54) Mr Kerr o?i a Simple, Cheap, and Accurate Method of 
gases. Tliese obstacles, it is hoped, the following contrivance 
will do much to remove. 
A glass tube from 6 to 12 inches in length, and from 2 to 5 
lines wide, so as to be capable of holding from 2 to 6 drachms, is 
to be hermetically shut at one end, and bended a little beloAV its 
middle, so as to form two branches, of which the shut branch 
will be somewhat shorter than the other, diverging from each 
other nearly at a right angle. The vertex of the tube should be 
widened on the concave side, and this done more toward the shut 
than the open branch, as is represented in Plate II. Fig. 3. The 
vertex A of the convex side of the curvature does not correspond 
with B, that of the concave side, but is beneath the shut branch. 
The gas evolved from the mutual action of two bodies, of 
which at least one is a liquid, may be collected in the shut 
branch in the following manner. Let the tube be held in the 
hand by the open end, so that this be the highest, and the shut 
end the low^est part of the tube ; then the liquid is to be poured 
in till it begins to ascend above the vertex. The shut extremi- 
ty is now to be elevated as high as the open end, while the ver- 
tex is depressed, so as to be the lowest part of the tube. In this 
position the shut branch will remain full, the liquid within it 
being supported by the pressure of the atmosphere on the small 
portion of the fluid, that is above the vertex, in the open branch. 
If a solid body, of greater specific gravity than the liquid, be now 
introduced into the open end of the tube, it will fall down to 
the vertex, and any gas evolved from its surface will rise 
through the liquid, and be collected in the shut branch, while 
the liquid will ascend in that which is open. Owing to the 
vertex of the convex side being beneath the shut branch, while 
that of the concave is nearer the open one, the whole, or almost 
the whole of the gas evolved, from any bit of solid matter rest- 
ing on the most depending part of the curvature, will ascend 
into the shut branch. In this manner the gas is collected un- 
mixed with common air, and if the experiment requires the ap- 
plication of heat, the bent tube may be placed in a sand-bath ; 
so that by means of such tubes, experiments may be performed 
on small quantities of gases, not only more economically, but, it 
is hoped, more accurately than is commonly done on larger 
quantities, with a more costly apparatus. 
