object looked at is not in the way to act as a screen, and shut out all else from sight; the result is, that two or more objects may seem to exist before the eye at once, being visually superposed. The schoolboy experiment of seeing both sides of a whirling halfpenny at the same moment,—the appearances produced by the thaumatrope,—and the transparency of the revolving cog- or spoke-wheels referred to, in consequence of which other objects are seen through the shaded parts,—are all effects of this kind; two or more distinct impressions, or sets of impressions, being made upon the eye, but appearing to the perception as one.
So it is in the appearances particularly referred to in this paper: they are the natural result of two or more impressions upon the eye, really, but not sensibly, distinct from each other. If; whilst the eye is stationary, a series of cogs like those represented by the continuous outline (fig. 9) pass rapidly before it, they produce a uniform tint to the eye: and for the purpose of following out the description, let it be supposed the cogs are in shade between the eye and a white background; the tint is then a hazy, semitransparent grey. If another series of cogs, represented by the dotted outline, and close to the first, so as to give no sensible angular difference in the dimensions of the cogs, pass with equal velocity in the same direction, it will produce its corresponding tint. If the two sets of cogs be visually superposed in part, as in the figure, there will be no alteration in the uniformity of the tint. If the cogs of one set be more or less to the right or left of the other, then the superposed part will approach more or less to the tint of the shaded and uncut part of the cardboard wheel, and be less tran parent. But if, instead of the motion being equal, the velocities are unequal, then total changes of the appearance supervene; the spectrum (if I may so call it) of the superposed parts becomes alternately light and dark, and the alternations take place more or less rapidly as the velocities of the two sets of cogs differ more or less from each other.
When the cogs move in opposite directions, the uniform tint which each alone can produce is soon broken up in the superposed parts into lighter and darker portions, and when the velocities of both are equal, the spectrum is resolved into a certain number of light and dark alternations, which are per-
