fmall, and that of the fixed air incomparably 

 more than fufficient for faturating the alkali. In the prefent 

 cafe, the laft of thefe conditions never could take place, the 

 quantity of acid added at once being far too fmall to detach 

 enough of air, even although the alkali had been originally fa- 

 turated with air, which, it certainly was not ; it appeared rather 

 to be in a cauflic ftate, or very nearly cauftic. This reafoning 



fuggefted 



