The eye is a light-bearer, "made of that part of elemental fire which does not burn, but sheds a mild light, like the light of day.... When the light of the day meets the light which beams from the eye, then light meets like, and make a homogeneous body; the external light meeting the internal light, in the direction in which the eye looks. And by this homogeneity like feels like; and if this beam touches any object, or any object touches it, it transmits the motions through the body to the soul, and produces that sensation which we call _seeing_.... And if (in sleep) some of the strong motions remain in some part of the frame, they produce within us likenesses of external objects,... and thus give rise to dreams.... As to the images produced by mirrors and by smooth surfaces, they are now easily explained, for all such phenomena result from the mutual affinity of the external and internal fires. The light that proceeds from the face (as an object of vision), and the light that proceeds from the eye, become one continuous ray on the smooth surface."] Now we have to divide the segment which represents intelligible things in this way: The one part represents the knowledge which the mind gets by using things as images--the other; that which it has by dealing with the ideas themselves; the one part that which it gets by reasoning downward from principles--the other, the principles themselves; the one part, truth which depends on hypotheses--the other, unhypothetical or absolute truth. Thus, to explain a problem in geometry, the geometers make certain hypotheses (namely, definitions and postulates) about numbers and angles, and the like, and reason from them--giving no reason for their assumptions, but taking them as evident to all; and, reasoning from them, they prove the propositions which they have in view. And in such reasonings, they use visible figures or diagrams--to reason about a square, for instance, with its diagonals; but these reasonings are not really about these visible figures, but about the mental figures, and which they conceive in thought. The diagrams which they draw, being visible, are the images of thoughts which the geometer has in his mind, and these images he uses in his reasoning. There may be images of these images--shadows and reflections in water, as of other visible things; but still these diagrams are only images of conceptions. This, then, is _one_ kind of intelligible things: _conceptions_--for instance, geometrical conceptions of figures. But in dealing with these the mind depends upon assumptions, and does not ascend to first principles. It does not ascend above these assumptions, but uses images borrowed from a lower region (the visible world), these images being chosen so as to be as distinct as may be.