112 
MRS. SOMERVILLE ON THE ACTION OF RAYS 
the experiment is more successful when the liquid has been kept a night. The action 
of the surface in changing the colour of the lavender rays may be illustrated by 
passing the spectrum over paper coated with nitrate of silver brought to a clear yel- 
low-brown by exposure to the sun ; one half being washed with the liquid in question, 
and the other half with a solution of sulphate of quinine in dilute sulphuric acid, on 
the first half the lavender rays became vivid apple-green, while on passing to the 
other half they instantly change to an equally vivid lavender colour. These rays 
often darken the surface throughout their whole length ; sometimes they acquire a 
powerful bleaching action, and sometimes they have no effect, as evidently appears 
from the following experiments. 
There was reason to suspect that the action of the spectrum at the junction of the 
lavender with the violet rays, is in some cases different from what it is on either of 
these colours separately, a second dark image having appeared at the extremity of 
that which terminated with the violet, indicating a break in the continuity of action 
and giving the idea of a secondary spectrum. 
The juice of fresh-gathered petals of double flowering pomegranate in alcohol 
afforded an example of this. Paper washed with this juice became rich crimson, and 
on being exposed wet to the condensed spectrum, a narrow line of deep crimson was 
formed at the junction of the green and yellow rays, or perhaps in the most refrangible 
yellow, surrounded by a whitish lozenge-shaped border (Plate III. fig. ].). On again 
washing with the juice, instead of the white border, which had vanished, there was a 
crimson flame-shaped image, curved at the lower edge of the yellow rays and taper- 
ing upwards to the violet. Its colour was darker than that of the ground, though 
paler than the narrow line, which maintained its intensity ; and although the latter 
increased in width, it did not become as broad as the image in question. At the 
upper end of the violet another little dark image was formed, apparently owing to 
the action of the lavender rays, having exactly their form. The orange and red 
rays, especially the red, had no effect, though at the distance of about half the length 
of the spectrum beyond the red, two distinct spots were formed of deep crimson, 
which I believe to be the heat spots which you discovered. After some time, a 
bleaching appearance surrounded the whole image from the red upwards, probably 
owing to rapid evaporation from the heat of the spectrum (fig. 2.). Exterior bleach- 
ing frequently took place in the course of the experiments, permanent in some in- 
stances, while in others it vanished as the surface dried. When water was used with 
the juice instead of alcohol, the general character of the image was similar to that 
described, except that the small figure beyond the violet was more distinct, and 
seemed to bear the same proportion to that formed by the rest of the spectrum which 
the length of the lavender rays bears to the length of the sum of The others. The 
bleached part round the whole was more extended, and a faint crimson haze encom- 
passed the dark spots, which were very distinct (fig. 3.). 
This experiment is one of many instances in which I have observed the powerful 
