Plutarch, De facie quae in orbe lunae apparet
Harold Cherniss and William C. Helmbold, Ed.

Apollonides was delighted. ‘What an original and absolutely novel contrivance the hypothesis is,’ he said, ‘the work of a man of daring and culture; but how did you proceed to bring your counterargument against it?’ ‘In the first place,’ I said, ‘in that, although the outer ocean is a single thing, a confluent and continuous sea,¹ the dark spots in the moon do not appear as one but as having something like isthmuses between them, the brilliance dividing and delimiting the shadow. Hence, since each part is separated and has its own boundary, the layers of light upon shadow,² assuming the semblance of height and depth, have produced a very close likeness of eyes and lips. Therefore, one must assume the existence of several outer oceans separated by isthmuses and mainlands, which is absurd and false; or, if the ocean is single, it is not plausible that its reflected image be thus discontinuous. Tell me whether for in your presence it is safer to put this as a question than as an assertion whether it is possible, though the inhabited world has length and breadth, that every visual ray when reflected from the moon should in like manner reach the ocean, even the visual rays of those who are sailing in the great ocean itself, yes and who dwell in it as the Britons do, and that too even though the earth, as you say,³ does not have the relation of centre to the orbit of the moon. Well, this,’ I said, ‘it is your business to consider; but the reflection of vision either in respect to the moon or (in general) is beyond your province and that of Hipparchus too.⁴ Although Hipparchus was industrious, still many find him unsatisfactory in his explanation of the nature of vision itself, (which) is more likely to involve a sympathetic compound and fusion⁵ than any impacts and rebounds such as those of the atoms that Epicurus invented.⁶ Moreover, Clearchus, I think, would refuse to assume with us that the moon is a body of weight and solidity instead of an ethereal and luminiferous star as you say⁷; (and) such a moon ought to shatter and divert the visual ray so that reflection would be out of the question. But if anyone dismisses our objections, we shall ask how it is that the reflection of the ocean exists as a face only in the moon and is seen in none of all the many other stars, although reason requires that all or none of them should affect the visual ray in this fashion. But <let us have done with this; and do you>>,’ I said with a glance at Lucius, ‘recall to me what part of our position was stated first.’