348 Analysis of the Hot Springs at Bath. 



It has been observed, that one of the most active tests of 

 oxide of iron does not in this water produce any appearance 

 of its presence; and the slight metallic taste which it coni- 

 niimicates \\ heu hot and fresh has been unnoticed by some 

 analysts. This taste is lost by cooling, even in well stopped 

 bottles; and every method which I have tried to restore it 

 has been unsuccessful. It has also been mentioned that the 

 action of infusion of galls is in some cases lost. 



The proofs that this water contains oxide of iron thus ex- 

 isting under singular circumstances, and liable to cease, it is 

 not surprisinti that they have been assisted by collateral evi- 

 dence. It has been asserted that the vi'ater deposits " a pale 

 yellow ochrey precipitate;" but this is certainly an error, 

 the precipitate being perfectly white. Another circumstance 

 which has been adduced to prove the precipitation of oxide 

 of iron is, that the sides of the king's bath become encrusted 

 with it : this observation, as to fact, is correct ; but the 

 oxide appears not to be deposited from the water, but derived 

 from the stone, by the increased oxidation of the iron con- 

 tained in it, by the alternate application of air and w^ater. 

 Having procured a specimen of oolite similar to that of 

 which the sides of the bath are constructed, I added a drop 

 or two of nitric acid to it; by this the iron became instantly 

 and completely oxygenated, aflbrding an appearance similar 

 to that which has been supposed to be deposited from the 

 water. 



Another fact has been noticed equally deceptive with the 

 above stated, which is, that the clothes of the bathers be- 

 come stained with iron moulds. It is indeed true, that the 

 bright yellow colour of the substance of which these clothes 

 are made chances to a brown ; but this change is not partial, 

 nor has it any resemblance to iron moulds : it seems to be 

 effected merely by the decay of the colouring matter, and I 

 find that solutions of iron do not change tiie vellow colour. 



The obi^ervation which has occasioned most discussion 

 respecting the oxide of iron, is the loss of power of infusion 

 of galls to detect it. The following experiments will show 

 under what cii-cumstances this occurs. 



{a) About one pint of the water was exposed, while hot, 



to 



