reagents and fluxes. In many cases they will replace the more expensive platinum crucibles, and where easily reducible metals are under treatment, they must be used in preference to platinum.

Crucibles in order to be perfect and capable of being used indifferently for any operation, ought to possess the four following qualities: firstly, not to break or split when exposed to sudden changes of temperature; secondly, to be infusible; thirdly, to be only slightly attacked by the fused substances they may contain; fourthly and lastly, to be impermeable, or nearly so, to liquids and gases. But as it is very difficult to unite all these qualifications, various kinds of pots are made to fulfil one or more of them.

In order to render crucibles capable of withstanding changes of temperature without breaking, a certain proportion of substances infusible by themselves, is mixed with the pasty clay; sand, flint, fragments of old crucibles, black-lead, and coke, are used for this purpose. They are reduced to a state of division more or less fine, according to the grain of the clay paste. For ordinary pots, the powder ought not to be very fine; but for porcelain crucibles it ought to be as fine as flour. The choice of these various bodies depends upon the use for which the crucible is intended.

The most refractory crucibles are those made with the pure clays, or such as contain little or no oxide of iron, and especially free from calcareous matters. Amongst those clays, the best are those which contain most silica; nevertheless, these are not absolutely infusible, and in the high temperature of a wind furnace they sometimes soften so much as actually to fall into a shapeless mass. This defect, as before stated, can be in some measure diminished by mixing with the clay a quantity of graphite or coke; either of these substances forms a kind of solid skeleton, which retains the softened clay, and prevents its falling out of shape.

Coke and black-lead are more efficacious than sand, because they have no action on clay, whilst sand forms a fusible compound with it. If too large a quantity of blacklead or coke be employed, it gradually consumes in the fire, and the pots become porous, and break at the least move

ment. Wood charcoal can be used instead of black-lead or coke, but is not so good, as it burns more readily.

Black-lead crucibles are generally composed of 1 part of refractory clay, and from 2 to 3 of black-lead. These pots withstand all possible changes of temperature without cracking, and their form is rarely changed by the heat, not because they are absolutely infusible, but because they are supported by the skeleton of graphite.

Crucibles into whose composition carbonaceous matters enter, reduce any oxides that may be heated in them, and hence are inconvenient in certain cases. They can, nevertheless, be employed in all cases by giving them a lining of clay, which must be tolerably thick, and well dried before

use.

Earthen crucibles, which have not been baked at a white heat, are more or less permeable to liquids and gases, according to the grain. In order to render them impermeable to liquids, they must be heated to such a temperature as will suffice to fuse the outside. When treated in this way, however, they are very liable to crack with sudden changes of temperature: the best method, therefore, of rendering them capable of containing water, &c., is to coat them with the mixture of borax and lime described as Willis's lute.

In order that crucibles may resist the corrosive action of the fused substances contained within them, they must be as compact as possible, and the substance of which they are made must have little or no tendency to combine with the fused contents. The metals and their non-oxidised compounds neither attack clay nor black-lead; but there are, nevertheless, some metallic substances, galena, for instance, which, without exercising any chemical action on earthy matters, have the property of filtering through their pores.

The readily reducible oxides gradually corrode black-lead crucibles and those pots in the composition of which coke enters, by burning the carbonaceous matter. The greater number of these oxides, the alkalies, earths, and glasses (which are fusible silicates, borates, &c.), act more or less powerfully on the earthy base of all crucibles; so that these substances are most difficult to keep in fusion for any

length of time. They attack the crucible layer by layer, dissolving the substance of which it is composed, and after a lapse of time rendering it so thin that it cannot withstand the pressure of the molten mass within it; and the fracture of the pot, and consequent loss of contents, is inevitable.

Under the same circumstances, all those crucibles whose texture is loose, are more readily corroded than those with a firm, compact body; because the corrosive substance filters to a certain depth in the former crucibles, and, in consequence, has a larger surface to act upon than when it is contained in a compact pot.

Earthen crucibles may be assayed by noticing the time they will contain fused litharge, which exercises a very corrosive action on them, honey-combing them in all directions; and those pots which contain it longest without undergoing much damage, may be considered the best. However, this method of assay is not exact, even by taking into account the thickness of the pot, for litharge runs through crucibles; firstly, because it is very fusible, and easily filters through their pores; and secondly, it has the property of forming fusible compounds with all the silicates by combining with them. From these remarks, it will be evident that a crucible whose grain is loose will readily allow litharge to pass through it, however slightly its substance may be fusible or acted on; or, on the contrary, it may be very easily acted on (even when infusible) when it has an extremely fine grain; so that the promptitude with which a crucible is traversed by litharge bears no relation to its fusibility. A crucible of pure quartz will be very readily attacked by litharge, because the latter has much affinity for silica, and the simple silicates of lead are all very fusible; whilst a crucible composed of silica, alumina, and lime, which by itself is very fusible, would be corroded less rapidly, because the oxide of lead has much less affinity for the earths than it has for the silica; moreover, it forms less fusible compounds with the earths than with silica alone. The assays of crucibles with litharge, if not of use in ascertaining their degree of fusibility, fulfils perfectly its object when it is wished to prove the resistance a crucible has to the corrosive action of

various bodies in a state of fusion; for of all fusible substances, none exercise such a powerful action on earthy matters as litharge.

Crucibles ought not only to resist the corrosive action of those bodies they may contain, but also that of the ash produced by the combustion of the fuel in which they may be placed. These ashes being often calcareous, alkaline, or ferruginous, act on the clayey part of the crucibles exactly as the fluxes. Whence it follows, that those crucibles which contain litharge longest, will also resist the action of the fluxes best.

In order to ascertain the fusibility of a crucible, a direct experiment must be made, either by heating a piece in a crucible lined with charcoal, and scertaining if its angles be rounded, if its substance has become translucid, &c.; or, better still, by heating the crucible to be assayed with another whose properties are well known.

As to permeability, it may approximately be ascertained by filling two crucibles with water, and noting what length of time is required to empty them completely; the crucible which contains it longest being, of course, the least permeable.

To prove if a crucible be able to sustain great changes of temperature without breaking, introduce it, perfectly cold, into a furnace full of lighted coal: take it out when reddish white, and expose it to a current of cold air produced by a bellows or otherwise: if it stand these trials, it may be heated afresh and plunged red-hot into water, and if it be not broken, placed immediately in the fire. The best pots support all these operations without breaking; but it often happens that they are filled with innumerable small fissures, through which fused matters can pass. This can be ascertained by fusing rapidly in the assay pot a quantity of litharge if these be present, the fused oxide will readily filter through them.

CHARCOAL CRUCIBLES.-As all oxidised matters act readily on clay pots, and a great number of the metals and their compounds adhere to them, they have long since been replaced, under certain circumstances, by charcoal crucibles,

which do not possess these disadvantages. The older assayers used merely a piece of charcoal, with a hole made in it, and then bound round with iron or other wire. The use of these has, however been abandoned for some time, and earthenware crucibles lined with charcoal have been substituted

(see fig. 59, a, b, and c). These may be considered as charcoal pots enveloped with refractory clay; they are solid, always free from cracks, and easy of preparation, and they have the same properties as the solid charcoal crucibles without their inconveniences.

In order to prepare these crucibles, the charcoal must be chosen carefully, so as to contain no foreign substances; it must be pulverised and passed through a sieve; the powder moistened with water, mixed with a spatula, and then kneaded with the fingers until it just adheres, and forms adhesive lumps without being sufficiently wet to adhere to the hand. Some advise the addition of gum to the water with which the charcoal is moistened. The crucible is moistened slightly by being plunged into water, and withdrawn as speedily as possible, and about half an inch in depth of the charcoal paste, prepared as above, placed in it; the paste is then pressed firmly down, by means of a wooden pestle the blows are to be slight at first, and then increase in force until it is as firm as possible: another layer is then applied and pressed as before, and the process repeated until the crucible is quite full, taking great care to render all