pressure, until when the temperature is lowered to the boilingpoint of the gas it passes into the liquid state without the applica tion of any external pressure. The following table contains the most recently determined physical constants of a number of common gases :

From the figures given in this table it will be seen that the critical pressure (which is the pressure required to liquefy a gas at the highest temperature at which pressure can possibly cause liquefaction) is in most cases comparatively small. In only one instance, namely, ammonia, is it over 100 atmospheres, and falling in the case of hydrogen as low as 15.3 atmospheres. The enormous pressures, therefore, amounting often to many hundred atmospheres, which some of the earlier experimenters employed in attempting to effect the liquefaction of the so-called permanent gases, are thus seen to have been efforts in an entirely wrong direction. It was not greater pressure that was required, but the means of cooling the gases to a sufficiently low temperature.

In ordinary language such a gas as chlorine is spoken of as an easily liquefied gas, while oxygen would be described as a difficult! liquefied gas. Strictly speaking, however, and considering them from a comparable standpoint, it would perhaps be more correct to regard them in exactly the opposite light. Thus, taken at their respective critical temperatures, oxygen is liquefied by a pressure of 58 atmospheres ; while at the critical temperature of chlorine this

gas requires a pressure of 84 atmospheres to reduce it to the liquid state. At o° it is true chlorine may be liquefied by a pressure of only 6 atmospheres, but it must be remembered that o° is 141 degrees below the critical temperature of this gas. Long before oxygen has been cooled 141 degrees below its critical temperature, which would be down to -254°, it not only passes into the liquid state without the application of any external pressure at all, but is frozen to the solid state.

Diffusion of Gases.-If a jar filled with hydrogen be placed mouth to mouth with a jar of air, the hydrogen being uppermost, it will be found that after the lapse of a few minutes some of the hydrogen will have passed into the bottom jar containing air, and some of the air will have made its way up into the hydrogen jar. The light gas hydrogen does not, as might have been supposed, remain floating upon the air, which is 14.44 times as heavy, but gradually escapes into the lower jar; and the heavier gas finds its way, in opposition to gravitation, into the upper jar. This process goes on until there is a uniform mixture of air and hydrogen in both jars, and the gases never separate again according to their densities.

This transmigration of gases will take place even through tubes of considerable length: thus, if two soda-water bottles be filled one with hydrogen and the other with oxygen, and the two bottles be connected by a piece of glass tube a metre in length, the system being held in a vertical position with the light hydrogen uppermost, it will be found after an hour or two that the two gases have become mixed. Some of the hydrogen will have descended through the long tube into the lower bottle, and in like manner a portion of the oxygen, although nearly sixteen times as heavy as hydrogen, will have travelled up into the top bottle. That the gases have so mixed may be readily shown by applying a lighted taper to the mouth of each bottle, the detonation which then takes place proving that the bottles contain a mixture of oxygen and hydrogen. This passage of one gas into another is called the diffusion of gases. It was observed by Graham that when the two gases were separated from each other by a thin porous septum, such, for instance, as a piece of unglazed porcelain (socalled "biscuit "), or plaster of Paris, the pressure of the gas on the two sides of the porous partition did not remain the same during the process of diffusion that is to say, one gas made its way through the partition faster than the other, and it was noticed

:

F

that the lighter the gas the more rapidly was it able to transpire or diffuse through the porous medium. This fact, viz., that a light gas diffuses more rapidly than a heavier one, may be observed in a variety of ways. The apparatus seen in Fig. 8 is a modified form of Graham's diffusiometer. It consists of a long glass tube with an enlargement or bulb near to one end. Into the short neck of this bulb there is fastened a thin diaphragm of stucco, or other porous material. If the apparatus be filled with hydrogen by displacement, the short neck being closed by a cork, and the long limb be immersed in water, it will be seen, upon the withdrawal

of the cork, that the water rapidly rises in the long tube. The hydrogen diffusing out through the diaphragm so much more rapidly than air can make its way in, a diminution in pressure within the apparatus results, and this causes the water to ascend in the tube. The same phenomenon may be seen even more strikingly by means of the apparatus, Fig. 9, which consists of a tall glass U-tube, upon the end of one limb of which there is fastened, by means of a cork, a porous cylindrical pot, such as

See Experiments Nos. 350-359, Newth's "Chemical Lecture Experiments,"

new ed.

is used in an ordinary Bunsen battery. The U-tube is half filled with coloured water. Under ordinary circumstances air is continually diffusing through the porous pot, but as it passes at an equal rate in both directions, there is no disturbance of the pressure, and consequently the coloured water remains level in the two limbs. If now a beaker containing hydrogen be brought over the apparatus, as seen in the figure, the hydrogen will stream through the porous pot so much more rapidly than the air in the pot can make its way out, that there will be an increase in the total amount of gas inside the apparatus, which will be instantly rendered evident by the change of level of the liquid in the U-tube, the water being forcibly driven down the tube which carries the porous pot. Upon removing the beaker the reverse operation will at once take place; the hydrogen inside the apparatus now rapidly diffuses out, and much more quickly than air can pass in, consequently a reduction of pressure within the apparatus results, which is indicated by a disturbance of the level of the water in the tube, in the opposite direction to that which occurred at first.

The Law of Gaseous Diffusion.-Graham established the law according to which the diffusion of gases is regulated, and it may be thus stated: The relative velocities of diffusion of any two gases are inversely as the square roots of their densities.

The density of hydrogen being 1, that of air is 14.44, the velocity of the diffusion of hydrogen, therefore, as compared with that of air, will be in the ratio of √14.44 to √. √14.44 = 3.8, √1 = 1. Therefore hydrogen diffuses 3.8 times faster than air ; or 3.8 volumes of hydrogen will pass out through a porous septum, while only I volume of air can enter.

If d = the density of a gas, air being unity, and 7 = the volume of the gas which diffuses in the same time as I volume of air, then

The following table gives in the last column the results obtained by Graham, which will be seen to accord very closely with the calculated numbers demanded by the law of diffusion :

The property of diffusion is sometimes made use of in order to separate gases, having different densities, from gaseous mixtures. This process of separation by diffusion is known as atmolysis. The principle may readily be illustrated by causing a mixture of

FIG. 10.

oxygen and hydrogen, in

proportion to form an explosive mixture, to slowly traverse tubes made of porous material, such as ordinary tobacco pipes. Two such pipes may be arranged as shown in Fig. 10, and the gaseous mixture passed through in the direction indicated by the arrow. On collecting the issuing gas over water in a pneumatic trough, it will be found to have so far lost the hydrogen, by diffusion through the tube, that a glowing splint of

wood when introduced into it will be reignited.

From the rate of diffusion of ozone, in a mixture of ozone and oxygen, Soret was able to calculate the density of this allotropic form of oxygen, and so confirm the result he had previously obtained by other methods (see Ozone).

Attempts have been made to utilise this principle in order to obtain oxygen from the air. The relative densities of oxygen and