of its composition. We cannot attach any sense to the assertion, that 'soda or baryta has a metal for its base,' except we know what a metal is, or at least what properties it implies. It may not be, indeed it is not, possible, to define the kinds of bodies by words only; but the classification must proceed by some constant and generally applicable process; and the knowledge which has reference to the classification will be precise as far as this process is precise, and vague as far as this is vague.

There must be, then, as a necessary supplement to Chemistry, a Science of those properties of bodies by which we divide them into kinds. Mineralogy is the branch of knowledge which has discharged the office of such a science, so far as it has been discharged; and, indeed, Mineralogy has been gradually approaching to a clear consciousness of her real place, and of her whole task; I shall give the history of some of the advances which have thus been made. They are, principally, the establishment and use of External Characters, especially of Crystalline Form, as a fixed character of definite substances; and the attempts to bring into view the connexion of Chemical Constitution and External Properties, made in the shape of mineralogical Systems; both those in which chemical methods of arrangement are adopted, and those which profess to classify by the natural-history method.

CRYSTALLOGRAPHY.

CHAPTER I.

PRELUDE TO THE EPOCH OF DE LISLE AND HAÜY.

OF a

F all the physical properties of bodies, there is none so fixed, and in every way so remarkable, as this; that the same chemical compound always assumes, with the utmost precision, the same geometrical form. This identity, however, is not immediately obvious; it is often obscured by various mixtures and imperfections in the substance; and even when it is complete, it is not immediately recognized by a common eye, since it consists, not in the equality of the sides or faces of the figures, but in the equality of their angles. Hence it is not surprising that the constancy of form was not detected by the early observers. Pliny says, 'Why crystal is generated in a hexagonal form, it is difficult to assign a reason; and the more so, since, while its faces are smoother than any art could make them, the ругаmidal points are not all of the same kind.' The quartz crystals of the Alps, to which he refers, are, in some specimens, very regular, while in others, one side of the pyramid becomes much the largest; yet the angles remain constantly the same. But when the whole shape varied so much, the angles also seemed to vary. Conrad Gessner, a very learned naturalist, who, in 1564, published at Zurich his work, De rerum Fossilium, Lapidum et Gemmarum maxime, Figuris, says, 'One crystal differs from another in its angles, and consequently in its figure.' And Cæsalpinus, who, as we shall find, did so much in establishing fixed characters in botany, was led by some of his general views to

1 Nat. Hist. xxvii. 2.

2 p. 25.

Thus

In his

disbelieve the fixity of the form of crystals. work De Metallicis, published at Nuremberg in 1602, he says, 'To ascribe to inanimate bodies a definite form, does not appear consentaneous to reason; for it is the office of organization to produce a definite form;' an opinion very natural in one who had been immersed in the study of the general analogies of the forms of plants. But though this is excusable in Casalpinus, the rejection of this definiteness of form a hundred and eighty years later, when its existence had been proved, and its laws developed by numerous observers, cannot be ascribed to anything but strong prejudice; yet this was the course taken by no less a person than Buffon. 'The form of crystallization,' says he, 'is not a constant character, but is more equivocal and more variable than any other of the characters by which minerals are to be distinguished.' And accordingly, he makes no use of this most important feature in his history of minerals. This strange perverseness may perhaps be ascribed to the dislike which Buffon is said to have entertained for Linnæus, who had made crystalline form a leading character of minerals.

It is not necessary to mark all the minute steps by which mineralogists were gradually led to see clearly the nature and laws of the fixity of crystalline forms. These forms were at first noticed in that substance which is peculiarly called rock-crystal or quartz; and afterwards in various stones and gems, in salts obtained from various solutions, and in snow. But those who observed the remarkable regular figures which these substances assume, were at first impelled onwards in their speculations by the natural tendency of the human mind to generalize and guess, rather than to examine and measure. They attempted to snatch at once the general laws of geometrical regularity of these occurrences, or to connect them with some doctrine concerning formative causes. Thus Kepler,5 in his Harmonics of the World, asserts a 'formatrix facultas, which has its seat in the entrails of the earth, and,

3 p. 97.

4 Hist. des Min. p. 343.

5 Linz. 1619, p. 161.

after the manner of a pregnant woman, expresses the five regular geometrical solids in the forms of gems.' But philosophers, in the course of time, came to build more upon observation, and less upon abstract reasonings. Nicolas Steno, a Dane, published, in 1669, a dissertation De Solido intra Solidum Naturaliter contento, in which he says, that though the sides of the hexagonal crystal may vary, the angles are not changed. And Dominic Gulielmini, in a Dissertation on Salts, published in 1707, says, in a true inductive spirit, 'Nature does not employ all figures, but only certain ones of those which are possible; and of these, the determination is not to be fetched from the brain, or proved à priori, but obtained by experiments and observations.' And he speaks with entire decision on this subject: 'Nevertheless since there is here a principle of crystallization, the inclination of the planes and of the angles is always constant.' He even anticipates, very nearly, the views of later crystallographers as to the mode in which crystals are formed from elementary molecules. From this time, many persons laboured and speculated on this subject; as Cappeller, whose Prodromus Crystallographic appeared at Lucern in 1723; Bourguet, who published Lettres Philosophiques sur la Formation de Sels et de Cristaux, at Amsterdam, in 1792; and Henckel, the 'Physicus' of the Elector of Saxony, whose Pyritologia came forth in 1725. In this last work we have an example of the description of the various forms of special classes of minerals, (iron pyrites, copper pyrites, and arsenic pyrites;) and an example of the enthusiasm which this apparently dry and laborious study can excite: Neither tongue nor stone,' he exclaims,9 'can express the satisfaction which I received on setting eyes upon this sinter covered with galena; and thus it constantly happens, that one must have more pleasure in what seems worthless rubbish, than in the purest and most precious ores, if we know aught of minerals.' Still, however, Henckel 10 disclaims the intention of

arranging minerals according to their mathematical forms; and this, which may be considered as the first decided step in the formation of crystallographic mineralogy, appears to have been first attempted by Linnæus. In this attempt, however, he was by no means happy; nor does he himself appear to have been satisfied. He begins his preface by saying, 'Lithology is not what I plume myself upon.' (Lithologia mihi cristas non eriget.) Though his sagacity, as a natural historian, led him to see that crystalline form was one of the most definite, and therefore most important, characters of minerals, he failed in profiting by this thought, because, in applying it, he did not employ the light of geometry, but was regulated by what appeared to him resemblances, arbitrarily selected, and often delusive. Thus he derived the form of pyrites f from that of vitriol;12 and brought together alum and diamond on account of their common octohedral form. But he had the great merit of animating to this study one to whom, more perhaps than to any other person, it owes its subsequent progress; I mean Romé de Lisle. Instructed,' this writer says, in his preface to his Essais de Crystallographie, by the works of the celebrated Von Linnée, how greatly the study of the angular form of crystals might become interesting, and fitted to extend the sphere of our mineralogical knowledge, I have followed them in all their metamorphoses with the most scrupulous attention.' The views of Linnæus, as to the importance of this character, had indeed been adopted by several others; as John Hill, the King's gardener at Kew, who, in 1777, published his Spathogenesia; and Grignon, who, in 1775, says, These crystallizations may give the means of finding a new theory of the generation of crystalline gems.'

The circumstance which threw so much difficulty in the way of those who tried to follow out this thought was, that in consequence of the apparent irregularity of crystals, arising from the extension or contraction of particular sides of the figure, each kind of substance

11 Marx. Gesch. p. 97.

12 Syst. Nat. vi. p. 220.