684 PYLUS. £jreat harbour of Navarino. Upon this hypothesis Col. Leake observes, that in itself il is perfectly admissible, inasmuch as there is scarcely a situation in Greece on the low coasts, near the mouths of rivers, where, by the operation of waters salt or fresh, or botli united, some cliange has not taken place since the times of ancient history; and that in the present instance, therefore, there is no great diificulty in imagining that the lagoon may be an ancient harbour converted into a lagoon by an ac- cumulation of sand which has separated it from the sea. But, among the many difficulties which beset this hypothesis, there are two which seem quite fatal to it; one of which has been stated by Mr. Grote and the other by Col. Leake. The former writer remarks that, if the peninsula of Old Nava- rino was the real ancient Sphacteria, it must have been a second island situated to the northward of Sphagia; and that, consequently, there must have been two islands close together and near the scene. This, as Mr. Grote observes, is quite inconsistent with the narrative of Thucydides, which presup- poses that there was only one island — Sphacteria, without any other near or adjoining to it. Thus the Athenian fleet under Eurymedon, on first ar- riving, was obliged to go back some distance to the island of Prote, because the island of Sphacteria was full of Lacedaemonian hoplites (Thuc. iv. 13); whereas, if the hypothesis of Dr. Arnold were ad- mitted, there would have been nothing to prevent them from landing on Sphagia itself. It is true that Xenophon (^Ilell. vi. 2. § 3) speaks of ^(payiai in the plural, and that Pliny (iv. 12. s. 25) mentions " tres Sphagiae;" but two of them appear to have been mere rocks. The objection of Col. Leake is still more fatal to Dr. Arnold's hypothesis. He calls attention to the fact that the French Com- mission observed that the walls of the castle of Old Navarino stand in many parts on Hellenic founda- tions, and that in some places three courses of the ancient work remain, consisting of a kind of ma- sonry which seems greatly to resemble that of Mes- sene. Besides these remains of middle Hellenic antiquity, some foundations are traced of a more ancient inclosure at the northern end of the penin- sula, with a descent to the little harbour of Voidho- Kilid by means of steps cut in the rock. Remains of walls of early date are to be seen likewise towards the southern extremity of the hill, among which is a tumulus; — all tending to prove that the entire peninsula of Navarino was occupied at a remote period of history by an ancient city. This penin- sula could not, therefore, have been the ancient Sphacteria, which never contained any ancient town. The only way of reconciling the account of Thucy- dides with the present state of the coast is to sup- pose, with Mr. Grote and Curtius, that a great change has taken place in the two passages which separate Sphagia from the mainland since the time of Thucydides. The mainland to the south of Na- varino must have been much nearer than it is now to the southern portion of Sphagia, while the north- ern passage also must have been both narrower and clearer. (Leake, Morea, vol. i. p. 401, .seq., Pelo- ponnesiaca, p. 190, seq.; Arnold, Appendix to Thu- cydides, vol. ii. p. 400, seq.; Grote, Greece, vol. vi. p. 427, seq.; Curtius, Peloponnesos, vol. ii. p. 173, seq.; Boblaye, Richer ches, p. 113; Expedition Sdentifique de la Moree, vol. i. pi. vii.) It is unnecessary to relate here the events which followed the erection of the Athenian fort at Pylus, PYLUS. and which terminated with the capture of the Spartans in the island of Sphacteria, as they are given in every Grecian history. The following ex- tract from Col. Leake illustrates the description of Thucydides in the most satisfactory manner: " The level and source of water in the middle where the Lacedaemonians encamped, — the summit at the northern end to which they retired, — the landing- places on the western side, to which the Helots brought provisions, — are all perfectly recognisable. Of the fort, of loose and rude construction on the summit, it is not to be expected that any remains should now exist; but there are some ruins of a signal-tower of a later age on the same site. The summit is a pile of rough rocks ending in a peak; it slopes gradually to the shore on every side, except to the harbour, where the cliffs are perpendicular, though here just above the water there is a small slope capable of admitting the passage of a body of men active in climbing among rocks and difficult places. By this pass it is probable the Messenians came upon the rear of the Lacedaemonians on the summit; for just at the southern termination of the pass there is a passage through the cliffs which border the greater part of the eastern shore of the island, .so that by this opening, and along the pass under the rocks to the northward of it, the Jlesse- nians had the means of passing unseen from the centre of the island to the rear of the Lacedaemo- nians on the summit. Though this hill slopes gradually from its rocky peak to the shore on every side except towards the harbour, it does not admit of a landing at its foot, except in the calmest weather; nor is it easily assailed on any side by land, on account of the ruggedness of the summit, except by the means to which the Messenians re- sorted; so that the words of Thucydides respecting it are perfectly accurate (e/c Siaa.crar)s aTr6Kp7jij.vov Kal 6K Trjs yrjs tJkio'to eTri^uaxo). The southern extremity of the island is rocky, steep, and difficult of access, and forms a separate hill ; in every other part the ground slopes from the cliffs on the side of the harbotir to the western shore, which, though rocky, is low; so that when the weather is calm it is more easy in face of an opponent to land, and to make way into the island on that side than on the eastern shore, where the cliffs admit of an easy access only in two places, one towards the northern end, the other in the middle of the island, where an opening in the cliffs leads immediately into the most level part of it; exactly in the opening stands a small church of the Panaghia. There are also two small creeks adjacent to each other, near the south- ern end of the eastern side of the island, opposite to Neokastro: near these creeks there is a welL The principal source of water is towards the mid- dle of the island, at an excavation in the rock 20 feet deep, which seems to be more natural than artificial; for below a shallow surface of soil, in which there is a circular peristomium of modern masonry, the excavation in the rock is irregular and slanting. In one or two places there are gi'oves of high bushes, and there are low shrubs in every part of it. It often happens, as it did in the seventh summer of the Peloponnesian war, th.at a fire, oc- curring accidentally or of intention, clears the face of the island during the droughts of that season: the northern hill exhibits at this moment recent marks of a similar conflagration." (^Morea, vol. i. 408, seq.) The peninsula of Coryphasium is a precipice on
