968 REPORTS ON THE STATE OF SCIENCE. 
is led through the outer space, while a current of hydrogen is led through 
the inner tube in the opposite direction. It is found that hydrogen 
diffuses outwards so rapidly that it can be lighted at the exit of the CO, 
tube, while very little or none escapes from the end of the inner porous 
tube. ‘Thus by virtue of ‘‘ endosmose ’’ the two gases have changed 
places, traversing in opposite directions, the porous wall separating them. 
These phenomena, forming a striking and instructive lecture experiment, 
are in perfect agreement with the observations already made by Graham 
and by Jamin.’ * 
After this preliminary investigation of the action of his apparatus 
when cold, Deville proceeds to study by its means the action of heat on 
steam, using CO: as the inert gas. Steam being led in by the outer tube, 
the products collected from the inner one, after absorption of the CO:, were 
found to be strongly explosive. In several experiments the average 
amount of explosive gas collected was 1 c.c. per gram of steam passed in. 
(2) Dissociation of Steam.—-A porcelain tube, five or six cm. diameter, 
was filled with fragments of very clean porcelain previously heated to 
redness. A rapid current of CO. was led through a flask of water main- 
tained at 90° or 95° C., and thence through the porcelain tube, heated to 
full redness in ‘ a furnace using a blast of air.’ It is easily seen that a 
small quantity of steam is broken up into its elements. The gas passing 
out of the tube is collected over strong potash solution. After two hours 
25-30 c.c. of an explosive mixture was obtained, which gave, on analysis : 
Sih Waa a) ete ty? ty MUG 
His, od ee ek. eo ee 
Omens elnreier re fiwicine), Siqed ee 
i, ewe? Wa bation o th bimini oe 
the speed of the gas current averaging seven or eight litres of CO. per 
hour. The quantities of gas obtained are only about a quarter of those 
got in a similar time by the first method. A blank test on the carbonic 
acid gas gave 1°6 c.c. of residue, instead of 25-30 c.c. 
Dissociation of CO».—In another paper, describing the same kind of 
experiment on CO; alone, the author adds some further details. He says: 
‘I have proved the dissociation of water at a moderately high tempera- 
ture by separating it into its elements by the action of a solvent (some 
experiments not discussed here) or by the action of a mechanical 
pheromenon. I have succeeded still more easily with CO., because of 
the resistance to combination shown by oxygen and carbonic-oxide 
when they are disseminated through the mass of an inert gas. For- 
tunately for the rigour of my demonstration the gas may be CO. itself. 
I take the porcelain tube filled with porcelain fragments, as previously 
described. This apparatus is carried to a temperature which I estimate 
at 1300° C.’ 
An analysis of the residual gas gave: 
O,=30 
CO=62 paverage of several samples. 
v= 8 
+ Analyses of the gases passing into the collecting vessel did not always give 
the same result, thus casting some suspicion on the experiment, particularly on 
the gas-tightness of the porcelain tube. Deville would not admit that his porce- 
lain was porous, but says that his joints were not all secure against small leaks of 
hydrogen. 
