if one of them
were increased in size and the other two diminished (as in Fig. 3), we
should have one long steady illumination and two short flashes at each
revolution. Again, the number of sectors may be increased; and by
varying both their number and their relative size, a number of
additional characteristics are attainable.

[Illustration: Fig. 2.]

Colour may also be introduced as a means of distinction. Coloured glass
may be set in the alternate spaces; but it is necessary to remark that
these coloured sectors will be inferior in power to those which remain
white. This objection is, however, obviated to a large extent
(especially where the dioptric spherical mirror is used) by such an
arrangement as is shown in Fig. 4; where the two sectors, WW, are left
unassisted, while the two with the red screens are reinforced
respectively by the two sectors of mirror, MM.

[Illustration: Fig. 3.]

[Illustration: Fig. 4.]

Another mode of holophotally producing the intermittent light has been
suggested by my father, and is shown in Fig. 5. It consists of alternate
and opposite sectors of dioptric spherical mirror, MM, and of Fresnel's
fixed light apparatus, AA. By the revolution of this composite frame
about the burner, the same immediate alternation of light and darkness
is produced, the first when the front of the fixed panel, and the
second when the back of the mirror, is presented to the eye of the
sailor.

One advantage of the method that I propose is this, that while we are
able to produce a plain intermittent light; an intermittent light of
variable period, ranging from a brief flash to a steady illumination of
half the revolution; and finally, a light combining the immediate
occultation of the intermittent with combination and change of colour,
we can yet preserve comparative lightness in the revolving parts, and
consequent economy in the driving machinery. It must, however, be
noticed, that none of these last methods are applicable to cases where
more than one radiant is employed: for these cases, either my
grandfather's or Mr. Wilson's contrivance must be resorted to.

[Illustration: Fig. 5.]

FOOTNOTE:

[44] Read before the Royal Scottish Society of Arts on 27th March
1871, and awarded the Society's Silver Medal.

ON THE THERMAL INFLUENCE OF FORESTS[45]

The opportunity of an experiment on a comparatively large scale, and
under conditions of comparative isolation, can occur but rarely