154 DR WILSON ON THE SOLUBILITY OF 



It seemed well worth while to seek for fluorine iu one of these waters. I 

 obtained accordingly from Mr Campbell's brewery, a portion of the abundant 

 deposit, consisting chiefly of sulphate and carbonate of lime, which collects with 

 great rapidity in the boilers. It was treated with nitric acid, the dissolved portion 

 poured off", neutralized with ammonia, and precipitated by nitrate of baryta. The 

 precipitate, after being washed and dried, was warmed with Nordhausen sulphu- 

 ric acid, in a lead basin ; a square of Avaxed plate-glass, with characters traced 

 through the wax, being laid as a cover over it. In this, as in all other experiments 

 of the kind, a wall of wax was raised on the edges of the upper side of the glass, 

 so as to retain a portion of water sufficient to keep the plate cool, and condense 

 the hydrofluoric acid on it. This simple, but useful device, I borrowed from Dr 

 Dauben y.* Three squares of glass were very distinctly, though not deeply, etched 

 in this waj'. 



Fluorine, then, was present in this water ; and the fact has an interesting 

 relation to the circumstance pointed out to me by Mr Rose, that the well from 

 which it was obtained is sunk through a bed of sandstone, containing much mica, 

 a mineral in which RosE,f Tubner, Gregory, t and other analysts, have found 

 between 1 and 2 per cent, of fluorine. In reference to the corrosion of the brewery 

 thermometers, however, I wish it to be distinctly imderstood, that I do not seek to 

 refer the whole abrasion of the glass to the action of a fluoride dissolved in the 

 water in which they are immersed. The well-known experiments of Lavoisier, 

 made iu the end of last century, proved that even distilled water, if long boiled 

 upon glass, can corrode it. Every chemist is familiar Avith the rapid action of 

 solutions of the fixed alkalis, and of phosphate of soda on flint-glass. The in- 

 ferior kinds of bottle-glass, especially when containing too little silica and excess 

 of lime, have been shewn by Faraday || and Warrington to suffer corrosion by 

 the action of wine, and of diluted hydrochloric, sulphuric, and tartaric acids ; 

 and it would be rash to suppose that these are the only re-agents that can act 

 upon artificial silicates, especially upon those which contain excess of basic 

 oxides. 



On the other hand, it is impossible not to connect the fact that the thermo- 

 meters are coiToded, with the cu-cumstance that the water which occasions this 

 corrosion contains fluoride of calcium. The other constituents of the brewery 

 water are chloride of calcium and sodium, sulphate of lime and of soda, cai-bo- 

 nate of lime and of magnesia, silica and organic matter ; no one of which is known 

 to have any action on glass. 



In connection with this fact I may mention, that Mr Stevenson finds the 

 thermometers used in the breweries in the valley of the Cowgate much more 



* Chemical Society's Memoirs, vol. ii., p. 101. t Poggendorf's Annaleii, vol. i., p. 80. 



1 Brewster's Jourual of Science. || Chemical Society's Memoirs, vol. ii., p. 247. 



