Col. William Leake, Travels in Northern Greece (v. 1-4)

§ 3.434 from the shore. This distance accords with that of the heights of Ayan from the sea, as well as with the relation which the same historian has left us of the capture of Pydna by Cassander. Towards the close of the year B.C. 316, Olympias, the mother of Alexander, retired into Pydna with a large army, attended by cavalry and elephants. Cassander being unable to besiege the place on account of the season, encamped around it, formed a circumvallation terminating at either end at the sea, and blockaded the port with his ships. Olympias resisted until the spring, when her supplies totally failing, the horses and beasts of burthen having been devoured, the elephants having died, great numbers of the men having perished of disease and starvation, and others having de-serted, the queen herself attempted to escape by sea but was taken prisoner. The fall of Pydna was followed by the surrender of Pella and Amphipolis to Cassander, who was not long in confirming his claim to the Macedonian throne, by marrying the sister of Alexander, by putting his mother to death, and by shutting up his widow and young son in Amphipolis, where a few years afterwards they were murdered.

§ 3.435 No remains are distinguishable from Ayan or Kitro of the port of Pydna, but the coast has doubtless undergone a considerable change by means of the alluvion of Olympus, and the Pierian mountain. As Methone is named in the Periplus of Scylax —as it was one of the Greek colonies established in early times on this coast, then considered a part of Thrace, and as it was possessed by Athene when she was mistress of the seas, there can be little doubt that it was upon or very near the shore. Elefthero-khori is so advantageous a situation that we can hardly suppose it to have been neglected by the ancients; and it is for this reason principally, that I conceive it to have been the site of Methane, for its distance from Ayan is certainly greater than the 40 stades which the epitomizer of Strabo places between Pydna and Methone. The epitome, however, is not much to be depended upon in this passage, as it names the Haliacmon in the place of the river of Katerina and an Erigon in that of the Haliacmon; whereas the only Erigon known from ancient history was a branch of the Axius, which joined it 80 miles inland.

§ 3.436 As Alorus is stated to have been situated between the Haliacmon and Lydias by Scylax, whose correct enumeration of the other places between the Peneius and Thessalonica entitles him to confidence in this particular, it seems to have stood not far from Kapsokhori, the position of which, opposite to the innermost part of the Thermae gulf, agrees with the description of Alorus given by Stephanus. Perhaps Palea-khora, near Kapsokhori, may have received its name from its preserving some remains of Alorus. Dec. 23.—The wind being “from the Vardar,” according to the local phrase, and consequently fair for the City, I descend over rich hills and through small woods of oaks, and embark at the skala of Elefthero-khori, which is a little more than half an hour distant from the village where the hills terminate, and the great plain begins, which is watered by the Vistritza, Karasmak, and Vardhari, and occupied in great part by the lake of Iannitza, or Pella. Elefthero-khori seems thus to be the natural frontier of Pieria and Bottiaea. Besides the lake of Pella, the maritime part of the plain contains a long succession of lagoons, beginning near Elefthero-khori and reaching nearly as far as Saloniki.

§ 3.437 Of these lagoons, Herodotus has noticed that between the Axius and the Echidorus. They produce an abundance of fish and salt. Of the latter, large heaps are seen near the extremity of the heights of Elefthero-khori on the water-side. A gentle breeze carries us at the rate of five miles an hour along the coast; in an hour and a half we arrive at a projecting cape formed by the alluvion of the Haliacmon. In the time of Herodotus this river was joined by the Lydias, or discharge of the lake of Pella, but a change has now taken place in the course of the latter, which joins not the Haliacmon but the Axius. The Haliacmon itself appears to have moved its lower course to the eastward of late, so that in time, perhaps, all the three rivers may unite before they join the sea. In all the large rivers of Greece, similar changes of direction in the lower parts of their course are observable. The new soil which is brought down by the water, and distributed along the shore by the sea, acted upon by prevailing winds and currents, produces a continual change of obstacles and of relative levels in the maritime plain, which speedily gives a new course to the waters, even in the land which is not of the latest formation. The joint stream formed by the Lydias and Axius is still navigable into the lake, and probably up to Pella, as it was in ancient times. After having passed Cape Karasmak, which is exactly opposite to the outer extremity of Cape Karaburnu, the wind

§ 3.438 heads us a little, and we proceed more slowly than before, but in half an hour, at 6 o’clock Turkish, arrive at a second point, about midway between the Vistritza and Vardhari, where numerous monoxyla belonging to Kulakia are employed in catching shell-fish and octopodhia, while at no great distance from them some large squadrons of wild swans are floating lazily on the gently-swelling surface, and appear to enjoy the fine weather. To the right, the clifis of Karaburnu extend for three or four miles in length. The cape seen from Saloniki is the westernmost point. This conspicuous promontory seems, from Dionysius of Halicarnassus, who consulted some early Greek writers, to have been once the platform of a temple of Venus, said to have been founded by Aeneias. There cannot be a more beautiful situation for such a building. At 6.25 we are opposite the mouth of the Vardhari, which now joins the sea in a bay between the last cape which we passed and another called Kazik-burnu, which we pass at 6.51. It is not improbable that the former was produced by the Lydias and the latter by the Axius, at some period when they fell separately into the gulf. From hence the wind falling and coming more a-head, we do not reach Saloniki till 9.

§ 3.439 CHAPTER 31 MACEDONIA Having been prevented by the occurrence of hostilities between England and the Porte from prosecuting my travels in Macedonia, I can here only offer a few remarks on the comparative geography of those parts of that celebrated province of Greece which I have not visited, illustrated by such an imperfect delineation as oral information can supply. I have already remarked, that between Saloniki and the Vardhari a river called Galliko crosses the road. This is evidently the Echidorus of Herodotus, and as in the Tabular Itinerary, Gallicum is the name of a place situated 16 M.P. from Thessalonica, on the Roman road to Stobi;

§ 3.440 it would seem that in this, as in some other instances which might be mentioned, the ancient name of the river had fallen into disuse, and had been replaced by that of a town which stood upon its banks. Hence also we perceive that the road to Stobi followed the valley of the Echidorus, and not that of the Axius. Next to Gallicum on this route occurred Tauriana, to which the modern Doghiran, or Doiran, corresponds so nearly in name that we can hardly doubt of the identity, the more so as the road thither from Saloniki led in the direction of the course of the Galliko. Nor is the distance of Doiran from Saloniki very different from the 33 M.P. which the Table places between Thessalonica and Tauriana. Doiran has been described to me as a town situated on a small lake which discharges itself into another lake, and that into the Axius. Kilkitj being nearly midway from Saloniki to Doiran, seems to occupy the site of Gallicum. Stobi, upon which the road was directed as being a Roman colony and municipium, and consequently the capital, in those ages, of the north-western part of Macedonia, appears to have been already a place of some importance under the Macedonian kings, though probably it had been greatly reduced by the incursions of the Dardani, when Philip had an intention of founding a new

§ 3.441 city near it, in memory of a victory over those troublesome neighbours, and which he proposed to call Perseis, in honour of his son. At the Roman conquest, Stobi was made the place of deposit of salt for the supply of the Dardani, the monopoly of which was given to the third Macedonia. Some vestiges probably still exist to prove its exact site, although I have not been able to obtain any account of them. According to the Tabular Itinerary, it stood 47 M.P. from Heracleia of Lyncus, which was in the Via Egnatia, and 55 M.P. from Tauriana; and as the sum of the Tabular distances from Heracleia to Stobi, and from Stobi to Serdica, now Sofia, is not greater than the real distance from the site of Heracleia near Filurina to Sofia, we may infer that Stobi was in the direct road from Heracleia to Serdica. Hence its position appears to have been on the Erigon, ten or twelve miles above the junction of that river with the Axius, a situation which agrees with Livy, inasmuch as he describes Stobi as a town of Paeonia, in the district Deuriopus, which was watered by the Erigon. Strabo, indeed, who names three towns of Deuriopus, and adds that they were all situated on the Erigon, has not noticed Stobi, but possibly he may have considered the lower part of that river as in Pelagonia, for the respective confines of these districts were very uncertain, especially after the Roman conquest.

§ 3.442 On the road in the Tabular Itinerary from Tauriana to Stobi occur the following distances and names:—20 M.P. Idomene, 12 M.P. Stena, 11 M.P. Antigoneia, 12 M.P. Stobi; where the Stena or Straits are evidently the pass now called Demirkapi, or Iron gate, where the river Vardhiri is closely bordered by perpendicular rocks, which in one place have been excavated for the road. Idomene consequently stood on the Vardhari, 12 Roman miles below the Demirkapi, and probably on the right bank, as it is included by Ptolemy in Emathia, a province bounded eastward by the Axius, which river may be supposed to have formed in remote times a protection to the Einathian towns from the barbarians of Paeonia and Thrace. These evidences as to the situation of Idomene, although not yet confirmed by the discovery of any ancient remains, already furnish a valuable illustration of Thucydides, whose narrative of the invasion of Macedonia by the Thracians, under Sitalces king of the Odrysss, in the third year of the Peloponnesian war, contains some incidental remarks on the geography of Macedonia, which are among the most useful to be found in the ancient authorities.

§ 3.443 The expedition of Sitalces having been undertaken in concert with the Athenians, who had several subject cities on the Thracian coast, the king was accompanied by Agnon of Athens, as well as by a pretender to the Macedonian throne, in the person of Amyntas, a nephew of Perdiccas the reigning monarch. As the authority of Sitalces extended from the shores of the Euxine and Propontis to the frontiers of Macedonia, where even the Paeonian tribes to the left of the Strymon were subject to him, he was enabled to enter Macedonia with no less than a hundred and fifty thousand men, one third of whom were cavalry. His route from Thrace into Macedonia crossed Mount Cercine, leaving the Paeones on his right, the Sinti and Msedi on his left, and descended upon the Axius at Idomene; from thence he moved by Gortynia, Atalanta, and Europus, into the maritime plain, but instead of proceeding to Cyrrhus and Pella, he turned to the left and ravaged Mygdonia, Crestonia, and Anthemus, without entering Bottiaea, still less Pieria, both of which were within Cyrrhus and Pella. From a previous knowledge of the relative situations of Sintice, Idomene, and Pella, it may confidently be inferred, that the Thracians invaded Macedonia from the plain of Serres, then considered a part of Thrace, and that crossing the mountains which close that plain to the westward, and separate it from the valley of the Axius, they

§ 3.444 entered the latter not far below the straits of Demirkapi passing near Doiran. Hence the mountains at the extremity of the Sirrhaean plain are identified with Cercine, and Doberus appears to have been not far from Doiran. This is in some measure confirmed by Hierocles, who names Dioborus next to Idomene among the towns of the Consular Macedonia under the Byzantine empire. From Idomene the Thracians evidently descended the valley of the Axius, until arriving in the great maritime plain, a little to the eastward of Pella, they turned from thence to the left towards Saloniki. As Gortynia and Europus, which occurred between Idomene and the plains of Cyrrhus and Pella, are placed by Ptolemy together with Idomene in Emathia, it is probable that like Idomene they stood on the right bank of the Axius below that city. Not far above the entrance of the great maritime plain, the site of Europus may perhaps hereafter be recognized by that strength of position which enabled it to resist the invaders. We have the concurring testimony of Ptolemy and Pliny, that this Europus of Emathia was different from Europus of Almopia, which latter town seems from Hierocles, who names Europus as well as Almopia among the towns of the consular Macedonia, a provincial division containing both Thessalonica and Pella, to have been known in his time by the name of Almopia only; and hence we may

§ 3.445 infer that it was the chief town of the ancient district Almopia. As Almopia was one of the earliest acquisitions of the Temenidae it was evidently contiguous to the original seat of the Macedonian monarchy about Berrhoea and Edessa. The other districts were Pieria on the south, Bottiaea on the east, and Eordaea on the west. Almopia, therefore, was on the north; being the same country now called Moglena, which borders immediately upon the ancient capital of Macedonia to the N.E. And this accords sufficiently with the intimation given by Thucydides, that the next conquests of the kings were in Anthemus, Crestonia, and Bisaltia: that is to say, after having obtained all the country to the right of the Axius, they crossed that river, and increased their dominions as far as the Paeones and Sinti; though they were still excluded from the greater part of the sea coast by the Greek colonies of Pieria and Mygdonia, and those which occupied the whole of the Chalcidic peninsula. Homer, whose writings are long anterior to the Argive colony of the Temenidae, alludes only to two provinces beyond the Greek

§ 3.446 cities of Thessaly; lying between them and Peeonia and Thrace—namely, Pieria and Emathia By the first he probably intended the country between the Peneius and Haliacmon, or as Hesiod describes Pieria, around Mount Olympus; by the latter that beautiful region beyond the latter river, and on the eastern side of the Olympene ridge, which protected on all sides by mountains or marshes, at a secure but not inconvenient distance from the sea, gifted with three magnificent positions for cities or fortresses in Verria, Niausta, and Vodhena, blessed with every variety of elevation and aspect, of mountain, wood, fertile plain, running water, and lake, was admirably adapted to be the nursery of the giant monarchy of Macedonia, where its wealth and power might thrive, and increase, until the time came for the augmentation of its territory on every side.

§ 3.447 at the head of the Thermaic Gulf for a colony, which could not venture to establish itself in a maritime site. It appears from Justin, that a portion of Emathia was occupied by the Bryges, who were expelled from thence by the Temenidae; and Herodotus, in stating that the gardens of Midas, who was their king, were situated at the foot of Mount Bermium, seems to show that their situation was around Berrhoea. It is not surprising that Emathia in later times should have had more extensive boundaries than those which Homer may have understood, or that Ptolemy should have advanced its limits to the right bank of the Axius. Polybius, indeed, and Livy, his transcriber in this place, assert, contrary to the tendency of Homer’s notice of Emathia and Paeonia, that Emathia was formerly called Paeonia’; but this may be reconciled by supposing that Emathia, before its colonization, was inhabited by the Paeonian race; whereas Pieria, the other provipce mentioned by Homer, is acknowledged to have been occupied by a Thracian people before its conquest by the Temenidae, whence Orpheus was called a Thracian, and Pydna and Methone in Pieria were described as Greek colonies on the coast of Thrace. It is not easy to reconcile the situation of the Maedi, as indicated in the passage of Thucydides descriptive of the march of Sitalces, with other testimonies as to that people. They there appear to have dwelt, together with the Sinti, to the left

§ 3.448 of the route of the Thracians over Mount Cercine into Macedonia; whereas, according to other authors, as will be seen more fully hereafter, the Maedi occupied the country at the sources of the Axius and Margus (now Vardhari and Morava) as well in the reign of Philip, son of Demetrius, as under the Roman emperors; nor does any author but Thucydides notice any Maedi near Lower Macedonia. Possibly they had become extinct in the course of the two centuries intervening between the reigns of Perdiccas and Philip, or had migrated to Mount Scomius, like the Pieres to Mount Pangaeum, and the Bottiaei into the Chalcidic peninsula. It is clear, at least, that the Maedi could not have occupied any great extent of territory to the south of the route of Sitalces; for in the country which is bounded northward by that line, southward by the ridge of Mount Khortiatzi, eastward by the Strymonic plain, and westward by that of the Axius, and which is a space not more than equal to a square of forty geographical miles the side, we have to place Mygdonia, Orestoma, Anthemus, and Bisaltia. Mygdonia comprehended the plains around Saloniki, together with the valleys of Klisali and Besikia, extending westward to the Axius, and comprehending the lake Bolbe to the east. Crestonia adjoined Mygdonia to the northward; for the Echidorus, which flowed through Mygdonia into the gulf near the marsh of the Axius, had its sources in Crestonia.

§ 3.449 The pass of Aulon, or Arethusa, was probably the boundary of Mygdonia towards Bisaltia, which latter extended to the Sintice northward, and eastward to the Strymon, on the right bank of which it included Euporia. The maritime part of Mygdonia formed a district called Amphaxitis, a chorographical distinction first occurring in Polybius, who seems to divide all the great plain at the head of the Thermaic Gulf into Amphaxitis and Bottiaea, and which is found three centuries later in Ptolemy. The Amphaxii coined their own money; but as no mention of a town of Amphaxia occurs in history, and the silence of Ptolemy is adverse to the supposition, those coins were probably struck at Thessalonica.

§ 3.450 Anthemus appears to have been a city of some importance, as well from the mention made of it in ancient history, as from its having given name, like some of the other chief cities of Macedonia, to a town in Asia. As Thucydides shows its territory to have bordered upon Bisaltia, Crestonia, and Mygdonia, there seems no situation in which it can be placed but to the south-east of Crestonia. Probably it comprehended, therefore, the vale of Laugaza, with the surrounding heights. As to the towns of Mygdonia, which possessed the fertile plain included between Mount Khortiatzi and the Vardhari, their population was undoubtedly absorbed in great measure by Thessalonica on its foundation by Cassander, and it cannot be expected, therefore, that many remains of them should now exist. Nor are the ancient references sufficient to fix their sites. One of them would seem from the inscriptions which I found at Khaivat to have stood in that situation, and others probably occupied similar positions on the last falls of the heights which extend from Khaivat nearly to the Vardhari. One in particular is indicated apparently by some large tumuli, or barrows, situated at two-thirds of that distance. Sindus, ac-cording to Herodotus, was a maritime town between Therme and Chalastra, which latter stood to the right of the mouth of the Axius.

§ 3.451 Altus was a place near Thessalonica, and Philerus and Strepsa appear to have occupied inland situations in the same part of the country. The Crossaea, Crusaea, or Crusis, was sometimes considered a portion of Mygdonia, but is distinguished from it by Herodotus, who describes the Crossaea as comprehending all the maritime country on the Thermaic Gulf, from Potidaea to the bay of Therma, where Mygdonia commenced. The cities of Crossaea were Lipaxus, Combreia, Lisaea, Gigonus, Campsa, Smila, and Aeneia. Of these, Gigonus and Aeneia alone are noticed by later writers: of Aeneia, coins are still extant with a type referring to the reputed foundation of the city by Aeneias after the Trojan war. The situations both of Aeneia and Gigonus may be presumed from their having been situated near two capes, and from there being no promontories worthy of

§ 3.452 notice on this coast, except the little Karaburnu, the great Karaburnu, and the cape of Apanomi, the first of which is so near to Thessalonica, and so inconsiderable compared with the great Karaburnu, that it can hardly enter into the question. Of the two others, the great Karabarnu being about 10 G.M. in direct distance from Thessalonica, seems to be sufficiently identified by this circumstance with the Cape Aeneium of Scymnus, as we learn from Livy, that the town of Aeneia was fifteen Roman miles from Thessalonica. He adds, indeed, that it was opposite to Pydna, which, if it were correct, would imply an error in the distance just stated, as the two conditions are incompatible, and would lead us to place Aeneia and Cape Aeneium at Apanomi, which is nearly opposite to the site of Pydna. It is evident, however, from the order of names in Herodotus, that Gigonus was the more southern of the two capes, and from another fact which occurs in history, that its situation was nearly that of Apanomi. We learn from Thucydides, that in the year before the beginning of the Peloponnesian War, an Athenian force which had been employed against Perdiccas marched in three days from Berrhoea to Gigonus, from whence they proceeded against Potidaea. Gigonus, therefore, was not more than an ordinary day's march from Potidaa, which can hardly be said of Karaburnu;

§ 3.453 whereas, placing Gigonus at Apanomi, we have four days’ march of about twenty miles each, the second to Saloniki, and the third to Apanomi. Stephanus also favours the more southerly situation of Gigonus by intimating that its territory confined upon that of Pallene, which was probably true in later times, when the intermediate places mentioned by Herodotus having fallen to decay, the maritime country was divided between Thessalonica, Aeneia, Gigonus, and Cassandreia. Still, however, I am inclined to defer to Livy’s words adversus Pydnam, so far as to look for Aeneia on the southern rather than the eastern side of Cape Karaburnu, the former better answering moreover to the same author’s 15 M.P. from Thessalonica. In illustration of the great number of towns which in the time of Herodotus occupied Pallene and Crossaea, it may be worthy of remark that this is now considered the most fertile and best cultivated part of Macedonia, and the advantage of the harbour of Apanomi, added to that of a rich surrounding territory, will equally account for that place having retained its pre-eminence both in ancient and modern times. Cissus was a mountain (with a town of the same name) which a comparison of Xenophon and Lycophron seems to identify with Khortiatzi, the former by mentioning it among the mountains which produced beasts of prey, the latter by describing it as a lofty summit not far from Rhaecelus, which appears from Lycophron to have been the name

§ 3.454 of the promontory where Aeneias founded his city. I cannot learn, indeed, that the Frank merchants or consuls, many of whose country houses are on or near Mount Khortiatzi, or that the villages near it, are ever disturbed by the formidable inhabitants of Mount Cissus enumerated by Xenophon, such as the lion, ounce, lynx, panther, and bear; but Khortiatzi is the only high mountain within a moderate distance of the site of Aeneia which we can conceive to have been the haunt of those animals. That the town Cissus was not far from Saloniki, seems evident from its having contributed, together with Aeneia and Chalastra, to people Thessalonica. Although it has been generally found convenient to apply the name Chalcidice to the whole of the great peninsula lying southward of the ridge of Mount Khortiatzi, in consequence of the influence which the Χαλκιδικόν γένος, or people of Chalcidic race, enjoyed in that country in the meridian period of Greek history, the original Chalcidice did not comprehend Crusaea nor the districts of Acanthus and Stageirus, which were colonies of Andros; nor that of Potidaea, a colony of Corinth; nor even Olynthus, or the territory around it to the northward, which was occupied by a people who had been driven out of Bottiaeis, westward of the Axius, in the early times of the Macedonian monarchy, and who, as it appears from their coins,

§ 3.455 were in subsequent times written Βοττιαίοι, and their country Βοττική, to distinguish them from the Βοττεάται, or inhabitants of Βόττεια, or Βόττια, a district and town to the westward of the Axius. The principal possession of the Chalcidenses, in the earliest time of their migration, seems to have been the peninsula of Sithonia, and their port and fortress to have been Torone; from thence they extended their power inland, until at length they occupied all the part of Mygdonia to the southward of the ridges which stretch westward from Nizvoro, together with the Crusaea. The Chalcidenses were indebted to the Persians for the acquisition of Olynthus. Artabazus, on his return from the Hellespont, whither he had escorted Xerxes after his defeat at Salamis, having reduced Olynthus together with some other places in this quarter which had revolted from his master, slew all the Bottisei, who had garrisoned Olynthus, and gave up the place to the Chalcidenses. The Bottisei after this period seem to have been the humble allies of the Chalcidenses, with whom we find them joined on two occasions.

§ 3.456 Spartolus, which was at no great distance from Olynthus to the northward, belonged to them, and was perhaps their capital. Scolus, another town near Olynthus, was of sufficient importance to be mentioned, together with Spartolus, in the treaty between Sparta and Athens, in the tenth year of the Peloponnesian War. Angeia and Miacorus, or Milcorus, are two other names which may be assigned to the interior of Chalcidice. Proof is wanting of there having been a town of Chalcis in any part of the country occupied by the colonists of Euboea. Stephanus, who enumerates five cities of that name, is silent as to any such in the Thracian Chalcidice, and Eudoxus, whom he cites, merely describes Chalcis as the coast lying between Athos and Pallene. Aristotle also, who knew Macedonia well, employs Chalcis or Chalcidice of Thrace, as the name of a district, not a town. Nevertheless, it can scarcely be doubted that before the time when Olynthus became subject to the Chalcidenses, and at length obtained the supremacy over their other towns, there was a chief city of the Chalcidenses where the most ancient of those beautiful coins were struck which have

§ 3.457 the head of Apollo on one side, and on the reverse his lyre with the legend Χαλκιδίων; for that these were the coins of the Thracian Chalcidenses, and not of the Euboean, I can have no doubt, having found several of them in or near the country of the former people, and not one in any other part of Greece, while those of Chalcis in Euboea bearing the eagle and serpent on one side, and a female head on the other, are everywhere extremely numerous. The coins of the Chalcidenses of Thrace were the produce perhaps of the mines of Sidherokapsa, to the possession of which the colony may have been in great measure indebted for its prosperity. The Acanthii may have derived the silver of their fine coins from the same source. The name of the ancient capital of Chalcis I conceive to have been Apollonia, in conformity with that worship of Apollo which is recorded on the coins; for that there was an Apollonia of Chalcidice different from Apollonia of Mygdonia, is clearly shown by Athenaeus and Xenophon: an author cited by the former remarks that two rivers flowed from Apollonia into the lagoon Bolyca, near Olynthus; from the latter we learn that Apollonia was only ten or twelve miles from Olynthus; whence

§ 3.458 it is evident that the Apollonia intended by these two authors was on the southern side of the ridges which intersect the Chalcidic peninsula from east to west. Apollonia of Mygdonia, on the other hand, as the indubitable testimony of St. Luke and the Itineraries demonstrate, stood to the northward of the same mountains, on the direct road from Thessalonica to Amphipolis, by the pass of Arethusa. In fact, the ruins of this Apollonia are still to be seen exactly in that line to the south of Pazarudhi, at a place preserving the ancient name in a corrupted form, and nearly at the proportionate distance between Thessalonica and Amphipolis indicated by the Itineraries.

§ 3.459 The distance of the Chalcidic Apollonia from Olynthus, stated by Xenophon, and the circumstance of its not being in the direction of Acanthus, which his narrative also indicates, combine to place it at or near Polighero, which, like Apollonia of old, is now the chief town of the Chalcidice. Spartolus would seem from the transactions related by Thucydides not to have been so far from Olynthus as Apollonia was, which is somewhat confirmed by Isaeus, who describes it as Spartolus of the Olysia or territory of Olynthus. It was in consequence of the complaints of the Apollonians of Chalcidice and of the Acanthii, that the Lacedemonians sent an army against Olynthus, which, after losing two of its commanders, succeeded in the fourth campaign, B.C. 379, in reducing the city to submission. When Olynthus became a part of Chalcidice, it is not surprising that its maritime situation should have caused it gradually to eclipse the ancient capital. It was particularly after the Peloponnesian War, that it became one of the greatest cities in Greece, made successful war with Macedonia, took Pella from Amyntas, and was of such importance to the league which it headed, that when

§ 3.460 reduced by Philip, it was followed in its submission by thirty-two other towns. Nor can there be any difficulty in conceiving, that when Chalcidice had been between three and four centuries subject to Rome, the received chorography of the country should have been different from that which prevailed in the time of its freedom. Ptolemy appears to have divided the whole peninsula into two parts, Chalcidice and Paralia; for thus I read the word which in all the printed copies of his works is Paraxia. Paralia contained all the maritime country between the bay of Thessalonica and Derrhis the Cape of Sithonia: thus the western coast of Sithonia was at that time included in Paralia, and the eastern in Chalcidice, together with Acanthus, the entire peninsula of Acte, and all the maritime country adjacent to the Strymonic Gulf, as far north as Bromiscus, with the exception of Stageira. Livy mentions an Antigoneia of Crusis between Aeneia and Pallene: it was perhaps one of the towns of that coast noticed by Herodotus, which had been repaired by one of the Antigoni. By Ptolemy it is surnamed Psaphara, probably in order to distinguish it by this adjunct from another Macedonian Antigoneia on the road from the Stena of the Axius to Stobi. As Chaetae and Moryllus are placed by Ptolemy together with

§ 3.461 Antigoneia Psaphara in Paralia, and their names do not occur in the periplus of the fleet of Xerxes, they were places perhaps in the bay of Thessalonica, between the city and Cape Aeneium, or Karaburnu. Ptolemy has not noticed either this cape or the city Aeneia. On the road from Thessalonica to Apollonia of Mygdonia, a Melissurgi occurs in two of the Itineraries: this place still preserves its ancient name in the usual Romaic form of Melissurgus, and is inhabited by honey-makers, as the word implies. It was 20 or 21 M.P. from Thessalonica. The third, or Jerusalem Itinerary, seems to have followed a different line from Apollonia to Thessalonica, leaving probably the summit of Khortiatzi to the right, whereas the two others seem to have passed on the opposite side of it. But both roads evidently crossed that mountain, the Romans having seldom allowed such an obstacle to divert them from their direction. The modern barbarians, on the contrary, have found a circuit by the pass of Khaivat, which avoids the ridge entirely, more convenient for the caravan route to Constantinople; and in consequence of this change, they follow the northern shore of the lakes, instead of the heights on the southern side of them, which was the direction of the ancient road. These routes reunite in the pass of Arethusa, now called that of Besikia, and by the Turks the Rumili Boghazi, as being one of the most important defiles on this great line of communication. In the list of Greek bishoprics as arranged by the emperor Leo the philosopher, Lete, conjointly with Rendina,

§ 3.462 was the see of a bishop subordinate to the metropolitan of Thessalonica, and styled ό Αητκ και PevrivtK. Rendina having been at or near the pass of Besikia, it would seem that Lete was not far from thence, which agrees with the intimations derived from the ancients as to the position of Lete, the lake of Besikia having been in Mygdonia, and Lete being named by Ptolemy next to Apollonia of Mygdonia On the other hand, it seems difficult to find a place for Lete in the Mygdonium valley, if Stephanus is right in asserting the existence, of a town Bolbe, since in that case this valley seems sufficiently occupied by Bolbe, Apollonia, and Anthemus. Possibly Mavrovo may be the site of Lete, or Sokho, if we place Ossa at Lakhana. I shall now offer a few remarks on Paeonia, a geographical denomination, which prior to the Argolic colonization of Emathia, appears to have comprehended the entire country afterwards called Macedonia, with the exception of that portion of it which was considered a part of Thrace. As the Macedonian kingdom increased, Paeonia was curtailed of its dimensions on every side, though the name still continued to be applied in a general sense to the great belt of interior country which covered Upper and Lower Macedonia to the N. and Ν. E., and a portion of which was a monarchy nominally independent of Macedonia until fifty years after the death of Alexander the Great.

§ 3.463 The banks of the Axius seem to have been the centre of the Paeonian power, from the time when Pyraechmes and Asteropaeus led the Paeonians to the assistance of Priam, down to the latest existence of the monarchy. When the Temenidae had acquired Emathia, Almopia, Crestonia, and Mygdonia, the kings of Paeonia still continued to rule over the country beyond the straits of the Axius, until Philip, son of Amyntas, twice reduced them to terms, and they were at length subdued by Alexander, after which they were probably submissive to the Macedonian sovereigns. The coins of Audoleon, who reigned at that time, and who adopted after the death of Alexander the common types of that prince and his successors, prove the civilization of Paeonia under its kings. Diodorus informs us that Cassander assisted Audoleon against the Autariatae, an Illyrian people, and that having conquered them, he transported 20,000 men, women, and children, to Mount Orbelus, whence we may infer that regal Paeonia lay between the Autariatae and Mount Orbelus.

§ 3.464 From a comparison of Appian and Strabo, as well as from an incident in the life of Alexander the Great, to which I before adverted, it is evident that the Autariatae bordered to the eastward upon the Agrianes and Bessi, to the south upon the Maedi and Dardani, and in the other directions on the Ardiaei and Scordisci. Upon the whole, therefore, it is consistent with history and the general chorographv of the countries to the northward of Macedonia, to conclude that regal Paeonia comprehended all the central and most fertile part of the more extended Paeonia, and that it was situated above, the straits of the Axius, occupying all the countries on the upper branches of that river, with the exception of those districts towards the sources of the Erigon, which had been united with Upper Macedonia. Bylazora, although described by Polybius as the chief city of Paeonia, was not the capital of the kingdom, perhaps on account of the inconvenience of its proximity to the Dardani. The royal residence, as we learn from Polyaenus, was situated on the river Astycus, evidently the

§ 3.465 Vravnitza, or river of Istib, which, next to the Erigon, is the greatest of the tributaries of the Axius. Of the tribes on the Thracian frontier of Paeonia which were subject to Macedonia, as early at least as the reign of Philip, son of Amyntas, I have already shown reasons for believing that the Odomanti occupied the whole of Mount Orbelus from above the Stena of the Strymon near the modern Demirissar to Zikhna inclusive, where they confined on Mount Pangaeum. Thus their north-western portion lay to the right of Sitalces as he crossed Mount Cercine: and their general situation accords with the description of Thucydides, according to whom they dwelt beyond the Strymon to the north; that is to say, to the northward of the Lower Strymon, where alone the river has such an easterly course as can justify the historian’s expression. It is observable, that the Panaei, whom Thucydides couples with the Odomanti, are stated by Stephanus to have been a tribe of the Edones. These authorities agree, therefore, in confirming the situation of the Odomanti just indicated. Between Meleniko and Petritzi, above Demirissar and the Strymonic straits, the main branch of the Struma, or Strymon, is joined by a large tributary named Strumitza, upon which stands a town of the same name, situated a day's journey beyond Petritzi, in the road from Serres to Velesa. Strumitza

§ 3.466 I am disposed to identify with the ancient Astraeum, to which Philip sent his son Demetrius, when he gave directions for his death to Didas, governor of Paeonia, though it was not there that Didas executed his orders, but at Heracleia (Sintica) having invited Demetrius thither on the occasion of a festival during which poison was administered to the prince. Didas, in return for his services, was favoured by Perseus when he came to the throne; and hence we find Didas, at the beginning of the Persic war, commanding a body of 3000 men, who consisted of Paeones, Paronei, Parstrymonii, and Agrianes. The Paeonian monarchy was then extinct, and its territory, with the exception probably of a part occupied by the Dardani, had been united to the Macedonian kingdom; from which fact, and the names of the people who were governed by Didas, it seems evident that the Paeonian province, at that period of the Macedonian monarchy, comprehended the valleys of the Upper Strymon and Upper Axius, with the intermediate mountains, and including the country of the Agrianes, who dwelt near the sources of the Strymon. Astraeum seems to have been a central position in this country, and the provincial seat of government. The site of Strumitza was well adapted to be the chief fortress of such hardy tribes: its strength is particularly attested by Nicephoros Gregoras, when he was sent in A.D. 1326 to Skopia

§ 3.467 on a mission to the Kral of Servia from the Emperor Andronicus the elder: he relates, that after having travelled half a night and one day from a ferry of the Strymon, he arrived at Strumitza, a fortress so lofty that the men on the walls looked from the plain like birds. Ptolemy, in assigning to the Aestraei Doberus as well as Aestraeum, shows those two places to have been at no great distance from one another; which is true, on the supposition that Aestraeum or Astraeum, was at Strumitza, and Doberus near Doghiran.

§ 3.468 Strymon, Struma, Astraeus, and Strumitza, seem to be all dialectic modifications of some original word of Macedonia, meaning river. The name Astraeus, as I have already remarked, was applied to the lower part of the Haliacmon, and Vistritza seems to be nothing more than the corruption, or modern Bulgaric form of Astraeus. The town of Strumitza, therefore, as well as ite predecessor Astraeum, I conceive to have taken its name from the river on which it stood, as being the position of greatest importance upon that great branch of the Strymon, and the natural capital of its valley. The name implies the lesser Strymon. In the north-western part of Paeonia, the principal place under the Romans, as I before hinted, was Stobi. From this point four roads are drawn in the Tabular Itinerary 1 One proceeded northwest to Scupi, and from thence north to Naissus, a position on the great south-eastern route from Viminacium on the Danube to Byzantium,—the second north-eastward to Serdica 100 M.P. southeast of Naissus, on the same route —the third

§ 3.469 south-eastward to Thessalonica, and the fourth south-westward to Heracleia; the last forming a communication with that central point on the Via Egnatia, or great Roman road from Apollonia to Thessalonica, leading through Stobi from all the places on the three former routes. In the valleys which are watered by the confluents of the Upper Axius, and which were traversed by the two roads branching northward from Stobi, there are three considerable towns, of which the modern names sufficiently resemble the ancient, to lead at once to a presumption of identity. These are Skopia, Velesa, and Istip. In regard to the first there can be no question, as the name which in Ptolemy and Hierocles is Σκούποι, is still found in the same form in the history of Nicephorus Bryennius at the beginning of the twelfth century, though Skopia, the present Greek form, is used by Anna Comnena at an earlier period, and at a later by Nicephorus Gregoras, who has exactly described Skopia as situated on the banks of the Axius, which was then, as it is now, called Βαρδάριον. It may be objected, perhaps, that the number of M.P. between this place and Stobi is much greater in the Table than the real distance, from Skopia to the supposed site of Stobi; but as the Table often fails in the accuracy of its numbers, particularly in excess; and as there can be no doubt as to Scupi, we are fully authorized in this instance in preferring to that authority the

§ 3.470 evidence derived from the agreement of the supposed site of Stobi with all the other requisites derived from ancient testimony. The identity of Velesa, or Velesso, with Bylazora, besides the similarity of sound in modern Greek pronunciation, is supported by the circumstantial evidence of history. Advantageously placed on the Upper Axius, in the midst of the fertile country watered by that river and its branches, and on the edge of the mountains which here separated Paeonia from Illyria, Bylazora was well qualified by situation to become “the greatest city of Paeonia,” while the situation of Velesa exactly illustrates the further remark of Polybius, that Bylazora was near the passes leading from the Dardanice into Macedonia; that is to say, through Paeonia, for which reason it was taken and fortified by Philip, son of Demetrius, as a barrier against the Dardani, previously to his descent into Greece in the last year of the Social War. As the Paeonian power was then extinct, it was probably from the Dardani that Philip took the city, and it may have been upon the ground of their temporary possession of the western part of Paeonia that the Dardani, on the division of Macedonia into four regions at the Roman conquest, claimed Paeonia of the Senate of Rome, as having formerly belonged to them.

§ 3.471 It may be thought, perhaps, an objection to this position of Bylazora, that the name is not found in the Tabular Itinerary on the road from Stobi to Scupi, although Velesa lies exactly in that line: I am inclined to believe that it does occur under the very corrupted form of Anausara. Bylazora is again mentioned in the history of that eventful year, B.C. 168, when Perseus, not long before the battle of Pydna, endeavoured to obtain the mercenary services of 20,000 Gauls, who in the expectation of being employed by him, had advanced in equal numbers of horse and foot as far as Desudaba in Maedica. Perseus with the view of drawing them into Macedonia, moved with half his army from the river Enipeus in Pieria to Almana on the Axius,-which was 75 miles distant from Desudaba. Having ordered supplies to be in readiness on the intended route of the Gauls, he sent a messenger to Desudaba, requiring the Gallic army to advance to Bylazora, and inviting their chiefs to visit him at Almana, where he gave them to understand by the messenger that he had prepared some rich presents for them, by these means hoping to obtain the services of the Gauls without farther expence. But they were not a people to be so duped: they refused to move beyond Desudaba until they should receive the stipulated present of ten pieces of gold for each horseman, five for each foot soldier, and 1000 for each chief, and such an advance of treasure being more than (he avaricious monarch could consent to advance, the Gauls returned to the Danube, ravaging the parts of Thrace through which they passed.

§ 3.472 As Perseus had left a most formidable enemy in Perrhaebia and Pieria on this occasion, we cannot suppose that he advanced farther up the Axius than was absolutely necessary. Almana, therefore, was probably below the straits of Demirkapi, between the Stena and Idomene, and Desudaba having been 75 M.P. distant from thence, on the direct route to the Danube by the valley of the Margues, will fall at or near Kumanovo, on one of the confluents of the Upper Axius. This indeed is nearly the greatest southern extent that can be given to Maedica towards Paeonia and the respective situations of Desudaba, Bylazora and Almana, as just indicated, will then perfectly agree with the circumstances stated by the historian, and the more so as Perseus had undertaken to furnish the Gauls with provisions, and as Bylazora, the intermediate station, was in the middle of the most fertile part of Paeonia. Maedica thus placed accords also with the remark of Strabo, that the Maedi bordered eastward on the Thunatae of Dardania, for the Dardani extended to Skopia,

§ 3.473 and the Thunatae therefore we may suppose to have been a tribe of the Dardani, possessing the modern Katzaniki. If the southern boundary of the Maedi was near Kumanovo that people must have possessed the sources of the eastern branch of the Morava, or Margus, and its upper valleys, in one of which Vrania, or Ivorina, has very much the sound of Jamphorina, the capital of the Maedi, which was taken by Philip, son of Demetrius, in the year B.C. 211. On this occasion the king, whose design it was by previous intimidation to keep his troublesome neighbours quiet, while he should be employed in Greece against the Aetolians, had first assaulted Oricus and Apollonia, from whence he marched into Pelagonia, took a city of the Dardani, which had facilitated the entrance of that people into Macedonia on the side of Pelagonia, and then passed through Pelagonia, Lyncus, and Bottiaea, into Thessaly. The situation of the Maedi is farther illustrated by the fruitless excursion of the same king of Macedonia to the summit of Mount Haemus in the vain expectation of beholding from thence at once the Adriatic and Black Sea, the Danube and the Alps. He arrived at the foot of the mountain in seven days from Stobi, passing through the country of the Maedi; after a laborious ascent of three days, and a

§ 3.474 descent on his return of two, he rejoined his camp in Maedica; thence made an incursion into the country of the Dentheletae for the sake of provision, re-entered that of the Maedi, where he received the momentary submission of a place named Petra, and from thence returned into Macedonia. It seems evident from the number of days’ march, that the mountain visited by Philip, and named Haemus by the historian, could have been no other than that which by two of the best authorities is denominated Scomius, or Scombrus, being that cluster of great summits between Ghiustendil and Sofia, which sends tributaries to all the great rivers of the northern part of European Turkey; for this, in fact is the most central point of the continent, and nearly equidistant from the Euxine, the Aegaean, the Adriatic, and the Danube. The Dentheletae would seem from the circumstance mentioned by the historian to have bordered on the Maedi towards the south-east. Haemus itself was chiefly occupied by the Bessi, who from their fastnesses defied the power of Rome until the reign of Augustus, and according to Pliny extended as far to the southward and eastward as the Nestus.

§ 3.475 Astibon, the third of the ancient towns of Paeonia, the names of which still subsist in a corrupted form, was on the road from Stobi to Serdica. It is now by the Turks called Istib, and stands exactly on that line, at a distance from each of those ancient sites which, as well as our present imperfect geographical materials admit of judging, sufficiently corresponds with the numbers in the Table. It occupies probably the site of the capital of the kings of Paeonia, which appears from Polyaenus to have been situated on a river named Astycus. The modern Djustendil or Ghiustendil equally accords with the Pautalia of the Table, and the situation of Ghiustendil at the sources of the Strymon is remarkably in accordance with the figure of a river god, accompanied by the legend Στρύμων on some of the autonomous coins of Pautalia, as well as with the letters ΕΝΠΑΙΩ, which on other coins show that the Pautaliotae considered themselves to be Paeonians, like the other inhabitants of the banks of that river. On another coin of Pautalia the productions of its territory are alluded to, namely, gold, silver, wine, and corn, which accords with Ghiustendil.

§ 3.476 In the reign of Hadrian, the people both of Pautalia and the neighbouring Serdica added Ulpia to the name of their town, probably in consequence of some benefit received from that emperor. This title in the case of Pautalia would seem at first sight to warrant the supposition, that it was the same place as Ulpiana, which, according to Procopius, was rebuilt by Justinian, with the name of Justiniana Secunda, and the modern name Ghiustendil lends an appearance of confirmation to this hypothesis by its resemblance to Justiniana. But there is an insurmountable objection to this hypothesis. Both Procopius and Hierocles notice Ulpiana and Pautalia as distinct places, to which we may add, that Ptolemy as well as Hierocles ascribes Ulpiana to Dardania, which seems never to have extended far to the eastward of Scupi, or Skopia. A further argument against the identity arises from a comparison of the Tabular Itinerary with a passage in Jornandes, who relates that Theodemir being at Naissus, sent a body of troops, under his son Theodoric, through Castrum Herculis to Ulpiana, where Castrum Herculis is evidently the same as the Ad Herculem of the Table, which was on the road from Naissus to Scupi, and consequently very far to the westward of Ghiustendil. Ulpiana, or the Second Justiniana, therefore, was probably situated in one of the valleys of the branches of the Morava, northward of Skopia, but not in the route from Scupi to Naissus, as it is not mentioned in the Table.

§ 3.477 From a place named Hammeno, which was in that road, at an uncertain distance from Scupi, but evidently in a N.W. direction, and probably not very far from it, there was a branch to the westward leading to Lissus, now Lesh, near the mouth of the Drilon. Of the ancient places on this route, Theranda bears some similitude in sound to the modern Prisrend, though it must be admitted that the proportion of distances on the route, even without any addition for the interval between Scupi and Hammeno, would place Theranda farther westward. As Ulpiana does not occur either on this road nor on that from Scupi to Naissus, it lay probably between them in the country to the northward of Prisrend, which is watered by the western branch of the Morava, perhaps at the modern Pristina. Beyond Theranda the route to Lissus seems to have fallen into the valley of the Mathis, where I should be disposed to look for Gabuleus, Crevenia, and the other names in the route of the Table; for on the more direct line occurred the “solitudes of Scordus,” which mountain being described incidentally by Livy as lying in the way from Stymbara to Scodra, and again as giving rise to the Oriuns which flowed through the lake Labeatis to Scodra, seems clearly to have comprehended the great summits on either side of the Drilon, where its course is from east to west. The important position of Scupi at the debouche from the Illyrian mountains into the plains of Paeonia and the Upper Axius, caused it in all

§ 3.478 ages to be the frontier town of Illyria towards Macedonia. There is no evidence of its ever having been possessed by the kings of Macedonia or of Paeonia. Under the Romans it was ascribed to Dardania, as well in the time of Ptolemy, as in the fifth century, when it was the capital of ducal Dardania. The position “ad fines,” which in the Tabular Itinerary stands at 35 M.P. beyond Anausara (Bylazora) on the road from Scupi to Stobi, would seem to indicate that the Romans had there fixed the boundaries of Dardania and Macedonia, and consequently that they had given Bylazora to Dardania, thus yielding in part to the demand which the Dardani had made, on the establishment of the tetrarchy of Macedonia after the conquest by Aemilius. Scupi was probably seldom under the complete authority of Constantinople. In the reign of Michael Paleeologus it was wrested from the Emperor by the Servians, and became the residence of the Kral. Here Nicephorus Gregoras met the court of the αρχών των ΎρφάΧΧων, as he learnedly denominates the Kral, whose successor (in 1342) afforded protection and hospitality to John Cantacuzenus when he retired before Apocauchus. By the treaty afterwards made between Cantacuzenus and the king of Servia, the latter obtained a temporary authority over a great part of Macedonia, the Romans, as they called themselves, giving up to him Zikhna, Pherae

§ 3.479 (Serres), Meleniko, Strumitza, and Kastoria, and retaining Servia (the town), Berrhoea, Edessa, Gynaecocastrum, Mygdonia, and the towns on the Strymon, as far as the district of Serres and the mountains of Tandessano. It may be not unworthy of remark, that in the histories of Anna Comnena, Gregoras, and Cantacuzenus, several other existing names occur, as:—on the Illyrian frontiers, Dibra (Δευρη), Velesso (Βελεσσός), Prillapo (Πρίλλαπος), Morava (Μοράβα), and Pristino (Πρίστηνος), which last Cantacuzenus describes as a small town without walls (κώμη ατείχιστος):— towards Thessaly, Servia (Σερβία), Kastri (Καστριον), Lykostomi (Λυκοστόμιον), and Platamona (Πλαταμών πόλις παραθαλασσία):—to the eastward, Rendina (Ρεντινα) and Dhrama (Δράμα), besides Zikhna (Ζίχνα) and Meleniko (Μελενίκος);—and near Edessa and Berrhoea, Ostrovo (Όστροβος), Notia (Νότια), and Staridhola (Σταριδολα), with some others which might probably be found by

§ 3.480 a diligent search. Soskos (Σωσκος) appears from Anna Comnena to have been between the lake of Ostrovo and Servia. I shall here subjoin, as containing a compendious view of Macedonian geography, the edict for the division of Macedonia into four regions, issued by the authority of the Roman Senate B.C. 167, the year after the conquest. It was read at Amphipolis to the assembled Macedonians by L. Aemilius Paullus, and then explained to them in Greek by Cn. Octavius the praetor:— The first would embrace the district between the Strymon and the Nessus, and in addition, beyond the Nessus to the east, the forts, towns and villages which Perseus had held, with the exception of Aenus, Maronea and Abdera, and beyond the Strymon to the west the whole of Bisaltica together with Heraclea, which district the natives call Sintice. The second canton would be bounded on the east by the Strymon, exclusive of Sintice, Heraclea and Bisaltica; and on the west by the Axius, including the Paeonians, who dwelt to the east of the Axius. The third division would be the district enclosed between the Axius on the east and the Peneus on the west; the Bora range shuts it in on the north. This canton was increased by the addition of the part of Paeonia which extends westwards beyond the Axius; Edessa and Beroea were assigned to this division.

§ 3.481 The fourth canton lay on the other side of the Bora range, bordering Illyria on the one side and Epirus on the other. Aemilius then designated the capital cities where the councils were to be held in the different cantons; Amphipolis was fixed for the first, Thessalonica for the second, Pella for the third, and Pelagonia for the fourth. There the councils for each canton were to be summoned, the tribute deposited, and the annual magistrates elected....He gave permission to those cantons whose frontiers were contiguous to those of the barbarians to maintain armed forces on their borders. [Livy 45.29] By this celebrated decree the Macedonians were called free, each city was to govern itself by magistrates annually chosen, and the Romans were to receive half the amount of tribute formerly paid to the kings, the distribution and collection of which was probably the principal business of the councils of the four regions; for none but the people of the extreme frontiers towards the barbarians were allowed to defend themselves by arms, so that the military power was entirely Roman. In order to break up more effectually the national union, no person was allowed to contract marriage, or to purchase land or buildings, but within his own region. They were permitted to smelt copper and iron on paying half the tax which the kings had received; but the Romans reserved to themselves the right of working the mines of gold and silver, and of felling naval timber, as well as the importation of salt, which, as the Third Region only was to have the right of selling it to the Dardani, was probably made for the profit of the conquerors on the shore of the Thermaic Gulf. No wonder that the Macedonians compared this division of their country and interruption of the mutual intercourse

§ 3.482 between the several parts of it to the laceration and disjointing of an animal body, or that they should have been ready to join a few years afterwards in the revolt of Andriscus. The historian then remarks:— The first section includes the Bisaltae, a nation of warriors living on the other side of the Nessus and around the Strymon and contains many special kinds of fruit and minerals and the city of Amphipolis, which is so conveniently situated, commanding as it does all approaches from the east. Then again, the second division comprises the populous cities of Thessalonica and Cassandrea and also the rich corn-growing district of Pallene. Facilities for sea-borne traffic are afforded by numerous harbours: some at Torone under Mount Athos, and at Aenea and Acanthus, others facing Thessaly and Euboea, and others again easily accessible from the Hellespont. The third canton includes the famous cities of Edessa, Beroea and Pella, the warlike tribe of the Vettii and also a large population of Gauls and Illyrians who are devoted to husbandry. The fourth canton is peopled by the Eordaei, the Lyncestae and the Pelagones; adjoining them are Atintania, Tymphaeia, and Elimiotis. The whole of this strip of country is cold and unkindly and difficult of cultivation, and the character of the peasants corresponds to that of their country. Their barbarian neighbours make them still more ferocious by sometimes familiarising them with war, and in times of peace introducing their own rites and customs. [Livy 45.30] After all that has been offered on the situation of the districts and places here mentioned, scarcely any explanation is necessary beyond a reference to the Map at the end of this volume.

§ 3.483 Macedonia Prima comprehended all the former possessions of Perseus in Thrace to the eastward of the Nestus, with the exception of the three principal maritime cities between that river and the Chersonese; and it contained all the country between the Nestus and Strymon probably as far as the sources of those rivers, together with Sintice and Bisaltia, to the right of the Strymon. Amphipolis, the capital of this region, is justly described as the great defence of Macedonia from the eastward; and we have an illustration of the allusion made by the historian to the mines of Mount Pangaeum, which Amphipolis commanded, in the numerous existing silver coins of the time of the tetrarchy bearing the head of the Amphipolitan deity Diana Tauropolus, with an obverse representing the club of Hercules within a garland of oak, and the legend Μακεδόνων πρώτης: these coins were evidently struck at Amphipolis. The second Macedonia comprehended all the country between the Strymon and Axius, except the Sintice and Bisaltia, and extended as far towards the sources of both rivers as the boundary of the Macedonian kingdom had reached. The eastern turn of the Strymon below Serres shows at once why the Sintice and Bisaltia were excepted from the countries between the Strymon and Axius,

§ 3.484 and placed in the first instead of the second Macedonia. The second region was the richest and most populous of the four, no part of Macedonia being comparable in fertility and other advantages to Mygdonia, Chalcidice, and the three contiguous peninsulas, where the historian especially notices the productive Pallene, and the convenient havens of Torone and Athos. The name Aeneia, which Livy attaches to the harbour of Athos, is not found I believe in any other author, nor is it certain to which of the ports of Acte it applies. The third region is very clearly described as bounded by the sea, by the Axius, and by the Peneius, on three sides—as containing the cities Pella, Edessa, and Berrhoea, and as extending northward to Mount Bora, where its limits were such that it was the only one of the three provinces not in contact with the Barbarians, the nearest of whom were the Dardani. Hence Mount Bora, which is not noticed by any other author, appears to have been the summit northward of Vodhena, now called Nitje, one of the chief links in the Olympene or eastern chain, of which the others are Bermius, Pierus, Olympus, Ossa, and Pelium. This great ridge terminates in a northerly direction at the fork of the Erigon and Axius. Here, therefore, the Third Region terminated, and thus Paeonia was interposed between the northern extremity of the Third Region and the Illyrians. The Paeonians to the westward of the Axius, were an exception to the definition otherwise given of the extent of the Third Region, as they lay beyond Mount Bora to the N.W.;

§ 3.485 and hence the particular mention of the Paeonians in the edict, which refers undoubtedly, like History at this period in general, not to the original Paeonia in its fullest extent, but to the limited portion of it which had formed a monarchy, until, about a century before the Roman conquest, it was incorporated with Macedonia. The portion of Paeonia separated from the rest of that country, and attributed to the Third Macedonia, while the remainder of it was attached to the Second, was situated on the lower Erigon around Stobi, and this city was decreed to be the place of deposit for the salt, sold to the Dardani, the monopoly of which was given to the Third Macedonia. To the fourth division remained every thing beyond the district of Stobi to the west and southwest, as well as all the country beyond the crest of the Olympene range, as far as Illyria and Epirus. The historian enumerates the following districts as composing it: namely, Pelagonia, Lyncestis, Eordaea, Elimiotis, and Atintania, where he has obviously omitted Orestis, which lay between Atintania and the rest of Upper Macedonia. Thus it appears that the Fourth Macedonia extended nearly to Berat and Tepeleni, and included Konitza. To the southward its limits were nearly those of the modern districts of Grevena and Trikkala, where Upper. Macedonia confined upon Upper Thessaly.

§ 3.486 The warlike nation of the Vettii, mentioned together with Pella, Edessa, and Berrhoea, as forming part of the third region, are evidently the Bottiaei, and this allusion to them, showing that they were still of some importance, accords with the apparent date of their coins. The Chalcidic Bottiatae had probably been long extinct. Numismatic evidence, therefore, concurs with Polybius and Strabo, in showing that the great maritime plains after the Roman conquest were divided between the Bottiaei and Amphaxii. The chief place of the latter, as we learn from Ptolemy, was Thessalonica, that of the former probably Alorus. The strength of the “bellicosa gens” of Bottiaea was derived from the intersection of rivers and marshes, natural defences which have maintained in the same position some unmixed Greeks to the present day in the midst of surrounding Bulgarians and Turks. There exists a silver tetradrachm with the legend Μακεδόνων δευτέρας, coined probably at Thessalonica, of which city no money bearing its name has been found more ancient than the Roman empire. The silver of the mines of Nizvoro may have supplied the coinage of the Second Macedonia. No silver money of the Third and Fourth Macedonia has been discovered, nor is it known that either of those regions possessed mines. The only other coin bearing an allusion to the tetrarchy, besides those I have mentioned, is a small one in bronze, so rare that I met with only one. It is inscribed Μ. τετάρτης, and presents on one side the Dioscuri on horseback, on the other the head of Minerva. But there is another coin of the Fourth Macedonia in the Caesarian Museum, bearing a head of Jupiter, and on the obverse the common

§ 3.487 Macedonian type of a club within a garland of oak, with the legend Μακεδόνων τετάρτης. We are to infer from Livy that these were struck at Pelagonia. The rarity of all the money of the Macedonian tetrarchy, except that which was coined at Amphipolis, is to be attributed to the shortness of its duration. Only 18 years after the edict of Amphipolis, Andriscus, calling himself Philip, son of Perseus, reconquered all Macedonia but was defeated and taken in the following year, by Q. Caecilius Metellus, after which the Macedonians were made tributary, and the country was probably governed by a praetor, like Achaia after the destruction of Corinth, which occurred two years afterwards, B.C. 146. From that time to the reign of Augustus, the Romans had the troublesome duty of defending Macedonia against the people of Illyria and Thrace, and during that time they established colonies at Philippi, Pella, Stobi, and Dium.

§ 3.488 CHAPTER XXXII. FOURTH JOURNEY PREVYZA, March 1809.—Since my visit to this place in 1805, the Porte having found that very little accrued to it from Prevyza and the other ex-Venetian places, after paying the expences of the residents and their little garrisons, was tempted to sell them to Aly Pasha, as a malikiane or farm for life, for the sum of 800 pnrses, thus virtually violating the treaty of 1800, by which the Sultan engaged to maintain these places in their Venetian laws and privileges, and liable only to a fixed duty on commerce and land, to be paid to a resident Bey; instead of which, he now gives

§ 3.489 them over to a man whom he cannot control, and who has already treated them with every kind of vexation. Prevyza has been the principal sufferer. Its alliance with the French when the place was taken by assault in 1798, furnished the Vezir with an excuse for extortion and cruelty, which has lasted ever since, and the population is now reduced to less than half its number at that period. In 1807, when the war broke out between Russia and the Porte, the Prevyzans were obliged to labour at an entrenchment across the peninsula near two miles in length, to construct which the Pasha sent for men from all parts of his territory, as far as Katerina beyond Mount Olympus, and gave them nothing but a ration of koromana, or black bread. In this light soil, with few palisades to support it, this entrenchment is already falling to ruin. Afterwards his new fortress and serai were constructed in the same manner, by an angaria or compulsory labour. Since his bargain with the Porte, Aly considers himself absolved from the necessity of keeping any measures with the Prevyzans, giving away their land to his Albanians, sending whole families to people new tjiftliks in unhealthy situations, and quartering his soldiers upon those whom he has allowed to remain. But notwithstanding the forced labour, which has given him materials as well as construction at little cost, Prevyza has been very expensive to the Vezir: Albanian soldiers must be paid, the fortress armed, and the palace furnished from his own pocket, and the maritime situation has required the aid of some small vessels, which

§ 3.490 could only be obtained by purchase. Parga, moreover, though he has paid for it, he has not much prospect of obtaining. Nevertheless, his bargain is an excellent one, on account of the military importance of the places, and the facilities which they give him in making further acquisitions in Tzamuria. The only part of the ancient privileges of Prevyza now remaining, is its system of taxation. The present revenue of the Crown is as follows:— Livaria (fisheries) - 22,000 Dhekatia (tithe of produce of land) - 10,000 Dogana (custom-house) - 15,000 Monopoly of bread - 2000 Monopoly of tobacco - 5000 Monopoly of butchers’ meat - 3000 Monopoly of raki (brandy) - 6000 Monopoly of playing cards - 500 Monopoly of τό πέραμα (the ferry to Punta) - 1500 Monopoly of sealing tanned leather - 800 Monopoly of gunpowder - 1000 Monopoly of statiri, the public weighing - 800 Monopoly of the retail of oil - 1200 τό Νόμιστρον, a capitation tax upon cattle fed in the pastures of Prevyza - 600

§ 3.491 Total - 69,400 All the articles are farmed except the dogana, which is collected by a person named by the Vezir. Under the Venetians the same practice obtained, and the different heads of revenue were sold by auction every six years. The amount was then about 18,000 or 20,000 piastres a year. The increase has been chiefly owing to the debasement of the coin, and to the great increase of late years in the produce of the fisheries. For the same reason, the livari of Vutzintro, which, united with some other branches of revenue, produced, in 1805, only fifty-five purses, is now alone let to the same συντροφιά, of which the bishop of Ioannina is the head, for eighty purses. The revenue of Vonitza consists of the same articles, and amounts in value to 20,000 piastres a year. That of Parga to 10,000. So that deducting the latter, the Vezir has given 800 purses for a life annuity of 200 at the age of sixty, and having as good a prospect of keeping his head upon his shoulders as any man in his station in Turkey. The excavations which have been made at Nicopolis for the purpose of obtaining materials for the fortress and palace of the Vezir at Prevyza, have not led to any interesting discoveries, partly it seems because the city having been hastily built, more in the Roman than Greek manner, little more was found than fragments of walls formed of tiles, mortar, and broken stones, unfit for the purpose of the masons, and which did not much encourage them to persevere. By order of the Vezir, the sculptured pieces were set apart, but the only

§ 3.492 result has been two inscriptions, which have been placed at the gate of the Serai. One of these is a dedication to Augustus by the Mallotae, or people of Mallus, a great maritime city of Cilicia; the other, which from the form of the letters seems to be of a later period of the Empire, was in honour of a praetorian praefect of Macedonia, who was tribune of the first legion surnamed the Minerria Pia Fidelis, procurator of the corn of Epirus, procurator of the province of Pontus and Bithynia, and procurator of the dismissions of the Emperor. The monument was raised agreeably to a decree of the council (of Nicopolis ) by Mnester, a freedman of the Emperor, in token of his gratitude to the praefect, of whom he was the assistant. Several of the letters in the inscription require to be supplied, particularly in the praefect’s name, which seems to have been Lucius Ofellius Maius.

§ 3.493 March 15.—At 3 p.m. we make sail for Vonitza in a large sakkoleva belonging to the Vezir, which has a covered deck and cabin, and is rigged with two high latine sails and a small sail aft. A fresh maestrale soon carries us past Punta; and along the side of a woody plain, on the southern shore of the Gulf of Prevyza, beyond which towards the lake of Vulkaria are heights clothed with larger trees. On the northern side of the gulf the coast is higher, and forms a peninsula in which is a hamlet of five or six houses, called Skafidhaki, and below it a lagoon, communicating, by a small opening, with the sea, and having a fishery which belongs to Arta. Having crossed the Gulf of Prevyza to its southeastern extremity, I land at the ruins of Anactorium, for such I shall venture to denominate a circuit of Hellenic foundations, surrounding a rocky promontory between two bays, and following the crest of some heights which embrace a little plain on the shore of the smaller or southern bay, where a small church of Aios Petros gives name to the place. The distance of these ruins from Punta accords exactly with the forty stades placed by Strabo between Actium and Anactorium.

§ 3.494 The circumference of the town was less than two miles. In most parts foundations only are traceable; but to the southward there are remains of several towers: the interior wall of the acropolis in part subsists also, and between it and a marsh in the middle of the plain are some foundations, apparently those of the peribolus of a temple. From the vestiges of a gate at the eastern angle of the town, a walk of an hour across the heights which fall north-eastward to the commodious little harbour of St. Mark, leads me in a south-easterly direction to the limeni, or limni of Vonitza, from whence there are two roads to the town; that to the right by a stone causeway along the southern side of the limni, at the foot of a steep hill covered with brushwood, from the foot of which issues a body of water so large as to render the limni almost fresh. I follow the northern shore, passing for a mile through a wood of bramble, myrtle, mastic, dwarf oleaster, and ilex, to Myrtari, at the entrance of the limni, from whence I cross in the ferry-boat to Vonitza, to the house of Kyr Κ., with whom I lodged on my former visit to this place. My host, in conjunction with Kyr G. of Prevyza, has lately purchased of the Vezir for one year, for 95,000 piastres, the farm of the salt works and fisheries of Arta, in which is included the sole right of fishing throughout the gulf, except within the district of Prevyza.

§ 3.495 March 16.—A strong easterly wind prevents me from leaving Vonitza until 3.30 p.m., when, embarking in the sakkoleva with Messrs. K. and G. and our vice-consul of Prevyza, we follow the coast for two hours as far as a bay between the capes Volimi and Khaliki, where a paleokastro called Ruga induces me to land. It is a Hellenic fortress, about half a mile in circumference, surrounded on three sides by a lake about 500 yards wide, beyond which are heights covered with thick woods. The lake communicates at its two extremities with the sea in seasons of rain, but at present is separated by a narrow beach. The walls are more or less preserved in the whole circuit; and in one or two places there are some foundations of rectangular towers of the ordinary kind. Near one of them a piece of wall, which is standing to the height of twelve feet, is a complete specimen of the second or polygonal kind of Greek masonry: the stones being of various shapes, accurately fitted to one another without cement, and none of them rectangular. On the summit of this wall are a few other masses which seem to show that the upper courses of the walls were of more regular masonry. Perhaps these, as well as the towers, were repairs or additions to the original work. The inclosed space, which is one of the very few ploughed spots on this woody shore, is not much above the level of the sea. From Ruga we follow the coast, with a favourable breeze, and soon pass Cape Khaliki, which is a sandy point projecting from a low woody cape. The wind falls as we approach Nisi, and in consequence of the turn of the coast becomes contrary. So that it is eight in the evening before we anchor opposite to this tjiftlik of the Vezir, which is built in the usual manner of this part of the country;— that is to say, the dwellings inclose a quadrangle

§ 3.496 into which all the doors and windows open; thus the outside presents only bare walls, and serves as a fortress against the robbers of Xeromero and Valto. Whenever there is any suspicion of danger, the cattle and other stock are collected at night within the square, the only entrance into which is a large strong gate. Some boats of Kyr K. of Vonitza having been very successful in spearing in the hay of Nisi, we have a plentiful supper of fish, and keep out the cold with a large fire, though not without some inconvenience from smoke, as there is no chimney. My companions sleep upon a carpet by the fire; I spread my mattress in the further part of the cottage where is a raised floor made of a few planks. The tjiftlik of Nisi possesses some cornfields among the velani oaks which cover the heights between it and Cape Khaliki; in the other direction there is a marshy bottom, grown with ashes, oaks, and other trees, and frequented by wild hogs. Those who hunt them say that the animal generally makes directly at the man who wounds him, and if the hunter is not very alert, the hog by his strength and quickness seldom fails to inflict a most severe wound with his short thick reverted tusk. No wonder the ancients without fire-arms held these animals in so much respect. The forests extend from hence, with a few intervals only of cultivated country, as far as Lefkfidha, and besides swine, abound with three species of deer, the ελαφι, πλατώνι, and ζαρκάδι, which by the description of them are the red deer, the fallow deer, and roe.

§ 3.497 March 17.—The wind being contrary for Lutraki, and the passage round the inner curve of the bay of Nisi being muddy for loaded horses, we cross the bay in boats, and ride up to Palim, or Balim Bey. This operation, as we have an escort of thirty Albanians besides our own baggage, takes us till 10.30. Palim-bey is another farm of the Vezir, having a few kalambokki and corn-fields and flocks belonging to it, in the midst of the woods. It differs only from Nisi in having a larger house, by way of a serai or pyrgo, and a garden of fine lemon and cypress trees attached to the house, with a few katyvia on the outside of the quadrangle. We had intended to pass the last night here had the wind been more favourable. The level which separates the farm from the sea is covered with large plane trees, together with some oaks, both common and velani, wild pears, paliuria, and other shrubs. In the most marshy parts ashes are numerous; this tree, which is not very common in Greece, is generally called by its ancient name Melia, but is here known by that of Fraxo, an abbreviation of the Latin Fraxinus. The hills behind the tjiftlik are clothed with oaks, velanidhies, and pimaria; beyond them, three miles from Palim-bey, formerly stood Aghius Saranda, and beyond it Tersova and Vustri. Beyond a peaked snowy summit, 2 hours to the southward of the summit of the mountain of Pergandi, was the monastery Robo, reckoned 4 hours from Palim-bey.

§ 3.498 These and twenty other villages or monasteries in this part of Acarnania are now deserted and ruined. On the western side of the mountains were Zaverdga, Sklavena, Runisi, Synodhi, and Bogonia, formerly all large villages, but now reduced to insignificance, or totally deserted. Having remained at Palim-bey until our Albanians have dined, we leave it at 12.20, and at Ι cross a stream shaded by large planes, and flowing from the southern side of the summit of the mountain of Varnaka. After passing over a root of the same hills, we proceed along the side of them until, at 1.20, they slope into the narrow harbour of Lutraki, where on its western side are a Dogana and Kula surrounded with a wall; from the head of the harbour we proceed through a narrow gorge, called Dhafnies, from the numerous bay trees which grow here, and which are mixed with hushes of Paliuri and wild Kharub; the hills on either side are covered with thick underwood. This is a strong pass, and like those of Amvrakia and Kekhrenia, may be considered one of the gates of Acarnania. At Lutraki, and in a halt for our Albanian infantry by the way, we lost 15 minutes. Having entered the valley, we begin at 2.20 to skirt the marsh, on the opposite side of which I passed when coming from Amvrakia to Lutraki, on June the 19th, 1805. Little streams ooze from the foot of the hills on our right, and flow into the marsh. At 2.45 we are opposite to the end of the marsh, and to the hollow on the slope of the opposite mountain,

§ 3.499 through which leads the road to Amvrakia. Leaving this to the left, we mount the hills through ravines shaded with olives, and at 3.55 arrive at Katuna, at the house of Mr. George Mavromati. Katuna is situated on a fertile range of hills, which are divided by an elevated valley from Mount Bumisto; on one side of this valley, beyond a hill, is the river which, taking its rise near Komboti, Joins the sea between Palim-bey and Lutraki. The heights of Katuna extend southward, with a little inclination to the east, for a distance of about 12 miles. This ridge consists of hard limestone, covered with a stratum of fertile soil, which feeds numerous sheep and oxen, and has some intervals cultivated with wheat and barley. These and the produce of the velanidhies scattered in the woods, once supported a considerable population in the towns of Katuna and Makhala, which are now mere villages, and in several subordinate places now abandoned. The Aetolian plains, though still cultivated to a considerable extent, and better peopled than Acarnania, have declined nearly in the same proportion, and among the Beys of Vrakhori, some of whom formerly derived 3000L a year from their landed property, not one has now a third of that income. In Katuna there remain not more than forty inhabited houses; seventy were abandoned in the course of the last year, chiefly in consequence of the excessive expence attending the quartering of Albanians, who all pass through this derveni in their way to or from Aetolia, or the south-western parts of Acarnania.

§ 3.500 This grievance has particularly pressed upon them since the death of Yusuf Aga, the Valide Kiayassy, when Aly obtained the Mukata of Karlili, and immediately sent his Albanians into the country. He is now making his first visit in person. His chief object is to substitute his own Albanians for the Greek armatoll, who under the command of their captains were in the service and pay of the villages. As soon as his intentions were known, many of the armatoli fled into the islands, and returned from thence as robbers. The individual among them whose enterprize and knowledge of the country renders him at present most formidable, is named Dhrako Griva, first cousin of the Katziko-Iannis, two celebrated characters of the same stamp, whom the Vezir succeeded in destroying. Griva began his career at an early age, like most of these heroes, by entering into a band of robbers, to whom he recommended himself by his activity, hardiness, and cruelty. It was his practice to tie every Musulman who fell into his hands, or any unfortunate Christian who had given him offence, to a tree, to be fired at by his followers as a mark. Having rendered himself the terror of the villages of Karlili, and long defied the efforts of the Vezir as Dervent Aga, he was at length, at the Pasha’s suggestion, taken into the service and pay of the district as captain of armatoli, to keep the country clear of thieves.

§ 3.501 He was afterwards disgraced by the same influence, and superseded by Katziko Ianni, because he could not, or would not (as it is said), murder Mitjo Mavromati of Katrina for His Highness. Griva had then no other resource than to enter into the Russian service in the Islands, from which he passed into that of the French, and in both has succeeded in tormenting the Vezir by continual depredations on his territories. Varnaka, a village once of 400 families, hut now deserted, is at present the principal resort of the thieves. To the spoliation of the kleftes is to be added that of the Albanians sent against them; these, together with the similar effects of the collection of troops in 1807 against the Russians at Lefkadha, and of those now assembled to observe the French, have almost depopulated the entire country around Mount Bumisto, or between the Ambracian gulf and Leucadian sea. The Vezir, when he halted here the day before yesterday, lodged at the house of the son of the very Mavromati whom he had formerly put to death. Mitjo was a man of considerable property, and much beloved in Acarnania, where he long acted as agent of Kurt Pasha in the management of the armatoli and police of this province.

§ 3.502 His friendship with Kurt was a crime in the eyes of Aly, which Mitjo’s riches rendered unpardonable. Conscious of the injury he had done to the family, the Vezir ordered the house to be searched before he entered it, though when he announced to Mavromati at Prevyza that he intended to lodge with him in passing through Katuna, he pretended never to have heard that his old friend Mitjo, as he called him, was dead. After dining at Katuna he went forward to Makhala, accompanied by 1500 Albanians, whose pay is four months in arrear. Scarcely any chieftain but Aly could take such a liberty with these men, as there is nothing on which Albanians are so sensitive. In the meantime, rove κνβφνάη, he quarters them on the places which he passes through, and thus they can live without pay, which they have no doubt of receiving in the end, that being a point in which Aly dares not deceive them. March 18.—The view from Katuna, though confined by the mountains to the west and east, commands some distant objects through the openings to the north and south—namely, Mounts Olytzika and Tzumerka in the former direction, with the mill above Arta, which was one of my former stations. To the southward beyond Aetolia appear the great summits of Voidhia and Olono in the Morea. Our escort of thirty Albanians from Prevyza is joined by ten more from Vonitza by direction of Kyr Κ., who, as Hodja-bashi of that place, has the direction of these troops within his own district.

§ 3.503 The necessity of this reinforcement shows how insecure the country is supposed to be beyond the range of the Albanian muskets. Half a mile below the lowest houses of Katuna, a little on the left of the road to Makhala, is the upper extremity of a Hellenic fortress which occupies the slope of the ridge of Katuna on its eastern side. The valley into which it descends is a continuation of that which we followed yesterday coming from Lutraki, being the same as that I described on the 18th of June, 1805, as included between the parallel ridges of Amvrakia and Katuna. The existing remains consist of foundations of regular masonry belonging to an acropolis which surrounded a theatre-shaped piece of ground at the head of a water-course: vestiges of the town walls are seen also on the descent towards the valley, and I am told there are others quite at the foot of the mountain. The city, therefore, was large as well as important by its position, which commanded the principal passage from Epirus through Acarnania into Aetolia.. It is supposed by the learned of this part of the country to be Conope, because there is a small village, situated a few miles to the south-west, named Konopitza. Conope, however, was certainly beyond the Achelous, in Aetolia; and Konopitza no more indicates the position of Conope than Amvrakia does that of Ambracia. Our road continues to follow the crest of the ridge over heights remarkable for their variety of form, and for many immense circular cavities, covered within with trees, and at the bottom of some of which are deep pools of water.

§ 3.504 The trees are chiefly pirnaria, and the rocks, as generally in this part of Acarnania, a hard yellow limestone, or marble, which is very handsome when polished. The heights are uncultivated, but produce a fine herbage, affording an excel-lent pasture for sheep. At 3 p.m., after having ridden 50 minutes from the Paleokastro, the monastery of Agrilio is a mile and a half to the left, on a point of land on the western side of the lake of Valto or Amvrakia, opposite to that part of the mountain of Kekhrenii which I descended on the way from that village to Amvrakia. The projection on which Agrilio stands, is an abrupt termination of the ridge beginning on the eastern side of Lutraki, and upon which stand the villages Sparto, Amvrakia, and Stanu. Below Agrilio is the narrowest and deepest part of the lake. In dry summers nothing remains but a circular pool in that part, all the rest being dry or muddy. It is the opinion at Katuna, that by means of a few canals of drainage, and at the expence of about 60 purses, all but the pool near Agrilio might be made capable of bearing maize or any other kind of grain in abundance. Some parte of the edges of the lake when dry are now cultivated in that manner, as I witnessed below Amvrakia on my former journey. We now leave the few dispersed houses which form the village of Konopitza, or Konopina, two miles on the right, and at 3.35 pass through the ruins of the village of Anino, from whence came the family of that name which is now one of the principal in Cefalonia.

§ 3.505 On the opposite slope of the mountain of Kekhrenia were formerly Alpitza and Makri, from which latter came the Makris of Zante. At 3.50 we halt for a quarter of an hour at a large well of ancient construction resembling another which I observed close to the walls of the Paleokastro of Katuna. Several others, all probably works of the ancient Greeks, are said to exist in this ridge, which is totally deficient in superficial sources. All the larger houses at Katuna and Makhala are provided with cisterns for collecting rain water. At 4.45 we pass Papadhates, or Papalates, standing on the crest of the ridge, and now containing only a few cottages, and there arrive in sight of a valley to the westward, included on one side by the mountains which protrude from Bumisto towards Tragamesti, and on the other by the ridge which, trending westerly from that of Makhala, borders the great plain of the Achelous to the north-west. In an opening between the two ranges the sea appears. On the opposite side of the valley, at the foot of the hills towards Tragamesti, are the villages of Babini, Makhera, and Khrysovitzi, lying in that order from North to South. At 5.15 we arrive at the highest point of the ridge, where stands a ruined windmill, visible from Katuna, and which is a conspicuous object to all the surrounding country. Immediately below it begin the houses of the village of Makhala, which are dispersed over a slope falling towards the plain of the Achelous.

§ 3.506 It may be a question, whether the lake of Agrilio, or the marsh between Katuna and Lutraki, was the scene of a transaction of the year B.C. 391, which is related by Xenophon. The Achaeans, who were in possession of Calydon, finding themselves greatly annoyed by the Acarnanians, who were assisted by some Athenians and Boeotians, craved the succour of the Lacedaemonians, who sent Agesilaus, with two morse and some allies, to join the Achaeans. Agesilaus, previously to entering the hostile territory, sent a message to Stratus, threatening to destroy the whole country unless the Acarnanians quitted their alliance and joined that of Sparta; but they disregarded his menaces, retired into their cities, and drove their cattle to a distant part of the country. Agesilaus then entered Acarnania, and destroyed every thing within his reach; but marched not more than ten or twelve stades each day, by which mode of proceeding, at the end of fourteen or fifteen days, he had thrown the Acarnanians so much off their guard, that many of them resumed their rural employments. He then made a sudden march of 160 stades in one day to a lake surrounded by mountains, where the greater part of the cattle of the Acarnanians was collected, and thus captured a great quantity of horses, oxen, and sheep, besides men, all which he sold the next day. In the evening he was attacked by the Acarnanians and forced to descend from his position on the heights, into a plain and meadow on the bank of the lake, from whence there was only

§ 3.507 a narrow and difficult outlet across the mountains. By this pass Agesilaus attempted to retreat on the following day, but the Acarnanians had occupied the mountains on either side of it, from whence their light armed annoyed their opponents by missiles, easily escaping into shelter when pursued by the cavalry or hoplitae. The Acarnanian hoplitae, with the greater part of their peltastae, were posted on the summit of the mountain to the left of the enemy's line of march; and this mountain happened to be the more accessible of the two to horsemen and hoplitae. Agesilaus, therefore, after sacrificing, during which operation many of his troops were wounded, ordered an advance upon the height to his left. All the hoplitae who had arrived at 15 years beyond the age of puberty ran forward, preceded by the cavalry, and followed by Agesilaus himself with the remainder of the forces. In this manner they reached, and slew or put to flight the Acarnanians on the declivity of the mountain, by whom they had been annoyed. Nor did those on the summit of the ridge wait for the encounter, though the peltastae had slain some of the horsemen and horses of the enemy in the ascent. The loss of the Acarnanians on this day was about 300. Agesilaus then continued to ravage the country, and even presented himself at the request of the Achaeans before some of the cities, but none sur-

§ 3.508 rendered to him; and as the autumn was advancing, he decided upon retiring from Acarnania, replying to the Achaeans, who requested him to remain, so long at least as to prevent the Acarnanians from sowing their corn, that the more they sowed the more inclined they would be to peace. His retreat through Aetolia, adds the historian, was by passes through which it would have been impossible for any numbers to have found their way, had the Aetolians, who hoped for his assistance in the recovery of Naupactus, been desirous of preventing him. Although the Valto, or lake of Agrilus, may seem better to deserve the description of a λίμνη in the present time of the year than that between Katrina and Lutraki, there is probably little difference in their dimensions in the season of the expedition of Agesilaus. Both are surrounded by mountains, that of Valto more closely; but for that reason it was less adapted to the assemblage of the Acarnanian cattle than the lake of Lutraki, which has at all times a greater extent of pasture around it. The latter had also the advantage of being farther removed from the frontier of Aetolia, whereas the southern extremity of the lake of Agrilio is not many miles from Stratus and the Achelous. The ravine therefore by which I ascended from the marsh of Lutraki to Katuna seems to have been the defile in which the Acarnanians opposed the Spartans.

§ 3.509 In this case the hill of Katuna itself was the position of the Acarnanian hoplitae and peltastae, which was taken by the bold charge of the Spartans. There is indeed another opening conducting from the marsh of Lutraki, which leads towards Amvrakia; but as this would have cauried Agesilaus farther from the frontier, and would have obliged him to march along one side or the other of the lake of Agrilio, both difficult routes, and that to the westward defended by the fortified town near Katuna, it is not probable that he should have ventured into so hazardous a situation. It seems evident, moreover, that he retreated by the same route by which he had arrived, that is to say, into the plain of Aetos; for on this side of the ridge of Katuna lay the principal extent and the more fertile parts of Acarnania, through which he had made his fifteen marches, probably in various directions, as convenience or plunder prompted. The last day’s march of 160 stades, by which he surprised the Acarnanians, would seem from the distance to have been begun from a positiou on the Achelous. It is almost unnecessary to point out how perfectly the geography of Aetolia justifies the remark of Xenophon, as to the difficulty which Agesilaus would have found in retreating through that country to Calydon, had the Aetolians been adverse to him, his only routes being along one side or other of the lake of Apokuro, or through the passes of Zygos, or if he entered the maritime plains from Acarnania, along the borders of the lagoons of Anatoliko and Mesolonghi.

§ 3.510 March 19.—The ruined windmill behind Makhala commands a magnificent prospect. As at Katuna, I recognize to the north Mount Olytzika, near Ioannina, and to the south-east the mountains Voidhia and Olono, in the Morea, to which are here added, the Sandameriotiko of Elis and the Mavra Vuna near Dyme. But the principal objects are the Aetolian plains, with their noble river and lakes, the positions of Stratus, Thermos, and Conope, the great summits called Arakhova and Vitina, and Mount Rigani, near Naupactus. In the midst of the basin which lies to the westward of the ridge of Makhala, and which is surrounded on the other sides by Mount Bumisto, by the mountains towards the sea coast, and by that which slopes on the opposite side to the right bank of the Achelous, rises an insulated height, surrounded by Hellenic wails, on the western point of which stands a monastery, called Porta, properly ή Παναγία στην Πόρταν. In the lower part of the inclosure a ruined tower is conspicuous, having eight courses of regular masonry still standing, and oa either side of it some walls of polygonal masonry, which have an appearance of a more remote antiquity than the tower. Makhala, to judge by the ruins disposed over the hill on which it stands, was once a considerable town; there are now not more than 50 families. It is said to be the healthiest position in the interior of Kariili. Katuna, although nearly as high, does not enjoy such good air in summer, because the day breeze which draws through the opening of Lutraki passes over the marshes.

§ 3.511 In the winter and spring it suffers from cold, and in the end of the summer and autumn from the vicinity of the Valto of Agrilio and the effluvia of the mud which is continually stirred up by the wild hogs. Makhala, Katuna, Zavitza, Tragamdsti, Katokhi and Stamna, now reduced to inconsiderable villages, were all flourishing towns in the time of Kurt Pasha. Katuna was considered the richest and most polished. Many families have migrated from these places to the islands. From Makhala to Skortus takes us an hour and twenty minutes, with our Albanians on foot. At Skortus there are only two families left. The rains of the village stand at the foot of a small height, surrounded with an ancient Greek wall, of which there remain in some places two or three courses of regular masonry. From hence I proceed to the summit of the hill of Lygovitzi, which rises immediately above Skortus, in search of some ruins which a woman of the latter place, probably for the sake of getting rid of us, described as a μεγα κάστρον, but where I find nothing, after an ascent through a thick wood of velanidhi oaks, and over difficult rocky paths, but the ruins of four or five churches among the trees, and on the summit some remains of a small castle, apparently of the same date as the churches. I have since been informed, however, that among the woods on the south-eastern face of the hill, the walls are traceable of an ancient Hellenic city, which, by its position relatively to Oenope at Anghelo-Kastro, would seem to have been Metropolis.

§ 3.512 The persons left in charge of the monastery of Lygovitzi, on perceiving our approach, locked it up and fled into the woods, taking us for thieves. My Albanian Palikaria had not only climbed up the hill on foot, but found their way into the building before I could reach the summit on horseback. The monks have not occupied the house since the country has been tormented by the frequent incursions of robbers from the islands: its landed property is considerable, but not so large as that of Vlokho. There is a neat small church, a cistern, and several cells. The prospect from the monastery, which stands just below the summit, repays the trouble of ascending the hill. To the south are seen Kastro Tornese, and the plains of Elis and Achaia; to the north-eastward the mountains of Agrafa, from whence extends the hilly country which terminates in the plain of Vrakhori, bounded on the S.E. by the lake and hills of Apokuro, and the great ridge of Zygos or Aracynthus. Beyond the mountains of Apokuro are seen those of Kravari, ending to the south in Mount Rigani over Epakto. The great mountain Viena, which hides Velukhi, has its whole range extended before us. As well from its vicinity to the capital Thermits, as from its being the most extensive and central summit of Aetolia, this mountain seems exactly suited to the Panaetolium, which Pliny names as one of the mountains of Aetolia. No other author, I believe, has alluded to it, although one of the highest and greatest of the ridges of Greece.

§ 3.513 At the foot of the steep woody descent of the mountain is a large deep perennial lake, abounding in fish and wild fowl, and discharging a copious stream into the Achelous, the broad bed of which is separated only from the lake by a narrow plain. The junction of this discharge of the lake with the Achelous occurs a little below that of the river anciently called Cyathus, which flows from the lake of Vrakhori and joins the main river opposite to Anghelokastro. Two miles below the union of the discharge from the lake of Lygovitzi, the Achelous is joined by a second tributary on the right bank proceeding from a marsh, and between them on the same side by a third smaller stream. The. broad white bed of the Achelous, from which it derives the modern name Aspro, is widest between the site of Stratus at Surovigli and the lake of Lygovitzi. On the right bank, between Surovigli, the extremity of the mountain of Kekhrenia and the northern side of the lake of Lygovitzi, is a triangular plain, once the chief support of Stratus, but now almost entirely uncultivated, as it always has been in the memory of the present Acarnanians, though nothing inferior in natural fertility to that of Vrakhori. Having dined upon some provisions brought with us from Makhala, very much in the manner of the kleftes, whom we are taken for, we descend through woods of velanidhi, among which are a few corn-fields, and some horses belonging to the monastery, into the direct road from Skorttis to Prodhromo,—pass through some large flocks of sheep, which are attended by Vlakhiote Karagunidhes of Mount Pindus, and arrive at Prodhromo

§ 3.514 at half-past 4 p.m. The distance from Skortus is an hour and a quarter. Prddhromo stands exactly opposite to Khrysovitzi as Skortds does to Babini. In the valley between the two former, and about a mile in a direct line from Prodhromo, rises an insulated hill, the summit and one side of which are enclosed with the remains of Hellenic walls, the summit forming a separate inclosure. It appears to have been nothing more than a small fortified κώμη, like that at Skortus, and very inferior in importance to the cities which stood at Porta and near Katuna. Anciently it would seem that every village in Acarnania was walled, whence we may infer that their insecurity was almost as great as it is now. It may easily be conceived, indeed, that between the sea pirates of the adjacent islands, who were at all times ληίστορες άνδρες, and the semi-barbarous tribes of the Epirotic and Aetolian mountains, their position was one of continued vigilance. Its effects, however, had not injured their character; for Thucydides speaks favourably of the Acarnanians, and they seem not to have altogether degenerated when compared with other Greeks. The Proestos of Prodhromo, who is upwards of seventy years of age, remembers when there were 60 or 70 houses in his village: there are now only six. It is situated just on the skirt of the woods which occupy all the range of hills from Lygovitzi to where they terminate in the plains towards the mouth of the Aspro.

§ 3.515 The air is said to be very healthy. In the valley, and on the slopes adjacent to this side of it, the Prodhromites cultivate wheat and barley, and they gather vallonea, gall-nuts, and a seed or berry used in dyeing, called μερζόσπορος, on the hills. The soil is a dark-coloured friable mould, like that of the greater part of Acarnania. The grinia wheat, is sown from November to January, whenever there is an interval of dry weather favourable to it:—the dhiminio from the 10th of February to the 25th of March (old style.) If the spring be very dry this yields no more than 3, 4, or 5, to 1; but it usually gives 10, while the grinia never more than 6 or 7. The latter would perhaps yield as much as the dhiminio if it were carefully cleared of weeds, but this is seldom done in Greece. Barley is sown in the same season as grinia: the harvest is in the middle of June (old style.) Upon the kind of weather which leads to a good harvest they have this proverb— Χαρά στα χριστόγενα στεγνά, Τα φώτα χιονισμένα Με την λαμβρήν βρεχούμενην Τα μπάρια γιομισμένα “Joy to a dry Christmas, a snowy Epiphany, and a rainy Easter, then the barns will be filled.” The Sicilians say—Gennaro sicco borghese ricco. Prodhromo, like all the smaller villages of Karlili, is a Spahilik, and pays two fifteenths of the crop to the Spahi. The rest belongs to the Prodhromite, who is his own labourer, and pays all the expences of cultivation.

§ 3.516 His condition, which from this statement would seem to be independent, is quite the reverse. The Hodja-bashi, or Proestos of Tragamesti, or of any other place upon the coast where the Prodhromite carries his corn or other produce for sale, prevents him from communicating with the islanders, who would give him a good price, and forces himself in as an intermediate purchaser, at a much lower: hence the current price of wheat here at present is not more that 31 piastres the kilo of 22 okes, which is equivalent to about 3s. 6d. the bushel. The velanidhi, which being procured for the trouble of gathering would be a great advantage to the peasant, is monopolized in the same manner by the Proesti, who give him for the small sort, called γαμάϋα, 20 piastres the milliaja of 1000 lire grosse Venete. One of my companions tells me that he has himself lately bought a quantity from the primates of Karlili for 37 piastres and sold it for 50. The large inferior kind of velanidhi, called κάχλα, sells at 12 piastres the milliaja. Kikidhi, or gall-nuts, are sold by the gatherers for 15 paras the oke, and merzdsporo the same. The surrounding hills upon which these productions are gathered abound in stags, deer, roebucks, and wild boars, as well as in jackals, which make a dismal howling at night. Another disadvantage of which the Prodhromites, in common with the other small villagers of Acarnania, complain is, that although surrounded with pasture, they are unable to have any flocks, which all belong to the Vezir and his sons, or to rich Turks, or to other persons who pay the Vezir for permission to feed their flocks in this part of

§ 3.517 the country, all which are in the care of Vlakhiotes, or of Albanians from Mount Pindus. But even this oppression, or that which prevents the industrious man from employing his means in the most advantageous manner, or from carrying the fruit of his labour to the best market, is less grievous than the direct taxes and extortions which often deprive him at one blow of his scanty earnings. The kefaliatiko, or kharatj, is 7 piastres for every male above ten years old, in which is included half a piastre for the expences of the Proestos of Tragamesti, the chief town of the district, or of the persons whom he sends here to collect it The vostina, which is paid to the Spahi, is a capitation tax of 60 paras for every married, and of 30 paras for every unmarried man. Τα χρεη, or the dues, as the taxes are denominated collectively, amount at Prodhromo to near 500 piastres a year for each family, a large part of which consists of the share of an arbitrary imposition laid upon the village by the Proestos of Tragamesti in acquittance of the demand which the Vezir makes upon Karlili, to defray the expence of troops, or journeys, or wars, or upon any other pretence, and for the amount of which he is supposed to be accountable to the Porte, but does not account to any one. The Hodja-bashis assemble and divide the burthen among the different districts, according to their population. Each of them afterwards adds to the sum the expences which he himself incurs, or pretends to have incurred, in journeys to attend the Vezir, or for entertaining and lodging Turks and soldiers, or for horses in the public service,

§ 3.518 or upon any other plausible pretext. The imposition upon the village being as arbitrary as that of the Vezir upon the district, the Proestos enriches himself quickly, unless he should happen to be a man of extraordinary humanity, of whom there cannot be many in a country where honour and honesty are so little encouraged. In the territory of the Vezir they are particularly rare; for it is his usual policy to appoint the worst men to be primates, that he may make them disgorge when they are full of plunder; after which he often allows them to begin their extortions anew. In the smaller villages where the chief is styled protoghero, or chief alderman, he arranges in like manner the mode of payment of the khrei among the families, and generally in the Vezir's territories, or at least in those where his authority is firmly established, one person is charged with this office, or at most two in the large towns, whereas, in the Elefthero-khoria of Greece, it is the common custom for all the primati, or arkhondes, to meet and allot the taxes. If there be jealousy among them, as frequently occurs, so much the better for the great body of contributors, unless, which too often happens, one party complains to the Turkish authorities, and probably bribes them for the sake of the delightful advantage of triumphing over some hated opponent, and of acting the Turk over his fellow Christians.

§ 3.519 But the most dreadful of all evils to the Acarnanian peasant is the konakia, or lodgings which he is obliged to give to the Albanian soldiers, although it is only upon such extraordinary occasions as the present progress of the Vezir that small villages situated so far out of the route as Prodhromo feel the inconvenience in its highest degree by the actual presence of the detested palikaria. Musta Bey, of Konitza, who was quartered upon Makhala, after having been supplied with provision and forage for himself and 250 followers, insisted upon a present of 100 piastres at departure, but was contented with 45. This was an unpardonable extortion, even by the laws of the Aly code, and would meet with punishment if it were made known to him, as he only allows the chief armatolos to demand presents in this manner. The poor Makhalidtes, however, stand probably too much in awe of the resentment of the Albanians to complain of the injury.

§ 3.520 March 20.—From Prodhromo to Bodholovitza, distance 4 hours 7 minutes, with Albanians on foot. We set out at 9.25, ascend the pass which lies immediately at the back of Prodhromo, and in less than half an hour arrive at the summit of the ridge, when there appears before us a vast extent of velanidhi woods, frequented only by robbers, or by Karagunidhes with their flocks, and traversed by winding paths difficult for a horse, and much more so for baggage. This is called the forest of Manina. I had taken a path to the left of the direct road, with a view of finding my way to some ruins on the bank of the Aspro, called Paleo Mani, but now perceive that it cannot be effected with the baggage horses. As the bolu-bashi of our Albanian escort declares at the same time that we are too few to be separated in these perilous times and places, we regain the common route from Prodhromo, having lost about 8 minutes by the detour. Our guide from Prodhromo points out a place where three Turks were murdered two years ago, by robbers who came from the Islands, then occupied by the Russians. During a halt which we make, from 11.40 to 12.30, to dine at a well in a little opening in the midst of the forest, some families of Karagunidhes pass us; they consist chiefly of women and children, walking by the side of the horses, which carry the tents, maize, barley, and all the domestic furniture. The infants are in baskets slung over the shoulders of the women, who with their bodies bent forward and a hurried step, drag along a horse, or a string of two. or three horses, and are employed at the same time in spinning wool. These persons are Vlakhiotes from the mountains of Kalarytes, and are on their way to the plains of Katokhi, where the men have preceded them with their flocks. The forest consists entirely of the velani oak, which never grows to a great height, but is sometimes broad and spreads into a great number of branches. The little underwood there is, consists chiefly of the paliuri and wild kharub. The khrysoxylo (Cotinus) used as a yellow dye, is also found here.

§ 3.521 Half an hour from Podholovitza, we emerge from the forest and enter on the plain which extends along the banks of the Aspro to the sea. Though generally inundated in winter, it is now dry. The soil, consisting of a stiff white clay, is now under the plough for the reception of kalambokki, which they have not the means here of irrigating artificially. Podholovltza consists only of a tower and a quadrangular inclosure of cottages surrounded by some wicker katyvia: it is situated at the foot of a small height, surmounted by a church, on the right bank of the Aspro, which, being now collected into a narrower bed than in the plain of Vrakhori, and augmented by the tributaries which join it near Anghelokastro, may be compared to the Thames at Staines. In summer it is very shallow, and may be crossed on foot at Podholovitza; but a quarter of a mile lower down, where a projecting rocky bank on the opposite side narrows the river to fifty yards, it is never fordable. Here is the ordinary ferry, and the only one except that of Katokhi. We are informed by the people of Podholovitza that an epidemic disorder now reigning in Karlili has lately carried off six persons in the village. We therefore cross to Guria, which is situated about the same distance below the ferry that Podholovitza is above it. Here I find that the λοιμική, as they call the sickness, was much exaggerated at Podholovitza, in order to frighten us away from thence, and that it has been worse here, though in neither place does it appear to be of a very malignant nature; for though hardly a Greek house in this village out of 30 or 40 has escaped it, two or three persons only have died.

§ 3.522 It is said to begin with head-ache and fever; but if the patient is blooded, which is almost their only remedy, he generally recovers in fifteen days. There are a few Turkish families at Guria, and a little mosque without a minaret. Below Guria the river spreads over a large space, and has some sandy islands in it. It then takes a long bend to the left towards the extreme point of the hills which slope from Stamna into the plain. In the opening between this point and some heights towards the mouth of the river, appears the village of Magfila, on a small eminence in the plain, and Palea Katuna at the foot of the hills to the right. Katokhi is hid by a projection of them. Our Albanian escort consists partly of Mahometans and partly of Christians, who are all from the country near Berat and Kolonia. Since we got rid at Makhala of a bolubashi who had persuaded some of the Mussulmans that it was beneath their dignity to march before ghiaure, we have had no difficulties with any of them, and have kept them in perfect good humour by presenting them with a sheep or two every evening for their supper. Unlike the lazy, proud Turk, or the poor Greek peasant often deprived of all energy by the effects of continued misery and oppression, these Albanians are remarkable for their indefatigable activity. Every commanding height near the road I find occupied by one or more of them, by the time I come in sight of it, and it seems to be an object of emulation who shall arrive first. They answer all questions upon the topography with remarkable intelligence and accuracy, and permission to look through my telescope is an ample reward.

§ 3.523 Nothing can be more dissimilar than the Albanian manners and those of the Osmanlis, the most indolent and phlegmatic of human beings, unless when roused by some extraordinary excitement. In one respect, however, the two people accord,— namely, the love of gaming, though it is forbidden by the religion of Mahomet. As the Albanian soldier seldom burthens himself with provisions, he commonly solaces himself at a halt upon the road with a pinch of snuff and a draught of water. On arriving at a village, the first thing they generally do is to form a party at cards with heaps of paras, while those who do not play look on. A young man, who particularly distinguishes himself by his activity, named Alms, informs me, that in his younger days, like many of the Albanian soldiers, he attended cattle in his native mountains, and that at Arza, a place on Mount Trebusin, two hours from Klisura to the north-eastward, five hours from Tepeleni, and eight from Premedi, he was often in the habit of finding ancient coins of silver and copper.

§ 3.524 March 21.—Having procured some horses at Gurii for some of the escort, and mounted others on the post-horses which we brought for the baggage from Prevyza, I cross the ferry with twelve of the palikaria, and proceed in an hour and a half to Palea Mani. The road is a horse-path, which, after crossing the little plain of Podholovitza, follows a narrow level on the bank of the Achelous, along the edge of the forest at the foot of the lowest slopes of the hills of Minina. Palea Mani is the modern name of a Hellenic fortress standing upon one of the points of these hills, in the thickest part of the woods. As in the ruins of Stratus, one of the gates stood very near an arm of the Achelous, which is separated from the main stream by a portion of its broad gravelly bed. This gate is eight feet wide, diminishing towards the top, which is formed by two opposite stones hollowed into a curve, but not quite meeting, and covered in the middle with a single quadrangular stone ten feet in length, three feet and a quarter in height, and two feet and a quarter in the lower dimension or soffit. I remarked the same kind of construction in a small gate at Kamarina. Beams similar to the upper stone of the gate covered the passage in its whole length of eighteen feet; but of these only two remain in their places. This gateway leads into a small court of an irregular pentagonal form, which was defended externally on the side to the right in entering by a tower open to the court. Nearly opposite to the tower, a small gate leads from the court into the principal inclosure of the town or fortress. This inner gate standing on a slope, the beams of stone above the door project beyond one another like steps, and there are probably some corresponding steps below, which are now buried in the ruins and earth. The natives call the outer gate the Avloporta, being in fact the entrance of a sort of αυλή, or ante-chamber, of the fortress, which formed a good protection to the inner gate.

§ 3.525 I have never seen any similar example of this kind of outwork. From the inner gate the two walls of the principal inclosure mount the height to a small quadrangular acropolis at the summit of the hill, the wall to the right more directly, that to the left embracing a larger portion of the height, but both in curved lines, and that to the left in the upper part, forming a second curve, concave towards the exterior. The acropolis has an outer inclosure flanked by towers: both this and the Avloporta are obviously posterior additions to the original work, being of more regular masonry, while that of the body of the place was entirely polygonal, without towers, and of an irregular plan, bearing strongly the character of a rude people, who possessed little of the science of military architecture as it existed in the more civilized parts of Greece. Such, in fact, was the condition of Acarnania before the age of Alexander. The original walls are in some parts near eleven feet in thickness, but are formed in the middle of rubble and are faced only with large uncemented masses. Among the posterior additions are the remains of a tower at the lower part of the citadel, of which ten or twelve courses of regular masonry remain on one side, and a small part of the adjacent side. The thickness of the wall here consists of single stones, not more than two feet and a half or three feet thick.

§ 3.526 In the mid-height of the remaining courses there is a loop-hole, or window, with a course of masonry narrower than the rest, and projecting a few inches; there is a similar projection also at the foot of the wall. The defence of the acropolis on the lower side towards the town is partly formed by a perpendicular excavation of the rock, upon which a wall has been built consisting of irregular blocks exactly fitted to the rock and to one another. The ruins are in no part more than eight or ten feet high, except at the avloporta. The inclosed space is so extremely rugged that one is surprised how such a place could ever have been inhabited, nor is there a single excavated foundation to be found. The greatest length, which is from the acropolis to the Avloporta, is about 600 yards. In position this ruin seems to accord perfectly with Old Oenia, which Strabo describes as a deserted place situated on the Achelous, midway between Stratus and the sea. It is not to be inferred, however, that the Old Oenia so called in the time of Strabo, was the same city which was founded by Alcmaeon after the Trojan war, and named Oeneia in honour of Oeneus; for Thucydides clearly indicates that place as identical with

§ 3.527 the famous city of the Oeniadae near the mouth of the Achelous It would seem, therefore, that the ruins at Paleo Mani are those of a small and very ancient city of the Acarnanes, which, having been deserted long before the age of Strabo, and its history forgotten, had improperly received in his time the name of Old Oeneia, as often occurs in the instance of ruins and deserted sites. It may possibly have been Erysiche, mentioned by the poet Alcman, which Stephanus improperly confounds with the city of Oeniadae, as seems evident from Apollodorus, whom Strabo cites to show that the Erysichaei were an inland people of Acarnania. In later times, in consequence of the commanding situation in the pass leading along the right bank of the Achelous from the upper to the maritime plain, the original work may have been repaired and furnished with towers to serve as a fortress. Some part of the remains at the acropolis consists of Roman tiles, mixed with small stones and mortar, built on the Hellenic wall. As the pass naturally divided the territory of the Oeniadae from that of the Metropolitae, to one of those two people probably the fortress belonged. At present there is no road to the northward beyond Palea Mani; the wide branching bed of the Achelous, the marshes and lake at the foot of the steep woody mountain of Lygovitzi, and the thick forest

§ 3.528 between the latter and Palea Mani, being impassable, except to the shepherds and peasants of the neighbourhood. The woods around the ruin consist of oak, ilex, maple, and various kinds of underwood, festooned with wild grapes. On the opposite side of the river stands a small tjiftlik and pyrgo called St. Elias, around which the lower falls of Zygos reach to the river side, and are covered with the cultivated fields belonging to Stamna. This village, distant three or four miles to the south-eastward, is situated upon a ridge, sloping on one side into a narrow plain on the bank of the Aspro, and on the other to the lagoon of Anatoliko, on the border of which Stamna has a skala and some magazines. At Anghelokastro, which is two or three miles to the north-eastward of St Elias, is a ruined castle of middle times, standing upon the lowest heights of Zygos, with a small village below it in the corner of the Aetolian plain. The mountain above Anghelokastro and Stamna is separated from the highest woody summit of Zygos, upon which stands Khierasova, by the pass of Klisura, already described as leading directly through the lofty ridge of Aracynthus, by a narrow rocky cleft forming a natural gate of communication between maritime Aetolia and the great interior plain. Having returned from Palea Mani to Guria, we proceed in the afternoon to Anatoliko, over a plain of the same clayey white soil before remarked, and producing maize, wheat, barley, and flax.

§ 3.529 It is marshy in some places, and near Anatoliko is artificially drained. In the parts most distant from the mountains dhiminio is not sown, as little rain falls in the spring, and they have not the means of irrigation. The distance from Guria to the ferry of Anatoliko is two hours menzil pace; but in a direct line much less, because the road makes a great turn to avoid the ridge, which, sloping from Stamna, ends in a point at which stands a hamlet called Mastu, where we arrived in forty minutes from Guria. In approaching Anatoliko we pass through some of its gardens and olive plantations, at the foot of a hill which is quite unconnected with the heights of Stamna, and borders the lagoon on the west almost as far as the outer sea. Having crossed the lagoon in a monoxylo, we proceed to the house of an iatros, who is brother-in-law of my travelling companion K___. The island of Anatoliko is about three miles distant from the northern extremity of the lagoon at the foot of the ridge of Stamna, and a mile distant from the bank on either side to the east and west The island is so small as to be entirely covered with the town, which contains about 400 houses. Though some of these are large, the place is not at present in a flourishing state. Being, like Mesolonghi, supported chiefly by the profits of its ships and maritime commerce, it has suffered by the war, and many of the lower orders are deprived of their employment as sailors. The territory extends three or four miles along either shore of the lagoon, and produces corn for about two months’ consumption, wine rather more

§ 3.530 than sufficient for the place, with a quantity of oil which admits of an export to the value of 40,000 piastres in the alternate years, when the fall olive crop occurs. The fresh and salted fish from the lake furnish a traffic with Zakytho and other neighbouring places. The Vezir takes 46 purses a year for the fishery and other revenues of the crown from the proesti of Anatoliko, who share the farm with other principal persons of the place. These 23,000 piastres include 700 kharatjes, together with the imposts of the two villages of Magula and Neo-khorio, near the mouth of the Aspro. My host the Iatros says, that during the six years he has lived here he has been five years ill; while the natives have not such bad health—a melancholy state of affairs for the doctor, but which would be much more so were it not that according to the common custom in Greece he receives a fixed stipend. From the looks of the inhabitants I should not have supposed the place healthy: indeed, the narrowness of the lagoon in this part and the woody mountains which inclose it on three sides, seem far less favourable to health than the open and well-ventilated situation of Mesolonghi, where the people in every sense of the word are a well-looking race. The small quantity of salt held in solution by the water at Anatoliko, as I was surprised to find on tasting it, may also affect the quality of the air: the lagoon towards Mesolonghi, on the contrary, is as salt as the sea. This shows that all the northern part of the lake is chiefly formed by springs from the surrounding mountains, of which indeed there are several to be seen on the neighbouring shore, particularly one near Klisura, and another near a fresh-water marsh opposite to the town to the eastward. Though the water of the former is considered much the bettor, the monoxyla are more frequently sent to the latter because it is nearer. In the town there are only cisterns for rain water.

§ 3.531 March 22.—From Anatoliko the ruined mill above Makhala is visible to the N.N.W.; and a little to the left of it is seen the hill of Lygovitzi, then Mount Bumisto in a line with Stamna, and a pointed height to the southward of that village on the same ridge, called St. Elias. In all other directions the view is much circumscribed by the neighbouring part of Mount Zygos and by the height on the western side of the lagoon. The distance in a direct line from Anatoliko to Mesolonghi is about 6 o. m. With a monoxylo it is almost double the distance, on account of a long low cape which separates the lagoon of Anatoliko from that of Mesolonghi, leaving only a communication between them half a mile broad, between the extremity of the cape and the ramma or thread of land which separates all the lagoons from the open sea. Having landed on the eastern shore at 3 p.m. we proceed to Mesolonghi by land. Already have the post-meridian thunder-showers, which characterize the Grecian spring, commenced. Both yesterday and to-day the clouds collected on the mountains about noon, and fell afterwards in rain accompanied with lightning. After an hour’s ride, we are obliged to take shelter from one of these storms in a tower at the Aliki,

§ 3.532 or salt-works, which are situated to the right of the road, on the narrow point of land. These salt-works belong to Mesolonghi, and produce 28000 piastres a year. Instead of repeated supplies of water being let into the salt-pans, as at Lefkadha, by which each pan produces a thickness of a foot or two of salt, and only the lower part of the salt is impure, it is here gathered as fast as each admission of water is evaporated; the consequence of which is, that a great quantity of earth is mixed with the salt, and only small portions of it are white and pure. There is another salt-work in the lagoon of Bokhori. As soon as the weather clears we proceed, and soon enter the olive-grounds, gardens, and marshy ditches of Mesolonghi. In the town I find the Vezir Aly and all his court.

§ 3.533 CHAPTER 33, AETOLIA, ACARNANIA. MARCH 25.—Kurt-aga, the site of Calydon, is a ride of 1 hour and 35 minutes from Mesolonghi. Midway, opposite to the eastern termination of the lagoon of Mesolonghi, at a ζευγαλάτι, or farm belonging to Stathaki, one of the proesti of Mesolonghi, are some remains of ancient buildings, resembling Roman baths. Two chambers subsist which have curved and arched niches in the walls, and on the outside several holes, one of which is partly filled with indurated sediment formed by a long continued course of water.