nitrogen in the air, it is, of course, necessary to guard against any error arising from the presence of the water, carbonic acid, and ammonia. With this view, Dumas and Boussingault, to whom we are chiefly indebted for our exact knowledge of the composition of the air, caused it to pass through a series of tubes (A, fig. 14) containing potash, in order to remove the carbonic acid, then through a second series (B), containing sulphuric acid, to absorb the ammonia and water; the purified air then passed through a glass tube (C) filled with bright copper heated to redness in a charcoal furnace, which removed the whole of the oxygen, and the nitrogen passed into the large globe (N).

Both the tube (containing the copper) and the globe were carefully exhausted of air and accurately weighed before the experiment; on connecting the globe and the tube with the purifying apparatus, and slowly opening the stop-cocks, the pressure of the external air caused it to flow through the series of tubes into the globe destined to receive the nitrogen. When a considerable quantity of air had passed in, the stop-cocks were again closed, and, after cooling, the weight of the globe was accurately determined. The difference between this weight and that of the empty globe before the experiment, gave the weight of the nitrogen which had entered the globe, but this did not represent the whole of the nitrogen contained in the analysed air, for the

Fig. 14.-Exact analysis of air.

tube containing the copper had, of course, remained full of nitrogen at the close of the experiment. This tube having been weighed, was attached to the air-pump, the nitrogen exhausted from it, and the tube again weighed; the difference between the two weighings furnished the weight of the nitrogen remaining in the tube, and was added to the weight of that received in the globe. The oxygen was represented by the increase in the weight of the exhausted tube containing the copper, which was partially converted into oxide of copper, by combining with the oxygen of the air passed through it.

The calculation of the result of the analysis is here exemplified :

2573

Add nitrogen received into the globe,

Exhausted tube (C) with oxidised copper (at the conclusion),
metallic copper (at the commencement), 2550

Oxygen in the air analysed,

23

The ratio of the oxygen to the nitrogen, therefore, is that of 23N: 770, or 1N: 3.3470. 100 parts by weight of the air purified from water, carbonic acid, and ammonia, contain 77 parts of nitrogen and 23 parts of oxygen.

20. The nitrogen remaining after the removal of the oxygen from air in the above experiments, was so called on account of its presence in nitre (saltpetre KO. NO,). In physical properties it resembles oxygen, but is somewhat lighter than that gas, its specific gravity being 0.9713.

This difference in the specific gravities of the two gases is well exhibited by the arrangement shown in fig. 15. A jar of oxygen (O) is closed with a glass plate, and placed upon the table. A jar of nitrogen (N), also closed with a glass

plate, is placed over it, so that the two gases may come in contact when the glass plates are removed. The nitrogen will float for some seconds above the oxygen, and if a lighted taper be quickly introduced through the neck of the upper jar, it will be extinguished in passing through the nitrogen, and will be rekindled brilliantly when it reaches the oxygen in the lower jar.

It might at first sight appear surprising that oxygen and nitrogen, though of different specific gravities, should exist in uniform proportions in all parts of the atmosphere, unless in a state of chemical combination, but an acquaintance with the property of diffusion (see Hydrogen) possessed by gases teaches us that gases will mix with each other in opposition to gravitation, and when mixed will always remain so.

That air is simply a mechanical mixture of its component gases is amply proved by the circumstance that it possesses all the properties which would be predicted for a mixture of these gases in such proportions; whilst the essential feature of a chemical compound is, that its properties cannot be foreseen from those of its constituents.

Fig. 15.

The absence of active chemical properties is a very striking feature of nitrogen, and admirably adapts it for its function of diluting the oxygen in the atmosphere. There is no direct test by which nitrogen gas can be recognised, so that the chemist is obliged to prove that the gas under examination does not possess the characters of any other gas with which he is acquainted before he can pronounce it to be nitrogen.

The chemical relations of air to animals and plants will be more appropriately discussed hereafter. (See Carbonic Acid, Ammonia.)

HYDROGEN.

21. Unlike oxygen, hydrogen is very rarely found uncombined in nature. In combination it occurs abun

dantly in water and in all animal and vegetable substances. All varieties of fuel contain hydrogen. It is always procured from the first of these

sources.

Water is composed of the two elements, hydrogen and oxygen, held together by chemical attraction. To separate these elements, that is, to decompose or analyse water, we have to overcome the chemical attraction between them, which may be effected by causing the particles of water to transmit a current of voltaic electricity.

Fig. 16.-Electrolysis of water.

is represented in fig. 16.

20

ANALYSIS OF WATER.

The glass vessel A contains water, to which a little sulphuric acid has been added to increase its power of conducting electricity, for pure water conducts so imperfectly that it is decomposed with great difficulty. B and C are platinum plates bent into a cylindrical form, and attached to stout platinum wires, which are passed through corks in the lateral necks of the vessel A, and are connected by binding screws with the copper wires D and E, which proceed from the galvanic battery G. H and O are glass cylinders with brass caps and stop-cocks, and are enlarged into a bell-shape at their lower ends for the collection of a considerable volume of gas. These cylinders are filled

Fig. 17.

with the acidulated water, by sucking out the air through the opened stop-cocks; on closing these, the pressure of the air will of course sustain the column of water in the cylinders. G is a Grove's battery, consisting of five cells or earthenware vessels (A, fig. 17) filled with diluted sulphuric acid (one measure of oil of vitriol to four of water). In each of these cells is placed a bent plate of zinc (B), which has been amalgamated or rubbed with mercury (and diluted sulphuric acid) to protect it from

corrosion by the acid when the battery is not in use. Within the curved portion of this plate rests a small flat vessel of unglazed earthenware (C), filled with strong nitric acid, in which is immersed a sheet of platinum foil (D). The platinum (D) of each cell is in contact, at its upper edge, with the zinc (B) in the adjoining cell (fig. 18), so that at one end (P, fig. 16) of the battery there is a free platinum plate, and at the other (Z) a free zinc plate. These plates are connected with the wires D and E by means of the copper plates L and K attached to the ends of the wooden trough in which the cells are arranged. The wire D (fig. 16), which is connected with the last zinc plate of the battery, is often called the "negative pole ;" whilst E, in connexion with the last platinum plate, is called the "positive pole."

Fig. 18.

When the connexion is established by means of the wires D and E with the "degalvanic current" is commonly said to pass along the composing cell" (A), the " wire E to the platinum plate C, through the acidulated water in the decomposing cell, to the platinum plate B, and thence along the wire D back to the battery.

22. During this "passage of the current" (which is only a figurative mode of expressing the transfer of the electric influence), the water intervening between the plates B and C is decomposed, its hydrogen being attracted to the plate B (negative pole), and the oxygen to the plate C (positive pole). The gases can be seen adhering in minute bubbles to the surface of each plate, and as they increase in size they detach themselves, rising through the acidulated water in the tubes H and O, in which the two gases are collected.

Since no transmission of gas is observed between the two plates, it is evident that the H and O separated at any given moment from each plate do not result from the decomposition of one particle of water, but from two particles, as represented in fig. 19, where A represents the particles of water lying between the plates P and Z before the "current" is passed,

ELECTROLYSIS OF WATER.

21

27

and B the state of the particles when the current has been established. P is (the positive pole) in connexion with the last platinum plate of the battery, and Z is (the negative pole) in connexion with the last zinc plate.

The signs and made use of in B refer to a common mode of accounting for the decomposition of water by the battery, on the supposition that the oxygen is in a negatively electric condition, and therefore attracted by the positive pole P; whilst the hydrogen is in a positively electric condition, and is attracted by the negative pole Z.

The decomposition of compounds by galvanic electricity is termed electrolysis. When a compound of a metal with a non-metal is decomposed in this manner, the metal is usually attracted to the (negative) pole in connexion with the zinc plate of the battery, whilst the non-metal is attracted to the (positive) pole connected with the platinum plate of the battery.

Hence the metals are frequently spoken of as electro-positive elements, and the non-metals as electro-negative.

23. If the passage of the "current" be interrupted when the tube H has become full of gas, the tube O will be only half full, since water contains hydrogen and oxygen in the proportion of two volumes of hydrogen to one volume of oxygen. When the wider portions of the tubes (fig. 16) are also filled, the two gases may be distinguished by opening the stop-cocks in succession, and presenting a burning match. The hydrogen will be known by its kindling with a slight detonation, and burning with a very pale flame at the jet; whilst the oxygen will very much increase the brilliancy of the burning match, and if a spark left at the extremity of the match be presented to the oxygen, the spark will be kindled into a flame. The oxygen will be found to smell strongly of ozone, and will impart a deep blue tinge to the iodised starch paper (see Ozone).

Another method of effecting the decomposition of water by electricity consists in passing a succession of electric sparks through steam. It is probable that in this case the decomposition is produced rather by the intense heat of the spark than by its electric influence.†

For this purpose, however, the galvanic battery does not suffice, since no spark can be passed through any appreciable interval between the wires of the battery, a fact which electricians refer to in the statement that although the quantity of electricity developed by the galvanic battery is large, its intensity is too low to allow it to discharge itself in sparks like the electricity from the machine or from the induction-coil, which possesses a very high intensity, though its quantity is small.

24. The most convenient instrument for producing a succession of elec

• "HACKтрOV (amber-root of electricity); Xów, to loosen.

That a very intense heat is capable of decomposing water into its elements has long been known. When globules of melted platinum are dropped into water, bubbles of hydrogen and oxygen are disengaged.

22

DECOMPOSITION OF STEAM.

tric sparks is the induction-coil, by the aid of which the electric influence of even a single cell of the galvanic battery may be so accumulated as to become capable of discharging itself in sparks, such as are obtained from the electrical machine.*

Fig. 20 represents the arrangement for exhibiting the decomposition of steam by the electric spark.

A is a half-pint flask furnished with a cork in which three holes are bored; in one of these is inserted the bent glass tube B, which dips beneath the surface of the water in the trough C.

D and E are glass tubes, in each of which a platinum wire has been sealed so as to project about an inch at both ends of the tube. These tubes are thrust through the holes in the cork, and the wires projecting inside the flask are made to approach to within about inch, so that the spark may pass easily between them.

The flask is somewhat more than half filled with water, the cork inserted, and the tube B allowed to dip beneath the water in the trough; the wires in D and E being connected with the thin copper wires passing from the induction-coil F, which is connected by stout copper wires with the small battery G.

The water in the flask is boiled for about fifteen minutes, until all the air contained in the flask has been displaced by steam. When this is the case, it will be found that if a glass test-tube (H) filled with water be invertedt over the orifice of the tube B, the bubbles of steam will entirely condense, with the usual sharp rattling sound, and only insignificant bubbles of air will rise to the top of the test-tube. If now, whilst the boiling is still continued, the handle of the coil (F) be turned so as to cause a succession of sparks to pass through the steam in the flask, large bubbles of incondensable gas will accumulate in the tube H. This gas consists of the hydrogen and oxygen gases in a mixed state, having been released from their combined condition in water by the action of the electric sparks. The gas may be tested by closing the mouth of the tube H with the thumb, raising it to an upright position, and applying a lighted match, when a sharp detonation will indicate the recombination of the gases.

25. In the preceding experiments, the force of chemical attraction holding the particles of oxygen and hydrogen together in the form of water, has been overcome by the physical forces of heat and electricity. But water may be more easily decomposed by acting upon it with some element which has a sufficiently powerful chemical attraction for the oxygen of water to draw it away from the hydrogen.

*For a description of the induction-coil, see Miller's "Elements of Chemistry," Part I. p. 532.

†The end of the tube B should be bent upwards and thrust into a perforated cork with notches cut down the sides. By slipping this cork into the neck of the test-tube, the latter will be held firmly.

With a powerful coil, a cubic inch of explosive gas may be collected in about fifteen minutes.