separated into its constituents by making use of a physical property of these gases called rate of diffusion, which property

p. 30. Experiment describing diffusion of oxygen and hydrogen should be so arranged that the tubes are covered with a bell-jar filled with carbon dioxide.

is a characteristic mark of each gas when unmixed with other substances.

If now a mixture is made of equal volumes of hydrogen 39 and chlorine, and this mixture is exposed to diffused sunlight for some time, a new gas will be formed; the new gas is colourless; chlorine is yellow, hydrogen is colourless-: the new gas fumes much in the air-neither chlorine nor hydrogen fumes in air; it has an intensely acrid smell, quite different from the smell of chlorine. This gas is called hydrogen chloride. The weight of hydrogen chloride formed is equal to the sum of the weights of the hydrogen and chlorine which have combined to form it.

If hydrogen chloride is passed through a dry clay-pipe under conditions similar to those already described it can be proved that no separation into hydrogen and chlorine has occurred, but that the gas which issues at a is identical with that which issues at b, and that both are hydrogen chloride.

The compound of hydrogen and chlorine (hydrogen chloride) cannot be separated into its constituents by making use of a certain physical property-viz. rate of diffusion-which belongs to, and characterises, each of its constituents when these are uncombined with other kinds of matter.

The results of experiments such as those considered in 40 pars. 33 to 39 shew that the substances comprised in the class Not-Elements may be divided into two groups; compounds and mixtures.

41

Each constituent of a mixture retains in the mixture the properties which characterise it when unmixed with other substances: the properties of the mixture are, broadly, the sum of the properties of the constituents. No constituent of

a compound retains in the compound the properties which characterise it when separated from other substances: the properties of a compound are not the sum of the properties of the constituents; the compound is a definite kind of matter, as distinct from each of its constituents as these are from one another, and yet formed by the combination of these constituents.

To say of a mixture, that it contains the bodies by mixing which it has been produced, is to use an expression which conveys a correct notion of the relations of the properties of the mixture to those of its constituents. But it is not so correct to say that a compound contains each of those kinds of matter by the interaction of which it has been formed. Thus, a mixture of iron and sulphur contains iron and contains sulphur; inasmuch as, not only is the mass of the mixture the sum of the masses of the mixed iron and sulphur, but the properties of the mixture are also the properties of iron added to those of sulphur. The mass of a compound of iron and sulphur is certainly the sum of the masses of the iron and sulphur which have combined to form it; but the properties of the compound are quite distinct from the properties by which iron or sulphur is marked off from other kinds of matter.

The formation of a mixture is a physical process.

The properties of every mixture probably differ slightly. from the sum of the properties of its constituents; some change occurs in the formation of the mixture; nevertheless the properties of each kind of matter in the mixture are so slightly modified by the presence of the other kinds of matter that it is always possible, and generally easy, to recognise each of these kinds of matter by some, or all, of the properties which distinctly mark it off from other kinds of matter.

The formation of a compound is a chemical process.

The properties of each of those kinds of matter which combine are so largely modified by the presence of the other combining substances that it is impossible to recognise any of them by the properties which belong to it when uncombined. Iron sulphide is as distinct and definite a kind of matter as iron or sulphur; ammonium chloride is as distinct and definite

a kind of matter as ammonia or hydrogen chloride; butylene bromide is marked off from other kinds of matter by properties as distinct and definite as those which characterise butylene or bromine; hydrogen chloride, so far as its physical properties indicate, is as homogeneous and as little formed of unlike parts as either hydrogen or chlorine.

Iron sulphide, or ammonium chloride, or butylene bromide, 42 or hydrogen chloride, can be separated into unlike parts; but this separation is accompanied by the disappearance of all the distinctive properties of the compound, and by the production, in each case, of two kinds of matter-iron and sulphur, ammonia and hydrogen chloride, butylene and bromine, hydrogen and chlorine-so unlike the compounds from which they have been produced that the only expression to be used regarding the occurrence is that each compound has ceased to exist and has been replaced by two new kinds of matter. Neither iron nor sulphur has yet been separated into unlike parts; the methods which succeed in separating iron sulphide into iron and sulphur fail to separate iron or sulphur into kinds of matter different from iron or sulphur. Bromine likewise refuses, at present, to reveal its composition, if composition it has in the sense in which it may be said that butylene bromide is composed of butylene and bromine. But ammonia and hydrogen chloride, which are produced by separating ammonium chloride into unlike parts, can, each, be further separated into two kinds of matter totally unlike either ammonia or hydrogen chloride. Ammonia is formed by the union of, and can be resolved into, two colourless, odourless, gases-nitrogen and hydrogen; hydrogen chloride is formed by the union of, and can be resolved into, hydrogen, and another, yellowishgreen, badly smelling, gas, chlorine. All attempts to separate nitrogen, or hydrogen, or chlorine, into unlike parts, have hitherto failed.

A mixture is separated into its constituents by making 43 use of some property or properties of each constituent which belong to that substance when it exists apart from other kinds of matter. Thus the mixture of iron and sulphur was separated by making use of the fact that iron is attracted by a magnet while sulphur is not attracted; or of the fact that iron sinks in water, while sulphur floats, at least for a time; or of the fact that sulphur is soluble, while iron is not soluble, in carbon disulphide. The mixture of ammonia and charcoal was separated by taking advantage of one of the properties of ammonia viz.

M. E. C.

3

44

that it is a very volatile gas. The property possessed by hydrogen of diffusing four times more rapidly than oxygen through a porous plate gave us a method for approximately separating a mixture of hydrogen and oxygen into its constituents.

But if a compound is to be separated into unlike parts it is necessary either to act upon it by some natural agency, or form of energy, such as heat, light, or electricity—or in some cases mechanical energy—or, and this is the more usual method, to cause it to interact under suitable conditions with some other

kind, or kinds, of matter. Thus the compound water was separated into hydrogen and oxygen by passing an electric current through the water (s. experiment in par. 9). Similarly ammonia may be separated into nitrogen and hydrogen by passing electric sparks through it.

Copper oxide was separated into copper and oxygen (s. par. 28) by causing it to interact with hydrogen at a high temperature; the results of this interaction were copper and water; but the results of a previous experiment shewed that water is produced by the combination of hydrogen with oxygen.

As we proceed in our study we shall learn more of the methods employed for separating compounds into the different kinds of matter by the combination of which they are produced; meanwhile it is important to observe that the method does not consist in making use of the physical properties belonging to these different kinds of matter. The formation and decomposition of a compound are chemical processes.

We have already learned that the chemist puts in one class all those distinct kinds of matter which he has not been able to separate into unlike parts, and calls them Elements.

We now learn that certain Not-Elements are distinct kinds of matter, each marked by its own definite and characteristic properties, yet each capable of being separated into parts, totally unlike each other, and unlike the original. These not-elements the chemist puts in one class, and calls them Compounds. One marked characteristic, viz. the constancy of composition, of compounds will be dealt with later. (pars. 58 and 59.)

All other substances belonging to the group Not-Elements are classed together and called Mixtures. An infinite number of these exists, or may be formed, by mixing elements with elements, or compounds with compounds, or elements with

compounds, or mixtures of any of these with other mixtures; they are all marked off from elements and compounds by the. facts that their properties are, broadly, the sums of the properties of their constituents, and their constituents exist in the mixtures each with its own properties scarcely, if at all, modified by the presence of the other constituent parts.

Chemistry deals with certain parts of the phenomena 45 presented in the changes of elements into compounds, and of compounds into simpler compounds or into elements. Chemistry concerns itself but little with the formation of mixtures or the resolution of mixtures into their constituents.

By the composition of an element is meant the element 46 itself; so far as our knowledge goes at present, each kind of matter placed in the class element is entirely homogeneous.

By the composition of a compound is meant, primarily, a statement of the elements by the combination of which the compound is formed and into which it can be resolved, and also a statement of the mass of each element which goes to form, or can be obtained from, a specified mass of the compound. By the composition of a compound is frequently meant a statement of certain less complex compounds, and of the masses of these, which interact to produce a specified mass of the compound in question, or which can be obtained from a specified mass of this compound. Thus, experiment has shewn us (par. 35) that the compound ammonium chloride is produced by the interaction of ammonia and hydrogen chloride; experiment (par. 39) has also told us that hydrogen chloride is itself a compound of hydrogen and chlorine. It may also be proved that ammonia is a compound of nitrogen and hydrogen. The composition of ammonium chloride may be expressed by either of the following statements:

(1) 100 parts by weight of ammonium chloride are formed by the combination of 31-77 parts by weight of ammonia and 68.23 parts by weight of hydrogen chloride;

(2) 100 parts by weight of ammonium chloride are formed by the combination of 26 17 parts by weight of nitrogen, 7·48 parts by weight of hydrogen, and 66-35 parts by weight of chlorine.

We have now gained a clearer conception of chemical 47 change. We now regard such a change as, either the change of a specified mass of a compound into fixed masses of two or more compounds or elements, or the interaction of fixed masses of two or more elements or compounds to produce