108 THERMAL WATERS OF ASIA MINOR. 



The observation of the above fact has led me to adopt the 

 following method of estimating the silica in mineral waters : 

 Take a certain quantity of the water, evaporate almost to dry- 

 ness, add hydrochloric acid, a little more than is required to 

 saturate the carbonates present; continue to evaporate to 

 complete dryness, and then add water acidulated with a little 

 hydrochloric acid; filter and wash the silica that remains on the 

 filter ; in this way we are sure to have the silica perfectly free 

 from any silicate. 



The method adopted for estimating the alkalies will be 

 mentioned in a few words, as more details of it will be given 

 in a paper devoted especially to that subject ; the method has 

 particular reference to the separation of the alkalies from 

 magnesia. 



SEPARATION OP THE ALKALIES FROM MAGNESIA. 



Take the solution filtered from the silica, evaporate to 

 dryness to drive off the excess of acid, add a little water to 

 redissolve, then add pure lime-water and filter, when the 

 chlorides of the alkaline metals and calcium with excess of lime 

 will pass through, the magnesia, alumina, oxide of iron, etc., 

 remaining on the filter. Separate the lime with carbonate 

 of ammonia, or still better with oxalate of ammonia, evapo- 

 rate to dryness, and heat to drive off the ammoniacal salts, 

 when nothing but the chlorides of alkaline metals will be left, 

 which can be separated in the ordinary way. 



This completes the description of all the thermal waters 

 of Asia Minor which have as yet come under my notice, with 

 the observations that the investigation have given rise to. 



