CRYSTALLIZATION.

vi. The dodecaedron with isoceles triangular faces.

25. These primitive forms, by further mechanical analysis may be

Integral ele reduced to three integral elements.

i. The parallelopiped, or simplest solid, having six surfaces, parallel two and two. (24. i.)

ii. The triangular, or simplest prism, bounded by five surfaces.

iii. The tetraedron, or simplest pyramid, bounded by four surfaces. (24. ii.)

26. The secondary forms are supposed to arise from decrements of condary forms. particles taking place on different edges and angles of the primitive forms. Thus a cube, having a series of decreasing layers of cubic particles upon each of its six faces, will become a dodecaedron, if the decrement be upon the edges; but an octoëdron, if upon the angles ; and by irregular, intermediate, and mixed decrements, an infinite variety of secondary forms would ensue, as the annexed figures show.

27. But in crystallography we meet with appearances which Haüy's Objections to theory but imperfectly explains. A slice of fluor spar, for instance, ob- Haüy's theotained by making two successive and parallel sections, may be divided

into acute rhomboids; but these are not

the primitive form of the spar, because by the removal of a tetraedron from each extremity of the rhomboid an octoëdron is obtained. Thus, as the whole mass of fluor may be divided into tetraëdra and octoëdra, it becomes a question which of these forms is to be called primitive, especially as neither of them can fill space without leaving vacuities, nor can they produce any arrangement sufficiently stable to form the basis of a permanent crystal.

28. To obviate this incongruity, Dr. Wollaston (Phil. Trans. 1813,) Wollaston's has very ingeniously proposed to consider the primitive particles as theory. spheres, which, by mutual attraction, have assumed that arrangement which brings them as near as possible to each other. When a number of simi

lar balls are pressed together in the
same plane, they form equilateral trian-
gles with each other; and if balls so
placed were cemented together and af-

terwards broken asunder, the straight lines in which they would be
disposed to separate, would form angles of 60°
with each other. A single ball, placed any where
on this stratum, would touch three of the lower
balls, and the planes touching their surfaces would
then include a regular tetraedron. A square of
four balls, with a single ball resting upon the cen-
tre of each surface, would form an octoëdron; and
upon applying two other balls at opposite sides of

25. These primitive forms, by further mechanical analysis may be

Integral ele reduced to three integral elements.

i. The parallelopiped, or simplest solid, having six surfaces, parallel two and two. (24. i.)

ii. The triangular, or simplest prism, bounded by five surfaces.

iii. The tetraedron, or simplest pyramid, bounded by four surfaces. (24. ii.)

26. The secondary forms are supposed to arise from decrements of condary forms. particles taking place on different edges and angles of the primitive forms. Thus a cube, having a series of decreasing layers of cubic particles upon each of its six faces, will become a dodecaedron, if the decrement be upon the edges; but an octoëdron, if upon the angles ; and by irregular, intermediate, and mixed decrements, an infinite variety of secondary forms would ensue, as the annexed figures show.

27. But in crystallography we meet with appearances which Haŭy's Objections to theory but imperfectly explains. A slice of fluor spar, for instance, ob- Haüy's thsotained by making two successive and parallel sections, may be divided

into acute rhomboids; but these are not

the primitive form of the spar, because by the removal of a tetraedron from each extremity of the rhomboid an octoëdron is obtained. Thus, as the whole mass of fluor may be divided into tetraëdra and octoëdra, it becomes a question which of these forms is to be called primitive, especially as neither of them can fill space without leaving vacuities, nor can they produce any arrangement sufficiently stable to form the basis of a permanent crystal.

28. To obviate this incongruity, Dr. Wollaston (Phil. Trans. 1813,) Wollaston's has very ingeniously proposed to consider the primitive particles as theory. spheres, which, by mutual attraction, have assumed that arrangement which brings them as near as possible to each other. When a number of simi

lar balls are pressed together in the
same plane, they form equilateral trian-
gles with each other; and if balls so
placed were cemented together and af-

terwards broken asunder, the straight lines in which they would be
disposed to separate, would form angles of 60°
with each other. A single ball, placed any where
on this stratum, would touch three of the lower
balls, and the planes touching their surfaces would
then include a regular tetraedron. A square of
four balls, with a single ball resting upon the cen-
tre of each surface, would form an octoëdron; and
upon applying two other balls at opposite sides of

this octoedron, the group will represent the acute
rhomboid. Thus the difficulty of the primitive
form of fluor, above alluded to, is done away, by
assuming a sphere as the ultimate molecula. By
oblate and oblong spheroids other forms may be
obtained.

A sphere the ultimate molecule.

29. The subject of crystallization has more lately engaged the attention of Mr. J. F. Daniel (Quarterly Journal of Science and the Arts, Vol. Confirmation i.) and his researches have produced some singular confirmations of theory. Dr. Wollaston's hypothesis. If an amorphous piece of alum be im

of Wollaston's

mersed in water and left quietly to dissolve, at the end of about three weeks we shall observe that it has been unequally acted upon by the fluid: the mass will present the forms of octoëdra, and sections of octoëdra, as it were carved or stamped upon its surface, as seen in these figures:

This appearance is produced when the attraction of the water for the solid is nearly counter-balanced by its mechanical texture. The crystals formed by this species of dissection are highly curious, from