OF ASPECT, ROTATION, LIBRATION, ETC. 143

great circle passing through the sun and moon.* Asthe moon is always near the ecliptic, this amounts tosaying that the line joining the cusps is always nearlyat right angles to the ecliptic. It follows, of course,that as the angle at which the ecliptic is inclined tothe horizon is variable, so the position of the linejoining the cusps varies with respect to the horizon.As respects the gibbous moon (or moon more thanhalf-full), these variations are not much noticed; butin the case of the crescent moon, generally observedrather near the horizon, they are very noteworthy.For instance, let the time of year be such that thepart of the ecliptic near the western horizon, soonafter sunset, is inclined at nearly the greatest possibleangle to the horizon,—that is, let the season some-what precede the vernal equinox,—the time, as weknow, when the zodiacal light is most conspicuous inthe evening. Then in our latitudes, the inclinationof the ecliptic to the horizon is about sixty-twodegrees, and supposing the moon on the ecliptic, andyoung, as shown at M x , in fig. 49, Plate XII., the line

* This will perhaps seem obvious to most readers. The proofof the proposition is comprehended in the following considerations:—The circle bounding the illuminated half of the moon necessarilyhas its plane at right angles to the line joining the centres of thesun and moon ; the circle bounding the moon’s visible hemispherenecessarily has its plane at right angles to the line joining thecentres of the earth and moon : thus the intersection of these circlesor the lunar cusps, must lie on a line at right angles to the planecontaining the three centres,—that is, to the plane of the greatcircle through the sun and moon.