MM. Deville and Troost on Vapour densities. 207 
Inversely, when the tube is full of air and a column of water 
is forced into it by pressure, it fills the several chambers suc- 
cessively, destroys the beads, and annuls their resistance ; finally, 
it fills the whole tube and commences to overflow. 
These consequences may be applied to porous bodies in which 
we may assume the existence of canals alternately narrow and 
broad. When a porous vessel, such as is used in the battery, 
or an alcarraza, or a plaster statuette, or any other cavity formed 
in a porous mass is filled with water, any pressure exerted upon 
this water causes the same to filter through the mass; on the 
other hand, however, a perfect vacuum may be formed in the 
interior, atmospheric air being incapable of penetrating through 
the walls when moistened. 
When both surfaces are immersed in water, and a pressure is 
exerted in the interior by means of compressed air, the latter, in 
the first place, expels all the water ; but when this is accomplished, 
the air does not filter through the walls: the pressure, indeed, 
may be increased to two, three, and in some cases even to four 
atmospheres without causing the least air-bubble to traverse the 
porous sides; and this pressure, too, may be maintained for an 
indefinite period, exactly as if the sides of the vessel were not 
traversed by capillary fissures. 
XXVIII. Chemical Notices from Foreign Journals. By K. AvK1n- 
son, Ph.D., F.C.S., Teacher of Physical Science in Cheltenham 
College. : 
[Continued from p. 126.] 
N continuation of previous researches, Deville and Troost* 
have made some determinations of the specific gravities of cer- 
tain vapours at high temperatures. The vessels used were por- 
celain flasks with narrow necks, of 280 cubic centims. capacity. 
These flasks are loosely closed by means of a small porcelain cy- 
linder of 1 or 2 millims. diameter, which fits im the neck. At the 
termination of the experiment, the projecting end of this cylinder 
is fused by the oxyhydrogen lamp and closed effectually. The 
bath employed was that previously described, in which high con- 
stant temperatures are obtained by means of metallic vapours. In 
these experiments, Deville and Troost used cadmium vapour, the 
temperature of which is 860° C., and zinc vapour, the temperature 
of which is 1040° C. 
The determinations were made by taking the density of the 
substance operated upon, and that of iodine under the same 
circumstances. Deville and Troost obtained in this manner a 
* Comptes Rendus, vol. xlix. p. 239. 
