Page:Transactions of the Geological Society, 1st series, vol. 1.djvu/382

 crystal, and which probably deceived the Abbé Haüy. On looking at the crystals of this substance, through their terminal faces opposite to the light, lines are observed in the interior of several, perfectly distinct, and in the direction of the two diagonals of these terminal faces. The intersection of these lines, instead of taking place perpendicularly, so as to be at right angles, and thus forming squares, as in figs. 3 and 4, appears to be made obliquely, fig. 5, so as to form thumbs, as shown at fig. 6. If, in order to find the measure of the angles of these rhombs, different angles be placed on them till their sides apparently coincide, an angle of 100°, or nearly so, appears to agree very well with the obtuse angles, so that the rhombs have apparently 100° and 80° for the measures of their angles. But then if a natural, or an artificial fracture, the latter of which is very difficult, be made according to the natural diagonal joints of the primitive crystal of this substance, the plane produced should make with the adjacent planes of the prism on one side an angle of 140° 3′, and on the other an angle of 129° 5′; which never is the case, for these two angles are always very exactly 135° being what the same section must produce on the supposition of the sides of the terminal faces being equal. Yet if we compare the direction of this face, with the lines traced in the interior of the crystal, according to the natural joints, it appears to be perfectly parallel with that of these lines.

Such are the reasons which determined me to consider the base of its primitive rectangular tetrahedral prism as a square: yet, as I have observed above, there is in this substance, on account of the difference between the appearance of the angles formed by the meeting of the interior lines, indicating the natural joints; and the correspondent ones formed by the planes parallel to these joints, something very peculiar, which would seem owing to some illusion dependent on refraction, for which I cannot account.