346 Analysis of the Hot Springs at Bath. 



it as in Experiment (K). The precipitate was carbonate of 

 lime, or of magnesia, or both. 



As earthy carbonates are not precipitable by alkaline car- 

 bonates, the precipitate formed by carbonate of ammonia (O) 

 indicates the presence of some other earthy salt. 



From Experiment (R) it appears that no alumina or mag- 

 nesia exists in the water, and that the precipitate obtained in 

 Experiment (K) was carbonate of lime. But according to 

 Dr. Gibbes, sulphate of alumina is present. It appears, how- 

 ever, that Dr, Saunders is perfectly correct in supposing that 

 it forms no part of the sahnc contents of the water. In ad- 

 dition to the cxperinjcnts already stated, it may be shown to 

 be incompatible with the s;dts actually existing in it ; for 

 the addition of a solution of sulphate of alumina occasions 

 immediate precipitation. 



As Dr. Gibbes has insisted on this point, I shall here 

 mention what appear to have been the causes of his mistake. 

 In reply to Dr. Saunders's supposition, that the precipitate 

 which he took for alumina was carbonate of lime, he has 

 stated that it was precipitated from the water by ammonia 

 after oxalic acid had ceased to produce any further etfect. 

 This method is subject to error j for the oxalic acid appears. 

 to have been employed without previous combination with 

 an alkali ; and, as oxalate of lime is soluble in oxalic acid, 

 it is evident that, if more of the latter were employed than 

 was sufficient to precipitate the lime, it would dissolve a 

 portion of oxaUte of lime. The acid in combination with 

 the lime previous to the addition of oxalic acid, not having 

 an alkali to combine with, would dissolve a further portion 

 of oxalate of Wme; and unless the carbonic acid was expelled 

 or saturated, it also would increase the error. From these 

 circumstances I suspect that the precipitate afforded by am- 

 monia was oxalate of hme, this compound being precipitable 

 from acids by alkalies. 



Experiment (S) determines the presence of lime. 



Experiment ( T) shows that sulphuric acid exists in the 

 water. 



The effect, produced by nitrate of silver (U) results from 

 the action of muriatic acid. 



As 



