393 



less also present, but were not sought for. Hydrochloric acid boiled 

 upon the powdered stone yielded a solution in which protoxide and 

 peroxide of iron, alumina, Hine, and magnesia, were found in marked 

 quantity ; and traces of manganese and cobalt, along with potassa, 

 soda, and silica, in small quantities. 



From those results it will be seen that the purest water can dissolve 

 a certain amount of substance from Craigleith sandstone ; that if 

 charged with carbonic acid it will disintegrate it further ; and that 

 if containing free mineral acids, as the rain-water of towns occasion- 

 ally does, it will decompose it still further. 



In connection with those results, it is important to notice the ex- 

 tent to which the stone absorbs and retains water, points on which 

 Mr Napier has already made valuable observations. The specimens 

 selected for the following trials had an average sp. gr. of 2'443. 



A piece weighing 3506*1 grains, which had been received from 

 the quarry in the month of November 1855, and remained for about 

 a month in a room without a fire, was kept at 212°, till it ceased to 

 lose weight ; the loss was equivalent to 5-7 fluid ounces per cubic 

 foot. 



A similar piece, weighing 4597-95 grains, was immersed in dis- 

 tilled water at 58°, till it ceased to gain weight. The surface- 

 moisture was then allowed to evaporate, and the stone weiohed. The 

 gain was equivalent to 3-8 imperial pints per cubic foot. 



According to Mr Napier, a sandstone acquires much more moisture 

 if allowed to absorb it by capillary attraction from one part of its 

 surface, than if entirely immersed in water, but upon making the ex- 

 periment in the way he describes, the difference, by capillary attrac- 

 tion, was comparatively small ; the whole gain being 4-2 imperial 

 pints on the cubic foot. On the other hand, when the stone was im- 

 mersed in water under the bell jar of the air-pump plate, and the 

 air withdrawn, the ultimate gain in weight amounted to 6-2 imperial 

 pints per cubic foot. The error of those who hope to render build- 

 ings dry by constructing their walls of solid sandstone, will be suffi- 

 ciently apparent from these facts. The numerical results obtained 

 by Mr Bloxam are added in full. 



