in Chemistry, a general name for all permanently elastic fluids, which are obtained by chemical processes, as azotic gas, muriatic acid gas, nitrous gas. It is derived from the German gascht or gast, signifying the ebullition attending the expulsion of elastic fluids from substances in a state of effervescence. Gas-Light.

Light, whether obtained from natural or artificial sources, is so necessary for the correct and successful execution of almost every operation of human industry, that whatever is calculated to simplify the means of procuring it, or to increase its intensity, cannot fail of being attended with the most beneficial consequences to civilized society. For every purpose to which it is applied, it must be admitted that the light of day, when it can be enjoyed freely and without interruption, is, by far, the most suitable; but in large and crowded cities, as well as in less favourable situations in point of climate, where the sun is sometimes shrouded for days together in dense and impenetrable clouds, it becomes expedient to compensate for the absence of his rays by artificial substitutes, which, however inferior in brilliancy and general usefulness, may nevertheless answer sufficiently well in those cases where a less ample supply of light is requisite.

Some substances, denominated phosphorescent, have the property of absorbing the solar rays, on being exposed for a short time to their influence, and of emitting the light which they thus imbibe when they are afterwards placed in the dark; but the feeble and transient illumination which they shed, though sufficient to indicate their luminous condition, is totally unfit to afford such a supply of light as is necessary for conducting any of the operations of art, which must be performed with care and precision.

During the process of combustion, however, a variety of inflammable substances, both of animal and vegetable origin, are found to give out light, as well as heat, when they undergo that species of decomposition; and hence, from the earliest periods of human society, it has been customary to burn substances of that description, for the purpose of obtaining artificial light. These substances, which were generally of a fatty or oleaginous nature, are found, when analyzed, to be composed chiefly of carbon and hydrogen. When they are exposed to a certain pitch of temperature they are resolved into some of the compound gases which result from the union of these elements, particularly carburetted and bicarburetted hydrogen, or olefiant gas, both of which are highly inflammable, and yield, during their combustion, a fine white light. To facilitate the decomposition, and carry on the combustion with due economy, a quantity of some fibrous substance, in the form of a wick, is connected with the oleaginous matter, for the purpose of causing it to burn slowly and effectually. Accordingly, if the flame be suddenly extinguished, the inflammable gas which is formed by the decomposition of the matter in immediate contact with the wick is observed to escape from it, and may be again set on fire by the application of a lighted taper.

When the light obtained from these substances must be frequently conveyed from place to place, no arrangement can be more convenient for their decomposition than that which is effected by means of the wick; but if it be to remain in a permanent position, it will frequently be more advantageous to resolve the oleaginous matter into gas, and then transmit it, in that state, through pipes, to the various points where it is to be consumed.

Though the different substances which have been used, from the earliest times, for yielding artificial light, have always been actually resolved into gas before they underwent the process of combustion, that fact was entirely unknown till pneumatic chemistry unfolded the properties of the aerial bodies, which perform so many important functions in the economy of nature, as well as in the processes of the arts. The use of gas for the purpose of illumination is therefore of recent date; but though late in its origin, the successive improvements which the invention has received, from the joint labours of chemists and practical engineers, have carried it, in a few years, to a pitch of perfection to which little can be added, either to simplify the processes for producing the gas itself, or improving its quality.

Besides matters of an oleaginous nature, it has been ascertained that pit-coal and other bituminous substances yield, when they are exposed in close vessels to a high temperature, a large quantity of aerial matter, which is found to consist of various gases, most of which are extremely inflammable. This fact seems to have first attracted the notice of the Reverend Mr Clayton, who gives an account of it in a memoir on the subject, which he published in the Transactions of the Royal Society for the year 1739; and as his experiments furnish the earliest evidence of the possibility of extracting from coal, by means of heat, a permanently elastic fluid of an inflammable nature, we shall give the account of the discovery in his own words: "Having introduced a quantity of coal into a retort, and placed it over an open fire," he states that "at first there came over only phlegm, and afterwards a black oil, and then likewise a spirit arose which I could noways condense; but it forced my lute, or broke my glasses. Once when it had forced my lute, coming close thereto in order to repair it, I observed that the spirit which issued out caught fire at the flame of the candle, and continued burning with violence as it issued out in a stream, which I blew out and lighted alternately for several times. I then had a mind to try if I could save any of this spirit; in order to which I took a tubulated receiver, and putting a candle to the pipe of the receiver whilst the spirit rose, I observed that it caught flame, and continued burning at the end of the pipe, though you could not observe what fed the flame. I then blew it out and lighted it again several times; after which I fixed a bladder, squeezed, and void of air, to the pipe of the receiver. The oil and flame descended into the receiver, but the spirit still ascending, blew up the bladder. I then filled a good many bladders therewith, and might have filled an inconceivable number more; for the spirit continued to run for several hours, and filled the bladders almost as fast as a man could have blown them with his mouth; and yet the quantity of coals distilled was considerable.

"I kept this spirit in the bladders a considerable time, and endeavoured several ways to condense it, but in vain; and when I had a mind to divert strangers or friends, I have frequently taken one of these bladders, and pricked a hole therein with a pin, and compressing gently the bladder near the flame of a candle till it once took fire, it would then continue flaming till all the spirit was compressed out of the bladder; which was the more surprising, because no one could discern any difference in the appearance between this bladder and those which are filled with common air."

It is evident from this narrative, related with so much simplicity, that an accident which happened to Mr Clayton's apparatus was the means of leading to the discovery of Mr Clayton alludes to the discovery of the gas he obtained from coal, in a letter to the Royal Society, dated May 12, 1835. Of the Site and general Arrangement of the Apparatus for Gas-Light.

In describing the site and general arrangements of a gas establishment, it is not easy to give directions respecting points, which must be regulated, in every case, by circumstances of a local nature; but when a choice of ground is in our power, a spot ought to be selected having a central situation with regard to the buildings, streets, &c. it is to supply with light, and standing as nearly as possible on a medium level with them. When the manufactory is placed considerably below that level, the gas is apt to be propelled with too much velocity through the burners; and when above it, an opposite inconvenience is experienced, the gas being in that case necessarily subjected to an extra pressure, by which the chance of its escape through any imperfection of the pipes is proportionally increased. Of the two evils, therefore, the least objectionable is that in which the situation of a gas-work is below the mean level of the streets.

But besides the conditions favourable to an equable and uniform distribution of the gas at the different points to which it may be conducted, there are other considerations scarcely less important, which, in selecting a proper site for the erection of the establishment, ought not to be disregarded. Among these may be reckoned a regular supply of water for the various manipulations of the work; and facility of access for the delivery of coal, and the removal of the coke, tar, and other products of the distillation. Nor, in fixing the situation of an establishment which is professedly erected for the public benefit, ought the comfort or the interest of individuals to be entirely overlooked; for though a gas-work may not prove, under proper management, a nuisance, it can never be considered to be any advantage to the neighbourhood in which it is placed.

The apparatus for the production and purification of coal-gas consist, in the first place, of suitable vessels for decomposing by heat the coal from which the gas is to be procured; secondly, of a series of pipes for conveying off the gas, and conducting it into proper receptacles, where it may be separated from the grosser products, which tend to impair the brilliancy of the light it is capable of yielding; thirdly, of the condensing apparatus, for removing more effectually the tar and other condensible substances that come over with the gas; fourthly, of the purifying apparatus, for abstracting the sulphuretted hydrogen, carbolic acid, &c., by which the gas is contaminated; and, fifthly, of the gasometer or gas holder, with its tank, into which the gas is finally received in a purified state.

Of the Retorts, or Vessels for decomposing the Coal.

The vessels employed for the decomposition of coal and other substances capable of yielding carburetted hydrogen, by their destructive distillation, are usually formed of cast iron, and termed retorts. Various shapes have been adopted in the construction of these vessels; nor have their forms been more varied than the modes in which they have been disposed, in the furnaces erected for their reception. In most instances, they have been constructed of a cylindrical shape, varying in length, and diameter. Those first employed were placed with their axis in a vertical direction; but experience soon showed that this position was extremely inconvenient, on account of the difficulty it occasioned of removing the coke, and other residuary matters, after the coal had been carbonized. Attempts were made to remedy this inconvenience, by enlarging the size of the retort, and introducing the coal enclosed in a proper grating of iron, having the form of a cage. The increased dimensions of the retort, from which the principal advantage to be derived from this arrange- Gas-Lightment was expected, were found, however, to present great obstacles to the complete carbonization of the coal; for though the disengagement of gas during the first stages of the process was sufficiently copious, it diminished rapidly the longer the distillation was continued; a crust of coke being formed next the heated metal, which not only opposed the transmission of the temperature to the internal mass of coal, but gradually prevented, by its accumulation, the extrication of the gas from the undecomposed portion of it.

The retorts were, therefore, next placed in a horizontal position, as being not only more favourable to the most economical distribution of the heat, but better adapted to the introduction of the coal, and the subsequent removal of the coke, after the carbonization was carried to a due extent. At first the heat was applied directly to the lower part of the retort, but it was soon observed that the high temperature to which it was necessary to expose it, for the perfect decomposition of the coal, proved destructive to it, and rendered it useless long before the upper part had sustained much injury. The next improvement was accordingly to interpose an arch of brickwork between it and the furnace, and to compensate for the diminished intensity of the heat by a more diffused distribution of it over the surface of the retort. This was effected by causing the flue of the furnace to return towards the mouth of the retort, and again conducting it in an opposite direction, till the heated air finally escaped into the chimney.

This arrangement, which on the whole has been found to be the best, and is now generally used, seems to admit of little improvement, unless with respect to the shape of the retort. The cylindrical form has the advantage of possessing great durability, but it is not so well fitted for rapid decomposition of the coal (on which depends much of the good qualities of the gas) as the elliptical shape. Retorts presenting a saddle-shaped section, when they are supposed to be cut transversely, appear to be preferable even to those of an elliptical form; but the greater expense of casting them, together with their inferiority in point of durability, has prevented them from coming into extensive use.

The temperature, best suited for the production of gas from coal, being what the workmen term a bright red, was found to be very destructive to the retorts, when they were exposed to the direct action of the fuel; and accordingly means were employed to protect them from the rapid oxidation which they suffered under these circumstances, by interposing between them and the furnace, a partition of fire tiles or arched bricks, with side flues for the admission of the heated air. This arrangement is represented in fig. 1, Plate CCLV., where \(a\) is the retort, \(g\) the grate of the furnace, \(e\) the fire-brick arch, and \(f\), \(f\) the side flues.

With the view of occupying less room, and saving the expense of fuel, several retorts are sometimes set together in an oven of brickwork, and heated by a smaller number of furnaces than there are of retorts. By this arrangement the fuel is certainly economized; but the plan is liable to the objection, that when any one of the retorts is worn out, those connected with it cannot be used till the faulty one is replaced; and though various expedients have been proposed for obviating that inconvenience, none of them can be said to have effectually answered the purpose. Fig. 2 and 3 represent varieties in the oven plan of setting retorts.

The fuel required for carbonizing a given quantity of coal may be stated to be, in general, about two fifths of its weight. It has been affirmed, indeed, that the decomposition has been effected at some establishments with less than twenty-five per cent.; but if the retorts are maintained at a proper temperature, and the carbonization is carried to a due extent, the fuel, however well economized, will rarely be less than thirty-five to forty per cent. of the coal it is employed to decompose.

The quantity of gas produced during the time the coal is undergoing decomposition is extremely variable. From one small retort, exposed for eighty-five minutes to a bright red heat, which was kept up with the utmost possible uniformity, the following results were obtained from eight pounds of the Wemyss coal:

| Cub. Ft. | Cub. In. | |---------|----------| | In 1st ten minutes | 6 | 235 | | 2d do | 8 | 980 | | 3d do | 8 | 1254 | | 4th do | 5 | 784 | | 5th do | 4 | 1450 | | 6th do | 3 | 313 | | Last twenty-five minutes | 6 | 1660 |

At the time the process was terminated the extraction of aciform matter had nearly ceased, so that the quantity of gas yielded by a pound of the coal was about five and a half cubic feet. The same coal carbonized on the large scale yielded, when the process was carried on for four hours, at the rate of four and one third cubic feet of gas per pound. The weight of the coke in the above experiment was 32,050 grains; and as the weight of the gas, the specific gravity of which was .65, must have been 15,026 grains, the tar and other residuary products, including the sulphuretted hydrogen abstracted by the process of purification, must have amounted to 8924 grams.

When the decomposition is effected on the large scale, the quantity of gas is found to vary with the quality of the coal, and the manner in which the operation is conducted. According to Mr Peckstone, a chaldron of Newcastle Wallsend coal yields 10,000 cubic feet, being at the rate of 370\(\frac{1}{3}\) cubic feet per hundredweight. At Edinburgh a hundredweight of cannel coal yields 430 feet and a similar return from coal of the same species is obtained at Glasgow, and other towns in Scotland.

The quality of the gas yielded by coal varies greatly at different periods of the carbonizing process. The first gas products, when the coal has not been previously well dried, consist almost entirely of aqueous vapour and carbonic acid; these are succeeded by light carburetted hydrogen, olefiant gas, and sulphuretted hydrogen, which gradually diminish in quantity till towards the close of the process, when almost the only products are carbonic oxide, and hydrogen. Hence if the process be carried on too long, the gases obtained in the latter stages of it will not only be useless for the purpose of yielding light, but the fuel employed for their production will be expended in wasting the retorts to produce substances which, we shall afterwards find, are calculated to impair the illuminating power of the gases with which they are mixed. In the case of cannel coal, the interval between the charges of the retorts should not exceed three and a half or four hours; nor in the case of the Newcastle coal, which is not so easily decomposed, ought that interval to extend beyond six hours.

**The Condensing Main and Dip Pipes.**

From the retorts the gas, after its production, ascends by bent pipes, called the dip-pipes, into what is termed the condensing main, which is a large cast-iron pipe, about twelve or fifteen inches in diameter, placed in a horizontal position, and supported by columns in front of the brickwork which contains the retorts. This part of a gas apparatus is intended to serve a twofold purpose: First, to condense the tar and grosser products of distillation; and, secondly, to allow each of the retorts to be charged singly without permitting the gas produced from the rest, at the time that operation is going on, to make its escape. To accomplish these objects, one end of the condensing main is closed by a flanch; and the other, where it is connected with the pipes for conducting the gas towards the tar vessel and purifying apparatus, has, crossing it, in the inside, a semi-flanch or partition, occupying the lower half of the area of the section, by which the condensing vessel is always kept half full of liquid matter.

The dip-pipe, represented by bedh, fig. 1, Plate CCLV., is connected by a flanch i, with a branch-pipe rising from the upper side of the condensing main; and as the lower end of it dips about two inches below the level of the liquid matter, it is evident that no gas can return and escape, when the mouth-piece m of the retort is removed, until it has forced the liquid matter over the bend cd; a result which is easily prevented by making dh of a suitable length. The upper part of the branch bc of the dip-pipe is generally furnished with a ground plug p, to allow the removal of the tarry matter, which is apt to accumulate, in a concrete state, at the lower part of the pipe where it is nearest the furnace.

Of the Tar Apparatus.

After emerging from the lower end of the dip-pipe, the gas, now bereft of a considerable portion of the vapour of water, tar, and oleaginous matter, which ascend with it from the retort, is conveyed by pipes, for the purpose of being completely freed from these impurities, into contrivances where a more perfect condensation takes place. As the subsequent purification of the gas depends, in no small degree, upon the perfect separation of the tar and other condensible products by which it is accompanied, the construction of the vessels best calculated for attaining that end, is a matter of the utmost importance; and indeed it may be justly affirmed, that unless that separation be effectually accomplished, the action of the chemical agents, to which the gas is afterwards exposed, must be limited and imperfect.

The first arrangements, employed for the purpose of condensation, were all constructed on the supposition that the object would be best attained by causing the gas to travel through a great extent of pipes, surrounded by cold water, and winding through it, like the worm of a still, or ascending upwards and downwards in a circuitous manner. One of these condensers is represented by figs. 4, 5, 6, and 7, Plate CCLV., where ABCD is the bottom of the condenser, and EGHF the top; IKLM and PQRO show the pipes in elevation. SKLT and SORT are sections of the vessel below the condenser, for the reception of the tar, and other condensible products. This vessel is divided by plates into a number of compartments which have no communication with each other, except at the bottom, for allowing the tar to be drawn off, by one discharging pipe abc. The gas is admitted at V, and after a repetition of ascents and descents, escapes at W, to be afterwards conveyed to the purifying apparatus.

A cylindrical pipe being the most capacious of all pipes having the same surface, an improvement upon the condenser as here described, was conceived to be gained by causing the gas to pass between plates of metal, forming a succession of narrow chests or galleries; an arrangement by which every portion of the gas being brought into more immediate contact with the sides of the condenser, it was concluded the condensation would be proportionally more complete. Such an apparatus is represented by figs. 8, 9, and 10, Plate CCLV., where ABDC represents the bottom of the condenser, E the entrance, and F the exit pipe of the gas, which passes between the plates, as indicated by the arrows. The bottom of the plates a, a, a have a slight inclination, so as to cause the condensed products to flow towards the openings a, b, c, where they are conveyed by pipes into the general receptacle below, and afterwards drawn off by the cock H.

These tar vessels, and others of a similar form to accomplish the same purpose, are constructed on the supposition that the tar and other condensible products are held in a state of suspension in the gas by heat; and that nothing more is necessary to produce their deposition, than exposure to a reduced temperature. As it appeared, however, that a large portion of the gas, after leaving the condensing main, was enveloped in thin films of tar and oleaginous matter, it occurred to the writer of this article that the disengagement of it, from the vesicles in which it was enclosed, might be more effectually promoted by interrupting it in its progress through the condenser, than by mere exposure to cold; and in conformity with this idea, he found that when the gas was forced to make its way through brushwood, placed in proper vessels for the purpose, the separation of the tar was more complete, than by the ordinary modes of condensation. ABCD, fig. 1, Plate CCLVI., represents a tar condenser constructed on that principle. abcd are large vessels, either of a square or cylindrical form, for containing the brushwood and loose materials through which the gas is made to pass; E being the pipe where it enters, and F the pipe where it escapes. ef, ef, pipes for conveying the condensed products into the horizontal pipe GH, which communicates with the tar receptacle M, by the bent pipe IK. The vessels containing the brushwood may be surrounded with cold water, or with the tar pumped up from the tar receptacle; but as the loose materials, through which the gas passes, seem to have the principal share in condensing the tar, it appears to be of little consequence whether the vessels containing it be surrounded with cold water or not, provided the apparatus be placed in a shaded situation. m, m, are man-hole doors for cleansing or removing the brushwood when it becomes clogged with tar; an operation which it may be necessary to perform once a year.

Of the purifying Apparatus for separating the Gases unfit for the purposes of Illumination.

With the two compounds of hydrogen and carbon, viz. Impurities olefiant gas, and light carburetted hydrogen, which are of coal-gas, yielded by coal during its destructive distillation by heat, several other products are obtained, which are not only useless for the purpose of illumination, but calculated to diminish the brilliancy of the light which is afforded by these gases, and even to prove a source of serious nuisance during their combustion. Among these products of a deleterious nature, are carbonic acid, and sulphuretted hydrogen; and, in smaller quantity, carbonic oxide, nitrogen, and hydrogen. The first two are by far the most objectionable of these impurities; and fortunately, their separation can be effected more easily than that of the others, whose presence is of less importance.

Carbonic acid is readily absorbed by any of the alkalies or earthy bodies, in a caustic state; and sulphuretted hydrogen, which possesses many of the properties of an acid, unites not only with the alkalies and alkaline earths, with which it forms a species of salts termed hydrosulphurates, but also with the metallic oxides, most of which it reduces.

The alkalies being too expensive to be used for separating carbonic acid and sulphuretted hydrogen from coal-gas, a more economical substitute, and which answers the purpose almost equally well, is found in quicklime. This purpose. Gas-Light: substance is accordingly used in every gas establishment on the large scale, in some form or another, for the last step of the purifying process to which coal-gas is submitted to render it fit for combustion. It is employed in two states; either in the condition of a thin paste, which the workmen call the cream of lime, or of a moistened powder, such as lime assumes when it is slaked with a little more than the usual quantity of water. The apparatus must therefore be accommodated, in its construction and arrangement, to these different conditions of the purifying material.

When the lime is used in a liquid state, the gas is made to pass through it so as to be as much as possible exposed to its action. One of the arrangements adopted for the purpose is represented by fig. 2, Plate CCLVI., where aa is an oblong close vessel, divided into a number of compartments formed by vertical partitions, which are fastened to the top and sides, but open at the bottom. The lower margin of the partitions is perforated with numerous small holes through which the gas passes from one compartment to another; the cream of lime being introduced in such quantity that its surface is above the level of the uppermost row of holes. The gas enters by the pipe c, and, after bubbling through the holes in the partitions, from cell to cell, makes its escape in a purified state by the pipe d. The liquid line is introduced from time to time at b, and run off, after being unfit for the purpose of purification, by the pipe e.

It being highly conducive to the success of the purifying process that a succession of fresh portions of the liquid lime should be brought in contact with the gas as it passes through it, the material is kept, at some establishments, in a state of constant agitation by means of machinery for the purpose. Fig. 3 represents an arrangement of that kind which is used at some of the London gas-works. aaa is a flat cylindrical vessel in which the purification is performed, the gas entering by the pipe c and escaping by the pipe d. In the inside of this vessel is another of a similar form ff, terminating at the bottom in a broad horizontal flanch or circular plate gg, and having revolving within it an agitator hh, which works in an air-tight stuffing box. The gas being forced by the pressure from the retorts to descend below g, is more effectually exposed in its progress to the action of the lime by the commotion produced by the agitator, and finally ascends, in a purified state, through the pipe d. Fresh portions of the purifying material are supplied from the vessel k by means of the connecting pipe b; and the cream of lime, after being saturated with the impurities, is withdrawn at e. In large establishments the gas is forced in succession through several vessels of the kind described, the cream of lime being changed in each of them at different times, to render its action more uniform and regular.

To supersede the necessity of using a series of purifiers, the following arrangement has been adopted, which seems well calculated to insure great regularity of action, with a simplification in the means of attaining it: aa, bb, fig. 4, is a cylindrical vessel, placed in a slanting position, and having a number of internal partitions h, h, rising nearly as high as its axis, in which there works a spindle kk, carrying round with it a number of arms of unequal lengths l, l. These arms act as agitators of the liquid lime contained in the different cells, formed by the partitions h, h. The cylinder aa, bb terminates at the upper extremity in a vessel aa, ff, which is surmounted by another vessel ooo, in which the cream of lime is prepared. Through the opening e a vertical axis descends, giving motion to the inclining one k by means of two bevelled wheels. To the vertical spindle ii are attached an inverted cup m, operating as a valve; and the arms n, n for agitating the cream of lime before it is introduced into the vessel aa, ff. The gas is introduced at c, and escapes at d; and the cream of lime no longer fit for the purpose of purification is removed by the bent pipe g. The gas, ascending upwards in this machine, has to pass in its progress among all the arms of the inclined axis; and being thus exposed to the perpetual spray of the liquid lime, the supply of which is constantly and regularly renewed, it finally escapes at d in a purified state.

One of the objections against the method of purifying by the cream of lime, or lime in a liquid state, is, that unless the gas be previously freed entirely from tar, that substance, enveloping it with a thin film of oleaginous matter, which has little tendency to unite with water, carries the gas along with it in rolling bubbles, so that the internal parts of it can thus scarcely ever come into contact with the purifying materials. In some of the arrangements, indeed, which we have described, mechanical contrivances are employed to agitate and disperse the gas, with the view of exposing every portion of it, more or less, to the action of the lime; but these modes of promoting the efficacy of the process cannot be resorted to without the aid of some moving power, which, in many cases must necessarily be attended with considerable trouble as well as additional expense. There is another objection to which this method of purification, even if it required not the assistance of machinery, must always be liable; namely, that the olefiant gas upon which the illuminating power mainly depends, is largely absorbed by water, insomuch that either oil or coal gas, standing a few days over that fluid, suffers a great deterioration of its quality, and becomes in every respect less fit for the purposes of illumination. When lime is used in the dry state, or rather in the state of a moistened powder, for purifying coal-gas, neither of these objections is at all applicable; and when the arrangements for that mode of purification are contrived with a due regard to the simplicity and convenience of the manipulations, the separation of the useless and noxious gases is effected more easily, and not less effectually, than by the method of liquid lime. The abstraction of the sulphuretted hydrogen becomes more perfect, by adding to the lime a small portion of the peroxide of manganese, which being a cheap substance, adds very little to the expense of the process. This method of purifying coal-gas having been adopted with the utmost success in many gas establishments, we shall now proceed to explain the nature of the apparatus by which it is carried into effect.

Figs. 1 and 2, Plate CCLVII., represent the elevation and plan of an apparatus for dry lime, with the valves for regulating the direction of the gas when the purifying materials require to be changed. PO and QR represent double chests, formed of plates of cast iron, and which are thus capable of being water-luted at mn, by means of an inverted vessel of the same form abcd, and which is made of sheet iron. This vessel, when the purifying process is going on, is kept down by means of cross bars g, g; and it is furnished with a stop-cock k, for allowing the common air to escape from it when it is let down into its position, and also for allowing the admission of that air, when it must be raised for the purpose of changing the purifying materials, without which the atmospheric pressure would render it extremely difficult to remove the vessel. x, x are bolt-eyes for hooking the tackle which is employed to lift the inverted chest abcd. The pipe T by which the gas is introduced rises above the level of the lime contained in the uppermost sieve, so that the gas, after issuing from it, spreads itself as pointed out by the arrows, and descends downwards through the several layers of lime o, o, till it reaches the bottom of the chest, where it makes its escape at V, to pass, if necessary, through a similar process in another chest. h, h represent sections of the valves, which are luted with mercury. Gasometers for receiving and containing the Gas before it is consumed.

As many disadvantages would be experienced by attempting to adjust the production of the gas to the rate of its consumption, it is found to be more convenient, as well as more economical, to store up such a portion of it during the day, as shall compensate for the deficiency of the supply that may be furnished during the time the gas is consuming, in the course of the evening. The capacity of the vessels used for this purpose, which are called gasometers, must be regulated by a regard to that consideration.

The form of the gasometer is generally that of an inverted cylindrical cup, the diameter of which, when economy is studied, ought to be double of its depth, or at least not more than two or three inches less. Gasometers are usually composed of sheet iron, varying in weight from two to three lbs. to the square foot, well riveted at the joints, and kept in shape by means of stays and braces formed of cast or bar iron. The sheet iron is made to overlap at the joints, a slip of canvass well besmeared with white-lead being interposed to secure perfect tightness. The prismatic shape, though occasionally adopted, is not so convenient as the cylindrical, partly on account of the difficulty of making it retain its form, and partly on account of the greater quantity of material, compared with the capacity, that is necessary for its construction.

The gasometer is furnished with a tank, of the same form Tank of itself, but a little larger in dimensions, for containing gasometer, the water, in which it is suspended at different altitudes, by means of a chain and counterpoise, moving over pulleys. The tank is sometimes built of stone, but more frequently it is constructed of cast-iron plates bolted together by flanges, with an interval between them of about three eighths of an inch, which is afterwards filled up with iron cement.

A gasometer, such as has been described, is represented Counter by fig. I, Plate CCLVIII., in which aaa is the gasometer, poise and dd the suspension chain, ee the pulleys, and e the counter-chain, poise; also bb is the tank, and ff the pipes for introducing, and conducting off the gas.

As the gasometer, when it is immersed in the water of the tank, suffers a loss of weight equal to that of the portion of fluid it displaces, it is evident that unless some arrangement be made to counteract the varying pressure which must thus result from the different depths to which it may be immersed, the gas contained in the gasometer will be expelled, at different times, with a varying force. If, however, the weight of the chain of suspension, or rather the weight of that portion of it whose length is the same as the height to which the gasometer ascends, be equal to half the loss of weight which the gasometer sustains by immersion in water, a perfect compensation will be made, and an equilibrium will hold between the gasometer and its counterpoise, at all altitudes. Thus, if the weight of the gasometer were five tons, or 11,200 lbs., and it lost by immersion a seventh part of its weight, or 1600 lbs., then the weight of that portion of the chain, equal in length to the highest ascent of the gasometer, would require to be 800 lbs., and the weight of the counterpoise 11,200 — 800, or 10,400 lbs.

For, the gasometer being immersed, its virtual weight is 11,200 — 1600, or ........................................... 9,600

Weight of portion of chain now acting with the gasometer ......................................................... 800

Sum is the weight of counterpoise ........................................... 10,400

Again,

The gasometer being elevated out of the water, its weight is .......................................................... 11,200

Weight of chain now acting in opposition to it ......................................................... 800

Difference is the weight of counterpoise ........................................... 10,400

Though the compensation, by this adjustment of the weight of the chain, answers the purpose in the most effectual manner, the following method is by some deemed preferable. Let the counterpoise (instead of having the form as shown at e in the figure) consist of a long cylindrical or

---

1 The following iron cement is recommended by Peckstone: Take iron turnings or borings, and pound them in a mortar till they are small enough to pass through a fine sieve; then, with one pound of these borings, so prepared, mix two ounces of sal ammoniac in powder, and one ounce of flowers of sulphur, by rubbing them well together in a mortar; and afterwards keep the mixture dry till it may be wanted for use. When it is so, for every part thereof, by measure, take twenty parts of iron borings, prepared as above mentioned, and mix them well together in a mortar or other iron vessel. The compound is to be brought to a proper consistency by pouring water gently over it as it is mixing; and when used it must be applied between the flanges by means of a blunted caulking iron. Gas-Light prismatic body, as represented by the dots \( g, g \), having the area of its horizontal section equal to the area of a similar section of the plates of the gasometer, and be allowed to descend into the water as the gasometer rises out of it. Also let the chain \( d \) be of a weight equal (length for length) to a column of water of equal bulk with the counterpoise. Then, if the weight of the gasometer be, as already supposed, 11,200 lbs., the weight of the counterpoise must be the same; but the weight of that portion of the chain which, by the above arrangement, was only equal to half the loss of weight sustained by the gasometer when immersed, must now be equal to the whole of that weight.

Then, the weight of the gasometer in the water is,

as before ........................................................................... 9,600

Weight of the chain now acting with the gasometer, 1,600

Weight of counterpoise, now out of the water .......... 11,200

Again,

The weight of the gasometer, out of the water, is .... 11,200

Weight of the chain, now acting in opposition to the gasometer .................................................. 1,600

Weight of the counterpoise, in water ......................... 9,600

Though we have only shown, in both these modes of compensation, that an equilibrium between the gasometer and its counterpoise holds, in the extreme cases, it would be easy to prove that the same thing must subsist, at all the intermediate elevations of the gasometer. At the same time it must be obvious that these contrivances, however well calculated they may be to secure the equilibrium alluded to, can have no effect in expelling the gas; and therefore, when it is wished that the contents of the gasometer shall issue from it under a certain pressure, the weight of the counterpoise must be diminished to a suitable extent. Thus, if it were required that the pressure employed for expelling the gas should be equal to that produced by a column of water three fourths of an inch deep, then it would be necessary to diminish the weight of the counterpoise by the weight of a column of water having the same diameter with the gasometer, and an altitude of three fourths of an inch. If the diameter of the gasometer above mentioned were, for example, 35 feet or 420 inches, the weight of a cylindrical portion of water having that diameter, and a depth of three fourths of an inch, would be, in grains,

\[ 420^2 \times 7854 \times \frac{3}{4} \times 252.5 = 26236876 \text{ grs. or } 3748 \text{ lbs.} \]

Hence it would be necessary to make the counterpoise 3748 lbs. lighter than it was supposed to be, according to the above-mentioned arrangements, in order that the gas might issue from the gasometer, under a pressure of three fourths of an inch of water. If the calculation were conducted with extreme accuracy, the specific gravity of the gas ought also to be taken under consideration; but the object to be attained is not of so delicate a nature as to require an attention to such minute circumstances. Besides, we shall afterwards find that the value of the arrangements we have described for obtaining an uniform and equable pressure is greatly diminished; and these are even entirely superseded by a contrivance called the governor, to be afterwards explained.

In situations where a sufficient depth of the tank cannot be easily obtained, an expedient may be resorted to, such as is represented in fig. 2. The gasometer consists of two parts, separable from each other; the one, \( abba \), having the form of the common gasometer, and the other, \( cccc \), being open at the top as well as the bottom, but connected with the other by a channel running around the bottom, which is filled with water. The lower part has a recurved form at the top; and this portion of it entering into the channel, becomes water-luted, by which the entire gasometer is rendered air-tight, at the line of junction. Various other contrivances have been proposed for saving room; but most of them answer the purpose so imperfectly, that they are scarcely deserving of being noticed. We shall, however, give a description of one of them, invented by M. Clegg, in which the gas is expelled by the partial revolution of the gasometer.

This kind of gasometer is represented in fig. 3, \( a \) being the cistern filled with water to the height of the dotted line; \( b \) the axle, which is hollow at each end, and works on friction sectors; \( ddd \), a vessel supported by arms proceeding from the axle, and formed of parts of two concentric cylinders, shut at each end, and also at \( g \), except where the entrance and exit gas-pipes passing from \( g \) to the hollow axis are connected (one of these only is shown in the figure, the other being directly behind it); the end \( k \) is left open, and when the gasometer is filled with gas it is just immersed in the water; \( i \) a pulley to which is attached a chain and weight \( h \), disposed as represented. The whole apparatus is constructed so as to be an equilibrium in every position, the framing being made heavy at that part of the circle to which the gas-holder does not extend, so as to counterbalance the matter opposed to it. The gas enters at one of the hollow ends of the axis, and passes through one of the tubes \( g \) into the gasometer; and it is discharged under any required pressure, obtained by means of the weight \( h \), through the other tube behind \( g \), into the opposite extremity of the axis. Though this kind of gasometer is more expensive than one of equal capacity of the common form, yet requiring a shallow cistern for working in, it may be resorted to in some cases with advantage.

The gas is commonly introduced and conducted off by pipes, passing under the bottom of the tank for containing the water, in which the gasometer floats; and as these pipes are in many cases considerably below the level of the street pipes with which they communicate, they are apt to be filled up in the course of time with the condensed water which passes off in a vaporous state with the gas. To remedy this inconvenience, it is necessary to place vessels for receiving that water, in connection with the entrance and exit pipes, so contrived that the accumulated water may be easily removed from them when required. One of these vessels, which are improperly termed syphons, is represented by fig. 9. When the water rises to the height \( d, e \), it is drawn off by means of a small suction pump, introduced into the pipe \( ae \), which reaches within about an inch of the bottom of the syphon. To prevent too much water being removed, and thus allowing the gas to escape at \( ae \), the lower end of the pipe of the suction pump should be perforated with small holes to as great a distance as the water is to be left in depth; or, if deemed preferable, the upper end of the pipe \( ae \) may be rendered air-tight by a screwed cap.

When access to the bottom of the tank is difficult, swivel or flexible-jointed tubes, disposed so as to rise and fall with the gasometer, are sometimes resorted to. Jointed tubes, connected by means of water-lutes, are also occasionally employed, and they are found to answer the purpose very well in situations where the water is not exposed to frost. One of these is represented in fig. 4, \( a \) being the pipe by which the gas is introduced, \( b \) a vertical pipe admitting of a small angular motion on the axle or support \( c \), and connected with \( a \) by means of a water-lute joint. The upper end of \( b \) is united in a similar manner with one end of the pipe \( d \), which is connected at the opposite end by means of a like entrance with the extremity of the pipe \( f \), attached to the top of the gasometer; \( g \) is a regulating radius bar, centred to a bracket, which is fastened to the pipe \( d \), and working on a fixed pivot at \( i \). The gasometer, as it rises and falls, carries \( f \) along with it, and Of the Main and Service Pipes.

The gas being duly purified and prepared for combustion, the next point to be considered is the transmission of it from the gasometer to the various places where it is to be consumed. As it must sometimes be conveyed, particularly in the case of large establishments, to the distance of several miles, it is evident that unless the diameters of the various pipes through which it is to be conducted have a due relation to the quantity of gas to be transmitted, there will be a danger either of incurring an unnecessary expense, by making the pipes too large; or, what is still worse, of being exposed to a deficiency of supply, by making them too small. The first object, therefore, to be ascertained by the engineer, is the probable number of lights that may be required in the various streets and lanes in which these pipes are to be laid; and these being known, the corresponding quantity of gas, according to the quality of it, may be afterwards computed. With regard to the relative dimensions of the pipes at different distances from the gaswork, the only general rule to be observed is, that the sum of the areas of the sections of the main-pipes proceeding immediately from the gasometer, should be equal to the sum of the areas of the sections of the various branch-pipes which they supply with gas; and this rule, with some little modification, should be followed in the case of the subordinate ramifications.

In the case of good coal-gas, we may safely reckon that one fourth of a cubic foot of it will furnish the light of a moulded candle for an hour, of which one pound will, when the candles are burnt in succession, last forty hours. On this supposition, and assuming that the pressure upon the gas in the gasometer is equal to three fourths of an inch of water, the diameters of pipes necessary for conveying various quantities of gas may be stated as follows:

| Diameter of Pipe in Inches | Quantity of Gas in Cubic Feet per Hour | Equivalent Number of Candles | |---------------------------|--------------------------------------|----------------------------| | 1 | 4 | 16 | | 2 | 20 | 80 | | 3 | 50 | 200 | | 4 | 90 | 360 | | 5 | 380 | 1,520 | | 6 | 880 | 3,520 | | 7 | 1,580 | 6,320 | | 8 | 2,480 | 9,920 | | 9 | 3,580 | 14,320 | | 10 | 4,880 | 19,520 | | | 6,380 | 25,520 | | | 8,090 | 32,360 | | | 10,000 | 40,000 |

This table has been deduced partly from theoretical considerations, and partly from the results of experiment. Peckstone affirms, in his work on Gas-lighting, that a pipe ten inches in diameter is capable of transmitting 50,000 cubic feet of gas per hour, under a pressure of one inch of water, while, according to the statement of Mr Creighton, such a pipe would scarcely convey the tenth part of that quantity, under a pressure of from four eighths to three fourths of an inch of water. It is impossible to reconcile these discordant statements, either by an allowance for the difference of pressure, or the difference of the specific gravities of the gases; for it ought to be kept in view that the discharge of gas is directly proportional to the square root of the height of the column of water by which it is pressed, and inversely as the square root of the specific gravity of the gas. Both of these propositions, however, must be greatly modified by friction, and consequently by the lengths of the pipes through which the gas is conveyed. In the supply stated to be furnished by pipes of different dimensions, we have deemed it safest rather to underrate the quantity, than overrate it.

The main-pipes are usually made of cast iron, joined together with socket joints, in lengths of three yards. The depths of the sockets vary in pipes of different sizes, from three to six inches, part of them being fitted with gasket to bring the centres of the pipes into line, and the remainder lead after the gasket has been driven home with suitable chisels or caulking irons. The depth of lead to secure a good joint, should not be less than an inch and a half; the interval between the spigot and socket being from three eighths to seven eighths of an inch, according to the diameter of the pipe.

As a considerable quantity of water is carried off by the Necessity gas in the state of vapour, which is afterwards condensed of giving in the pipes, some arrangement must be made for its collection and occasional removal; and accordingly, in laying the pipes, care must be taken to give them a regularity of declivity towards one or more points, where proper syphons, close vessels, and cocks must be placed, to receive and discharge the collected water. When these precautions are neglected, or when the levels are inaccurately taken, much annoyance is experienced; and as the evil can only be corrected by lifting and rejoining the pipes, the utmost attention should be paid to guard against it at first.

For stopping off the gas in large pipes, valves fitted with water or mercury, according to the degree of pressure to be resisted, are usually employed. One of the most simple forms of these valves is represented in fig. 8, Plate CCLVIII., where aa is the valve, having the shape of an inverted cylindrical cup, which is raised and depressed by a rod c passing through an air-tight stuffing-box; and bb an annular cavity or interval, partially filled with water or mercury, into which the valve descends when it is shut.

To convey the gas from the main-pipes, and distribute it through the various apartments of dwelling-houses, pipes made of block-tin are generally used; these being more durable and better adapted to the purpose than pipes composed of copper or any other metal. In arranging the interior fittings, the same precautions must be observed as were recommended in the case of the main-pipes, viz. to give the various branches a due degree of inclination, so as to cause all the condensed water to flow to one or more points, where proper cocks must be placed for its removal. Unless this be done, the lights will be apt to flicker, or be extinguished at times altogether. Nor is it of trivial moment to enjoin the workmen, when they are soldering the service-pipes, to avoid with the utmost care allowing any of the melted metal to find its way into the inside of the pipes; it being in a great measure to this circumstance, that the deficiency in the supply of gas, so frequently complained of, is chiefly owing.

Of the Governor or Regulator.

The quantity of gas consumed in large towns varying greatly at different times, it is evidently a matter of some importance to the public, as well as to the manufacturers of gas, that the supply of it should be duly adjusted to the consumption; so that when the lamps are once regulated to a proper height of flame, they may continue afterwards to burn with the same steady light throughout the whole of the evening. Any contrivance that can accomplish so desirable an object must save a great deal of trouble to the consumer of gas, and much unnecessary waste of it to the manufacturer; and such is the design of the governor or regulator. Fig. 5, Plate CCLVI., represents one of these con- Gas-Light-trivances, \(a\) being the pipe proceeding from the gasometer, by which the gas is admitted, and \(f\) the pipe by which it escapes; \(e\) is a valve of a conical form, fitted to the seat \(b\), raised and depressed by means of the rod \(d_g\), to which it is attached; \(ddd\) is an inverted conical vessel, formed of sheet iron, which ascends and descends in the exterior vessel \(ee\), in which water is contained to the level represented. The gas, entering at \(a\), passes through the valve, fills the upper part of the inverted vessel \(ddd\), which it thus partially raises, and escapes at \(f\). If the pressure from the gasometer be unduly increased or diminished, the buoyancy of \(ddd\) will be increased or diminished in like proportion, and the valve being by this means more or less closed, the quantity of gas escaping at \(f\) will be unaltered. And not only will the governor accommodate itself to the varying pressure of the gasometer, but also to the varying quantities of gas required to escape at \(f\) for the supply of the burners. Thus, if it were necessary that less gas should pass through \(f\), in consequence of the extinction of a portion of the lights, the increased pressure which would thus be produced at the gasometer would raise the governor, and partially shut the valve, till the state of it was duly adapted to the requisite supply of gas.

Fig. 6 is another contrivance for the same purpose, which acts in a similar manner. The gas enters at \(a\), and escapes at \(f\). The valve \(e\), with its stem, is attached to the inverted vessel \(d\), which is suspended in a cistern of water \(gg\), and moves upon a pivot at \(e\). The action of this apparatus is the same as that of the contrivance we have already described. Both of them must be regulated by weights applied to the vessels \(dd\); but these being once accommodated to the supply of gas, no further adjustment will afterwards be necessary.

The Gas-Meter.

The gas-meter is a simple but ingenious mechanical contrivance, the design of which is to measure and record the quantity of gas passing through a pipe in any given interval of time. The expense of these machines has hitherto formed the principal objection against their use; but already they are largely employed by several gas companies in England as well as in Scotland. They will probably be found, upon trial, to be no less advantageous to the consumer than to the manufacturer of gas, by allowing the former to use gas without any unnecessary waste of it, and securing to the latter a fair and regular price for the quantity of it actually consumed.

Fig. 7, Plate CCLVI., exhibits sections of one of these gas-meters, where \(ee\) represents the outside case, having the form of a flat cylinder; \(a\) is a tube which enters at the centre for admitting the gas, and \(b\) is another for conveying it off to the burners; \(g, g\) are two pivots, one supported by the tube \(a\), and the other by an external water-tight cup, projecting from the outside casing, and in which is contained a toothed wheel \(h\), fixed upon the pivot, and connected with a train of wheel-work (not shown in the figure) to register its revolutions. The pivots are fixed to and support a cylindrical drum-shaped vessel \(ddd\), having openings \(e, e, e, e\), internal partitions \(ef, ef, ef, ef\), and a centre piece \(ffff\), as represented in the figure. The machine is filled with water, poured in at \(k\) up to the level of \(i\), and gas being admitted under a small pressure at \(a\), it enters into the upper part of the centre piece, and forces its way through such of the openings \(f\) as are from time to time above the surface of the water. By its action upon the partition nearest in contact with the water (to the right hand of the figure, in which all the partitions or openings are shown), a rotatory motion is communicated to the cylinder; the gas from the opposite chamber being at the same time expelled by one of the openings \(e\), and afterwards escaping at \(b\), as already mentioned.

As the quantity of gas which passes through the machine in any given time depends not only upon its internal dimensions and the number of revolutions which it performs, but also upon the level of the surface of the water in which the cylinder revolves, due care must be taken to maintain the water at the same level, for the regular action of the meter. This is easily accomplished, by pouring in water when necessary, till the superfluous quantity is discharged by an orifice properly placed for the purpose.

Burners.

The most economical mode of consuming gas, so as to obtain from a given volume of it the greatest possible quantity of light, both in degree and duration, is a problem of no less importance than that of the most suitable arrangements for its production and purification. The presence of oxygen, in some form or another, being essentially necessary to produce ordinary combustion, it follows, that from whatever cause that principle may be deficient in quantity, the combustion must be imperfect; and when this is the case, it appears that the light yielded by the combustible body is also diminished in a proportional degree. On the other hand, if the quantity of oxygen brought into contact with the combustible body be more than sufficient for its entire combustion, the superfluous quantity of that gas, instead of augmenting the effect, can only lower the temperature, and diminish, it may be presumed, in a corresponding degree, the intensity of the light. This must be the consequence if the brilliancy of the light yielded by a combustible body depends at all upon the temperature to which it is exposed during its combustion; and that this is the case may be inferred from the simple fact of causing the flame of a jet of gas to play, first against a sheet of ice, and then against a bar of red-hot iron, when the difference of the light will be such as to leave no doubt of the influence of temperature upon its intensity. A similar result is obtained by bringing the flames of two separate jets into contact, when an obvious increase of light is perceived. From these simple facts it may be inferred, that though a certain quantity of common air must be brought into contact with the inflamed gas to produce the greatest intensity of light, whatever exceeds that quantity will not only be useless, but, by diminishing the temperature of the flame, must tend to impair the brilliancy of its light.

But though the immediate cause of the light is probably the high temperature to which the carbonaceous portion of the gas is exposed, the condition in which the carbon exists, at the time it is so exposed, is of the utmost importance to the effect. According to the opinion of Sir Humphry Davy, as adopted by Drs Christison and Turner, "a white light is emitted only by those gases which contain an element of so fixed a nature as not to be volatilizable by the heat caused during the combustion of the gas; and that in coal-gas this fixed element is charcoal, formed by the gas undergoing decomposition before it is burnt. The white light is caused by the charcoal passing into a state, first of ignition, and then of combustion. Consequently no white light can be produced by coal or oil gas, without previous decomposition of the gas."

"That the gas undergoes decomposition before it burns, and that the carbonaceous matter is burnt in the white part of the flame in the form of charcoal, is shown by

---

1 The regulator, which was originally invented by Mr Clegg, was afterwards improved by Mr Crosley, who enjoys, by patent, the exclusive right of furnishing it. The gas-meter was also invented by Mr Clegg, and is constructed with great accuracy by Mr Crosley. placing a piece of wire-gauze horizontally across the white part of the flame, when a large quantity of charcoal will be seen to escape from it unburnt. And that this previous change is necessary to the production of a brilliant white light, will appear, if we consider the kind of flame which is produced when decomposition does not previously take place. For example, if the gauze be brought down into the blue part, which always forms the base of the flame, no charcoal will be found to escape. Or, if the gauze be held at some distance above the burner, and the gas be kindled not below but above it, by which arrangement the air and the gas are well mixed previous to combustion, the flame is blue, and gives hardly any light. The reason is obviously, that in both cases the air is at once supplied in such quantity in proportion to the gas, that the first effect of the heat is to burn the gas, not to decompose it." (Edin. Phil. Journ. No. XXV.)

To these statements it may be added, that if a jet of oil or coal gas, burning with a fine yellow flame in common air, be suddenly surrounded with an atmosphere of oxygen gas, the colour instantly changes into a pale blue, yielding the most feeble light; nor does the flame recover its brilliancy till the oxygen is largely diluted with carbonic acid, when it burns for a short time with greater splendour than at first. For though the light is greatly enchelbed when the combustion of the gas takes place in pure oxygen, it becomes much more vivid when the combustion is carried on in air, which is more largely charged with oxygen than common air. Hence the brilliancy of the light appears to depend upon two conditions: 1st, The perfect combustion of a portion of the gas, in an undecomposed state; 2dly, the temperature produced by that combustion upon the residual part, in a decomposed state. When a large portion of the gas is consumed in the first condition, the temperature is higher; but the undecomposed part is then too small in quantity to yield an intense light, in consequence of the attenuated state of the carbon; and, on the other hand, when a small portion of the gas is consumed in the undecomposed state, the temperature produced is too feeble to raise the temperature of the now partially decomposed part to a sufficient pitch for the full ignition of the carbon.

The conditions which thus seem to be necessary for obtaining the greatest portion of light from the combustion of a given quantity of gas, while they are perfectly consistent with the most anomalous facts presented by that process, so they appear to afford the only sure principles upon which we can proceed in the construction of gas-burners. One of the most obvious conclusions deducible from these principles is, that whatever be the form of the gas-burner, its construction should be such, that while it admits as much air, it should never admit more than is barely sufficient for the perfect combustion of the gas.

According to the experiments of Drs Christison and Turner, the character best fitted for single-jet burners appears to be about one twenty-eighth of an inch for coal-gas, and one forty-fifth for oil-gas. As these dimensions, however, must vary with the quality of the gas, we consider one thirty-sixth of an inch to be more applicable to the gas obtained from cannel coal, if its specific gravity be not less than .65. Every form of burner composed of separate jets, in which the gas is made to issue in a horizontal or oblique direction, gives a consumption which increases in a much faster ratio than the light which it yields; and consequently, however beautiful such burners may be in appearance, they are far from being economical.

One of the most useful forms of a burner with single jets, is where there are two holes, and their directions are so inclined as to cause the streams of issuing gas to cross, and exhibit during their combustion a broad continuous flame. This burner, which is termed a swallow-tail, is well adapted for street-lights, as it gives a powerful light, and consumes a small quantity of gas. When the gas is emitted by a narrow slit at the top of the burner, the burner receives the name of a bat-wing. Fig. 11, Plate CCLVIII.

But of all the forms of the burner, that upon the Argand principle, in which the holes are arranged in a circle, so as to admit the air to have access to the flame, externally as well as internally, is the most economical, and the best calculated to secure the complete combustion of the gas. The diameter of the holes should, in this burner, be about the fortieth part of an inch for coal-gas of an ordinary good quality, and the distance between them should be such as to allow the separate flame of the different jets to unite together and form a continuous hollow cylinder of light. Fig. 12.

The construction of burners, and the most economical mode of consuming gas, having been examined with much philosophical precision by Drs Christison and Turner, we shall extract from their elaborate dissertation on the subject, the most valuable and important conclusions which these learned individuals deduced from their experiments; and this we do with greater confidence, because the results they obtained coincide very exactly with those we procured, when engaged in the same inquiry. The three leading points to which they directed their attention were, 1st, the length of flame most suitable for different burners; 2dly, the form, magnitude, and position of the orifices through which the gas is discharged; and, 3dly, the modifications of the light produced by the glass chimney of the Argand burner.

With regard to the length of flame which afforded the greatest light compared with the expenditure of gas, they found that, in the case of the jet, the best length for coal-gas was about five inches, and for oil-gas about four inches. When the flame was kept shorter, the quantity of gas consumed was greater in comparison of the light which it yielded; but no advantage was gained by increasing the length beyond that mentioned as the most suitable for each gas; the combustion becoming less perfect, and beginning to be accompanied with the escape of the carbon in the form of smoke. Thus they found that, in the case of coal-gas having the specific gravity .602, while the lights emitted from a two-inch and a five-inch flame were as 556 to 1978, the corresponding expenditures were to each other as 605 to 1437. But the light, in an economical point of view, must be estimated inversely as the quantity of gas from which it is obtained; and hence the ratio of the lights, in reference to the expenditure, was as 546 to 1978, being as 100 to 150.

In the case of Argand burners, the augmentation of the light, in a ratio greater than the expenditure, was exemplified in a still more remarkable degree. Thus the following results were obtained with coal-gas of the specific gravity .605, by elevating the flame of a five-holed burner, successively from half an inch to five inches.

| Length of Flame | Half-Inch | One-Inch | Two-Inch | Three-Inch | Four-Inch | Five-Inch | |-----------------|-----------|----------|----------|------------|-----------|----------| | Light | 18·4 | 92·5 | 259·9 | 308·9 | 332·4 | 425·7 | | Expenditure | 89·7 | 148 | 203·3 | 241·4 | 265·7 | 318·1 | | Ratio of light to expenditure | 100 | 282 | 560 | 582 | 582 | 604 |

\[ \text{Specific Gravity} = 0.605 \] Gas-Light. Hence the light is increased about six times for the same expenditure, by raising the flame from half an inch to three or four inches; but very little is gained by any additional increase of the flame beyond that length, in the description of burners with which the experiments were made.

These facts receive a satisfactory explanation from the general principles which we have already laid down with respect to the combustion of the luminiferous gases. When the flame is short, the supply of oxygen for the combustion is too great; almost the whole of the gas is thus consumed before any portion of it can undergo the decomposition which is necessary for the evolution of light; while the temperature of the flame being reduced by the superfluous air which brushes along its surface, the intensity of ignition, and with it the splendour of the light, is proportionally diminished. This explanation is well illustrated by partially shutting the central part of the burner, and thus interrupting the supply of air to the internal surface of the flame; the moment this is done, the length of the flame is increased, and a visible improvement of the light takes place, thus indicating that more air was previously brought in contact with the gas than was requisite for its perfect combustion.

The second point to which Drs Christison and Turner directed their attention was the construction of the burner itself, particularly the magnitude and position of the orifices at which the gas is emitted during its combustion. The same principles which explained the relation between the light and the expenditure, in the case of flames of different lengths, suggested the rule for regulating the dimensions of the orifices; and accordingly they justly inferred that, in a single jet, the diameter of the aperture ought to be such as to ensure the complete combustion of the gas, without rendering it more vivid than is necessary for that effect. If the orifice be too small, the greater portion of the gas is liable to be consumed without suffering a previous decomposition, and thus the light is extremely feeble; and, on the other hand, if the orifice be too large, the surface of the flame exposed to the action of the air being too small in comparison of the discharge of gas, the combustion is imperfect, and the carbon, after being separated from the hydrogen, either burns at a low temperature with a dusky flame; or, what is still worse, a large portion of it passes off in the state of smoke. In conformity with these views, they recommend, as we have already stated, a twenty-eighth of an inch for coal-gas, and a forty-fifth for oil-gas, as the most suitable dimensions for single jets. They acknowledge, however, that their experiments with coal-gas were too limited to justify them in using very confident language on the subject; and we have therefore the less hesitation in stating that we consider an orifice, varying in diameter from a thirty-second to a thirty-sixth of an inch, as better adapted to coal-gas of a specific gravity between .62 and .70.

In Argand burners the diameter of the orifices ought to be a little smaller. Drs Christison and Turner state that the diameter which appeared to answer best for coal-gas of the specific gravity .6, when the holes are ten in a circle of three tenths of an inch radius, was a thirty-second of an inch. We consider this, however, to be too great for coal-gas of a better quality, and would recommend, in preference, apertures varying in diameter from a thirty-sixth to a fortieth of an inch.

The distance between the jet-holes of Argand burners is a matter of no less importance than the diameter of the orifices, and must be regulated by the same principles. When they are so far asunder that the flames of the separate jets do not coalesce, no advantage is derived from the Argand form; but when they unite, and compose an uniform and unbroken surface of flame, the light is considerably greater, compared with the expenditure of gas, than is obtained from detached jets. In order to determine the most suitable distance at which the orifices of Argand burners should be placed, Drs Christison and Turner took a burner six tenths of an inch in diameter, which they caused to be drilled with eight, ten, fifteen, twenty, and twenty-five holes, a fiftieth of an inch in diameter; and having determined, with each of these burners, the light and expenditure, in the case of oil-gas, they obtained the following results:

| Burners | VIII | X | XV | XX | XXV | |---------|-----|---|----|----|-----| | Light | 360 | 360| 391| 409| 382 | | Expenditure | 367 | 318| 296| 289| 275 | | Ratio of light to expenditure | 98 | 113| 132| 141| 139 |

As the standard of comparison was a single jet, burning with a four-inch flame, the ratio of the light yielded by which to the expenditure was expressed by 100, it was inferred that no advantage is gained by giving the jets the Argand arrangement with a burner of the dimensions above mentioned, if the holes are only eight in number; and that the gain does not increase after the number reaches to twenty. In the former case the distance of the holes must have been .2356 inch, or nearly one fourth of an inch, and, in the latter, .0945; so that the most advantageous distance for jet-holes of a fiftieth of an inch in diameter would seem to be about \(\frac{1}{15}\) of an inch.

For coal-gas burners, however, the distance between the jet-holes ought to be increased in a ratio varying inversely with the quality of the gas, or directly as the diameters of the orifices themselves. Hence, if the coal-gas were of an ordinary quality, the jet-holes should not be less than one eighth nor more than one sixth of an inch from each other.

The distance between the orifices being once assumed, serves to determine the diameter of the circle of holes. Thus, in a burner of eighteen holes, each a seventh of an inch asunder, the circumference ought to be \(18 \times \frac{1}{7} = 2.57\) inches, and consequently the diameter of the circle of holes should be \(\frac{2.57}{3.1416} = .816\) inch. If the breadth of the rim be supposed a tenth of an inch, and perhaps it ought not to exceed that quantity, it may be proper, in the case of the larger burners, to contract the lower part of the central air-hole, on account of the supply of air to the inside surface of the flame increasing in a faster ratio than the number of jets.

The only remaining point to be considered with respect to the burner is the glass chimney, which serves at once to protect the flame from irregular currents of air, and to convey to the gas a due supply of it during combustion. When the interval between the chimney and the external part of the flame is too great, the tendency of the air to flow through the air-hole is diminished, and the flame contracts towards the top, where it yields a dusky light, and indicates a disposition to smoke. The diameter of the chimney should therefore be reduced until it is perceived that the upper part of the flame is enlarged, and acquires the same diameter as the lower part. When this is the case, the colour of the flame is improved in brightness, and none of the carbon is uselessly wasted in the formation of smoke. On the other hand, if the supply of air to the external surface of the flame be diminished beyond a certain extent, either by reducing the diameter of Apparatus for Preparing Oil-Gas.

When tallow or oleaginous matter of any kind is raised to a certain pitch of temperature, it is resolved into various gases, of which the compounds of carbon and hydrogen, viz. olefiant gas or bincarburetted hydrogen, and light carburetted hydrogen, are the principal, both in point of quantity and quality, for the purposes of illumination. As oil contains in its composition a portion of oxygen, existing most probably in union with hydrogen in the state of water, that substance also yields, during its destructive distillation, a considerable quantity of carbonic oxide, as well as traces of carbonic acid, hydrogen, and even nitrogen. With these products, all of which are of a determinate character, is found in greater or lesser abundance a quantity of a very inflammable vapour, which seems to be a compound of carbon and hydrogen.

Oil-gas owes its illuminating power chiefly to the proportion of olefiant gas which it contains, and the oleaginous vapour which is diffused through it; and as both of these ingredients vary exceedingly with the temperature at which the decomposition is effected, the quality of oil-gas is extremely fluctuating. When the temperature is too high, a portion of the olefiant gas and oleaginous vapour is resolved, by the deposition of carbon, into light carburetted hydrogen; and though the quantity of gas from a given portion of oil is thus increased, the quality of it is diminished in a still higher ratio. On the other hand, if the temperature be rather too low, a larger quantity of olefiant gas, mixed with a greater proportion of oleaginous vapour, is obtained; but as the latter is gradually and rapidly condensed when the gas is allowed to stand over water, the higher illuminating power of this richer gas is more than counterbalanced by the deficiency in its quantity, and the deterioration to which it is liable by keeping.

We are indebted to Dr Henry of Manchester for the first analysis of the aëroform compounds obtained by the decomposition of oil by heat; and though his elaborate researches can scarcely be said to have led to the determination of the precise products of that decomposition, they furnish data from which their true nature may be inferred, with a probability nearly as great as that which belongs to the results of direct experiment. The principal difficulty of the analysis consists in determining the condition in which the elementary principles of carbon and hydrogen exist in union with each other, and reconciling the various suppositions that may be made respecting the compounds thus formed, with the specific gravity which belongs to the original gas, supposed to be produced by Gas-Light.

The results of Dr Henry's first experiments were published in 1805; but it was not till about ten years after that period that an apparatus for decomposing oil, on a large scale for economical purposes, was constructed. In 1815 a patent was obtained by Mr Taylor, in which he was afterwards joined by Mr Martineau, for producing gas from oil, by means of an apparatus which was employed with little variation in all the oil-gas establishments, which were subsequently formed in various parts of the country. The general arrangements adopted in that apparatus will be readily understood from figs. 5, 6, and 7, Plate CCLVIII.

The retort aaaa, has the form of a horse-shoe, being recurved and furnished with two mouth-pieces. The grate of the furnace is represented by b. The oil which is contained in the cistern c, is conveyed by means of a tube t, furnished with a stop-cock for regulating the discharge, and allowed to drop into the retort upon fragments of brick or coke heated to redness, to promote the decomposition. The gas, after its production, ascends from the retort, at the opposite limb where the oil entered through the upright pipe ff, by which it is conveyed into a close vessel hh, which is surrounded with cold water in the cistern ii. From the bottom of the vessel hh, a worm or spiral tube kk ascends, and after a few convolutions around hh, terminates in the pipe l, which descends into the airtight chest mm. The gas leaving the vessel hh, is conducted by the spiral tube kk, depositing in its progress any portion of the oil that may have come over with it in a volatilized state, and escapes from the lower end of the pipe l, into the chest mm, which contains water to the depth represented. After quitting the pipe l, the gas is forced to traverse the inclined partition n, in a zig-zag manner, by means of diagonal ribs attached to its lower side, till gradually ascending, it escapes at the upper end, and rising through the water, finally passes off at g to the gasometer.

Oil being decomposed at a loss of nearly fifty per cent. Loss incurred in the conversion of it into gas, after a protracted, but ineffectual competition with coal, has been gradually abandoned on the large scale, even in those places where, from the interest they had in the whale-fisheries, there was the strongest inducement to foster the unfounded prejudices, which prevailed for some time against the use of coal-gas. The exaggerated advantages which it was pretended would be derived from compressing oil-gas, and thus rendering it portable, served to prolong the gross delusions on the subject, under which the public mind, led astray as it was, not less by novelty than by erroneous statements, so long laboured; nor were these delusions fully removed, until a demonstration was given of the failure of the scheme, in the decay of the costly edifices, and expensive apparatus, which, in defiance of all sober calculation, had been constructed for carrying it into effect.

Methods for determining the Illuminating Power of the Gases.

Having described at considerable length the nature of the various manipulations by which gas is prepared, both from coal and oil, we shall now explain the methods which seem best calculated to determine their respective illu-

---

1 The oleaginous vapour alluded to consists, according to the experiments of Mr Faraday, of two distinct compounds of carbon and hydrogen. One of these he terms bicarburetted of hydrogen, which, by his analysis, is composed of six proportions of carbon and three of hydrogen. The other compound, to which Dr Thomson has given the name of quadra-carburetted hydrogen, consists of four proportions of carbon and four proportions of hydrogen, existing in a different state of aggregation from that in which they exist in bicarburetted hydrogen, the elementary constituents of which are in the same proportion. Gas-Light.

The first and most obvious of these tests is to determine the intensity of the light which the gases are capable of diffusing during their combustion, upon a white and smooth surface, directly exposed to its emanations. The determination of that intensity is obtained with the greatest accuracy, not by a direct comparison of the degree of illumination shed on two separate surfaces, but by means of a happy contrivance, first proposed by Count Rumford, which allows the illuminated surfaces to be contrasted with each other on the same ground, and so closely adjoining, that the eye can readily detect the slightest difference between them. The contrivance, which is extremely simple and easily applied, is as follows: Let A and B, fig. 8, Plate CCLVI., be two luminous objects; EF a smooth and white surface, having the same inclination to the rays of light emitted by A and B; and CD an opaque cylindrical rod parallel to the surface EF; then it is evident that aa and bb will be the shadows of CD, in reference to the lights A and B. But the shadow aa being illuminated by the light B, and the shadow bb by the light A, it follows that if these shadows be perfectly the same, in point of intensity of shade, the light yielded by A and B must be the same in degree. If the shadows, however, be different, one of the lights must be removed either further from EF or brought nearer to it, until the shades seem to be exactly alike, when the light shed upon EF by A and B, must, in point of intensity, be, as before, the same. But the intensity of light, like that of other emanations proceeding in straight lines from a central point, being inversely as the square of the distance, the relative degrees of light emitted by A and B, must, in conformity with that principle, be proportional to the squares of their respective distances from the surface, on which the shadows are projected. Thus, if the light A was at the distance of fifteen feet, and the light B at the distance of twenty-five feet, their relative illuminating powers would be as the square of fifteen to the square of twenty-five; that is, as 225 to 625, or as 9 to 25. As the quantity of gas consumed in the same time to yield the supposed lights might be different, it is evident that a correct estimate of the absolute value of the gases, for the purpose of illumination, would not be duly determined, unless that circumstance were also taken into account. But the economical value of the gases, yielding equal degrees of light, being inversely as the quantities consumed, it follows that that value will be directly as the squares of the distances, at which the shadows are the same, and inversely, as the rate of consumption. Thus, if we now suppose that the gas yielding the light A consumed three cubic feet, in the same time that the gas yielding the light B consumed five cubic feet, the value of the former would be to that of the latter, as \( \frac{3}{5} \) to \( \frac{5}{3} \), or as three to five. In obtaining the necessary data for determining the ratio of the lights, it may be proper to add, that the screen on which the shadows are projected should be guarded with the utmost care from all extraneous light. See Photometer.

Though the determination of the intensity of light by the method of shadows is extremely simple, and capable, when the necessary experiments are carefully managed, of leading to all the precision which the nature of the problem requires, it has been proposed, by those more conversant with chemical researches, to ascertain the relative value of the gases used for illumination, by finding the quantity of olefiant gas which they contain under equal volumes; it being assumed that the illuminating power of the compound combustible gases derived from the decomposition of oil and pit-coal, is directly proportional to the quantity of that gas existing in their constitution. Though that supposition is by no means a matter of certainty, or even of probability, we shall nevertheless briefly explain the mode of analysis which is recommended by those who have adopted it. According to the experiments of Dr Henry, chlorine has no action upon any of the gases obtained from oil or coal, when the influence of light is carefully excluded, with the exception of olefiant gas; and as chlorine and olefiant gas unite together in equal volumes, this property affords an easy mode of determining the quantity of the latter which may exist in any compound gas of which it forms a constituent part. All that is required for the purpose is to add somewhat more chlorine than is absolutely necessary for uniting with the olefiant gas, and to allow the mixture to remain about fifteen minutes completely excluded from light. The extent of the absorption being thus observed, half the quantity of the gas which has disappeared of the whole mixture will be olefiant gas. Thus, if twenty parts of chlorine by measure were added to twenty-five of coal-gas, and if the mixture, after being allowed to remain a sufficient length of time in the dark, was found to occupy thirty-six measures, the absorption would be nine measures, and consequently the coal-gas must have contained four and a half measures of olefiant gas, or eighteen per cent. The quantity per cent. of olefiant gas is determined without calculation, by adding to fifty measures of the gas to be analysed an equal volume of chlorine; when the diminution of volume, in the graduated jar, is the quantity which the gas contains per cent. of olefiant gas. Dr Fyfe states that the illuminating power of the different specimens of oil and coal gas which he subjected to this test bore a pretty exact ratio to the quantity of olefiant gas they contained. One great advantage to be derived from this method of testing the quality of any species of carburetted hydrogen containing olefiant gas in its composition, is, that it admits of a comparison being made between gases in different places and at different times, without the necessity of transporting them to a distance, and making a simultaneous examination of their illuminating properties.

The specific gravity of oil and coal gas, and the quantity of oxygen which they require for their perfect combustion, have also been proposed as means of ascertaining their illuminating powers. The latter, however, even if it were a correct test, is determined with considerable difficulty; and that little reliance can be placed upon the quality, former, may be inferred from the fact, that some of the gases which are component parts of oil and coal gas have a great specific gravity, without possessing any illuminating power. This will readily be perceived from the subjoined table.

| Gases | Specific Gravity | Quantity of Oxygen for 100 Volumes | |----------------|------------------|-----------------------------------| | Olefiant gas | .970 | 300 | | Carb. hydrogen | .556 | 200 | | Hydrogen | .069 | 50 | | Carbonic oxide | .972 | 50 | | Carbonic acid | 1.588 | None |

Of these gases, carbonic oxide and carbonic acid possess the greatest specific gravity; while the latter is not only destitute of every illuminating property, but calculated, as we shall afterwards show, to deteriorate to a great extent the quality of the luminiferous gases with which it may happen to be mixed. Comparative Merits of Oil and Coal Gas.

Though the examination of the relative values of oil and coal gas for the purposes of illumination might seem to be an inquiry little calculated to awaken party feeling, or lead to any great diversity of statement, there is perhaps no subject of investigation that has been conducted with less candour and impartiality. Nor is this remark merely applicable to the individuals who had an obvious interest in supporting the superiority of either gas; for even men of science, to whom it was apparently a matter of indifference which of the gases ought to be preferred, lent in too many instances their authority to the delusive and erroneous statements, which for a season misled public opinion on the subject.

As experience, the test to which all speculative doctrines relating to economy and public utility must ultimately yield, has already decided that, in this country at least, no establishment for manufacturing gas from oil can compete with one in which it is prepared from coal, it is less necessary now than it was formerly to lay before our readers a minute account of the comparative advantages of the two systems; but as there may still be cases where, from the scarcity of coal and the abundance of oil, it might be doubtful which of them should be adopted, we shall briefly consider some of the leading points connected with the two modes of illumination. These may be discussed under the heads of; 1st, the illuminating power of the gases; 2d, their purity; 3d, their heating power; 4th, their safety; and, 5th, their economy.

1st, The Illuminating Power.—This property may be considered either in regard to the relative light yielded by the flames of the gases, when equal surfaces are taken, or with respect to the absolute quantity of light furnished by equal volumes of the gases. In the former case, though the light of oil-gas is certainly more brilliant than that of coal, the difference is so slight that it is difficult to detect it by the eye, or even by the method of shadows; but in the latter case, when the consumption of the gas, as well as the light which it yields, is taken into account, it cannot be denied that the advantage is considerably in favour of oil-gas. In consequence, however, of the great diversity of condition in which the two gases are prepared, both with respect to their production and purification, the most discordant statements have been given on this point. According to Mr Brande, oil-gas and coal-gas give equal degrees of light, when their volumes are as one to two and a half; according to Dr Tyte and Mr Leslie, when they are as one to one and a half; and according to Mr Dalton, when they are as one to two and a fourth. The greatest deviations were obtained by Messrs Phillips and Faraday, and by Mr Ricardo, the former making the ratio as one to three and a half, and the latter as one to four. The writer of this article, who made the comparison between the two gases when each of them was of the best quality, found that a very pure oil-gas, the specific gravity of which was 1.054, consumed at the rate of 2779 cubic inches per hour, and yielded the light of thirteen and a half tallow candles, being 266 cubic inches to give the light of a candle for an hour; while a coal-gas, the specific gravity of which was 0.675, consumed at the rate of 3900 cubic inches per hour, and yielded the light of thirteen candles, being 300 cubic inches to give the light of a candle for an hour. From these experiments the illuminating power of the oil-gas was to that of the coal-gas as 206 to 300. It may be proper to add, that the oil-gas, after being kept five days over water, consumed at the rate of 282 cubic inches per hour for the light of a candle, and that the coal-gas kept in a similar manner, and for the same length of time, consumed at the rate of 409 cubic inches per hour, and gave the same degree of light. The deterioration produced on Gas-Light, the oil-gas was therefore thirty-seven per cent, and that on the coal-gas about thirty-six and a third. According to the experiments of Drs Christison and Turner, when by keeping the oil-gas was of the best, and the coal-gas of average quality, their relative illuminating powers were as 100 to 100.

This result, which was obtained from data obviously unfavourable to coal-gas, cannot be admitted as exhibiting a fair estimate of the illuminating power of the two gases; and as it may be presumed that, in some of the other experiments which they performed with a view to determine the same point, the effect of keeping the gases for different lengths of time over water may have been overlooked, the results they obtained, which make the relative powers of illumination as 100 to 233, are of less value. It appears that oil-gas suffers more by keeping, even when it is not preserved over water, than coal-gas. This seems to be owing to the deposition of an ethereal oil, already noticed, which remains suspended in the gas for a short time after it has been prepared, but which gradually separates from it when it cools, or is allowed to remain in a state of repose. It is a singular fact, that though the specific gravity of oil-gas, after it has been conveyed for some distance through pipes, is materially diminished by the loss of this oleaginous vapour, scarcely any trace of the latter is to be found in the vessels for collecting the condensed matter, between the manufactory and the houses of the consumers of the gas. The deterioration, however, is undoubtedly, and in some cases it amounts to thirty per cent. We have even found that the best oil-gas, if kept for several weeks over water, becomes so much debased in quality as to require 600 cubic inches per hour to yield the light of one candle. Coal-gas suffers also by keeping, but in an inferior degree; the olefiant gas, which it contains less abundantly, being more largely absorbed, in the case of oil-gas, by the water over which it stands.

The mean result deduced by the writer of his article, Relative from a very considerable number of experiments performed with great care, and by means of suitable apparatus for the purpose, is that 100 cubic inches of oil-gas are equal in illuminating power to 160 of coal-gas, both being of good quality. The coal-gas, it may be proper to mention, was produced from cannel coal, and purified without coming in contact with water, by dry lime, or at least quicklime slightly moistened, and mixed with a small portion of the peroxide of manganese. The oil-gas was in most cases obtained by the patent apparatus of Taylor and Martineau.

2d, The Purity.—Under the head of purity we may consider the offensive smell belonging to the materials oil and coal from which the gases are obtained; to the products connected with their decomposition; and also to the gases themselves, as well as the tendency which the gases may have to produce, during their combustion, substances capable of being deleterious to health, or injuring any kind of household furniture exposed to their influence. With respect to the first of these nuisances, it cannot be doubted that oil is more apt to prove offensive to the neighbourhood than coal. The latter, if kept for some months, only loses the water it had imbibed in the bowels of the earth, and is thus rendered more fit for yielding a better gas, as well as one more abundant in quantity; whereas the former becomes rancid and fetid by keeping, and emits, by exposure, effluvia which are equally offensive and unwholesome. The mere emptying of oil from one vessel to another, and especially the melting of some kinds of it, which is necessary before it can be introduced into the retorts, is attended with a degree of nuisance which no caution can entirely prevent. To counterbalance these disadvantages, it must be admitted that the Gas-Light products which result from the decomposition of coal are not only of a more offensive nature, but more abundant in quantity than those which are obtained from oil; and though the extent of the nuisance which they occasion is greatly diminished in well-regulated coal-gas works, by confining them in close vessels, it is impossible altogether to prevent the annoyance to the neighbourhood, which their removal from time to time produces.

With respect to the smell of the two gases, they are so nearly the same when the gases are properly purified, that it would be difficult to pronounce which of them is least unpleasant. At any rate, when they are properly burnt, and consumed without any portion of them being permitted to escape, no disagreeable odour is perceptible; and, in this respect, either of them has a decided advantage over tallow or oil used for the purpose of yielding light, and burnt, according to the ordinary mode, with a wick.

The most serious objection, however, against the use of coal-gas, coal-gas, was grounded on the various impurities with which, in its crude state, it is more or less contaminated. These impurities consist, as already stated in a preceding part of this article, of tar, carbonic acid, sulphuretted hydrogen, and carbonate of ammonia, some of which are so offensive that, without being freed from them, coal-gas could never be brought into competition with oil-gas. When coal-gas was first introduced, neither the tar nor the sulphuretted hydrogen was so completely separated from it as to render it fit for being used in private dwelling-houses; and accordingly strong prejudices continued to operate against it, long after the cause which gave them birth had ceased to exist. As it is now generally admitted, however, that coal-gas is so effectually purified, in many establishments for its production, as not to exhibit the slightest trace of sulphur when exposed to the most delicate tests for detecting the presence of that substance, its value for the purpose of illumination, as contrasted with that of oil-gas, must rest chiefly on considerations of economy. It has been alleged, indeed, by Mr Brande, that though no indications of the existence of sulphur in coal-gas could be discovered in it before it was burnt, he could never procure any which did not afford traces "very minute" of sulphurous acid during its combustion. But if he had submitted oil-gas to the same ordeal he would have obtained a like result; a portion of sulphur being expelled from the coke which is used in the retorts for decomposing the oil, and which sulphur accordingly unites itself with the gas at the moment of its formation, and may be detected by exposing the condensed aqueous vapour which is formed during its combustion, to the action of the nitrate or acetate of barium. In this respect, therefore, the two gases must be admitted to be on the same footing.

Under the head of purity it may not be improper to notice the quantity of aqueous vapour which is produced during the combustion of the gases, by their union with the oxygen of the atmosphere, as the water thus formed may be conceived to possess in some measure the character of a nuisance. To obtain a correct estimate of the quantity of this water, generated in any case, it will be necessary to take into account the proportional relation between the volumes of the gases yielding equal degrees of illumination, and the quantities of hydrogen contained in these volumes. With respect to the former of these points, we have already stated that 100 volumes of oil-gas are equal to 160 of coal-gas; but the actual composition of the gases is still involved in some obscurity. According to the experiments of Dr Henry, 100 measures of oil-gas having the specific gravity .906 contained 38 of olefiant gas, of which, by the acknowledged constitution of that gas, Ga.76 must have consisted of hydrogen in a condensed state. The remaining 62 parts were composed, by the analysis of that able chemist, of:

| Substance | Parts | |-----------------|-------| | Nitrogen | 3.1 | | Carb. hydrogen | 46.5 | | Carb. oxide | 9.3 | | Hydrogen | 3.1 |

But 46.5 parts of carburetted hydrogen contained, in a condensed state, 93 parts of hydrogen; to which adding the simple hydrogen 3.1, and the portion of the same gas contained in the olefiant gas 76, we obtain 172.1 for the whole quantity of hydrogen in the oil-gas.

Again, Dr Henry states that 100 measures of coal-gas having the specific gravity .650 yielded thirteen of olefiant gas, which therefore contained in the condensed state 26 of hydrogen. The remaining 87 parts consisted, according to his analysis, of:

| Substance | Parts | |-----------------|-------| | Nitrogen | 1.3 | | Carb. hydrogen | 82.2 | | Carb. oxide | 3.5 |

But as 82.2 parts of carburetted hydrogen must have contained, in the condensed state, 164.4 parts of hydrogen; if to this be added the hydrogen of the olefiant gas 26, we obtain 190.4 for the whole quantity of hydrogen in the coal-gas. Combining the results thus obtained with the numbers expressing the ratio of the volumes of the gases, giving equal degrees of illumination, it appears that the quantity of water generated by oil-gas will be to that generated in the same time by coal-gas, as 100 × 172.1 to 160 × 190.4, being as 17,210 to 30,464, or as five to nine nearly. Hence coal-gas must produce nearly twice as much watery vapour in an apartment, during its combustion, as oil-gas yielding the same degree of illumination.

To determine how far the moisture thus generated may be regarded as a nuisance, we shall make an exact calculation in the case of an apartment, which we shall suppose to be twenty feet long, sixteen feet broad, and eleven feet high, and containing therefore 6,082.560 cubic inches. To light a room of these dimensions in the way commonly done with coal-gas would require three Argand burners, each yielding the light of ten candles, and consuming about 10,000 cubic inches of gas per hour. Now, as 100 cubic inches of coal-gas of ordinary quality contain, as we have already seen, 190.4 of hydrogen in a condensed state, the quantity of the latter gas consumed by the burners would be 19,040 cubic inches, which would yield 4,284 grams of aqueous vapour per hour. This quantity of water, diffused equally over the whole apartment, would add about 1/100th part of a grain to the moisture previously existing in each cubic inch. This may appear to be a very inconsiderable quantity; but as it is about a fifth part of the whole of the moisture capable of existing in the air of the atmosphere at the ordinary temperature of our rooms in winter, it would be sufficient to occasion perfect dampness in an apartment, at that season, in the course of four hours. In consequence, however, of the external cold, this state of things never takes place, the greater part of the floating vapour being reduced to the liquid form by condensation against the windows, long before the air of the apartment is generally affected by it.

3rd, The Heating Power.—As the heating power of gas may be regarded as being, in many cases, an advantage rather than a nuisance, we shall merely state the results pro-

---

1 Transactions of the Royal Society for 1821. duced by the combustion of certain volumes of oil and coal gas, in reference to the light which each of them yielded, when they were exposed to that process. The apparatus used for the purpose was extremely simple, and of the following construction: abedged, fig. 9, Plate CCLVI., represents a double vessel, composed of two concentric cylinders of tin-plate, which was furnished with a lid; the external cylinder, abed, was six inches in diameter, and the internal one, fghy, three inches, the former being twelve inches deep, and the latter six. From the bottom of the internal cylinder proceeded two tin tubes, ikl, the use of which was to carry off the air contaminated by combustion. When the apparatus was used, a quantity of water at a fixed temperature was poured into the space between the cylinders, till it stood at a mark nn in the inside; a burner m, consuming a known quantity of gas, and yielding a light equal to that of a certain number of candles, was then introduced, and allowed to remain exactly at the same distance from fgy, till the water reached a given temperature, the time of its doing so being carefully noted with a seconds watch. To guard against the errors arising from the dissipation of the heat during the progress of the experiments, equal portions of pounded ice were sometimes mixed with the water, and a thermometer was applied from time to time to the upper extremities of the tubes ikl, to ascertain whether any of the heated air escaped before it imparted the whole of its calorific influence to the water.

The following result was obtained with oil-gas having the specific gravity 970, the rate of consumption being 3054 cubic inches per hour, with the light of 10-55 candles:

Temperature of water at commencement..............49°-0 After ten minutes........................................85°-5 ... twenty ditto........................................115°-0 ... thirty ditto..........................................139°-5 ... forty ditto...........................................161°-0 ... fifty ditto............................................181°-0 ... sixty ditto............................................195°-5

Hence 3054 cubic inches of oil-gas raised the temperature of 270 cubic inches of water 146°½ in an hour.

Again, with coal-gas the specific gravity of which was 50, and which consumed at the rate of 3583 cubic inches per hour when it yielded the light of ten candles, the following results were obtained:

Temperature of water at commencement..............48°-0 After ten minutes........................................81°-0 ... twenty ditto........................................113°-8 ... thirty ditto..........................................140°-5 ... forty ditto...........................................163°-0 ... fifty ditto............................................181°-5 ... sixty ditto............................................196°-0

Hence 3583 cubic inches of this coal-gas raised the temperature of 270 cubic inches of water 148° in an hour.

As the heating power, without regard to the light, must directly as the temperature communicated, and inversely as the quantity of gas expended in the same time, the heating power of oil-gas is to that of coal-gas, according to above experiment, as $\frac{146}{3048}$ to $\frac{148}{3583}$, or as 100 to 86.

Then the relative illuminating powers of the gases are so considered, the heating powers seem to be nearly equal, but on the whole greater, in the case of coal-gas, the ratio of about 105 to 100.

4th. The Safety of using the Gases.—The explosive force of the gases, when they are mixed with various proportions of common air, and the limits within which the explosive range is confined in the case of each gas, may be considered under this head.

Among the other delusive statements brought forward so much confidence in favour of oil-gas, it was even maintained, that while coal-gas is capable, when mixed with common air, of producing the most violent explosions, Gas-Light, the former was almost exempt from that source of accident. Nor was this absurd and erroneous opinion confined to the ignorant and ill-informed upon the subject; for we find that Sir W. Congreve, in the report which he drew up for the instruction of the British Government, on the state of gas-light establishments in the metropolis, actually affirms in that document, that "the most important feature in the use of oil-gas is its safety; as, from the very narrow limits of the explosive mixture that can be formed of this gas and atmospheric air, it is scarcely possible that an accident can occur." Now, with regard to the limits of explosion thus adverted to, we have found, by the most precise experiments, varied in every conceivable way, that in general one volume of oil-gas will not explode with less than three and a half volumes of common air, nor with more than sixteen volumes; the explosion being most violent when the common air is ten times the volume of the gas; in which case we found that a third part of a cubic inch of oil-gas was capable of displacing with violence a cylindrical mass of iron upwards of forty pounds in weight. One specimen of oil-gas of the specific gravity 1-054, when it was newly prepared, and which, being kept two days, became of the specific gravity 0-955, could not be made to explode with less than seven and a half volumes of common air; but it continued to explode till it was mixed with twenty-one volumes, when that effect could no longer be produced. In the case of coal-gas the explosive range was more definite, and less subject to variation in the proportions of its admixture with common air, than oil-gas; the limits being between five and sixteen of common air to one of the gas, almost invariably. The violence of the explosion was also greatly inferior to that of an equal volume of oil-gas, as might readily be inferred from its constitution.

And not only are the explosions of oil-gas more formidable than those of coal-gas, but, in consequence of the specific gravity of the former approaching nearer to that of common air, they are, contrary to the assertion of Sir W. Congreve, more likely to occur; so that in both these respects oil-gas must be pronounced to be more dangerous than coal-gas.

As it must be evident from what we have stated, that a considerable degree of danger may be connected with the use of gas, whether obtained from oil or from coal, it may danger at not be improper now to examine a little more in detail the real extent of this evil, in order that, by a due knowledge of the sources from which it is most likely to arise, we may be prepared to guard against its consequences; or, if that be wholly impossible, to enable us at least to judge how far the economy, and other advantages of this mode of obtaining artificial light may be regarded as an adequate compensation, for the risk incurred in adopting it. On this subject, where the safety of the public is so extensively involved, we wish to use the language of caution, but not that of timidity; and while we admit that in a few instances accidents of the most fatal nature have occurred from the use of gas, we hesitate not to affirm that these accidents might generally be traced either to gross ignorance, or culpable negligence, on the part of those who were the victims of them. Accordingly, we find that since the nature of gas has become better known, the accidents from its use are few and trifling. The danger, however, was regarded by the British legislature to be of too serious a nature to be left entirely to the discretion of the manufacturers of gas; and an inquiry into the subject was accordingly instituted in 1814, by order of the Secretary of State for the Home Department. This led to a report from the Royal Society, which was followed by a more minute examination of the subject by Sir William Congreve, who was appointed by Government to prepare a report on the state of the gas-works, in and Gas-Light about the metropolis. In this report, which was laid before the public in 1824, Sir William Congreve arranges the dangers from the explosion of gas, as well as those arising from peculiar defects in the construction of the works, under three heads:

1st. That which may occur at the gasometer.

2ndly. That which may arise after the gas is expelled from the gasometer, in its passage through the main-pipes under the streets.

3rdly. That which accompanies the use of it in the building where it is consumed.

In considering the chance of explosions occurring at the gasometer-house, he begins with remarking, that, in the first filling of a new gasometer with gas, or in refitting one which, from any cause, may have been emptied, there is no contrivance for "exhausting the atmospheric air;" and accordingly accidents, he states, have actually occurred in the first filling. In proof of this assertion he mentions that "an explosion happened at Manchester, in consequence of the people carrying lights about the gasometer to ascertain if it was air-tight in the act of filling, and very serious mischief ensued." The gasometer was blown to pieces, the house destroyed; one man is said to have been killed, and others severely injured." Now, this source of danger is easily avoided, not by any "contrivance for exhausting the air," but simply by expelling it before the gas is introduced; precisely in the same manner as the gas itself is expelled by the pressure of the gasometer.

The next cause of explosion to which he adverts, is that arising from a gasometer not perfectly air-tight being prevented from sinking by some defect in its chain of suspension. This, however, is a very imaginary case, and so unlikely to happen, that it is quite unnecessary to examine the supposed consequences by which it would be followed, these being of so contingent a nature as to be barely within the range of possibility.

If the causes of explosion which we have noticed admit of being obviated by a proper construction of the gas apparatus, and a moderate degree of caution on the part of the workmen, there are others which Sir William Congreve has mentioned that are scarcely deserving of any consideration, such as the danger of explosion by lightning, and the chance of the counterpoise of the gasometer falling by the breaking of its suspension-chain. On the first of these sources of danger he remarks, "That it appeared to him such a combustion by lightning or otherwise would produce very destructive effects to the neighbouring buildings and gasometers, merely by the sudden rarefaction and expansion of the air caused by it; and would, in fact, produce a violent detonation and concussion, without supposing any combination of oxygen and hydrogen, but merely by the simple conflagration of the hydrogen."

Are we to infer from these observations, that Sir William Congreve thought it possible that hydrogen can undergo any species of conflagration, however "simple," without the presence of oxygen? or did he merely mean to affirm, that if a gasometer were destroyed by lightning, and its contents violently discharged into the air, the dreadful effects would be produced which he describes? As we cannot suppose he entertained the former opinion, we need merely say with respect to the second, that it is not only in the highest degree improbable, but calculated to excite alarm when no reasonable cause for it exists. For, granting that a gasometer were struck by lightning, the electric fluid would either glide silently along its metallic sides to the earth, or, if it penetrated the gas, produce no change on its condition beyond a slight and partial enlargement of its bulk.

With respect to the danger arising from the breaking of the suspension-chain of the gasometer, Sir W. Congreve stated, that the gasometers are made so heavy as to require sometimes a counterpoise of six or seven tons; and he adds, that "if one of these weights should be detached while the gasometer is at work, the pressure of that gasometer on the gas would, on a sudden, be so greatly increased, that the lamps connected with it would immediately flare up to a great height, and in many situations (in shops particularly) set fire to the premises before the lamps could be turned off." This seems to be a most alarming source of danger; and the importance of it weighed so much on the mind of Sir William, that he mentions it in more than one part of his Report. It may be proper, therefore, to examine how far it is calculated to produce the direful consequences which he dreaded from its occurrence. The sudden disengagement of a weight of six or seven tons, with the increase of a corresponding pressure on the gasometer, appears at first sight to be an event likely to be attended with some of the evils, which, it is so confidently stated, would result from it; but the case is quite a hypothetical one; and had Sir W. Congreve reflected for a moment that the pressure of even six or seven tons diffused over a surface of 120 square feet, would be no more than 550 grains on a square inch, or equal to the pressure of about two inches of water, he never would have introduced into his Report such a groundless cause for alarm. The truth is, that the falling of the counterpoise would increase the pressure upon the gas from three fourths of an inch to perhaps two and a half inches of water; and, by doing so, might excite a little momentary surprise on the part of the consumers of the gas, at the unusual size of their lights; but such an occurrence could not by any means produce the fatal consequences ascribed to it.

The danger to which the community is exposed from the risk of accidents, by the transmissions of the gas through the main-pipes, and its escape from these by leakage, into cellars and other subterraneous places, where it may accumulate in considerable quantity, in admixture with common air, is deserving of more serious examination. It is to this cause, indeed, that we must ascribe the principal accidents that have happened from the use of gas; and the utmost care ought to be taken to guard against their occurrence. The best protection from this source of danger is to be found by trying, with a forcing pump, all the pipes before they are laid, and executing the different joints, where they are connected, with a suitable quantity of lead, and in the most careful manner. When clay can be easily procured, additional security is obtained by beating it well into the tracks after the pipes are laid, so as to surround them with a coating of it to the extent of several inches. This precaution is not only a safeguard against the escape of the gas, but highly conducive to the durability of the pipes.

The last kind of danger to be considered is that which attends the use of gas, through ignorance or negligence, in the building where it is consumed. It has already been stated, that coal-gas will not explode unless it be mixed with more than three, but less than fifteen times its own bulk. If we take the last proportion, being the one from which danger is most to be apprehended in dwelling-houses, it appears that the air of an apartment could not be brought to a state which would place it within the range of explosion, unless it contained a quantity of gas, in admixture with it, equal to one fifteenth part of itself. The most probable way of such a state of things occurring, would be the neglect of turning the stop-cocks of the burners after the extinction of the light. If this were done in an apartment fifteen feet square and ten feet high, and containing therefore 2250 cubic feet, it would require at least 150 cubic feet of gas to be thrown into it before the adulterated air could explode. Now 150 cubic feet gas would afford a light equal to that of sixty candles burning for ten hours together—a quantity of light which, being ten times more than could ever be required for an apartment of the size specified, it may reasonably be concluded, that unless a room were of very small dimensions, and furnished with more gas burners than necessary, little danger could arise even from the culpable neglect of the top-cocks. At the same time it ought to be noticed, that if the air of a room were undisturbed, it is possible that a limited portion of it, sufficient to occasion a dangerous explosion, might be brought to a detonating condition in the immediate vicinity of the burner, though the rest of the air in the apartment were not in that state. On the whole, therefore, while we are decidedly of opinion that the risk of using gas is a great deal less than is generally imagined, perhaps the best security to the public will be found to result from an exaggerated representation of its dangers. Time, and the experience of its advantages, will gradually remove unfounded prejudices against its use, and ultimately render it the most general, as well as the most useful and economical, mode of obtaining artificial light.

5th, The Economy of Oil and Coal Gas.—Having examined at some length the more important of the physical properties of the two gases, so far as these are connected with their utility for the purpose of illumination, we shall, firstly, consider the relative advantages which they possess from a commercial point of view. From a due consideration of these properties, it can scarcely be doubted that, if the two gases could be procured equally cheap, so as to afford an equal extent and an equal degree of illumination at the same expense, most people would be disposed to give a preference to oil-gas. The reasons however for doing so are by no means of a kind to supersede the necessity of instituting a comparison between the costs of their production; and in this respect, indeed, we shall find reason to wonder by what sophistry and misrepresentation oil-gas could have maintained so long a doubtful competition with coal-gas.

Though the results of experiments on a small scale are frequently inapplicable to extensive establishments, and, in the hands of rash and speculative men, often lay the foundation of ruinous consequences, yet, when examined with due caution, they are far from being of no value, but, on the contrary, frequently furnish data from which such useful information is derived. Thus, we can ascertain from experiments on such a scale, what quantity of gas is yielded from a particular species of coal, and that quantity is obtained from a certain kind of oil. From experiments of that description, it has been found that a pound of cannel coal yields, at an average, from five to six cubic feet of gas, and a gallon of whale-oil from eighty to a hundred cubic feet of it; and hence we may safely infer, that sixteen pounds of that species of coal could produce as much gas as a gallon of whale-oil. Now, sixteen pounds of coal, at 15s. per ton, would cost 4s.; and therefore, if we reckon the oil at 2s. per gallon, it would appear that coal-gas costs little more than one twentieth part of an equal volume of oil-gas. It must not be inferred, however, from this result, that it would be proportionally more advantageous to employ coal in preference to oil, for the manufacture of gas; it being possible that the smaller expense of the process of decomposition, and the difference of the illuminating powers of the gases, may compensate, to a certain extent, for so great an inequality in the cost of the materials of production. It will be necessary, therefore, to consider these points in detail, before we can form a true estimate of the relative economy of the two modes of illumination.

Before entering upon this inquiry, it may be proper to mention that a gallon of whale-oil of the specific gravity .925 weighs almost exactly 53,950 grains. It will be necessary to keep this fact in view, to check the extravagant statements which have been given respecting the number of cubic feet of gas yielded by a gallon of oil, which in some cases have been rated so high that the weight of the gas would have exceeded the weight of the materials from which it was pretended to be derived. On this point we have it in our power to furnish the result of two years' experience, from an exact register of the quantity of gas yielded by the patent apparatus of Taylor and Martineau, at Kinfauns Castle, the residence of Lord Gray, from which it appears to be exactly 703 cubic feet. Hence the prime cost of a thousand cubic feet, if the oil were L.25 per ton, would be 28s., independently of the expense of manipulation, and the interest of capital. If we allow it to be only 26s., as stated by the patentees themselves, we shall find that, even on this admission, oil-gas cannot for a moment bear a comparison in point of economy with coal-gas. For if we compare this cost of oil-gas, which, we must remember, is exclusive of the expense of production, with the actual cost of a thousand cubic feet of coal-gas, which at an average may be reckoned 5s. 3d., including every expense of outlay, we shall find the one to bear to the other the ratio of about five to one. But as coal-gas is inferior in illuminating power to oil-gas, in the proportion of 100 to 160, by allowing for this difference of quality, it would appear that the prime cost of coal-gas is less than one third of the cost of the mere oil requisite to yield an equivalent quantity of oil-gas.

One of the delusive representations respecting oil-gas, the light by which the public were for a season misled, was that yielded by more light was obtained from oil reduced to the gaseous state, than that substance is capable of yielding obtained when it is burnt in the ordinary way. The incorrectness, from oil however, of this assertion is clearly demonstrated by the following exposition: a gallon of whale-oil consumed with the ordinary wick, in a good Argand lamp, we found, by the method of shadows, to yield the light of ten tallow candles for seventy hours, or that of seven hundred candles for one hour. But, according to the statement of Taylor and Martineau, one and a half cubic feet of oil-gas give the light of ten candles for the same time; and therefore a gallon of oil, yielding seventy and three fourths cubic feet, ought to afford the light of 472 candles for an hour. Hence the loss of light by decomposing the oil is in the ratio of 700 to 472, or about thirty-two and a half per cent. This result coincides very nearly with that which we obtain by comparing the weight of the gas with the original weight of the oil, as already stated to be obtained at Lord Gray's gas-work. For the weight of seventy and three fourths cubic feet of oil-gas having the specific gravity .92 (the average specific gravity we found it to have), ought to be 34,573 grams; and as a gallon of oil weighs 53,950 grams, the loss by decomposition is 19,377 grams, or nearly thirty-six per cent.

It appears from these statements, that while coal is con-

---

1 At the Perth gas-work it appears, by the result of a year's experience, that 6,179,679 cubic feet of coal-gas are sold for L.2696. 9s. 9d., being at the rate of nearly 8s. 9d. for a thousand cubic feet, or 14s. for an equivalent quantity of oil-gas. But the cost of oil which yields that quantity of gas being 26s., it is evident, that without taking into account the expense of the manipulations and interest of capital, a loss of 12s. would be incurred on every 26s. expended for oil, in order to enable oil-gas to be brought into competition with coal-gas. Gas-Light.

Advantages of coal-gas over oil-gas.

Gas-lighted by its decomposition at a gas-work into a substance greatly more valuable than the material in its original condition, oil is reduced by the same process into one of diminished value; insomuch that the manipulations and the capital expended upon an oil-gas establishment are actually applied to reduce to the extent of thirty per cent. the intrinsic value of the raw material, which, it was pretended, they improve in an equal degree. Add to this, that the loss of gas in the main pipes is found to be fully twenty per cent., and it follows that the light from oil-gas is obtained at twice the expense at which it may be procured immediately from the oil itself.

Hints respecting the Improvement of Coal-Gas.

Of all the combustible bodies having an elementary character, carbon and hydrogen are not only the most widely and copiously diffused throughout the three kingdoms of nature, but best adapted by far for the evolution of light during their combustion. It is only, however, when they are united together in due proportion that they answer the purpose most effectually; and, indeed, in a separate state, their illuminating powers are so feeble, that even when their combustion is accelerated and rendered more perfect by the presence of oxygen, the light they yield is yet unfit for many of the useful ends to which light is subservient. The substances in which carbon and hydrogen are united in the best proportion, for the production of light, are pit-coal in the mineral kingdom, and oils and fatty matter in the animal and vegetable.

The great abundance of coal, and the comparative cheapness at which it can be obtained, give it a decided advantage in point of economy over oleaginous matter, whether of animal or vegetable origin; while the processes of decomposing it, with the view of converting it into a volatile and elastic product, have been so much improved as to render the gas which it yields equally fit for the purpose of illumination with the more costly gases obtained from the oils.

The gas produced by the decomposition of coal and oleaginous matter at a high temperature is a compound of carbon and hydrogen, and consists chiefly of two gases, in which these elementary substances exist in definite proportions. One of these gases is termed carburetted hydrogen, and the other olefiant gas or bicarburetted hydrogen. The former contains one atom of carbon united with two atoms of hydrogen, and the latter an atom of each of these elements.

Light yielded by which contains the largest proportion of the latter element is found to yield, during its combustion, the most brilliant light, and that for a longer period of time. And, indeed, so great is the difference in these respects, that the hydrogen may not improperly be regarded as the mere solvent or vehicle of the carbon, acting the part of wick, and thus presenting that substance in a state sufficiently comminuted for its more perfect combustion. Accordingly, the more abundantly the hydrogen is impregnated with carbon, the greater may we expect to be its illuminating power, and the fitter in every respect for yielding artificial light. These views are fully supported by experiment; for not only is the brilliancy of the light modified by the quantity of carbon held in solution by the hydrogen, but the time which a given portion of the gas takes to consume away by combustion is affected by it in a still greater degree.

To determine in what ratio the illuminating power of the gases obtained both from oil and coal was reduced by diluting them in various proportions with hydrogen, we instituted a series of experiments, the results of which are of considerable importance, insomuch as they decidedly indicate, not only that the mixture is deteriorated, but that the same quantity of carbonaceous matter yields less light the more largely it is diluted with hydrogen.

In the first experiment we took a portion of coal-gas of the specific gravity .67, which we found consumed at the rate of 4400 cubic inches per hour, and yielded the light of eleven candles, being 400 cubic inches per hour for the light of one candle. This gas being diluted with a fourth part of its bulk of pure hydrogen, acquired the specific gravity .55, and wasted away at the rate of 6545 cubic inches per hour, yielding the light of ten candles. As a fifth part of the compound gas was hydrogen, the remaining four fifths, amounting to 5236 cubic inches, was the quantity of the coal-gas which, in its diluted state, gave the light of ten candles for an hour; so that 524 cubic inches of the original coal-gas were requisite to give the light of one candle for the same time. But, in its unmixed state, 400 cubic inches were sufficient to give the light of one candle for an hour; and consequently, the deterioration occasioned by the dilution was in the ratio of 524 to 400, or of 100 to 76, being 24 per cent.

It must be distinctly kept in view that the deterioration has been reckoned, not with respect to the whole volume of the mixture (in which case it would have been 39 per cent.), but simply in reference to the coal-gas itself, and therefore the experiment, so far as it goes, justifies us in adopting the conclusion, that had the hydrogen existed originally in union with the coal-gas, the latter would have improved in quality 24 per cent. by its abstraction; because the residual portion would not only have lasted longer, but yielded during its combustion a superior light.

In a second experiment, conducted in a similar manner, in which the proportion of hydrogen was one third of the quantity of the coal-gas, the deterioration was 27 per cent.; in a third experiment, the proportion of hydrogen being a half of the volume of the coal-gas, the deterioration amounted to 31 per cent.; and, in a fourth experiment, the quantity of hydrogen being exactly equal to that of the coal-gas, the deterioration extended to 36 per cent.

These results indicate a progressive deterioration in the quality of coal-gas by the admixture of hydrogen; and conclude the important conclusion to which they lead is, that the abstraction or removal of the latter, though diminishing the entire volume, would improve the nature of the residuary portion, not only in a higher ratio than the loss which the whole sustained in its bulk, but render that portion capable of yielding, for a longer period of time, a greater light than it could have done in its original state. Hence it may be inferred that the illuminating power of coal-gas, whether considered with respect to the cost of its production or the intensity of its light, admits of being improved; first, by impregnating the hydrogenous element more largely with carbon; secondly, by preventing the disengagement of hydrogen in a free state during the carbonization of the coal; and lastly, by detaching a portion of that gas from coal-gas when it already exists in admixture with it.

The first of these modes of improvement seems to be made practicable, at least to a certain extent, by thoroughly drying the coal before it is introduced into the retorts, and modifying the pressure under which the gas is generated. The second, by intermixing with the coal some substance which is at once capable of uniting with the hydroge,

---

1 The coal for making gas ought to be kept for several months well protected from moisture before it is used for the purpose. This is a point of the utmost importance in the manufacture of gas. Deterioration of Gas by keeping it after it is prepared.

Both oil and coal gas suffer, by keeping, a gradual loss in their power of illumination, which seems to increase in a more rapid ratio than the time they are kept. The deterioration, though greatest when the gases are allowed to stand over water, takes place in a considerable degree even when they are kept over oil, or in air-tight vessels. Hence it may be presumed that the carbon held in solution by the hydrogen, is separated from that element, partly by its own gravity, and partly perhaps by uniting with the oxygen of a portion of the water over which it is suspended.

To whatever cause the deterioration is owing, the fact itself is undoubtedly. Thus, an oil-gas which, when newly prepared, had the specific gravity 1.054, gave the light of a candle for an hour when it consumed 200 cubic inches; kept two days, it gave the same light with a consumption of 215 cubic inches per hour; and kept four days, it required, for the same light, 240 cubic inches per hour. In the case of a portion of coal-gas, which, when newly prepared, required 404 cubic inches to yield the light of a candle for an hour; the same gas kept two days required 430 cubic inches; and kept four days, 460 cubic inches to yield the same light. These results indicate a progressive deterioration in the quality of the gases, increasing with the length of time they are kept; and it is deserving of remark, that in both gases the diminution of the illuminating power decreases in a faster ratio than the time increases. After being kept three weeks, the oil-gas was so much debased in quality that it required 606 cubic inches of it to yield the light of a candle for an hour; and hence its illuminating power was reduced to one third of what it was when the gas was newly made. From these experiments it may obviously be inferred, that both oil and coal gas should be used as soon as possible after they are prepared.

Economy of Coal-Gas.

Among the advantages which have resulted from the introduction of coal-gas, we may reckon, 1st, its comparative cheapness; and, 2dly, its superiority to all the other modes of artificial illumination.

In forming a comparative estimate of the cost of coal-gas, and that of the other means employed for procuring artificial light, we may contrast it with the expense of economy wax, tallow, and oil, the ordinary substances used for the purpose. It deserves to be remarked, however, that while the price of coal, in consequence of the regular and abundant supply of that article, is liable to little fluctuation, the cost of wax, tallow, and oil, on account of the more precarious nature of the sources from which they are obtained, varies exceedingly in different seasons. The very extensive use, too, into which coal-gas has been brought, has produced a considerable effect upon the price of oil and tallow, as well as of wax; so that a comparative estimate of the expense of procuring the same extent of illumination from coal-gas and from these substances, must appear less favourable to the former than would have been the case, had the comparison been made when gas was first introduced. But assuming that a pound of tallow candles, which last, when burnt in succession, forty hours, costs nine pence; that a gallon of oil, yielding the light of six hundred candles for an hour, costs two shillings; that the expense of the light from wax is three times as great as from tallow; and that a thousand cubic feet of coal-gas cost nine shillings; we may state the relative cost of the same degree of illumination from these different substances, after making a suitable allowance for waste, wicks, &c. to be as follows:

| Substance | Cost per Unit | |-----------|--------------| | Coal-gas | | | Oil | | | Tallow | | | Wax | |

When carbonic acid was used instead of nitrogen, similar results were obtained; only the deterioration was considerably greater. Thus, when five volumes of the coal-gas were mixed with one volume of carbonic acid, the illuminating power was reduced from 100 to 30, whereas the case of the nitrogen it was from 100 to 69. It is therefore a fortunate circumstance that carbonic acid, which is so apt to be generated during the production of gas, and has so debasing an influence upon its illuminating power, is readily absorbed by a variety of substances; while nitrogen, the less injurious, as well as the abundant accompaniment, cannot be separated from the other gases with which it may exist in mixture, by any process yet known. But the light obtained from coal-gas is not only procured at a smaller expense; it is also more convenient for most purposes than the light yielded by other substances. In the ordinary mode of lighting by tallow and oil, the light derived from their combustion cannot be diminished in intensity without considerable disadvantage and trouble; whereas, in the case of gas, it may be reduced in an instant from the most perfect splendour to the feeblest degree of illumination, by the simple adjustment of the stop-cock. The advantages arising from this easy method of regulating the light of gas, when it is used in the chambers of the sick, and indeed in all apartments, where a variable, but uninterrupted supply of light must be kept up, can only be duly estimated by those who have experienced them. To every branch of manufacturing industry which requires a steady and powerful light, the benefits which have resulted from the introduction of coal-gas are not less important. In many operations the light may be conveyed by means of flexible pipes, connected together with ball and socket-joints, so as to be almost in contact with the fabric it is intended to illuminate, without the slightest risk of injury; and it may be kept in the same state for many hours in succession, or altered, as circumstances may render necessary.

For lighting churches, theatres, and other public buildings, where a strong and uniform light is required, gas answers the purpose more effectually than any other mode of illumination; partly, from the facility of its application, and partly, from the diversified and tasteful manner, in which the jets of flame may be exhibited in various kinds of burners.

As a street light, its superiority is universally admitted; and from that application of gas it cannot be doubted that the metropolis, and other large towns, have derived great additional security against the perpetration of nocturnal crimes, as well as the means of carrying on the ordinary business of life during the evening, with nearly the same convenience, as during the full light of day. (v. v. v.)