in Cavities of regular Crystals. 313 



sion of it ; the pressure was applied at k; when the crack was 

 in the pointed part of a fragment of glass, such as fig. 4, it was 

 found to succeed best.) Upon communicating these results, 

 several j'ears afterwards, to Mr. Sivright, of Meggetland, he 

 informed me, that upon laying by for a few days pieces of glass 

 in which he had formed cracks of considerable extent, by 

 means of a hot iron, he had often found them to disappear 

 again. 



From the above circimistance, the conclusion was obvious, 

 That water might enter the void interstices of crystals, when 

 aided by pressure, not only from the porous nature of their 

 })articles, but also from their temporary display of rents during 

 the application of a high temperature. Such an explanation 

 as this might be equally applied to either theory for the for- 

 mation of our globe. 



Recalled again to my fonner inquiry, by the above facts, I 

 renewed my endeavours to get the hints regarding the porous 

 nature of glass, if possible, established from the results of bot- 

 tles or globes sunk to various depths in the sea. At the same 

 time it is to be kept in view, that although the success of these 

 experiments would bear strongly in favour of the doctrines 

 which have been laid down, yet their failui'e by no means mili- 

 tates against the mode of crystallization proposed. 



I shall now, therefore, lay before the Society a few experi- 

 ments, for which I am indebted to Mr. John Grant. 1 give 

 the account as communicated to me : 



" Sir, — When becalmed and nothing stirring, it was a fre- 

 quent amusement of the officers to let a bottle, either empty 

 or full of fresh water, down to a considerable depth in the sea, 

 100, 150, or 200 fathoms, with a heavy weight attached. 



" The result obtained on various occasions they found to be 

 uniformly the same. 



(1.) " When the bottle was let down empty, and the cork 

 not })ro])erly secured, though it was driven in as hard as a cork 

 could be, and the upi)er or outer part projecting considerably 

 over the neck of the bottle ; it was forced in and the bottle was 

 full of salt-water. It made no sort of dilfei'ence whether the 

 bottle was sent down bottom or top undermost ; nor was it to 

 he expected it would, as the pressure on hauling up nmst have 

 been as unich as on letting it down. 



(2.) " If the cork was properly secured with sealing-wax, 

 and a number of folds of oil-cloth, and all covered with 

 painted canvass well fixed upon the bottle, — the cork came up 

 in statu (juo, but the eiiij)ty bottle was now full of salt-water. 



(3.) " If the bottle was sent down with the cork properly 

 secured as before, and full of fresh-water, it invariably came 



VoLeO. No. 29i. Oct. 1822. Rr up 



