376 Apparatus to illustrate Work hy Diffusion. 
pan and counterpoise, B, and at the other the porous jar, C, 
fitted with a cork, and a glass tube c which dips into a gas-jar 
D containing water or methylated spirit. Three or four glass 
jets, of which one is shown at E, are supported in a horizontal 
position, and the opening of each jet is placed as near as pos- 
sible to the porous vessel without touching it during the oscil- 
lations of the beam. These jets are connected by means of a 
flexible tube with a gas-holder containing hydrogen; the bell of 
the holder being loaded so as to force the hydrogen in a gentle 
stream against the sides of the porous vessel. The best position 
for the jets is found by trial; but usually I place them a little 
below the middle of the porous jar when the beam is horizontal. 
The action of the apparatus is simple. On turning on the hy- 
drogen, inward diffusion takes place, producing plus pressure 
within the jar; this pressure is resisted equally in all direc- 
tions but the vertical, and in this direction, owing to the little 
friction of the water, movement takes place, and the jar rises. 
When the jar has risen above the jets, inward diffusion dimi- 
nishes, or perhaps ceases, while outward diffusion of the hy- 
drogen commences ; a minus pressure is thus produced in the 
porous vessel, and the external pressure of the air causes it to 
descend. This descent brings the jar again opposite the jets, 
when the series of movements again begins. 
The work done with this arrangement is very small, and 
falls far short of the theoretical value*. For the best effect 
the jar should be surrounded by hydrogen for inward diffusion 
to take place, and subsequently the connexion with the hydro- 
gen should be completely cut off and air take its place. I 
have tried to devise some water-seal arrangement by which 
the flow of hydrogen could be turned off and on at the right 
moments by the movement of the beam; but have not suc- 
ceeded, as the friction thus introduced would be more than 
the movement of the beam could overcome. In the arrange- 
ment exhibited there is a considerable waste of energy due to 
the imperfect cut-off of the hydrogen, even when the flow 
of gas has been regulated so as to obtain the maximum effect. 
* "The work that may he done during the mixing of the volumes /;, and v 2 
of two different gases is the same as that which would he gained during the 
expansion of the first gas from volume i\ to volume i\-\-r.„ together with 
the work gained during the expansion of the second gas from v 2 to v x -\-v 2i 
the expansion being supposed to he made into vacuum.'' See a paper by 
Lord Kayleigh in the Phil. Mag. [4] xlix. p. 311. 
