If then a ray, A B, (fig. 1, plate 19.) fall perpendicularly on water, the attraction of the water acts in the same direction as the course of the ray: it will not, therefore, cause a deviation, and the ray will proceed straight on, to E. But if it fall obliquely, as the ray C B, the water will attract it out of its course. Let us suppose the ray to have approached the surface of a denser medium, and that it there begins to be affected by its attraction; this attraction, if not counteracted by some other power, would draw it perpendicularly to the water, at B; but it is also impelled by its projectile force, which the attraction of the denser medium cannot overcome; the ray, therefore, acted on by both these powers, moves in a direction between them, and instead of pursuing its original course to D, or being implicitly guided by the water to E, proceeds towards F, so that the ray appears bent or broken. _Caroline._ I understand that very well; and is not this the reason that oars appear bent in the water? _Mrs. B._ It is owing to the refraction of the rays, reflected by the oar; but this is in passing from a dense, to a rare medium, for you know that the rays, by means of which you see the oar, pass from water into air. _Emily._ But I do not understand why refraction takes place, when a ray passes from a dense into a rare medium; I should suppose that it would be less, attracted by the latter, than by the former. _Mrs. B._ And it is precisely on that account that the ray is refracted. Let the upper half of fig. 2, represent glass, and the lower half water, let C B represent a ray, passing obliquely from the glass, into water: glass, being the denser medium, the ray will be more strongly attracted by that which it leaves than by that which it enters. The attraction of the glass acts in the direction A B, while the impulse of projection would carry the ray to F; it moves, therefore, between these directions towards D. _Emily._ So that a contrary refraction takes place, when a ray passes from a dense, into a rare medium. [Illustration: PLATE XIX.] _Mrs. B._ The rule upon this subject is this; _when a ray of light passes from a rare into a dense medium, it is refracted towards the perpendicular; when from a dense into a rare medium, it is refracted from the perpendicular_. By the perpendicular is meant a line, at right angle with the refracting surface. This may be seen in fig. 1, and fig. 2, where the lines A E, are the perpendiculars. _Caroline._ But does not the attraction of the denser medium affect the ray before it touches it? _Mrs. B._ The distance at which the attraction of the denser medium acts upon a ray, is so small, as to be insensible; it appears, therefore, to be refracted only at the point at which it passes from one medium into the other.