the Mineral Waters of Dunllane and Fitcaithhj. 271 



muriate of barytes ; applying this reagent with a slight excess of 

 acid, so as to prevent any precipitation of carbonate. The 

 quantity of precipitate thrown down from a pint of the water, 

 amounted, after drying at a low red heat, to 6-1 grains, equiva- 

 lent to 3'5 grains of sulphate of lime. As the portion of sul- 

 phate of lime thus obtained, above tliat obtained by the evapo- 

 ration, would remain principally mixed with the muriate of soda, 

 the quantity of that ingredient falls to be reduced a little, and 

 may therefore be slated at 24 grains. 



It remained to ascertain the proportion of iron. The quan- 

 tity, ho\vever, was evidently so small as to present a difficulty. 

 Succinate of ammonia, and benzoate of soda, produced little or 

 110 effect on the water in its natural state. Infusion of galls 

 produced, after some hours, a dark colour, and a precipitate very 

 slowly subsided. This method has been emplbyed to ascertain 

 minute quantities of iron, and I endeavoured to avail myself of 

 it, — adding to the \vater infusion of galls, in small successive 

 portions, at the interval of a day or two, as long as the colour 

 appeared to be rendered deeper; leaving it exposed to the air 

 for a longer time, that the whole matter rendered insoluble might 

 subside ; and lastly, washing the precipitate, drying and calcin- 

 ing it, to consume the vegetable inatter, and obtain the oxide of 

 iron. The difficulty, however, attending this method, is that 

 of precipitating entirely the iron, the liquor never becoming co- 

 lourless. Ill one experiment, conducted with much care, the 

 quantity of the calcined product from two pints amounted to 

 0*4 grain ; but it consisted partly of carbonate of lime. To re- 

 move this, pure muriatic acid diluted was added in excess, and 

 a moderate heat was applied ; the precipitate was entirely dis- 

 solved, and the liquor acquired a deep yellow colour. Being 

 further diluted, a little pure ammonia %vas added to it, in a close 

 phial, to precipitate the oxide, of iron, while the lime should 

 remain dissolved. The quantity thus obtained, when dried, 

 amounted to 0"27 grain. 



This method being liable to the above objection, I employed 

 another : two pints of the water were evaporated : when re- 

 duced to aljout two ounces, a brownish coloured sediment was 

 deposited, which was removed ; the evaporation was carried to 

 dryness, and the dry mass was redissolved in distilled water. 

 The insolul)le residue was of a grayish colour, and to this the 

 ilcposite formed during the evaporation was added. It was 

 known by previous experiments, that the greater part of the iron 

 was separated in this way ; the insoluble matter, when digested 

 with muriatic acid, affording a liquor, when diluted vvith water, 

 which gave, after neutralization with ammonia, a deep colour 

 with tincture of galls. To eiisure, however, the more perfect 



separation 



