177. According to [42]Dr. KEILL, and other astronomical writers, it is entirely owing to the Atmosphere that the Heavens appear bright in the day-time. For, without an Atmosphere, only that part of the Heavens would shine in which the Sun was placed: and if an observer could live without Air, and should turn his back towards the Sun, the whole Heavens would appear as dark as in the night, and the Stars would be seen as clear as in the nocturnal sky. In this case, we should have no twilight; but a sudden transition from the brightest sunshine to the blackest darkness immediately after sun-set; and from the blackest darkness to the brightest sun-shine at sun-rising; which would be extremely inconvenient, if not blinding, to all mortals. But, by means of the Atmosphere, we enjoy the Sun’s light, reflected from the aerial particles, before he rises and after he sets. For, when the Earth by its rotation has withdrawn the Sun from our sight, the Atmosphere being still higher than we, has his light imparted to it; which gradually decreases until he has got 18 degrees below the Horizon; and then, all that part of the Atmosphere which is above us is dark. From the length of twilight, the Doctor has calculated the height of the Atmosphere (so far as it is dense enough to reflect any light) to be about 44 miles. But it is seldom dense enough at two miles height to bear up the Clouds. [Sidenote: It brings the Sun in view before he rises, and keeps him in view after he sets.] 178. The Atmosphere refracts the Sun’s rays so, as to bring him in sight every clear day, before he rises in the Horizon; and to keep him in view for some minutes after he is really set below it. For, at some times of the year, we see the Sun ten minutes longer above the Horizon than he would be if there were no refractions: and about six minutes every day at a mean rate. [Sidenote: Fig. IX. PLATE II.] 179. To illustrate this, let _IEK_ be a part of the Earth’s surface, covered with the Atmosphere _HGFC_; and let _HEO_ be the[43] sensible Horizon of an observer at _E_. When the Sun is at _A_, really below the Horizon, a ray of light _AC_ proceeding from him comes straight to _C_, where it falls on the surface of the Atmosphere, and there entering a denser medium, it is turned out of its rectilineal course _ACdG_, and bent down to the observer’s eye at _E_; who then sees the Sun in the direction of the refracted ray _edE_, which lies above the Horizon, and being extended out to the Heavens, shews the Sun at _B_ § 171. [Sidenote: Fig. IX.] 180. The higher the Sun rises, the less his rays are refracted, because they fall less obliquely on the surface of the Atmosphere § 172. Thus, when the Sun is in the direction of the line _EfL_ continued, he is so nearly perpendicular to the surface of the Earth at _E_, that his rays are but very little bent from a rectilineal course. [Sidenote: The quantity of refraction.]