of an arrangement that may not be allowed strictly to come within the meaning of the term macle, in the sense in which it is used by Romé de l'Isle, being simply two crystals of the last variety, by one of the four hexagonal faces of each crystal, so as to form in appearance but a single crystal. I possess also another, in which two crystals of the succeeding variety are in like manner attached.
Fig. 12 is a highly interesting variety of the primitive crystal, as it forms the passage of it, as fig. 13 will shew, into the acute rhomboid, fig. 15, of which fig. 14 is the intermediary stage. This rhomboid, I have not, from the minuteness of the crystals, which, though numerous, exceed not in size the extremity of the smallest pin, been able to submit to the goniometer. This form also exists in the Spinelle ruby and in the diamond; which, as well as the red oxyd of copper, have for their primitive form, the regular octahedron, and give an acute rhomboid of 60 and 120d.
THE FIRST MODIFICATION.
This modification consists in a decrease on the six solid angles of the primitive crystal, so that each is replaced by an equilateral quadrangular plane, perpendicular to the axis that passes through the angle, and forms the passage of the octahedron into the cube. The angle caused by the meeting of the faces P and 1, Fig. 16, is, according to Haüy, 125°, 15′, 52″ Fig. 24 shews the direction of the laminæ of the cube. Fig. 25 may rather be considered an accidental circumstance, than as forming an important part of this series. That the triangular face formed by the decrease of crystalline laminæ on the solid angle of the cube (fig. 24.) corresponds with the face P of the primitive crystal will be obvious. I could not therefore hesitate to delineate this interesting combination, of which several instances occurred on the same specimen. Fig. 29 represents the
E
