ascertained by reading the position of the mercury within the tube, and deducting the height of the column from 6 to the surface of the mercury a, from the barometric pressure at the time of making the experiment. The temperature of the water in the cylinder may be varied, and the coefficient of absorption at different temperatures can thus be determined.

Fig. 21 represents a more modern absorptiometer, being a modified form of Heidenhain and Meyer's apparatus. In this instrument the measuring tube and the absorption vessel are separate, and it admits of the use of much larger volumes of liquid. By means of the three-way cock a, the gas to be experimented upon

B

a

is introduced into A by first raising and then lowering B; and the volume is measured when the levels of the mercury in A and B are coincident. By means of the three-way cock b, the vessel C, of known capacity, and which is connected with A by means of a flexible metal capillary tube, is filled with the desired liquid. The vessels A and C are then put into communication, and by raising B and opening the tap ca definite volume of the liquid is run out into a measuring vessel, which represents the volume of gas that enters. The gas and liquid are then thoroughly agitated, after which the gas is passed back into A by lowering 5, and, when A and Care in communication, opening the tap beneath mercury. By measuring the diminution in volume suffered by the gas, the volume absorbed by the known volume of liquid is obtained. The measuring tube and absorption vessel are kept constant at any desired temperature by surrounding them by water, or with vapours at known temperatures.

FIG. 21.

Solubility of Mixed Gases.-When two gases are mixed together, the pressure exerted by each is the same as would be exerted if the other were absent and the entire space were occupied by the same mass of the one. Thus, if a mixture of two gases are in the proportion of two volumes of one and one volume of the other, the pressure exerted by the one present in

larger proportion will be twice as great as that of the other; this pressure is termed the partial pressure of the gas under the circumstances, and obviously the total pressure of the mixture will be the sum of the partial pressures of the constituents. As the solubility of a gas in a liquid is proportional to the pressure, the solubility of the gases in a gaseous mixture will be influenced by the proportions in which they are present in the mixture. This is known as Dalton's law of partial pressures, which may be thus stated: The solubility of a gas in a gaseous mixture is proportional to its partial pressure. For example, the atmosphere consists of a mixture of oxygen and nitrogen, in the proportion of four volumes of nitrogen to one volume of oxygen (in round numbers). The partial pressure exerted by the oxygen is therefore only one-fifth of the total atmospheric pressure, and consequently the amount of oxygen which a given volume of a liquid is capable of dissolving from the atmosphere is only about one-fifth of that which it will absorb from pure oxygen-in other words, will be one-fifth the absorption coefficient of oxygen for that liquid.

The application of the law of partial pressures will be seen in the solvent action of water upon the atmosphere. Taking the coefficients of absorption of oxygen and nitrogen for water as given by Bunsen—

and the proportion of oxygen to nitrogen in the air as one to four, by volume, we get

for the number of cubic centimetres of oxygen and nitrogen which will be dissolved from the atmosphere by I cubic centimetre of water at o°.

One hundred volumes of water, therefore, will dissolve 2.451 volumes of air, of which .823 volume is oxygen and 1.628 volumes is nitrogen; and if this dissolved air be again expelled from the water by boiling, the air so obtained will contain oxygen and nitrogen in the proportions—

If a mixture of oxygen and nitrogen in this proportion be once more dissolved in water, since the percentage of oxygen has risen from 20 to 33.6, and the partial pressure proportionately increased, the mixture of the two gases that will be dissolved will be still richer in oxygen; and after solution in water for the third time the boiled-out air will be found to contain as much as 75 per cent. of oxygen. It will be obvious that the partial pressure which determines the extent to which the separate gases in a mixture are dissolved is not represented by the proportion in which the gases are present before solution, but that in which they exist in the gaseous mixture after the solvent has become saturated.

Henry's law does not hold good in the case of such very soluble gases as ammonia, hydrochloric acid, &c. These gases appear to enter into a true chemical union with the water, and in most of these cases the act of solution is attended with considerable evolution of heat. In some of these instances the deviation from the law diminishes with rise of temperature; thus at temperatures above 40° the absorption of sulphur dioxide obeys the law, while in the case of ammonia conformity to the law is observed at 100°.

The gases dissolved by a liquid are not only expelled by boiling. but are withdrawn by placing the solution in a vacuum. This, indeed, follows from Henry's law, for if the solubility is proportional to the pressure, and the pressure is nil, the amount of gas dissolved must also be nil.

The molecules of gas dissolved by a liquid are regarded as being held by some attractive forces exerted between them and the molecules of the liquid; in the course of their movements, gas molecules are constantly leaving and entering the liquid, and equilibrium is established when the same number enter and escape from the surface of the liquid in the same time. When the pressure is increased, more gas molecules strike the surface in a unit of time, and consequently a greater volume is absorbed. When a solution of a soluble gas is placed in an atmosphere of another gas, the dissolved gas continues to leave the liquid until equilibrium is established between the pressure exerted by the gas so leaving and the amount remaining in solution. For this reason a solution of ammonia, when left exposed to the air, rapidly becomes weaker, owing to the escape of the dissolved gas into the atmosphere. This process is accelerated if a stream of a less soluble gas be caused to bubble through the solution.

Solubility of Liquids in Liquids. The solubility of liquids in

liquids may be divided into two orders. First, cases in which the degree of solubility of one in the other is unlimited; and second, cases where the extent of the solubility is limited, or where the liquids are said to be partially miscible. Two liquids whose solubility in each other is unlimited are said to be miscible in all proportions; thus alcohol and water are capable of forming a homogeneous mixture when added together in any proportion.

In the second class, where the solubility of two liquids for each other is limited, it is found that each liquid is capable of dissolving some of the other. Thus, if equal volumes of ether and water are shaken together, the liquids will afterwards separate out into two distinct layers, one floating upon the other. The heavier layer at the bottom is an aqueous solution of ether, containing about 10 per cent. of ether; while the upper liquid is an ethereal solution of water containing about 3 per cent. of water. The presence of ether dissolved in the water may be proved by separating the two layers and gently heating the aqueous liquid in a small flask, when the dissolved ether will be expelled and can be inflamed. The presence of the water in the ether is also readily proved, either by introducing into the liquid a small quantity of dehydrated copper sulphate, which will rehydrate itself at the expense of the water in the ether, and be changed from white to blue; or by placing in the ethereal liquid a fragment of sodium, which decomposes the dissolved water with the liberation of hydrogen.

Another illustration of two partially miscible liquids is seen in the case of a strong aqueous solution of potassium carbonate and strong ammonia, which is of special interest as being the only example at present known of two aqueous solutions of inorganic substances which exhibit this phenomenon.* Thus, when strong aqueous ammonia (sp. gr. o.880) is added to a concentrated solution of potassium carbonate, the two liquids separate from each other in two distinct layers, the upper layer consisting of ammonia which has taken up a certain amount of potassium carbonate, while the lower liquid consists of a solution of potassium carbonate which has dissolved a definite quantity of ammonia.

In most cases the solubility of liquids in liquids is increased by rise of temperature, although in some it is decreased. As an example of the former, the case of these two aqueous liquids may be quoted. If the temperature be raised then the solubility of cach of these solutions in the other steadily increases, and the

Newth, Trans. Chem. Soc., 1900, p. 775.