GAS 633 tion. 1. It may be received in an exhausted vessel by means of a tube and stopcock. This method usually requires that the vessel be sev- eral times filled with the gas and exhausted, to remove the residual air which always remains at the first exhaustion in consequence of in- ability to produce a perfect vacuum. Bags, which may be very nearly emptied of their contents, are often conveniently employed in this method of collection. 2. By displacement. This is done by fill- ing a bell glass with water in a pneumatic cistern, placing it on the shelf, and bring- ing the mouth of the tube delivering the gas beneath it, as FIG. 1. Collection by Displacement. represented in fig. 1. When the gas to be collected is easily ab- sorbed by water, some other liquid is chosen, usually mercury. A modification of this plan, often used in collecting gases slightly absorbable by water, as hydrogen and oxygen, for ordinary experimental purposes, is to use a gas-holder, consisting of a copper cylindrical vessel, A, fig, 2, open at the top, in which is received a cylinder, B, closed at the top and open be- low, and counterbal- anced by a weight attached to a cord ring over pulleys, filling the outer cylinder with water, opening the stopcock c in the upper one, and depressing it, all the air may be forced out. Then, by at- taching the deliver- ing pipe to the stop- cock d in the outer cylinder, the gas will ascend into the inner one, which will rise as the pressure is re- stored to its interior. When a strong jet is required for use, weights may be laid upon the inner cylinder and the counterbalance weights removed. Another form of gas-holder is represented in fig. 3. A drum of copper, A, has mounted upon it a shallow vessel, B, communicating by two tubes with stopcocks, g and A, one of the tubes passing to near the bottom of the cylinder, while the other only enters the top. A water gauge, ef, shows the height of water in the drum; an opening at t admits the end of the tube supplying the gas, and a stopcock at c is for its exit. To use the apparatus, open the stopcocks, close the opening t with a plug, and pour water into the vessel B until the drum is filled ; then close the stopcocks and remove FIG. 2. Gas-holder. the plug from the opening t. Atmospheric pres- sure prevents the water from flowing out. In- troduce the end of the tube supplying gas; it will ascend in the drum, displa- cing an equal volume of water, which flows out at t. When suf- ficient gas has been introduced, close the opening , and open the stopcock g. The gas in the drum will then receive a hydro- static pressure equal to the height of the FIG. 8. Gas-holder. column of water in the tube and upper vessel above the level of water in the drum. The stopcock c may then be connected with any apparatus to which it may be desired to deliver the gas. The forms of apparatus of this kind may be varied indefinitely, but these examples will suffice for illustration. When it is desirable to separate mixed gases, which are absorbable in different degrees by different liquids, or when it is desired to saturate a liquid with a gas, an apparatus called Woulfe's bottles (fig. 4) is often used. The gas is made to enter each bottle at a and to pass out at e. A safety and supply tube, a, passes through a middle neck to below the surface of the liquid. A cup at the upper end is for the purpose of receiving a portion of liquid which may be forced up the tube by any sadden expansion. The number of bottles employed may be varied according to FIG. 4. "Woulfe's Bottles. the requirements of the case. II. DIFFUSION OF GASES. All gases, when mingled together mechanically in any proportion, tend to diffuse themselves uniformly, regardless of their spe- cific gravities. Thus, if two bottles are con- nected together by an upright glass tube 10 or 12 inches long and about ^ of an inch in cali- bre, and the upper bottle is filled with the lightest of all gases, hydrogen, and the lower one with oxygen, whose specific gravity is 16 times that of hydrogen, or with carbonic acid, which is 22 times as dense, after the lapse of two or three days the two gases will be found to have the same proportion to each other in both bottles. This was the original experiment of Dalton, published in vol. xxiv. of the "Phil-
