By WILLIAM PHILLIPS, Esq.
member of the geological society.
IN a communication read before the Society about the middle of
last year, I detailed some reasons for concluding that the angles of
some primitive crystals included in the terms parallelepiped, as well
as some varieties of the octahedron, had not been accurately ascertained.
Since that time, further attention to the subject has confirmed
those observations. I proceed to lay before the Society the
results of investigations in regard to ten other substances, two
or three of which have been measured by the assistance of the
reflecting goniometer only upon their natural planes, on account
either of their not yielding to mechanical division with sufficient
freedom, or not yielding to it at all. The rest have been fractured
with exactness enough to allow the use of that instrument; and for
that reason, the results allow of more complete confidence, than if
there had been a necessity for relying on their natural planes.
It would have spared me much time and difficulty, if to the other labours of the Abbé Haüy and the Count de Bournon, they had added some account of the means by which the mechanical division of
