knowledge of the exact distances between places on various parts of the coasts of Greece, notably on the coast of Achaia and on the wild inhospitable coast of Laconia. The _Coasting Voyage_ which he used may, like the extant _Coasting Voyage_ of Scylax, have comprised a description of the whole coast of Greece, and from it Pausanias may have borrowed his estimates of distances and perhaps other features of his description as well. This is Mr. Heberdey’s theory, and it is a perfectly tenable one, though in the absence of direct evidence it must remain only a more or less probable hypothesis. Yet when we remember that Pausanias’s topographical indications are nowhere more full and exact than in Arcadia, where by the nature of the case he cannot have used a _Coasting Voyage_, the hypothesis that he used one in other parts of his work seems superfluous, if not improbable. It is quite possible that he described the coast of Hermionis from notes he had made for himself in sailing along it, and that either he failed at the time to take in the natural features correctly or that afterwards in redacting his notes at home he misunderstood what he had written on the spot. Perhaps I may be allowed to say that having repeatedly sailed along the coast in question I can testify from personal experience how difficult it is to identify by sight the places from a ship, so bewildering is the moving panorama of capes, islands, bays, and mountains. It would be no great wonder if Pausanias’s head swam a little in this geographical maze. [Sidenote: Roads from Lepreus.] Another passage where error and confusion of some sort seem to have crept in is the mention of the three roads that led from Lepreus to Samicum, Olympia, and Elis. Here, again, Pausanias may have used and misunderstood some literary source, or he may have blundered on the spot, or his notes may have been lost, or his memory may have played him false. Any of these explanations is possible. To attempt to decide between them in the absence of any positive evidence would be fruitless. [Sidenote: The Enneacrunus episode.]