Three hours' ride across a range of hills to the south lies Dodona, at the foot of Mount Tomarus, which at the time of our visit (in May) was heavily capped with snow. This oldest oracle and sanctuary of Greece, famous before Delphi was born, has a situation only a little less imposing than Delphi. Until about twenty-five years ago one was still in doubt in just what part of this region to look for the old famous sanctuary. But at that time Carapanos, a wealthy Greek, settled that question with the spade. Procuring from the library, by the kindness of the professor of French in the Lyceum, the two beautiful volumes in which Carapanos has given the results of his work, I took them with me to the spot where he found Dodona, and followed in his footsteps. This library, by the way, contains a good many valuable books, new and old, but they are huddled together in very close quarters, as the librarian sadly admitted, employing the English phrase "pell-mell." But worse than the disorder was the dampness. Carapanos's book of plates was mildewed and fallen apart. Rich Greeks probably think their benefactions more safely bestowed in Athens. Otherwise somebody would have given a new library building to Joannina. The walls that Carapanos discovered all have a singularly late appearance, and there is a striking lack of pottery in the ground; but the inscriptions and dedicatory offerings leave no doubt that this is Dodona. The temples and the finely built theatre, one of the largest in Greece, doubtless supplied the place of older and ruder buildings of the time when all Hellas came to hear the voice of Father Zeus in the rustling oak-leaves. At the southeastern end of the lake of Joannina is a hill called Kastritza, crowned by fortifications indicating the existence of just such a city as one would look for to match this great plain and lake. Here are walls about three miles in circuit, and, in a considerable part of their extent, twenty feet high and twelve feet thick, of the finest polygonal work. And yet, so little is known of the history of Epirus, that even the name of this city cannot be ascertained with certainty. Leake thought it was Dodona itself, placing the sanctuary, however, at a little distance, on the spot now occupied by Joannina. Kastritza might still be regarded as a possible candidate for Dodona, inasmuch as sanctuaries in Greece were often at quite a remove from the nearest town, as in the case of Epidauros and Oropos. But the fact that very little is known of any _city_ called Dodona, as well as the fact on the other side that the sanctuary excavated by Carapanos has a small acropolis connected with it, makes it extremely improbable that this great city, Kastritza, was Dodona.