Mr Dunlop's Experiments with Bottles sunk in the Sea. 319 



with mercury, all of which would burst with a less temperature 

 than 100^ of Fahrenheit ; also five small glass globes hermeti- 

 cally sealed by the blowpipe, two of which were vacuum (or as 

 nearly so as I could make them), other two were suffered to 

 cool, previous to their being sealed, and the fifth contained a 

 small globule of mercury to enable me to detect any damp, as 

 an experiment on the porosity of glass ; three glass phials, well 

 corked, and firmly secured by leather coverings, tied round 

 the necks, and further secured by a coating of sealing-wax, were 

 also put into the canister. After letting them remain ten mi- 

 nutes at the depth of 80 fathoms, the line was hauled in, and 

 the experiments examined. The porter bottle was nearly filled 

 with water, and the cork floating inside ; the covering of canvas 

 and pitch was pressed concave into the mouth of the bottle, but 

 the pitch was not cracked or broken. The four thermometers, 

 and also the small glass globes came up unbroken. I examined 

 the one which contained the small globule of mercury, and it 

 gave not the slightest indications of damp having penetrated 

 through the glass. The three phials came up full of water : of 

 one of them the cork was forced in, and swimming in the wa- 

 ter ; in another, the cork was forced about half an inch into the 

 neck ; and the cork of the third was not apparently affected or 

 displaced in the least degree, although the phial was full of wa- 

 ter, and also several pieces of the sealing-wax lying in the bot- 

 tom, which by no means could have got into the bottle, but by 

 the cork being driven in. The wax on the top of each was 

 broken or cracked in regular concentric rings from the centre, 

 and the coverings of leather burst, as well that in which the 

 cork was not displaced as in the others. Indeed the hole in 

 the leather which covers the phial with the remaining cork is 

 larger than in the others, in which the cork is driven in ; which 

 in all probability may be accounted for, by considering this 

 cork to have been tighter fitted into the phial, and requiring a 

 greater force to displace it : there would be a greater rush of 

 the water into the phial, and the cork forced again into its neck. 

 I think it more than probable this has been the case, otherwise 

 the bits of sealing-wax could not have got into the phial had 

 the cork retained its situation ; neither could we account for the 

 bursting of the leather and wax which fastened down the cork. 



