590 MR J. Y. BUCHANAN ON THE 
As yet I have only experimented on glass, and only on one sort, namely, 
that made by Messrs Forp & Co. of Edinburgh. It contains 56°29 per cent. 
of silica, 29°5 per cent. of oxide of lead, 6°52 per cent. potash, 3°36 per cent. 
soda, 3 per cent. alumina, and 1:05 per cent. lime. I have observed its com- 
pressibility up to a pressure of 240 atmospheres, and before proceeding to 
higher pressures I intend to determine the compressibilities of other solids, 
especially metals, at pressures up to 240 atmospheres. The reason for taking this 
course is, that having got two glass tubes to stand this pressure, I am anxious 
to utilise them as far as possible before risking them at higher pressures. 
The pressure in these experiments was measured by a manometer, which 
consists simply of a mercurial thermometer with a stout bulb, which is im- 
mersed in the water under pressure, whilst its stem projects outside. The 
values of the readings of this instrument were determined by comparing it with 
a piezometer containing distilled water. This piezometer had been compared 
with others which had been subjected to the pressure of very considerable and 
measured columns of water on the sounding line. The mean apparent com- 
pressibility of water in glass was thus found to be 0:00004868 ;* or, multiplying 
by one thousand, to reduce the number of figures 0:04868 per atmosphere at 
temperatures from 1° to 4° C. 
The manometer (No. 2) was compared with this piezometer. The tempera- 
ture of the manometer was 12°5° C., while the piezometer was enveloped in ice 
in the receiver. The ice was thus melting under the same pressure as the 
instrument was undergoing, consequently the piezometer was not exposed 
really to precisely the same temperature at each succeeding experiment. For 
our present purpose, the effect of the possible variation in volume due to this 
thermic cause is negligeable, and we assume that the indications of our piezo- 
meter are comparable with those obtained in deep ocean waters. In a future 
communication I hope to return to this point. 
In Table I. we have in the first column the number of observations at each 
approximately identical pressure from which the average values of the mano- 
meter reading under A, and of the piezometer indication under H are com- 
puted. Manometer No. 2, when treated simply as a thermometer, showed at 
atmospheric pressure a rise of one division for a rise of 0°233° C. in tempera- 
ture. Piezometer K, No. 4, was filled with distilled water, and contained 7°74 
cub. centimetres at 0° and atmospheric pressure. It is made of Forp’s glass, 
though not drawn at the same date as the experimental rod. 
* Proc. Royal Society of London, 1876, p. 162. 
