It is owing to refraction, that we cannot judge so accurately of magnitudes and distances in water as in air. A fish looks considerably larger in water than when taken out of it. An object plunged _vertically_ into water always appears contracted, and the more so as its upper extremity approaches nearer the surface of the water. Every thing remaining in the same situation, if we take the object gradually out of the water, and it be of a slender form, we shall see it become larger and larger, by a rapid developement, as it were, of all its parts. The distortion of objects, seen through a crooked pane of glass in a window, likewise arises from its unequal refraction of the rays that pass through it. It has been calculated that in looking through the common glass of a window, objects appear about the one thirtieth of an inch out of their real place, by means of the refraction. Refraction likewise produces an effect upon the _heavenly bodies_, so that their apparent positions are generally different from their real. By the refractive power of the atmosphere, the sun is seen before he comes to the horizon in the morning, and after he sinks beneath it in the evening; and hence this luminary is never seen in the place in which it really is, except when it passes the zenith at noon, to places within the torrid zone. The sun is visible, when actually thirty-two minutes of a degree below the horizon, and when the opake rotundity of the earth is interposed between our eye and that orb, just on the same principle as, in the experiment with the shilling and basin of water, the shilling was seen when the edge of the basin interposed between it and the sight. The refractive power of the atmosphere has been found to be much greater, in certain cases, than what has been now stated. In the year 1595 a company of Dutch sailors having been wrecked on the shores of Nova Zembla, and having been obliged to remain in that desolate region during a night of more than three months--beheld the sun make his appearance in the horizon about sixteen days before the time in which he should have risen according to calculation, and when his body was actually more than four degrees below the horizon; which circumstance has been attributed to the great refractive power of the atmosphere in those intensely cold regions. This refraction of the atmosphere, which renders the apparent rising and setting of the sun both earlier and later than the real, produces at least one important beneficial effect. It procures for us the benefit of a much longer day, at all seasons of the year, than we should enjoy, did not this property of the atmosphere produce this effect. It is owing to the same cause that the disks of the sun and moon appear elliptical or oval, when seen in the horizon, their horizontal diameters appearing longer than their vertical--which is caused by the greater refraction of the rays coming from the lower limb, which is immersed in the densest part of the atmosphere.