[Illustration: Plate III.]

6. Place a circular piece of white paper, about four inches in diameter, in
the sunshine, cover the center of this with a circular piece of black silk,
about three inches in diameter; and the center of the black silk with a
circle of pink silk, about two inches in diameter; and the center of the
pink silk with a circle of yellow silk, about one inch in diameter; and the
center of this with a circle of blue silk, about half an inch in diameter;
make a small spot with ink in the center of the blue silk, as in Plate
III.; look steadily for a minute on this central spot, and then closing
your eyes, and applying your hand at about an inch distance before them, so
as to prevent too much or too little light from passing through the
eye-lids, and you will see the most beautiful circles of colours that
imagination can conceive; which are most resembled by the colours
occasioned by pouring a drop or two of oil on a still lake in a bright day.
But these circular irises of colours are not only different from the
colours of the silks above mentioned, but are at the same time perpetually
changing as long as they exist.

From all these experiments it appears, that these spectra in the eye are
not owing to the mechanical impulse of light impressed on the retina; nor
to its chemical combination with that organ; nor to the absorption and
emission of light, as is supposed, perhaps erroneously, to take place in
calcined shells and other phosphorescent bodies, after having been exposed
to the light: for in all these cases the spectra in the eye should either
remain of the same colour, or gradually decay, when the object is
withdrawn; and neither their evanescence during the presence of their
object, as in the second experiment, nor their change from dark to
luminous, as in the third experiment, nor their rotation, as in the fourth