particles of the mass be continued, another change immediately commences, in the progress of which the whole acquires a more stony texture, and a greater degree of tenacity. This is generally effected by a gradual formation of secondary spheroids in the heart of the jaspideous substance, whose centres are more distant, and dimensions greater than those of the above-mentioned globules: they are radiated with distinct fibres. When two of these spheroids come into contact by mutual enlargement, no intermixture of their fibres seems to take place; they reciprocally compress each other, and their limits are defined by planes, at which a distinct separation takes place. When several spheroids come in contact on the same level, they are formed by mutual pressure into prisms of tolerable regularity, whose division is perfectly defined: and when a spheroid is surrounded on all sides by others, it is compressed into an irregular polyhedron.

The transition from this fibrous state to a different arrangement, seems to be very rapid, for the centres of most of the spheroids become quite compact before they receive one quarter of their usual dimensions; the mass then becomes perfectly solid, very tenacious, and opake; and its hardness is somewhat inferior to that of the glass from which it is formed.

A further continuation of the temperature, favourable to arrangement, speedily occasions another change. The texture of the mass becomes more granular, and the brilliant points it exhibited in its former state become larger and more numerous, arrange themselves into regular forms, and finally, the whole mass becomes pervaded by thin crystalline laminæ, which intersect it in every direction, and form projecting crystals in the cavities.

It is thought that an equalized temperature would have rendered the whole mass at once similar to the substance last described; but then the interesting initial phenomena, from which the important inferences here announced are deduced, would not have been discovered.

These, and many more facts relating to the experiment, having been minutely detailed, the author proceeds to offer what he deems a partial explanation of the formation of the globules and of the radiated spheroids. It is well ascertained, he says, that heat is emitted by all bodies in their change from a gaseous to a fluid state; and it is reasonable to suppose that heat may also be emitted in those changes of arrangement which affect the internal texture of a body after it has attained an apparently solid state.

That a succession of such changes actually takes place, seems to be demonstrated by several of the appearances in the experiment, and particularly by the increase of specific gravity, which generally keeps pace with the internal changes of the substance. These changes, it is conjectured, may be caused by a gradual diminution of temperature, which permits certain laws to induce peculiar arrangements among the particles of the glass: when several of these particles enter into this new bond of association, they must form a minute point, from which heat will issue in every direction: that heat will

gradually propagate itself till the temperature of the glass is equalized, and then the recurrence of the circumstances which induced the first particles to arrange themselves will cause other particles to arrange themselves also; and these the attraction of aggregation will dispose round the point first formed. A second emission of heat in every direction will now take place; the temperature will again be equalized; and again another concentric coat of arranged particles will apply itself to the little globule. That these globules are formed of concentric coats does not clash with the circumstance of their being likewise radiated; as every one may have remarked the connexion that almost uniformly exists between the radiated structure and the formation by concentric coats; the more obvious instances of which are the hæmatites and the calcareous stalactites. In what manner this is likely to be effected is stated at some length in the paper.

Some curious remarks are next made on the observation of Mr. Smithson, that solution, far from being necessary to crystallization, effectually prevents its commencement; since, while solution subsists, crystallization cannot take place: and many of the phenomena being duly considered, it seems most probable that the particles of bodies apparently solid must be capable of some internal motion, enabling them to arrange themselves according to their crystalline polarity while they are in a solid state. Among the instances given, are the conversion of glass vessels into Reaumur's porcelain, the tempering of steel, and the process of annealing. This does not altogether disprove the crystallizations formed by molecules suspended in aqueous solutions; but it is in general insisted upon, that all crystallizations are dependent on heat, there being, in fact, no fluidity, and consequently no solution, which heat does not produce.

These observations tend to prove the analogy which exists between the igneous and aqueous formations, and to show that precisely the same order and kind of arrangement is followed in the generation of stony masses from water as from fire. Among the many instances that are adduced to justify this assertion, are, on the one hand, the phenomena exhibited by lavas, in which may be observed every step of the passage from the vitreous to the stony, from that to a porphyritic, and finally to the granitic state: on the other hand, we may select the formation of calcareous stalactites, in which the successive depositions of calcareous carbonate form a mass which at first is fibrous; a continuance of the process causes the fibrous structure to disappear, and the stalactite becomes irregularly spathose; after which the irregularities vanish altogether, and it becomes perfect calcareous spar, divisible into large rhomboids, with the form peculiar to that mineral.

Adverting now to the chief object of this paper,—the basalt,—the author observes, that should, in fact, the analogy between the aqueous and igneous formation appear founded, the transition from glass to stone can no way affect the great question which has so long divided geologists about the origin of basalt: for though it be synthetically demonstrated that basalt may be formed by fire, the proofs in favour

of its formation by water must be allowed to be at least of equal weight. In fact, while the frequent instances of petrifactions found in basalt support the aqueous hypothesis, the equally numerous indications that the heat emanating from it has manifestly changed beds of coal into coke and indurated strata of stony substances, strongly argue in favour of the igneous assumption.

The above sketch of the author's observations and reasonings, imperfect as it is, may however suffice for rightly comprehending the interesting part of the paper which relates to the wonderful regularity of the prismatic configuration of basaltic columns, and also for their articulations. If we suppose that a mass of fluid basalt has filled a valley to an indefinite depth and extent, the process of arrangement in its particles must be deduced from the removal of its heat or moisture, according as its solution is igneous or aqueous. This can only be done by the action of the atmosphere on its upper surface, and by the ground on which it reposes absorbing the heat or moisture from its under surface.

From the variations of the atmosphere, its action must be irregular; and from the perpetual change of the parts in contact with the heated or moist surface, its operations will always be nearly as active as at first, allowance being made for its variations: but the absorption of the ground will be regular, and regularly diminishing in activity, in proportion as the parts near the mass approach nearer to the same temperature or same moisture with the mass above; and thus absorption can only be carried on by the transmission of heat or moisture from the mass to the solid rocks below.

From these considerations, it seems evident that the arrangement of the part of the basaltic mass near the ground will be begun with more energy than it can be continued, and that the results will be more slow and regular, and that induced by the action of the atmosphere. After the first stage in the process of arrangement has been performed, and a stratum of the jaspideous substance is extended over the surface of the ground, there seems no reason to doubt that a number of radiated spheroids will be generated in it, having probably all their centres nearly at the same distance from the ground; and as the arranging power undergoes a gradual diminution of energy, it is not likely that two rows of them in height should be formed at once. In a word, it seems most probable, that in the arrangement of a mass of basalt, a single layer of radiated spheroids will be formed, reposing on the ground which supports the mass.

How these radiated spheroids, by coming in close contact, will compress each other, and form polyhedral and, generally, hexagonal prisms, will be understood from what we have said above of that operation in the experiment. If these prisms are resisted below, and there is no opposing cause above them, it is clear that they will extend their dimensions upwards into the undisturbed central mass of the fluid, till their structure is deranged by the action of the atmosphere on the upper surface of the basalt. According to this arrangement, the same cause that determines the concentric fractures of the

fibres of the spheroids, will produce convex articulations in the lower joints of the prisms. If the generating centres are not equidistant, the forms of the pillars will be irregular, and of different number of angles; and as the compression of the fibres will be greatest on the level of the generating centres, the lower part of the prisms will be most compact.

All the observations hitherto made on the great basaltic masses in nature, seem to confirm this simple theory; and the author bestows some pains in accounting for the appearances which seem at first sight not to agree perfectly with it. Those who shall peruse the paper will, if we are not much mistaken, be particularly gratified with the variety of information they will meet with in this part of the treatise.

Lastly, the author directs his attention to the many instances of other substances, besides basalt, which affect a columnar form, and which afford convincing proofs that their configuration is not confined to either the aqueous or igneous formation. Such are, certain lavas, columns of porphyry found near Dresden, a bed of gypsum at Montmartre, and other masses of various nature. Sandstone, clay, argillaceous iron ore, and many other substances, become prismatic by torrefaction; and prisms of starch formed in drying have often been considered as illustrative of basaltic formation. Some of these are probably to be attributed solely to contraction; and it is shown that they do by no means contribute toward any explanation of the process here in contemplation.

An Analysis of the magnetical Pyrites; with Remarks on some of the other Sulphurets of Iron. By Charles Hatchett, Esq. F.R.S. Read May 17, 1804. [Phil. Trans. 1804, p. 315.]

The substance which is the subject of this paper was, till lately, found only in some parts of Norway and Germany; but it now appears, from some specimens in Mr. Greville's collection, that it is likewise to be met with in considerable quantities near the foot of Snowdon in Caernarvonshire. The character by which this kind of sulphuret is chiefly distinguished from the other martial pyrites, is its magnetic property, by which, especially if it have been placed some time between magnetical bars, it will turn a needle completely round, attract and take up abundance of iron filings, and retain this addition to its original power for a considerable length of time.

After an accurate description of the external characters of this ore, Mr. Hatchett enters into a full account of the processes he instituted in order to discover its nature and component parts. In hopes of discovering the cause of the magnetic property which is peculiar to this species, he has entered into an analysis of the other kinds of martial sulphurets, not only natural, but also artificial, and has also paid particular attention to the experiments of others on this subject, particularly those of Mr. Proust, the learned Professor of Chemistry at Madrid, who has taken considerable pains in ascertain

ing the constituent parts of this ore, both analytically and synthetically. In the progress of this inquiry, a certain analogy soon presented itself, between these compounds of sulphur and iron and those of other inflammable substances, such as carbon and phosphorus, with the same metal, which suggested the idea of some experiments on the subject; from whence various facts were deduced, not only of much curiosity, but also likely to lead to some useful purposes, as will appear hereafter. The principal results of this laborious investigation are,

1. That the component ingredients of the magnetical pyrites are sulphur, and iron in the metallic state, the former being to the latter in the proportion nearly of 37 to 63.

2. That the chemical and other properties of this substance are very different from those of the common martial pyrites; which, however, are likewise composed of sulphur and iron, but varying in proportion from between 52 to 54 of sulphur, and 48 to 46 of metallic iron. Whence it appears, that the relative proportions in the composition of the magnetical pyrites, and of the mean of the common pyrites, is between 16 and 17.

3. That as the magnetical pyrites agrees in analytical results, as well as in all chemical and other properties, with that sulphuret of iron which hitherto has been only known as an artificial product, there is no doubt that it is identically the same, and that its proportions are probably subjected to a certain law, which, under certain circumstances, may be supposed to act in an almost invariable

manner.

4. That in the formation of the common martial pyrites, there is a deviation from this law, and that sulphur becomes a predominant ingredient, variable in quantity, but which, by the present experiments, has not been found to exceed between 54 and 55 per cent. ; a proportion, however, which possibly may be surpassed in other pyrites which have not as yet been chemically examined.

5. That iron, when combined naturally or artificially with 36 or 37 per cent. of sulphur, is not only still capable of receiving the magnetic fluid, but it is also rendered capable of retaining it, so as to become in every respect a complete magnet. And the same may in a great measure be inferred respecting iron which has been artificially combined with 45 per cent. of sulphur.

6. That beyond this proportion of 45 per cent. of sulphur, (at least in the natural common pyrites,) all susceptibility in iron of the magnetic influence appears to be destroyed; and although the precise maximum, which is capable of producing this effect, has not as yet been determined by actual experiment, it is certain that the limits are between 45 and 52; unless some alteration has taken place in the state of the sulphur or iron in the common martial pyrites, different from that which is conceived according to the present state of chemical knowledge.

7. That as carbon, when combined in a certain proportion with iron (forming steel), enables it to become a permanent magnet, and