232 Dr George Wilson on the extent to which 



contact with silica, was then evaporated to dryness in a pla- 

 tinum capsule, and the amount of residue ascertained. Six 

 careful trials were made, and gave as a mean, gr. 0-25166 as 

 the amount of fluoride of calcium soluble in 7000 grains or 16 

 fluidounces, i. e., a pint (Old Apothecaries' Measure.)* This 

 result approaches so closely to that previously obtained with 

 glass vessels, that the number found must be considered as 

 making a near approximation to the truth. 



A similar series of observations was made this summer, 

 but the fluor spar, which was of great apparent purity, as 

 furnished to me through the kindness of Mr Tennant of 

 London, was not subjected to any preliminary treatment with 

 hydrochloric acid, but simply boiled with distilled water, and 

 the solution collected and cooled as before, in a pewter 

 vessel. The liquid was allowed thus much contact with 

 silica that it was passed through a paper filter placed within 

 a tin funnel. Few, however, I think, will suspect that it can 

 have transferred to itself any silica from the saline constitu- 

 ents of the paper. Six trials were made in this way, the 

 mean of which gave gr. 0*26 as the quantity of fluoride of 

 calcium soluble in 16 ounces of water. The numbers, of 

 which this is the mean, like those obtained in the previous 

 determinations with metallic vessels, difi'er more from each 

 other than the numbers did in the first series of experiments, 

 where the solutions were made in glass flasks. This, how- 

 ever, was to be expected ; for the liquid employed in the first 

 series was prepared at once to the extent of many pints, and 

 the uniform composition of the whole secured, before any of 

 the solution was evaporated. In the case of the metallic 

 vessels, on the other hand, owing to their smallness, each 

 pint had to be prepared separately, and its evaporation com- 

 pleted before another was procured. The numbers, there- 

 fore, could not but diff'er more in the second and third deter- 

 minations than in the first. The highest number was 0-28, 

 the lowest 0*24. We may therefore consider 026 as suffi- 

 ciently nearly representing the true solubility of fluor spar, 



* In the report of the British Association for 1846 this amount of liquid was 

 inadvertently called the imperial, instead of the (old) Apothecaries' pint. 



