Habit.—The
most common habits are distinctly trigonal and pyramidal. Although
prism faces are almost universally present they are always
comparatively small, and no approach to a prismatic habit has yet been
observed. The largest crystal yet seen is about two and a half inches
across, but is of unusual size. The majority are less than one inch
across. The negative pyramid generally shows the largest faces, ρ commonly
not reaching to c, and thus giving a triangular outline to the basal
plane. The prisms appear as narrow bands. The basal plane may more
rarely be absent, and ρ is then quite small. This gives the two
habits shown in figures 1 and 2, plate 37. The positive pyramids reach
the basal plane in about 20 per cent, of the crystals and then a
hexagonal outline is produced, the edges generally being distinctly in
two alternate groups (pi. 37, fig. 3). Only one crystal was seen
wherein was produced a pseudo-hexagonal symmetry. A peculiar habit that
has been found in a few crystals is produced by a predominant basal
plane, the crystal being very thin, and, if growing from one side, has
the appearance of an orthorhombic table.
Only one crystal was found not showing prism faces. The positive prism m may
be narrower than the negative prism as in plate 38, figure 1, and in
about 3 per cent, of the crystals examined it was absent, as in plate
38, figure 4. On the other hand, it may be considerably broader than
the latter, even when its corresponding pyramid ρ is smaller than the negative pyramid π, as in figure 2.
The form r (10T2), has been found on considerably less than half of the crystals in which ρ does
not reach the basal plane, and then always as a narrow truncation of
the π edges. This is shown in figure 4, plate 37, together with
which is here
represented
as complete for the crystal, although actually it is not found in all
the sectants, and occurs only on a few crystals. In a certain number of
crystals the place of r is then taken by a strip of horizontal
striations or narrow planes due to oscillatory growth. The striations
are shown in plate 38, figure 3, and in figure 4 is represented a set
of coarser oscillations showing the origin of the striations as
alternations of c and p.
The prism of the second order
was observed in
