10 
Transactions of the 
rMonthly Microscopical 
L Journal, July 1, 1869. 
the siliceous tissue of the valve. The kettledrum with its double 
pencil of light is therefore, quoad hoc, a thing of the past. If the 
hemisphere on the stage were really the size which our powers 
make it, nearly half an inch in diameter, it would be seen by 
unassisted vision, and we should smile at a supposed necessity of 
forming its shadow by two sources of hght, just as an artist would 
smile if he were advised to have two windows in his studio at right- 
angles to each other, for the more artistic illumination of his sitter. 
The moon, as shown by the sun's illumination, is a fair illustration 
of diatom-illumination. Light, virtually parallel, falling obliquely 
on one side only of its mountains and craters, produces natural light 
and shade. Any other arrangement would fail, and for this reason 
right-angled apertures either with the kettledrum or the prism lead 
to illusions. The kettledrum, however, with one aperture properly 
placed, is still a serviceable condenser, and brings out the hemi- 
spheres remarkably well. Still, refracted light has not the power 
and purity of reflected light ; and converging rays, whether reflected 
from a convex prism or refracted through a lens, must yield the 
palm to parallel light, which is obtained by Newton's plane prism 
as from the sun. The truth of this remark will be obvious if we 
place the smaller hemisphere of the kettledrum at right angles to 
its present position, and use it for obtaining condensed, reflected, and 
convergent rays from its flat surface, as proposed by Brewster. In 
this case, the object, being in a cone of converging rays, is virtually 
under the influence of more than one source of light, and its cha- 
racter is lost amid the intense illumination. It would be easy, by 
means of a double concave lens placed within the focus of the con- 
verging cone, to produce an intense beam of parallel light without 
any assistance from the bull's-eye lens, and this might enable us to 
detect more accurately the structure of such unapproachable fine- 
ness as obtains in Aphijoleura pellucida. The direct light of the 
sun, when reflected by the plane prism, would thus be represented 
by a very close approximation. 
In the mechanical adjustment of the prism to the sub-stage, I 
would suggest a cradle above a ball-and-socket joint, as prisms are 
often mounted, with the addition of a jointed arm, as used for the 
extension of the mirror of our microscopes sideways, and, if neces- 
sary, a clamping-screw to keep the prism in position. At present, 
I fix the prism on the sub-stage with an india-rubber band. All 
that is required is the power of turning the prism on its axis, and 
also of placing it over any diameter or any chord of the sub-stage. 
In the latter position, the prism lying over a chord from 30° east 
of the vertex of the stage to 30° west of south, and its face slightly 
inclined to the upper stage, very efl'ective obliquity is obtained. 
The lamp, of course, stands to the west. We must rotate the 
valve by the circular motion of the upper stage, till the hemi- 
