the fact that if the thickness of the lines on the disk is much greater than 1 mm., or, more accurately, if it subtends at the eye a greater angle than about one-fifth of a degree, the red and some of the other colours appear only upon the borders of the lines, their inner portions remaining black or grey.
The true solution, at least as regards the red and the blue, is, I think, to be looked for in certain phenomena attending sudden changes of illumination, which, so far as I have been able to ascertain, have not hitherto been observed.
The following are a few out of a large number of experiments that have been made during the last four months. They are described, as far as possible, in logical and not in chronological order. Persons unaccustomed to visual observations will not easily perceive some of the effects mentioned.
Experiment I.
A circular aperture in. ( l-3 cm.) in diameter was made in a sheet of blackened zinc and was covered with thin white writing paper. Diametrically across the aperture a strip of tinfoil Ax in. (1 mm.) wide was attached to the paper. The aperture was closed by a shutter, which could be very rapidly opened by means of a strong spring. The sheet of metal was placed over a window in one side of alight-tight box, inside which at a distance of 1 ft. (30 cm.) from the aperture was an incandescent lamp of 8-candle power with a ground glass bulb. The observations were made at a distance of about 1 ft. from the box, the room being in darkness.
When the shutter was suddenly opened, several curious phenomena appeared simultaneously. The period of their duration was difficult to estimate ; it was probably more than one-twentieth of a second and less than one-tenth.
(1) Immediately after it was revealed, the small luminous diskfirst increased in size with extreme rapidity, and afterwards became somewhat smaller, being in its final condition still larger than at the moment of exposure. This effect was more easily seen when the tinfoil strip was looked a t; it seemed to become at first much thinner, then thicker again.
(2) At the moment when the disk was uncovered, a luminous halo, like a broad ring, appeared to start from its margin and spread outwards through a distance of more than an inch (2‘5 cm.) in every direction ; then it rapidly contracted and disappeared. The halo was blue or blue-violet in colour, and seemed more sharply defined upon its inner than upon its outer border.
(3) Contemporaneously with the existence of the halo, the disk was surrounded by a bright red corona, which, like the halo, expanded outwards, and then contracted. There was not, however,
