magnitude; to resolve this question, I viewed them first with the inside rays of the mirror, then with the outside, and found that in both cases the distance of the lunes remained without the least alteration. I viewed them also with the whole mirror open, but it occasioned no change.
I divided the aperture of the mirror into two parts, one from 0 to 4,4 and the other from 4,4 to 8,8 inches. When I measured the spurious diameter of a globule, the inside rays made it ,40; with all the mirror open it was ,31 ; and with the outside rays it was ,22.
(14.) From this we may conclude that the diameters given by the inside rays, by all the mirror open, and by the outside rays, are in an arithmetical progression; and that the inside rays will nearly double, the diameter given by the outside. It remains however to be ascertained whether this will hold good with spurious disks of various magnitudes.
It will not be necessary to carry the divisions of the aperture farther; for as the application of these experiments is chiefly intended for astronomical purposes, we can hardly do with less than half the mirror open; and on the other hand with a very narrow rim of reflection from the outside of the mirror, distinctness would be apt to fail.
22d Experiment. Trial of the Criterion on celestial Objects.
I viewed α Lyræ with the outside rays, and found its spurious disk to be small; with all the mirror open it was larger, and with the inside rays it was largest.