52 
GuY; on Crystals of Arsenious Acid. 
octaliedron^ well-formed triangular prisms^ terminated at 
either end by triangular facettes, also twin-crystals or mdcles, 
also equilateral triangles resting on half the adjoining 
triangle as a base. I will not take up your time further by 
specifying all the forms which at first puzzled and perplexed 
me. Suffice it to say that, in the full consciousness that I 
did not understand the things I saw, I determined to turn 
for awhile from nature on the small scale to art on the large. 
I procured octahedra of wood, and not being satisfied with 
them, prevailed on Messrs. Powell, of Whitefriars, to make 
me the crystals of glass now before you. By studying these 
large models, placing them in all sorts of positions, and viewing 
them from different points and in different lights, I was pre- 
pared to understand most of the appearances under the 
microscope. The broader shadows of the transparent glass 
crystals were reproduced in the small crystals of arsenious acid, 
and the several postures which I caused the large crystals to 
assume were recognisable under the microscope. I found 
that the sublimed crystals adhered to the flat surface of 
glass by their solid angles, by their edges, and by their 
faces, as well as in positions less easily described. I also 
inferred that the dark squares were crystals (octahedra) 
adhering to the glass by their solid angles, in which position, 
as my glass model taught me, the play of lights and shadows 
was such as to occasion confusion and possible darkness. 
This suspicion, which was strengthened somewhat when I 
examined the sublimates by reflected light, became certainty 
under the binocular microscope. Under that admirable 
instrument, with reflected light, there are no dark masses, 
and no obscure forms. The meaning of the dark oblong 
forms and of the dark lines which I at first identified with 
the acicular or prismatic crystals of authors did not occur 
to me till later in my inquiries. 
I have mentioned the frequent occurrence of the three-sided 
prism with bevelled extremities. I do not mean the figure 
sometimes described as a lengthened octahedron, but a figure 
having the deceptive appearance of a triangular prism. Was 
this a distinct crystalline form, or might it not be some aspect 
of the octahedron? It obviously could not be brought about 
by any attitude of the whole crystal ; but my wooden model, 
supplied by Professor Tennant, is cut in half by a plane parallel 
to, and equidistant from, two of its faces, and these two equal 
halves of the-crystal are made to rotate on each other, so 
as to show the twin-crystal, or made. Here, then, without 
supposing any new form of crystal, there was new material 
for speculation. I had seen the twin-crystal, or made, in 
