BLOWPIPE, a tube for directing a jet of air into a fire or into the flame of a lamp or gas jet, for the purpose of producing a high temperature by complete and rapid com-bustion. The blowpipe has been in common use from the earliest times for soldering metals and working glass; and since 1733, when Anton Swab first applied it to analysis of mineral substances, it has become a valuable auxiliary to the mineralogist and chemist, in the chemical examina-tion and analysis of minerals. Its application has been variously improved at the hands of Cronstedt, Bergmann, Gahn, Berzelius, Plattner, and others, but more especially by the two last-named chemists.
The simplest and oldest form of blowpipe (still used by gasfitters, jewellers, <fcc.), is a conical brass tube, about 7 inches in length, curved at the small end into a right angle, and terminating in a small round orifice, which is applied to the flame, while the larger end is applied to the mouth. Where the blast has to be kept up for only a few seconds, this instrument is quite serviceable; but in longer chemical operations inconvenience arises from the condensation of moisture exhaled by the lungs in the tube. Hence many blowpipes are made with a cavity for retaining the moisture. Cronstedt placed a bulb in the centre of his blowpipe. Dr Black's convenient instrument consists of a conical tube of tin plate, with a small brass tube, support-ing the nozzle, inserted near the wider end, and a mouth-piece at the narrow end. One of the most suitable forms of blowpipe is that shown in fig. 1. It is Oahn's instru-

Fia. 1,—Extremities of GaWs Blowpipe,—ordinary size, ment as improved by Plattner. The tube A is ground to fit accurately into a socket at the top of the water-trap B, as is also the jet-pipe C. The nozzle D, of platinum, is fitted in the same manner, so that it can be easily removed and replaced while hot; e.g. when it is desired to remove the crust of soot which deposits upon the point when an oil lamp or candle is used. The sizes of orifice recom-mended by Plattner are 04 and 0'5 millim. The trumpet mouthpiece, from the support it gives to the cheeks when inflated, conduces to a more steady and long-continued blast being kept up without fatigue than when the mouthpiece is inserted between the lips. Mr David Forbes has sug-gested the use of a double jet-pipe in connection with this instrument, so that a large or small orifice may be obtained without stopping the point; but it is doubtful whether the advantage gained is counterbalanced by the extra cost and complication. For the majority of blowpipe workers, there is probably no better instrument than Dr Black's, if pro-vided with a properly-shaped nozzle, if possible of platinum; but where it is much used, the large-sized trumpet-mouthed instrument of Plattner is to be pre-ferred. The instrument should be held with the first and fourth fingers passed round it, and the thumb laid along tthe side of the tube, the hold being steadied by resting the elbow on the table. The mode of blowing is peculiar, and requires some practice ; an uninterrupted blast is kept up by the muscular action of the cheeks, while the ordinary respiration goes on through the nostrils.
If the flame of a candle or lamp be closely examined, it will be seen to consist of four parts—(a) a deep blue ring at the base, (5) a dark cone in the centre, (c) a luminous portion round this, and (d) an exterior pale blue envelope. The blue ring is formed chiefly by combustion of carbonic oxide. In the central cone the combustible vapours from the wick, though heated, are not burned, atmospheric oxygen not reaching it. In the luminous portion the supply of oxygen is not sufficient for complete combustion; the hydrogen takes up all or most of it, and carbon is pre-cipitated in solid particles and ignited. In the exterior envelope, lastly, the temperature is highest, and combustion most complete,—sufficient oxygen being supplied to con-vert the carbon and hydrogen into water and carbonic acid.
In blowpipe work only two of thise four parts are made use of, viz., the pale envelope, for oxidation, and the luminous portion, for reduction. To obtain a good oxidizing flame, the blowpipe is held with its nozzle inserted in the edge of the flame close over the level of the wick, and blown into gently and evenly. A conical jet is thus produced, consisting of an inner cone, with an outer one com-mencing near its apex:—the former, corresponding to (a) in the free flame, blue and well defined; the latter, corre-sponding to (d), pale blue and vague. The heat is greatest just beyond the point of the inner cone, combustion being there most complete. Oxidation is better effected (if a very high temperature be not required) the farther the substance is from the apex of the inner cone, so far as the heat proves sufficient, for the air has thus freer access.
To obtain a good reducing flame (in which the com-bustible matter, very hot, but not yet burned, is disposed to take oxygen from any compound containing it), the nozzle, with smaller orifice, should just touch the flame at a point higher above the wick, and a somewhat weaker current of air should be blown. The flame then appears as a long, narrow, luminous cone,—the end being enveloped by a dimly visible portion of flame corresponding to that which surrounds the free flame, while there is also a dark nucleus about the wick. The substance to be reduced is brought into the luminous portion, where the reducing power is strongest.
The flame of an oil-lamp is the best for blowpipe opera-tions where gas is wanting; candle flame may be used when great heat is not required. The blowpipe lamp of Berzelius, supplied with colza oil, is probably the most suitable. The wick, when in use, should be carefully trimmed and clean, so as to avoid a smoking flame. The general introduction of gas has quite driven out the use of oil-lamps for blowpipe purposes in laboratories.
Various materials are used as supports for substances in the blowpipe flame; the principal are charcoal, platinum, and glass. Charcoal is valuable for its infusibility and low conductivity for heat (allowing substances to be strongly heated upon it), and for its powerful reducing agency by the production of carbonic oxide when ignited; so that it is chiefly employed in trying the fusibility of minerals, and in reduction. The best kind of charcoal is that of close-grained pine or alder; it is cut in short prisms, having a flat smooth surface at right angles to the rings of growth. In this a shallow hole is made with a knife or borer, for receiving the substance to be held in the flame. Platinum is employed in oxidizing processes, and in fusion of substances with fluxes with a view to try their solubility in them, and note the phenomena of the bead; also in observing the colouring effect of substances on the blowpipe flame (which effect is apt to be somewhat masked

by charcoal). Most commonly it is used in the form of wire, with a small bend or loop at the end. In flux experiments this loop is dipped when ignited in the powdered flux (e.g., borax), then held in a lamp flame till the powder is fused ; and the process is repeated, if neces-sary, till the loop is quite filled with a bead of the flux ; to this is now added a little of the substance to be examined. Platinum is also used in the form of foil and of spoons, and for the points of forceps. Metals and easily reducible oxides, sulphides, or chlorides should not be treated upon platinum, as these substances may combine with and damage it. Tubes of hard German glass, 5 to 6 inches long, about |th inch diameter, and open at both ends, are useful in the examination of substances containing sulphur, selenium, arsenic, antimony, and tellurium; these, when heated with access of air, evolve characteristic fumes. They are put in the tube near one end (which is held slightly depressed), and subjected to the blowpipe flame. The sublimates often condense on the cooler parts of the tube. Small tubes, closed at one end, are used, where it is required to detect the presence of water, mercury, or other bodies which are volatilized by heat without access of air.
The most important fluxes used in blowpipe analysis are carbonate of sodium, borax, and microcosmic salt. The first (which must be anhydrous and quite free from sulphates) serves chiefly in reducing metallic oxides and sulphides on charcoal, decomposing silicates, determining the presence of sulphur, and discriminating between lime and other earthy bases in minerals. Pure borax, or acid borate of sodium deprived of its water of crystallization by heating, is used for the purpose of dissolving up metallic oxides, when in a state of fusion at a red heat, such fused masses usually having characteristic colours when cold. In some cases the colour and transparency change on cooling. Microcosmic salts, or ammonio-phosphate of sodium, is used on platinum wire in the same way as borax ;x on heating, water and ammonia are given off. The following are some other reagents for certain cases—nitrate of potash, bisulphate of potash, nitrate of cobalt, silica, fluoride of calcium, oxide or oxalate of nickel, protoxide of copper, tinfoil, fine silver, dry chloride of silver, bone ash, and litmus and Brazil-wood paper.
It may be useful here to pass briefly under review a few of the effects obtained in qualitative examinations with the blowpipe. Beginning with the closed tube, organic sub-stances may be revealed by the empyreumatic odour given off, and by charring. Mercury condenses on the tube in minute globules. Selenium gives a reddish-brown, tellurium a grey, arsenic a black sublimate. Oxygen is sometimes given off, and will inflame an incandescent splinter of wood when introduced; while ammonia may be detected by red litmus paper, as also the acid or alkaline reaction of any liquid product. In the open tube, sulphur and sulphides give off pungent-smelling sulphurous acid gas. Selenium gives a steel-grey deposit, and an odour resembling that of horse radish. Arsenic, antimony, tellurium, yield their respective acids, forming white sublimates. The deposit from arsenic is crystalline, that from the others amorphous. In examination on charcoal, it is useful, in practice, to commence with pure materials and familiarize one's self with their phenomena. Most of the metals fuse in the heat of the blowpipe flame ; and in the outer flame they oxidize. The noble metals do not oxidize, but they fuse. The metals platinum, iridium, rhodium, and palladium do not fuse. The incrustations (when such occur) are in each case characteristic, both in aspect and in the effects they give before the blowpipe flame. Among the most com-mon oxides capable of reduction on charcoal alone, in the
1 In * paper to the Royal Society, Captain Ross points out that it is Ibetter to use boric acid and phosphoric acid, instead of borax and microcosmic salts, for various analyses.
PIPE
inner flame, are those of zinc, silver, lead, copper, bismuth, and antimony. The principal minerals that cannot be so reduced are those containing alkalies and alkaline earths, and the oxides of iron, manganese, and chromium. Many substances give a characteristic colour, when held by platinum forceps in the oxidizing flame. For example, arsenic, antimony, lead, colour the flame blue; copper, baryta, zinc, green; lime, lithia, strontia, red; potash, violet. Heated with borax, some bodies give a clear bead, both while hot and cold, except when heated by the inter-mittent oxidizing flame, or the flame of reduction, when the bead becomes opalescent, opaque, or milky white. The alkaline earths, tantalio and titanic acids, yttria and zir-conia are examples of this. The oxides of most of the heavy metals give coloured glasses with borax, similar to those obtained by their use in glass or enamel painting. Thus oxide of cobalt gives a showy blue, and oxide of nickel a reddish-brownish colour, both being very characteristic and delicate tests of the presence of these metals. Ferric oxide gives a feeble yellow colour, which is darker while hot; but when the bead so coloured is treated in the reduc-ing flame the iron passes into the state of ferrous oxide,giving an intensely green or nearly black colour. This reaction may be more certainly brought about by touching the bead while melted with a fragment of tin, when the ferric oxide is probably reduced at the expense of the metal. With manganese the reverse effect is produced. A bead containing a considerable quantity of manganous oxide, such as is produced by a clean reducing flame, is colourless, but when treated in the oxidizing flame the showy violet colour of the higher oxide is brought out. This reaction is a very delicate one, and is to be recommended to begin-ners as a test exercise in blowing a clean flame, the bead being rendered alternately coloured and colourless accord-ing as the oxidizing or reducing flame is used. Molybdic acid, which gives a black bead in the reducing, and a clear bead in the oxidizing flame, but requires more careful management, was usually recommended by Plattner to his students for this kind of exercise. Copper salts give a green bead in the oxidizing and a deep sealing-wax red in the reducing flame. This latter indication is of value in detecting a trace of copper in the presence of iron, which is done by reducing with tin as already described for iron. The effects obtained with beads of microcosmic salt, or as it is more generally called salt of phosphorus, are generally similar to those described for borax, but in certain cases it is to be preferred, especially in the detection of silica, which remains undissolved, and titanic acid, which can be made to assume the form of crystals similar to the natural mineral anatasebyparticular treatment and microscopic examination. Several new phenomena, due to the crystallization of titanic acid and similar bodies, have been described by Gustav Rose.
With carbonate of sodium as flux (a paste of which and the substance to be examined is made with water, and held on charcoal to the flame), three reactions may occur. The substance may fuse with effervescence, or it may be re-duced, or the soda may sink into the charcoal, leaving the substance intact on the surface. The first takes place with silica, and with titanic and tungstic acids. The oxides of tungsten, antimony, arsenic, copper, mercury, bismuth, tin, lead, zinc, iron, nickel, and cobalt are reduced. Lead, zinc, antimony, bismuth, cadmium, and tellurium are volatilized partially, and form sublimates on the charcoal. Mercury and arsenic are dissipated as soon as reduced. Silica and titanic acid are the only two substances that produce a clear bead. The bead in which silica is fused is sometimes rendered yellow by the presence of sulphur. Carbonate of soda, with addition of a little nitrate of potassa, is very useful for detecting minute quantities of manganese. The fused mass, when clear, has, from the production of man-

ganate of sodium, a fine green colour. (For particulars of the behaviour of different minerals before the blowpipe, see the detailed description in the article MINERALOGY.) Of late years the spectroscope has been successfully used in connection with blowpipe operations, in the detection of certain of the rarer metallic elements.
The blowpipe was first applied in the quantitative deter-mination of metals by Harkort in 1827, and was brought to a high degree of perfection by Plattner. The methods are substantially those adopted in the assay of ores on the large scale in the wind furnace or muffle, thin cap-sules of clay or cavities in charcoal blocks being substituted for crucibles, and steel basins faced with bone ash, for cupels, in silver and gold assaying. From the small size of the beads obtained, especially when the ores of the pre-cious metals are operated upon, the results are often such as cannot be weighed, and they are then measured by a tangent scale, and the weight computed from the observed diameter. This method, devised by Harkort, gives very accurate results when carefully used, but owing to the difficulty of sampling the minute quantities operated upon so as to represent the bulk of the mineral fairly, the quantitative blowpipe assay has not made much progress. Perhaps the most useful quantitative application is in the determination of nickel and cobalt. This depends upon the fact that when the compounds of these metals, as well as those of copper and iron, with arsenic, are melted in contact with an oxidizing flux, such as borax or salt of phosphorus, iron is first taken up, then cobalt, and next nickel, and finally copper; and as the oxides of these metals give very different colours to the flux, we are enabled by examining the slag to detect the exact moment at which each is removed. For the details of the process the reader is referred to Plattner's work.

Bellows.
Among the various arrangements which have been con-trived for supplying air to the blowpipe otherwise than with the mouth, we may select that represented in the annexed figure (2) as one which is generally sufficient for
FIG. 2.—Blowpipe
practical purposes It will be seen that the jet i is sup-ported on a slide which can be fixed by screwing in any direction and at any height on the rod s, which is jointed on the board b. The blast can thus be adjusted variously, according to the position given to the blowpipe lamp a, which is of the form devised by Berzelius. The bellows B, the tube k, and the reservoir R, are of vulcanized india-rubber, v and v' are valves. _ The bellows being alternately compressed (with hand or foot) and allowed to expand, air is driven into the reservoir, and a fresh supply admitted into the bellows through v. After a few trials a constant blast may thus be maintained through the nozzle
For glass-blowing ordinary coal gas is the best com-bustible, as the flame can be well controlled by a stop-cock, and requires no trimming. The nature of the apparatus will be understood from fig. 3, which shows the burner in horizontal section. The tube ab is screwed into another tube which is con-nected with the gas pipe ef. mn and op are two annular a disks which support the pipe ab; they have a series of open-ings round their edges, to admit a uniform flow of gas to the narrow annular mouth between the two tubes where it joins the blast. The stop-cock / regulates the supply of gas. The wind, supplied by double bellows fixed under the table, is sent" through a lead pipe on which brass nozzles of various width can be screwed, opening into ab; the finer nozzles being pushed up nearly to the end of this. Elastic tubing may sometimes be used with advantage for the connections. A modified form of the apparatus is suit-able for ordinary blowpipe researches of the mineralogist or chemist (see Plattner's work, 4th edition), and the appara-tus used in hand-soldering of metals and other operations of the workshop is on the same principle. With suitable trunnions the blowpipe may be made to point in any direction as re-quired.

The soldering lamp of tinners is an example of the seolipile, an instru-ment which deserves some notice here. The spirit lamp a (fig. 4) is inserted at the bottom of a sheet-iron cylinder M N, which is open on one side, as shown. The upper part of the cylinder supports a strong cup of hammered metal, with an opening for spirits at the top (closed by a screw or cock), and a bent tube coming down from its upper part, through a slit in the cylinder to the back of the flame. The weak spirits which are put in the cup are caused to boil by the heat of the lamp, and the vapour, escaping through the bent tube, pro-duces a jet of very hot flame. (The FlQ- 5.—Cap of fig. i. cup is shown separately in fig. 5). Similar advantage is gained by causing air to pass through a quantity of some soluble hydrocarbon before it goes to the nozzle of a blow-pipe.
There are several forms of apparatus in which water-pressure is utilized for supplying a steady blast to the blowpipe. One of these consists of a tin case, with an oblique partition reaching nearly to the bottom. The case is filled nearly three-fourths with water. Air is blown into the compartment which narrows upwards (and with which the nozzle is connected above) by a pipe reaching nearly to the bottom. This air rises through the water and accumulates above it, forcing the water up into the other compartment, which communicates freely with the outer air. The difference of water-level in the two cham-

bers thus sustains a continuous blast through thè nozzle. Blowpipes have also been made on the principle of the blowing-machine known as the trompe. Again, the blast is sometimes supplied from a chamber in which air is con-densed by means of a syringe.
The absorption of heat when an ordinary blast of cold air (with its large proportion of nitrogen) is sent into a flame is considerable; and this has suggested the employment of a hot blast for blowpipe work. Mr T. Fletcher has constructed an apparatus on this principle, which yields a very intense flame, sufficient to fuse plati-num wire. The arrange-ment is represented in fig. 6. It will be ob-served that the pipe con-veying the blast is coiled several times round the gas pipe (for ordinary coal-gas), and that both coil and core are heated by a row of burners placed below. The blast is furnished either with bellows.
The power of the blowpipe flame may be greatly in-creased by supplying oxygen in the place of atmospheric air ; and a still greater heat is obtained by the combina-tion of pure oxygen and hydrogen. In the latter arrange-ment, which constitutes the oxyhydrogen blowpipe, it is important that the oxygen and hydrogen be kept in sepa-rate reservoirs, and be only allowed to mix at the jet, otherwise explosion may occur through the flame running back through the jet to the reservoir of mixed gases. There are various methods of effecting this, which we do not stop to describe. The blue flame produced gives the most intense heat that is obtainable by artificial means, except by the electric current. Thick platinum wires are melted before it like wax in a candle flame ; and earths, such as lime, magnesia, or zirconia, are raised to intense incandescence. For the application of the oxyhydrogen blowpipe to the fusion of the more refractory metals, see PLATINUM.
The literature of the blowpipe is very extensive. The earlier notices of the subject will be found in Berzelius's original work, of which there are English translations by Children, published in 1822, and by J. D. Whitney (of a later edition), published in Boston in 1845. Themost complete work, however, is Plattner's Probirkunst mit dem Löthrohre, of which there are several editions; the fourth or latest, published since the author's death, has been edited by his pupil and successor, Professor Richter of Freiberg. An English translation, by Professor H. B. Cornwall, has been published in New York. For the use of the blowpipe in determining minerals, the best works are Scheerer's Löth- rohrbuch, translated by Professor H. B. Blanford, and a Manual of Determinative Mineralogy, with an Introduction to Blowpipe Analysis, by Professor G. J. Brush of Yale College. In addition to these works, notices, more or less extensive, will be found in most mineralogical handbooks and works on chemical analysis. (A. B. M.)