whereas if the moon sometimes hides the sun entirely, yet the eclipse does not last long and has no breadth; but a certain brightness is apparent around the rim, which does not allow the shadow to be deep and absolute. Aristotle,[342] I mean the ancient philosopher, after giving other reasons why the moon is more [Sidenote: C] often visibly eclipsed than the sun, adds this further one, that the sun is eclipsed by the interposition of the moon,[343] [the moon by that of the earth and of other bodies also.] But Posidonius gives this definition of what occurs: an eclipse of the sun is a concurrence of the shadow of the moon with our vision[344] ... for there is no eclipse, except to those whose view of the sun can be intercepted by the shadow of the moon. In allowing that the shadow of the moon reaches to us, I do not know what he has left himself to say. There can be no shadow of a star; shadow means absence of light, and it is the nature of light to remove shadow, not to cause it. XX. ‘But tell me’, he went on, ‘what proof was mentioned [Sidenote: D] next?’ ‘That the moon was eclipsed in the same way’, I said. ‘Thank you for reminding me’, he said. ‘But now am I to turn at once to the argument, assuming that you are satisfied, and allow that the moon is eclipsed when she is caught in the shadow, or do you wish me to set out a studied proof, with all the steps in order?’ ‘By all means,’ said Theon, ‘let us have the proof in full. For my own part, I still somehow need to be convinced; [Sidenote: E] I have only heard it put thus, that when the three bodies, earth, sun, and moon, come into one straight line eclipses occur, the earth removing the sun from the moon, or the moon the sun from the earth; that is, the sun is eclipsed when the moon, the moon when the earth, is in the middle of the three, the first case happening at her conjunction, the second at the half-month.’