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

§ 4.163 From the northern route leading into Ioannina from Skodra, Berat, and Premedi, the view of the city is also extremely beautiful, as the traveller approaches under an abrupt cliff in the hill of St. George, where the meadows and gardens at the head of the lake are on the left, and in front the island, with the kastro and town in profile, and the great ridges of Pindus in the background. In every view of the city and its immediate vicinity, the most remarkable object is Mount Mitzikeii, the gigantic proportions of which, resulting from its steepness and proximity, are a feature peculiar to this city. The lofty and graceful outline of the mountain, the deep furrows of its torrents, and the terrors of its precipices, are admirably contrasted with the plain and lake on one side, and on the other with the distant summits of Pindus, thus forming between the two an object of intermediate distance and comparison, which completes the harmony of the entire landscape. Although advanced considerably to the westward of the central line of Pindus, Mitzikeli is too near to it to be long free from vapours, when a southerly or westerly wind caps all the great summits, or when the regular diurnal changes in summer cover the Pindus with clouds. Mitzikeii then often throws the city into shade while the plain is in sunshine, and it is the cause of the frequent thunder-storms to which Ioannina is subject, especially in the spring and early summer.

§ 4.164 The clouds, extending from the higher ridges, first collect on the summit of Mitzikeii, and then gradually descend its western side, until suspended over the town, they are dissolved in torrents of rain, generally accompanied with lightning, discharged from so short a distance as often to cause fatal accidents, and accompanied with peals of thunder, which are reverberated from the mountain with tremendous violence, and are not exceeded in intensity even in the tropics. Subject as Epirus in general is to those atmospheric changes which generate thunder, and which caused Jupiter Tonans to be the presiding deity, there is no inhabited situation in the province to be compared to Ioannina itself, for its rapid transitions of temperature and the frequency of thunder-storms. These in the winter may often be witnessed accompanying a heavy fall of snow. It is said that the side of Mount Mitzikeli in face of the city was formerly covered with trees, instead of being, as at present, totally bare, except in the lower parts of the ravines near the lake, where some underwood still remains; and this supposition is the more probable, as the eastern side of the mountain, and its continuation Mount Dhrysko, are still well wooded like all the ridges of Pindus, except where they consist of rocky pinnacles, incapable of vegetation. The forests nearest to the town were naturally the first to disappear before the demand of so large a city; as the distance from whence the fuel is to be conveyed has increased, so has the prevalence of the use of charcoal, made in the woods, and brought to the town at a small expence. The court however still consumes a great quantity of wood; and one of the severest oppressions to which the villages in the district are subject is their obligation to deliver it in the city at their own expence. Every village in its turn, whether in the vicinity of a forest or not, is liable to this imposition.

§ 4.165 The kaza of Yenya (as the Turks call Ioannina) is divided into four nahiye, named Malakassi, Kurendo, Tzerkovista, and Zagori. 1. Malakassi, or Malakash, according to the vulgar pronunciation of the ending, comprehends all the southern part of the valley of Ioannina, beginning from Rapsista and Katzika inclusive, extending from thence southward to the boundaries of Arta, and eastward to those of Mεtzovo and Zagori. The principal places in Malakassi are Kalaryrtes and Syrako, then Kotzista of 100 houses on the northern face of the mountain of Syrako; Bozgoli, similarly situated; Khrysovitza, on the opposite side of the MOtzovo branch of the Arta; Klaziadhes, on the northern end of Mount Dhrysko, and Lozetzi, the largest of the Katzano-khoria. The latter villages, otherwise called Tomaro-khoria, are 12 in number, and are situated on the mountain which connects Dhrysko with Xerovuni, the greater part of them on elevated levels or slopes above the right bank of the Arta. The total number of villages in Malakassi is 49, and the houses 2350.

§ 4.166 2. Kurendo contains 71 villages, and 1870 houses; it comprehends the northern part of the valley of Ioannina, together with the hills which border it on the west and north, extending westward to the confines of Filiates and Dhelvino, and to the north-west as far as Zagoriani and Kalbaki included, where it borders on Zagori. Further south it is separated from the same sub-district by the ridge of Mitzikeli. The largest villages are Kurenda, consisting of several makhalas containing altogether 100 families; Perama and Zelova, near Ioannina; Zitza, already described; Kartzunista, Zelista, and Granitzopula; the three last lying in that succession, in the direction from Ioannina towards Filiates. 3. Tzerkovista contains only 22 villages and 460 houses; it comprehends the villages situated on the eastern slope of Mount Olytzika, together with those between its southern extremity and the boundaries of Arta. The largest is Sklivaneus, on the left of the road from Ioannina to Pendepigadhia near the latter, and containing 70 houses. 4. Zagori occupies all the highlands of Pindus included between the central ridge and the crest of Mitzikeli, together with two villages which lie to the westward of the line of Mitzikeli, namely, Ravenia and Mavrovuni. Zagori extends northward to the limits of Konitza, where Papingo is the northernmost village, and to the south borders on Metzovo and Malakassi. It consists chiefly of two parallel valleys, with a separating ridge named Paleovuni, which is intermediate between Mitzikeli and the central Pindus. In the valley next to Mitzikeii is the northern branch of the Arachthus flowing to the south.

§ 4.167 The eastern valley is watered by the Aous, flowing in the opposite direction, and which receives a western branch partly from the north-eastern extremity of Mitzikeli, and partly from Paleovuni. I have already remarked, that this latter seems to be the mountain, which by Livy is named Lingon. Zagori contains 3500 houses in 42 villages. The largest are Laista, of 200 houses, towards the source of the Viosa and Tzopelovo of the same number on the northern side of Mount Paleovuni. Dovra, on the northern extremity of Monnt Mitzikeli, has 150 houses. The upper and lower Sudhena, a little northward of Dovra, contain more than 300 between them. The chief villages in the valley of the northern Arachthus are Laskovetzi, Frangadhes, and Kalota, all on the side of Mount Mitzikeli. Twelve of the villages of Zagori are inhabited by Vlakhiotes, the remainder by Greeks, but the name of the district and of many of the villages show that at one time it was a great Sclavonian settlement. Like the Vlakhiotes and Christian mountaineers of Albania, the people derive their subsistence from trade, or as artisans in various parts of Turkey. The Vezir’s agent in the government of Zagori is Alexodhimos, son of Alexis, of Kapessovo, some of whose ancestors have held the same office. He has a large house also at Ioannina, is engaged in commerce there like several others of the principal Zagorites, is put to all sorts of expence by Aly, and in return extracts all he can from his government. With the exception of Kalarytes, Syrako, Matzuki, and the Vlakhiote villages of Zagori, all the others in the Kaza of Ioannina are peopled by Greeks.

§ 4.168 Some of them are Spahiliks in the hands of Turkish beys at Ioannina, which the Vezir loses no opportunity of obtaining possession of. The total number of houses in the four districts of Ioannina amounting to about 8,200, several of which have more than one family living in them, the rural population can scarcely be less than 50,000. In the city, where the families bear a much larger proportion to the number of houses than in the villages: the population, including that of a moveable kind, is probably not short of 30,000. I have already suggested the probability that the valley of Ioannina is the Dodonaea, and the ruins at Kastritza those of the city Dodona. Unfortunately, nothing more than an opinion can be advanced on this subject, as Dodona has neither been described by any ancient author, so as to be recognized by such description, nor have any remains or monuments been yet discovered tending to supply the deficiency. Hence Dodona is now the only Greek city of great celebrity, the situation of which is not exactly known by means of a comparison of ancient history with actual appearances: and hence an opinion upon the question of its site, cannot have any better basis than the negative argument, that there is no other situation in which Dodona can be placed so as to accord with the mention made of it in history, or so as to allow of a consistent adjustment of the several tribes of Epirus to the modern map. Whether the opinion that Dodona stood at Kastritza be correct or not, a review of the authorities upon which it is founded will at least be serviceable to those further researches which will probably end in leaving no doubt on the question:

§ 4.169 for it is not to be imagined that Epirus with the same language, religion, and manners as the rest of Greece, should have been destitute of those eloquent monuments which have frequently thrown light on the mythology, history, and topography of other parts of the country, or that the Epirotes were less accustomed than the other Greeks to employ lapidary writing for public and private memorials. Extant specimens prove that the arts were carried to as great perfection in Epirus as in any part of Greece, though in consequence of the desolation which followed the Roman conquest, very few of the productions of those subsequent ages, which have supplied ninetenths of the ancient monuments discovered in other parts of Greece are to be met with in this province. But for the same reason, they are likely to be as valuable as they are rare, being the productions of times when Greek art was in the greatest perfection. Of the Dodonaean temple in particular it is difficult to believe that some vestiges should not still subsist, or that some remains of the numerous dedications which had accumulated within its walls during the long ages of its sacred celebrity, should not be yet preserved below the surface of the soil, if we knew exactly where to explore, or, having that knowledge, could search in security.

§ 4.170 As there is some ambiguity in the allusion to Dodona by Homer, it may be proper to begin by reverting to the passages of his poems in which the name occurs. From the catalogue of the Greek forces in the Iliad, we learn that the Enienes of Dodona, and the Perrhaebi of the river Titaresius, accompanied Guneus of Cyphus, one of the Thessalian leaders, to the Trojan war. In the sixteenth book Achilles prays to Jupiter of Dodona, whose ministers were the Selli, “men of unwashed feet, sleeping on the bare earth.” In the Odyssey, Ulysses in his feigned speeches to Eumaeus and Penelope, pretends that, after having visited Pheidon, king of the Thesproti, he proceeded to consult the oracular oak of Jupiter at Dodona. It was the general belief of the ancient readers of Homer that there were two Dodonae, one in Thessaly the other in Epirus; the former situated in Perrhaebia, near Mount Olympus, the latter in a part of Epirus which, in the time of the Trojan war, was dependent on the king of Thesprotia: that from the former came the Enienes, who were joined with the Perrhaebi of the Titaresius under the command of Guneus;

§ 4.171 and that the latter was the place alluded to by Ulysses in the Odyssey. As to the prayer of Achilles opinions differed; some of the ancient critics having supposed that prayer to have been addressed to a Jupiter worshipped at Dodona in Thessaly, and who was chosen by Achilles as a -γεινιων θεος, or deity of his native country, in the same manner as Pandarus prays to Apollo Λυκηγένης, and Chryses to Apollo Σμινθεύς. But in this case, as Stephanus suggests, “how happened Achilles to have named the Selli in his prayer, who were particularly connected with Dodona of Epirus,” as several writers show, but particularly Hesiod, who places the Epirote Dodona in Hellopia, a country which received that name from its inhabitants, the Helli or Selli. Nor are the poet’s words τηλοθι ναιων altogether indifferent to the question. It is true that Achilles being at such a distance from Thessaly as Troy, might, without impropriety, apply them to the Thessalian Dodona; but it seems more consistent with Homeric diction, to conceive that τηλόθι ναιων was a customary form of address to the particular Jupiter to whom the hero was praying, and that it alluded to the separation of Epirus from the rest of Greece by distance and a chain of lofty mountains, which had caused it to be comparatively barbarous,

§ 4.172 and to be so unaffected by the political interests of the rest of Greece, that it contributed no forces to the Trojan expedition. We may remark also, in support of this view of the question, that proof is totally wanting of the existence of any sanctuary of Jupiter at the Thessalian Dodona. It seems evident, therefore, that the prayer of Achilles was addressed to the same Thesprotian Jove, whose worship and oracle had acquired great celebrity about that time, as the Odyssey shows, and which appear from Herodotus to have been more ancient even than the colonization of the Pelasgi in Epirus, having been established by some adventurers from Phoenicia and Egypt, who probably found the barbarism of Epirus favourable to their superstitious power. The ambiguity as to the two Dodonae in the Iliad appears to have arisen from several coincidences. The poet applies to both of them the epithet δυσχειμερος. They were both Pelasgic settlements; Pelasgiotis was the name of the northeastern part of Thessaly to the latest period of antiquity; and both the Dodonae had Perrhaebi dwelling in their vicinity. But these resemblances are no more than natural, if Dodona of Epirus received a colony and its name from Pelasgiotis of Thessaly, and if that colony was accompanied or followed by Perrhaebi, a people of the same country.

§ 4.173 As to the common epithet, both the Dodonae being situated near lofty mountains, it is not surprising that the poet, who so often repeats his epithets, should have attached the same to them both. An opinion appears to have prevailed among the ancients, which is supported by the Odyssey, that the Dodonaea appertained first to Thesprotia and afterwards to Molossis. This chorographical change is particularly noticed by Strabo, who supports his remark by a reference to the epithet Thesprotis applied to Dodona by Pindar and the tragic poets. Among the latter may be particularly mentioned Aeschylus, who, in describing the oracle of Dodona as that of the Thesprotian Jupiter on the Molossic soil, seems to allude exactly to the change which had taken place in the possession of the Dodonaea. It is evident, at least, that the Dodonaea bordered both on Thesprotia and on Molossis; to fix, therefore, the position and extent of those two divisions of Epirus, and to trace the changes which took place in their respective boundaries, is an essential preliminary in resolving the question of the situation of the Dodonaea.

§ 4.174 Thucydides, Scylax, Strabo, and Ptolemy, concur in assigning the sea coast of Epirus to the four tribes of Chaones, Thesproti, Cassopaei, and Molossi, as well as in placing them in that order from north to south, though they do not agree in the extent of each portion, which could not indeed be expected from authors of different ages in regard to boundaries, some of the variations of which are sufficiently explained by the history of Epirus. It was anciently believed, and apparently not without reason, that three or four centuries before the Trojan war two Pelasgic colonies began to civilize the barbarians of Epirus: that which came from the Peloponnesus by sea, very naturally made choice of the fertile plain near the mouth of the Acheron; the other from the Pelasgic settlements in the north of Thessaly crossed the Pindus and occupied Dodona. The former brought with them several Peloponnesian names, such as Dryopes, Thesproti, Ephyra, Acheron, Pandosia, and Pallantium, of which the three last were carried forward to Italy by colonists from Epirus, or by succeeding colonists from the Peloponnesus: the Thessalian colony introduced in like manner, into the more inland parts of Epirus, the names Dodona, Perrhaebi, Hellopia, Selli, and Achelous.

§ 4.175 This comparatively tardy introduction of the manners and language of Greece into Epirus, is strongly confirmed by the fact, that the two centres of Epirote civilization just mentioned are the only two places in this country alluded to by Homer; from whom it appears also that after the Trojan war they formed one kingdom. Ephyra, near the Glycys Limen, was the capital of Pheidon, king of Thesprotia, and Dodona was within his dominions. But about the same time a new colony was established in Epirus, Pyrrhus, or Neoptolemus son of Achilles, who was deprived of his paternal kingdom, having migrated to this country, accompanied by Helenus son of Priam. Buthrotum was supposed to have been the place which Pyrrhus occupied, a tradition deriving some support from the name of Phoenice, a city only twelve miles distant from Buthrotum, to which Pyrrhus may naturally have given that name in memory of his friend Phoenix, who was said to have accompanied him from Troy, and to have died on the road. Helenus inherited the possessions of Pyrrhus, while Molossus, son of the latter, obtained a settlement in the southern part of Epirus, to which his name was ever afterwards attached.

§ 4.176 As the Chaonian colony extended, Cestrinus son of Helenus founded a, city on the right bank of the Thyamis, at a distance of twelve or fifteen miles above its mouth; that frontier of Chaonia received the name of Cestrine, and its chief town, the ruins of which, now called Palea Venetia, still subsist, seems to have been named Ilium or Troja, in memory of the origin of its founders.

§ 4.177 In subsequent times commercial republics, colonized or augmented, and supported by the wealth and alliance of some of the powerful states of Southern Greece, occupied the entire coast of Epirus. Towards the Ambracic gulf, the descendants of Molossus were confined to the mountains by Ambracia, this noble position having attracted settlers from the Peloponnesus at a very early period, and having received, in the eighth century B.C., a second colony consisting of Corinthians, which people about the same time occupied several places in Acarnania. In the Peloponnesian war, all the maritime part of Epirus, including Thesprotia and Chaonia, was republican, while the mountainous districts of the interior, inhabited by warriors, pastors, and cultivators of the soil, still preserved the monarchical form of government. At that time there were kings of the Agraei, Atintanes, Paravaei, and Orestae, and we find the Athamanes governed by a king as late as the Roman campaigns in Greece.

§ 4.178 The poverty or patriarchal simplicity in which the Aeacidae were living in Molossis about the former period is strongly indicated by Thucydides, in the description which he has given of the reception of Themistocles by king Admetus. But it was to their monarchical union, which never ceased among the Molossi, that they were at length indebted for the preponderance which they acquired over the other tribes of Epirus. The extent and situation of the proper Molossia are clearly described by Scylax. After having stated that the coasts of Thesprotia and of Cassopaea were each half a day's sail in length, that the Cassopaei extended to the Anactoric, meaning the Ambracic gulf, and that the gulf was a little less than half a day’s sail in length from the στόμα, or strait of Prevyza, to the μυχός, or eastern extremity; he adds, that the Molossi bordered on the Cassopaei, that their sea-coast was 50 stades in length, and that next to it was the shore of Ambracia, extending 120 stades to that of the Amphilochi. The latter distance is confirmed by Dicaearchus;

§ 4.179 and as the entire length of the northern coast of the gulf from the shore of Lamari, which was the eastern boundary of the Cassopaea, to Makrinoro, which was the frontier of Ambracia, and Amphilochia, agrees sufficiently with the total of 170 stades, we can have no hesitation in assigning about three miles of the shore on either side of the mouth of the river of St. George as the position and extent of the Molossic sea coast. In the interior, the Molossi bordered to the eastward for a considerable distance upon the Ambraciotae, and beyond them upon the Athamanes, from whom they were separated perhaps by the river Arachthus. To the westward they confined upon the Cassopaei towards the sea, and further in the interior upon the Thesproti, who occupied the valleys of the Acheron and Cocytus, with all the country as far as the left bank of the Thyamis. Although the Molossi widened from their narrow maritime basis, and extended in the time of Scylax, as he remarks, to a great distance in the interior, by which we may suppose them to have then possessed the Dodonaea, the original Molossis was neither large nor productive, having contained little more than the mountainous region lying between the river Arta and the ridges of Olytzika and Suli, in which direction it comprehended the valleys of the river Luro and its tributaries, with those of the upper Acheron.

§ 4.180 But in process of time, the kings of Molossis, assisted by their hardy followers, added the Dodonaea to the northward, and the Cassopaea to the westward. From Pindar indeed it would seem that the former acquisition was made by Pyrrhus, son of Achilles, himself, but this can hardly be reconciled with Homer, or with the supposition of Chaonia having been the seat of the colony of Pyrrhus, as it would imply an almost entire conquest of Epirus. It is more easy to believe, that the extinction of the Thesprotian kingdom, the colonies of Elis in the former seat of that kingdom, and the encroachment of the Cassopaei upon Thesprotia, gradually gave the Molossi, who continued to be united under a kingly government, such a superiority over the Thesproti as was naturally followed by their acquisition of the Dodonaea froin the latter, though it is impossible to say at what period this event may have happened. Cassopaea proper, or the territory of Cassope, seems to have been added to Molossis before the fourth century B.C.; for Alexander, son of Neoptolemus, obtained at that time, by the assistance of his brother-in-law Philip, son of Amyntas, the cities of Pandosia, Buchaetium, and Elateia, all which were more distant from the bounds of Molossis than Cassope itself.

§ 4.181 The latter, therefore, had probably been an earlier conquest of the Molossic kingdom. The Dodonaea and Cassopaea were important additions to the poor Molossic mountains, and were sufficient under a few able monarchs to lead to the acquisition of all the other divided portions of Epirus. Tharypas, son of Admetus, was said to have been the first of the Aeacidae who encouraged science and literature, but Alexander, son of Neoptolemus, third in descent from Tharypas, was the prince who by his valour and talents, and not less by his double alliance with the royal house of Macedonia, brought all Epirus under a single head, and made it one of the leading states of Greece. Pyrrhus, who after two short reigns succeeded him, and made Ambracia his capital, was, like several princes recorded in history, destined by character and circumstance at once to raise his country and family to the height of their fortune, and to originate their decline. The Aeacidae were extinct in his fourth successor, after which Epirus was only a loose federacy of republics for about 50 years, when it fell under the Roman yoke, in the year B.C. 167.

§ 4.182 If Thesprotia and Molossis had respectively the extent and position just indicated, and if the Dodonaea bordered on the inland frontier of them both, there seems no possibility of assigning any other situation to it than that of the district of Ioannina. The journey of Aeneias to Dodona, as related by Dionysius of Halicarnassus, tends entirely to favour this opinion. After having founded Aeneia at the cape near Saloniki, now called Karaburnu, the Trojan hero visited Delos and Cythera, and at both places left memorials which still existed in the time of the antiquary. He then exchanged testimonials of a common origin with the Arcadians, and sailed to Zacynthus, where he founded a temple and established games, and where a dromus was still shown, called that of Aeneias and Venus. At Leucas he founded a temple of Venus, which still remained, in the time of Dionysius, on a small island between the Dioryctus and the city. From thence he proceeded to Actium and to Ambracia.

§ 4.183 At the former Dionysius describes a temple of Venus founded by him, and another of the Great Gods, and at Ambracia there was a heroum of Mneias near the small theatre, which contained his statue, served with sacrifices by female amphipoli. From Ambracia Anchises sailed with the fleet to Buthrotum, while ASneiae travelled by land in two days to Dodona. Here he consulted the oracle, and presented many Trojan offerings to the god, some of which, consisting of brazen vases, inscribed with the names of the dedicators in ancient characters, still remained at Dodona in the time of Dionysius. He then proceeded to join his fleet at Buthrotum: the journey occupying near four days: from Buthrotum he sailed to the port of Anchises, the name of which had undergone some change in the time of the historian, and from thence crossed over to Italy. It seems evident from the consistency of this relation, that whatever degree of confidence may be given to the facts, the narrative is that of a person well acquainted with the places, and is therefore equally entitled to consideration in a question merely geographical. Arta or Ambracia was exactly the place most convenient for landing, and that from whence the passage was easiest and shortest to Dodona, on the supposition of the latter having been near Ioannina;

§ 4.184 and if the journey between these two principal cities by easy passes required two days, (in fact, without considerable exertion the journey from Arta to Ioannina cannot be performed in one), the route from the latter to Buthrotum, across unfrequented districts intersected with mountains, may very well be supposed to have required not much less than four days. That Dodona was on the eastern frontier of Epirus is clearly shown by Pindar, who describes Epirus as beginning at Dodona, and extending from thence to the Ionian sea. The manner in which Aeschylus introduces the Dodonaean mountains as a part of the territory of Pelasgus, seems to connect them with the highest ridge, but their vicinity to it is still more strongly indicated by the epithets αιπύνωτος, attached to Dodona by the same poet, and that of δυσχειμερος by Homer. Aristotle and Strabo confirm the supposition that Hellopia, or the country of the Helli or Selli, whom Homer and Sophocles, to say nothing of later writers, place near Dodona,

§ 4.185 was in the vicinity of Ioannina, by stating that the Selli lived about the Achelous as well as Dodona, thus rendering it probable that the distance was not great between the city and the river. Ioannina, in fact, is only about 17 geographical miles in direct distance from the nearest part of the Achelous. The account which Polybius has given of the destruction of Dodona by the Aetolians, in the autumn of the first year of the Social war, B.C. 219, equally favours the belief, that Dodona was situated towards the south-eastern frontier of Epirus, and that it was exposed to Aetolia on that side. The expedition was headed by Dorimachus immediately after his election to the strategia. “He marched,” says the historian, “into the upper parts of Epirus, devastating the country, not so much for the sake of plunder as for that of injuring the Epirotes. Having arrived at the temple near Dodona, he set fire to the stoae, destroyed many of the dedications, and overthrew the sacred edifice itself.” The words ανω τόπους της Ηπείρου, which the historian employs in this passage,

§ 4.186 appear to be a mere synonym for the more common expression ανω Ηπειρον, or Upper Epirus, which, as in the instances of Upper Macedonia and Upper Thessaly, meant the part of Epirus most distant from the sea, or towards the central range of mountains. Among the passages of ancient history which prove this meaning, may be particularly cited the remark of Strabo, that Upper Thessaly contained Dolopia and Histiaeotis, that Pelasgiotis constituted Lower Thessaly, and that Upper Thessaly corresponded to and confined upon Upper Macedonia, as Lower Thessaly upon Lower Macedonia, which was the country near the Thermaic Gulf. In fact, Upper Thessaly, Upper Macedonia, and Upper Epirus, all met on the crest of Pindus. Dodona therefore was near this mountain, nor can the words of Polybius furnish an argument to prove that it was in the northern part of Epirus. The desultory nature of the expedition of Dorimachus, and its success without interruption, strongly indicate that Dodona was chiefly exposed to hostile invasion from the Achelous and Aetolia, as Aly Pasha now is to the kleftes from the same quarter. Dorimachus probably followed the valley of the Aspro from Lepenu or Stratus upwards, and having crossed the range of Pindus about Thodhoriana, entered the valley of the Arta to the northward of the Ambracian district, and followed that river until he was separated only by Mount Dhrysko from the plain of Ioannina.

§ 4.187 He thus avoided the hostile district of Ambracia, which city had recently been taken by Philip and delivered to the Epirotes, as well as the dangers of the pass of Pendepigadhia, or that still more hazardous which ascends from Arta by the valley of the Arachthus. The only author who has described the natural peculiarities of the Dodonaea is Hesiod, in a beautiful fragment of his lost poem the Eoeae, preserved by a scholiast of Sophocles, and in part by Strabo: “ Hellopia,” says the poet, “was a country of corn-fields and meadows, abounding in sheep and oxen, and inhabited by numerous shepherds and keepers of cattle, where on an extremity stood Dodona, beloved by Jupiter; here the god established his oracle in a wood of ilex, and here men received responses, when bearing gifts and encouraged by favourable omens they interrogated the god.” This description is accurately applicable to the valley of Ioannina, which, though producing corn, is more remarkable for that which the poet evidently intended to insist upon, namely, its abundant meadows, and the numerous flocks and herds which feed upon its pastures.

§ 4.188 Such being a summary of the testimony which tends to place Dodona near Ioannina, it may now be right to notice the objections which may be made to that opinion, as well as the arguments which may be adduced in favour of some other situations. The strongest objection is the silence of all antiquity as to a lake at Dodona. But when we consider that the only description of the place which has reached us, is contained in the poetical fragment of Hesiod, who may have alluded to the lake in the very next verses to those which are preserved, the objection loses the greatest part of its force, more especially as there is reason to believe that the existence of a lake in this part of Epirus was known to Pliny, who asserted that the Acheron flowed into the Ambracic gulf, and that it originated in the lake Acherusia at a distance of 36 miles from the coast, where the distance from the gulf accords so exactly with that of the lake of Ioannina, that one can hardly doubt of his having heard of the lake, though he has confounded it with the Acherusia. Eustathius proves also the existence of a lake in Molossia,

§ 4.189 by mentioning a tradition, derived probably from some ancient author now lost, that Neoptolemus, son of Achilles, when he introduced a Thessalian colony into Epirus, settled on the shore of the lake Pambotis in Molossia. As Pindar, who followed the same legend, considered Dodona a part of the domain of Neoptolemus, it seems clear that the lake Pambotis was at Dodona. Pambotis, moreover, is a word in exact agreement with the description of Hesiod, and is particularly suited to the pastures of the lake of Ioannina, which in fact is the only lake in the interior of Thesprotia or Molossis. But besides this strong presumption as to a lake at Dodona, we have direct evidence of the existence of marshes near the Dodonaean temple. It was supposed by some that the Helli, who were the ministers of the oracle, were so named from the ελη, or marshes round the temple, which though it may be etymologically erroneous, is an undeniable proof of the existence of the marsh. Again, Proxenus, as quoted by a scholiast on the Odyssey, stated that the oracular oak was found by a shepherd feeding his flock in the marsh Dodon. It is not impossible that there may have been more of a marsh and less of a lake in ancient times than there is now; the partial obstruction of the katavothra, and other causes, which in many parts of Greece have, in a long course of ages,

§ 4.190 increased or diminished the quantity of stagnant or running water, may have changed what was once a marsh, with pools in the deepest parts, into a continued lake. In this case the numerous sources which emerge from the foot of Mount Mitzikeli, close to the margin of the lake or below its surface, would have been more conspicuous, and more likely to have elicited the remark of Theopompus as to the hundred fountains at the roots of the mountain Tomarus, near which Dodona was situated. We may now proceed to inquire whether any other situation in Epirus will agree with the requisites of Dodona. One of the scholiasts of Homer supposes Dodona to have been near Dryiopolis, that is to say, near Arghyrokastro, which he places most erroneously in the ancient Thesprotia; another says that it was in the north of Thesprotia, meaning perhaps the same place; a third fixes it near Leucas. In like manner I have been referred for the site of Dodona by the learned of Ioannina, to Vutzintro, to Dei vino, to Arghyrokastro, to Vela, to Filiates, to Paramythia, and to Glyky, without ever meeting with one of them who imagined that it might have been at Ioannina itself.

§ 4.191 Of the modern places just mentioned, we may observe, that although the plains of Delvino, Vutzintro, and the lower Kalama, have an abundance of pasture in their lakes or maritime marshes, and were even famed in consequence for their breed of oxen, they are too near the sea to correspond to the evidence which has been adduced as to the site of Dodona, and we know them to have been occupied by the cities of Phoenice, Buthrotum, Cestria, and others. The valley of the upper Kalama is not liable to a similar objection, being surrounded by mountains, and containing a lake near Tzerovina, with some marshy tracts at the sources of the Thyamis. Both lake and marshes however are too diminutive for those of Dodona, and the latter are not permanent. The plain itself is too inconsiderable for the magnificent description of Hesiod, too distant from the Aetolian frontier of Epirus, and can scarcely be included either in Thesprotia or Molossis, according to their ancient boundaries. The valley of Dryiopolis or Arghyrokastro, although not unsuited in some respects to the picture of Hesiod, yet being marshy only in winter, cannot in this climate possess that distinguishing characteristic of abundant meadows and perennial pasturage, which the Dodonaea requires. This plain, moreover, is so near the Aous, that the name of Dodona could hardly have escaped mention in the history of the transactions which occurred in that part of the country, had it been so situated. It is evident, likewise, that the valley of the Dryno could never have been included either in Thesprotia or Molossis, having been a part of Chaonia or Atintania, possessed by the Argyrini and the city of Elaeus.

§ 4.192 Exclusive therefore of the district of Ioannina, there is none but the great valley watered by the ancient Cocytus, Acheron, and lake Acherusia, which can be taken for the Dodonaea. Here it must be granted that the lofty ridge called Kurila, which stretches northward from Kako-Suli to the Kalama, resembles Tomarus in the abundant sources which issue from its base, and that the name of Suli has every appearance of having been corrupted from that of the Selli, who served the Dodonaean temple and dwelt around it. Nor can it be denied that the noble plain of Fanari and Glyky, which extends 25 miles from Porto Fanari to Paramythia, fully deserves the description which Hesiod gives of Hellopia, especially as the borders of the Acherusian lake admit of some perennial pasturage, and that a farther argument in favour of Dodona having been here situated, may be derived from Pausanias, who, immediately after speaking of Cichyrus, or Ephyra, which was near Porto Fanari, mentions Dodona in a manner naturally leading to the presumption that it was not very distant from Ephyra. From these admissions, it would follow that Dodona was probably situated at or near Glyky, just at the foot of the mountain upon which the name of the ancient Selli is preserved in that of Suli. But there are strong objections to every part of the plains watered by the Acheron and Cocytus as the Dodonaaea. Even the most distant point Paramythia is too near the sea coast, and too far from the central ridge of Pindus, as well as from the inland frontier of Epirus and the Achelous, to conform to the combined testimony of Pindar, Aeschylus, Aristotle, Polybius, and Strabo.

§ 4.193 The olive-trees of Paramythia are alone a contradiction to that climate, and that elevation above the sea which is implied by the epithets of Homer and Aeschylus. Although the plain between the Acherusia and Glyky is marshy as late as the beginning of May, the only district of perennial pasture in these plains is around the Acherusia, or not more than from two or three to seven or eight miles distant from the sea. Above all, there appears no mode of explaining in what manner this valley could ever have formed a part of Molossis, being the centre of the original Thesprotia, which contained the cities Ephyra, Pandosia, and two others at Veliani and Paramythia, neither of which has the requisites of Dodona. As to Suli, it will hardly afford much assistance in the determination of this question. Suli is a common name in Greece, and naturally so, if we suppose it to be a corruption of Σελλοί, which was no more than a dialectic form of'Ελλοί, the people from whom the whole country derived its appellation of Hellas. The name is undoubtedly the more curious in the present instance, as being found in a district which we know to have been inhabited at a comparatively late period by Selli. But even here it can only be regarded as an accidental vestige of the people of Hellopia, who once occupied all the country around Dodona, and it is the less to be relied on as affording any proof of the exact locality of Dodona, there being another Suli on the Kalama, ten or twelve miles to the westward of Ioannina.

§ 4.194 Nor can any more precise inference be deduced from the juxtaposition of Cichyrus and Dodona in the passage of Pausanias, who there alludes to the early history of Thesprotia in the time of Theseus, when Ephyra was the capital of king Aidoneus, and the Dodonaea a part of his kingdom, and who introduced the names incidentally, as those of the only places in Epirus which he thought worthy of notice. It remains to be inquired whether the position of Dodona at Ioannina is consistent with the general arrangement of the tribes of Epirus on the modern map. Theopompus of Chios reckoned fourteen of them, and Strabo has named as many, to wit: the Chaones, Thesproti, Cassopaei, Molossi, Amphilochi, Athamanes, Aethices, Tymphaei, Paravaei, Talares, Atintanes, Orestae, Pelagones, and Elimiotae. Of the situation of several of these tribes, proofs have already been given. Three of them, the Orestae, Pelagones, and Elimiotae, were permanently united not long after the time of Theopompus to Macedonia, to which they naturally belonged, as being situated, the two latter entirely, and the first in great part at least, to the east of Mount Pindus. In the time of Strabo the Athamanes, Aethices, and Talares, were united in like manner with the Thessalians, though as all the three occupied the ridges of Pindus, and immediately bordered on the Molossi, they seem to have been considered by Strabo as properly Epirotic.

§ 4.195 To begin from the south-eastward, where the Molossi bordered upon Ambracia. Proceeding from thence northward, there is every reason to believe, from what has already been stated, that their next neighbours were the Athamanes, separated from them either by the Arachthus or the crest of the mountains beyond it; then the Perrhaebi and Tymphaei at the sources of the Arachthus, and the’ Talares of Mount Tomarus, who may be considered as a subdivision of the Molossi, as Dodona was situated at the foot of that mountain. Beyond the Talares and Tymphaei were the Paravaei, whose country was the northern part of Zagori and the district of Konitza. To the westward of these, the Molossis, considering the Dodonaea as a portion of it, confined on the southern extremity of Atintania, which I have before described as comprehending in general terms the mountainous country between the Mizakia and the valleys at the sources of the Dry no in one direction, and in the opposite, or from west to east, extending from the Lower Viosa and Dryno to the Uzumi and Upper Viosa. In the remainder of its boundary, Molossis (still considering the Dodonaea as a portion of it) bordered upon the north-eastern extremity of Thesprotia; that is to say, towards the Kalama and its tributaries, which flow from the ridges of Olytzika and Suli. And thus Molossis, together with the surrounding Epirotic tribes, forms a complete and consistent system of chorography, on the supposition that the Dodonaea was identical with the modern district of Ioannina.

§ 4.196 It can hardly be doubted by any person who has seen the country around Ioannina, and has examined the extensive remains at Kastritza, that the city which stood in that centrical and commanding position was the capital of the district during a long succession of ages. The fortresses at Velitzista and Gardhikaki were obviously no more than coma, though they may very possibly have been enumerated among the seventy cities of Epirus destroyed by order of the Roman senate. The hill of Kastritza, moreover, answers much better than either of those places to the εσχατιη, or extremity upon which Hesiod states Dodona to have been built, while the adjacent Mitzikeli will be found to correspond perfectly to the mountain, below which, according to Strabo, the temple stood. This commanding ridge, therefore, which in every point of view arrests the attention of the spectator, I conceive to have been the celebrated Tomarus. The numerous sources at its foot, which are the chief supplies of the lake, are in exact agreement with the hundred fountains issuing from the base of Tomarus, as described by Theopompus. Nor is the name Tomarus, though no longer attached to this mountain, quite obsolete, being still preserved in that of the Tomarokhoria, or villages situated on a part of the southern extremity of Dhrysko, which is a continuation of Mitzikeli.

§ 4.197 The temple of Jupiter Dodonaeus seems not to have been within the city of Dodona. Polybius describes it as περι Δωδώνην; in such a hasty invasion as that of Dorimachus, it could not have been so easily destroyed as the historian relates, had it stood within a fortress such as we cannot but suppose Dodona, like all the towns of Epirus, to have been, nor could there well have been in a fortified town, space sufficient for the temple, its courts, porticos, and dedications, its sacred grove, and the dwellings of its servitors. Strabo, in reporting from Ephorus the cruel treatment of one of the priestesses of the temple by the Boeotians, shows the unprotected condition of the building which Menedemon, an author cited by Stephanus, describes as being surrounded with tripods instead of walls. But in this inability to resist an enemy it differed not from other hiera of great celebrity in Asia and Greece, such as those of Samos, Branchidae, Sardes, Eleusis, Rhamnus, the Isthmus, Epidaurus, Olympia, Nemea, Abae, Ptoum, and that of Trophonius at Lebadeia. None of these were included within the fortifications of the neighbouring cities, but appear to have been inclosed by a simple peribolus, having been placed under the guardianship of their sanctity, the violation of which in the instance of Dodona, branded the Aetolians with the character of men regardless of the laws which governed the rest of mankind in peace and war.

§ 4.198 If the city of Dodona was at Kastritza, and the temple in some other part of the adjoining district, there seems no place so well adapted to it as the peninsula now occupied by the citadel of Ioannina. Such a situation equally accords with the good taste which the Greeks always evinced in the position of their sacred edifices, as with their mythology, which conceived the gods to delight in places rendered remarkable by natural causes, and with that tact which taught the priests to avail themselves of every thing which contributed to elevate the religious veneration of the people, and to promote their own influence through its means. This position is not, indeed, at the foot of Mount Mitzikeli, as Strabo may seem to require; but the commanding steepness and striking proximity of that mountain in face of the Peninsula, are perhaps still better adapted to his words. There is no place in all Greece more subject to thunder storms than Ioannina, none more worthy of having been the abode of the Thunderer, whose bolt was the type of Molossis and Epirus. Here, therefore, in place of the dirty streets and bazars of the modern town, we may imagine a forest, through which an avenue of primaeval oak and ilex conducted to the sacred peninsula. Within the porticos which inclosed the temple were ranges of tripods supporting cauldrons,

§ 4.199 the greater part of which had been contributed by the Boeotians in consequence of an annual custom, and which were so numerous and so closely placed, that when one of them was struck the sound vibrated through them all; many others had been dedicated by the Athenians, whose theoria or sacred embassy brought yearly offerings; but the most remarkable of the anathemata was a statue dedicated by the Corcyraei, holding in its hand a whip with three thongs loaded with balls, which made a continual sound as they were agitated by the wind against a cauldron. In a picture of the temple of Dodona which has been described by Philostratus, the prophetic oak was seen near the temple, and lying under it the axe of Hellus, with which he struck the tree, when a voice from it ordered him to desist. A golden dove, representing the bird of Egypt, which uttered the voice, was perched upon the tree; garlands were suspended from its branches, and a chorus from Egyptian Thebes was dancing around it, as if rejoicing at the recognition of the sacred dove from their native city. The Selli were seen employed in prayer or sacrifice, or in decorating the temple with fresh boughs and garlands, or in preparing cakes and victims, while the priestesses were remarked for their severe and venerable appearance.

§ 4.200 Whether this be the description of a real picture, or the ideas of Philostratus for the subject of one, it is probably a faithful portrait of the hierum of Dodonaean Jove in the height of its reputation, when it may easily be supposed that the temple, the porticos, the dedications, and the dwellings of the sacred servants, were sufficient to occupy the greater part of the peninsula. In the time of Strabo the oracle was already in an expiring state, though it may perhaps have partially recovered, like Greece in general, in the second century, as Pausanias still represents the temple of Dodona and the sacred ilex as the objects most worthy of a traveller’s notice in Epirus. Long after the introduction of Christianity, Dodona maintained its ascendancy among the towns of Epirus by means probably of its fertile district, strong site, and important position at the entrance of the passes leading into Thessaly. The names of several bishops of Dodona are found in the acts of the councils. The latest was in the year 516; and as the earliest record of a bishop of Ioannina occurs in 879, when a council was assembled at Constantinople for the restoration of Photius to the patriarchal throne, it is probable that between those two dates the peninsula of Ioannina,

§ 4.201 already cleared of its idolatrous worship, pagan edifices, and sacred grove, was chosen as a more defensible position than that of Kastritza, against the increasing invasions of the Sclavonic tribes, and thus became the capital of the Dodonaea. The new fortress was named probably in honour of the saint under whose protection it was placed, and whose church is known to have occupied the site of the mosque at the Ν. E. angle of the kastro. The sacred buildings of the hierum, if any remains of them had survived the ages of Christian warfare against paganism, were converted undoubtedly to the construction of buildings in the new fortress, where the continued existence of habitations from that time to the present will sufficiently account, as it does in so many other places in Greece, for the disappearance of all remains of Hellenic antiquity. It is by no means impossible, however, if the kastro of Ioannina be really the site of the Dodonaean temple, that some relics of architecture or sculpture may yet afford proof of this fact, and may even serve as a scale and elements whereby to form a judgment of the magnitude and architecture of the temple. Constantine Porphyrogennetus mentions Dodona as the chief town of the Theme of Nicopolis in the beginning of thet tenth century; but he seems to refer more to the ancient than to the actual Dodona; and little confidence, as Gibbon has remarked, can be placed in the imperial authority. Both Nicopolis and Dodona had probably been for some time extinct.

§ 4.202 The history of Ioannina is almost as obscure as that of Dodona; but enough remains to show that it gradually became the chief city of Greece to the westward of Mount Pindus, the only other place of importance having been Ambracia, which, about the same period of time, under the new name of Arta, recovered from Nicopolis that population and importance, which, during seven or eight centuries, had been absorbed by the Augustan colony, and became again the chief city of the country bordering on the Ambracic Gulf. Western Greece is so separated by nature from a government ruling at Constantinople, that a strong vice-regal power is required to maintain it in submission. If the authority of the Sultan is easily disputed here, it is not surprising that under the weaker sway of the Greek emperors their governors of Western Greece were generally independent or left to their own resources. From the seventh century to the eleventh, Northern Greece was a field of contention to the Byzantine Greeks, the Wallachians, and the Sclavonians, and large colonies of the two latter people settled in the country. To the Sclavonic tribes is to be attributed almost all the names of places in Southern Albania and Western Greece which are not of Greek derivation. Ioannina, however, seems to have maintained itself as a Greek city until a new race of adventurers made their appearance. It was taken by Bohemond, son of Robert Guiscard the Norman, in the year 1082.

§ 4.203 He intrenched his army among the vineyards which still occupy the heights of St. George, repaired the kastro, and under the walls of the town defeated the Greek emperor Alexius Comnenus. On the capture of Constantinople in 1204, Western Greece formed an exception to the provinces partitioned among the Frank conquerors, and continued, under a branch of the Comneni, to be a Greek principality, called the Despotate of Aetolia, or of the West, extending at first from Dyrrhachium to the Corinthiae Gulf, but gradually curtailed by Franks, Albanians, and Servians, until it contained only the central part of Old Epirus, to which condition, or nearly so, it was already reduced when it was conquered, about the year 1350, by Stephen Duscian, king of Servia, who placed his brother Siniscian,or Simeon, in the Despotate. On the death of Stephen, a few years afterwards, Nicephorus, son of the last Greek Despot, recovered for a short time the Despotate, but was defeated and slain in a battle on the Achelous, in which he was assisted by a body of Turks, against the Albanians, who had then made some acquisitions in Acarnania and Aetolia. Simeon having found sufficient employment in Servia and Thessaly, Ioannina was left to defend itself against the Albanians by its own resources for about eight years, when Simeon being applied to for aid, sent thither as governor, in 1367, his son-in-law Thomas Prelubo, under whom the Greeks had a complete foretaste of Mussulman cruelty and oppression. He was succeeded by Inico de Davalos, by the Greeks named Ιζάουλος or Ιζαού, an Italian noble, who had been made prisoner and taken into favour by Thomas,

§ 4.204 when the latter was assisting Spata, the Albanian lord of Arta, against the Franks, and who, on the death of Thomas, in the year 1385, married his widow. Izaulo was disliked by his subjects, maintained himself (like Prelubo) with difficulty against the Albanians, and when at length he formed an alliance with Spata, lost a part of his Aetolian possessions to Charles Tocco, second Count of Cefalonia of that name, Duke of Leucas, and who at length obtained both Arta and Ioannina, when to his other titles he added those of Duke of Ioannina and Despot of Western Greece. He was the most powerful of the Frank princes of the Islands, his continental possessions having comprehended a large portion of Acarnania and Aetolia, with some part of Achaia. Charles died at Ioannina in July 1429, leaving Arta and Ioannina to his nephew Count Charles III., and Aetolia in possession of two illegitimate sons. The disputes which arose between the cousins rendered the whole country an easy conquest to the Turks, to whom Ioannina capitulated in October 1431. The name of Karlili, or the country of Charles, is still attached by them to a large portion of Acarnania and Aetolia.

§ 4.205 CHAPTER 38 EPIRUS. JULY 29.—Setting out yesterday evening at sunset, I rode from Ioannina to the monastery of Eleokali on Mount Drysko, on the way to Kalarytes, and this morning at daybreak proceeded as on a former occasion, by the Khan on the left bank of the river Arta, and by that of Golfi, on the ascent to the church of St. George, where we arrived at 9.30, a.m. A wide portico round the church is the only accommodation for travellers. Our arrival having been announced to the town by the firing of musquets, a party of the principal inhabitants, headed by Kyr K. Turturi, acting as hodja-bashi for his brother, and accompanied by the Subashi, or Albanian governor, preceded by a gypsey band of music, arrive in an hour at the church.

§ 4.206 They bring with them a lamb roasted whole, wine, bread, and salad; and soon after their arrival we all sit down to dinner. A long cloth is spread in the portico, the meat is partly cut and partly torn in pieces by the servants, and spread about the cloth; some sit upon carpets, and some upon the pavement. The gypsies strike up their music, which consists of two drums, two violins, two tabors, a sort of oboe, with another wind instrument, and a fife out of which they produce the most piercing notes. The vocal performances with which they accompany it are equally distressing to the ears, to make as much noise as possible being the chief concern. After dinner we have some Kleftic songs, of which the exploits of the Suliote heroes are the subject, and those of the celebrated robber Kartz-Andonio, who slew the no less famous Bolu-bashi Vely Gheghe, sent against him by Aly Pasha. These heroic songs are followed by erotic, with a chorus of Po, Po, Po. The Albanian governor then rises, and leads the dance with bare feet.

§ 4.207 At 10 we descend the ridge through the wood before described; both Turks and Christians firing their musquets as we proceed, and causing a surprising increase of sound and echo amidst the precipices which overhang the winding river. It is remarkable, that at the river itself the same explosion produces no more than the ordinary sound, without any echo. After three quarters of an hour consumed in ascending the zig-zag path from the river to the town, we enter it among crowds of spectators. Kalarytes has increased since 1805 in riches and comfort, and is almost the only place that reflects any credit upon Aly’s government. Every year the return of the merchants to their native country produces some new houses. There are now upwards of 500; and 620 heads of families are enrolled in the taxbook, those of the first class are rated this year at 800 piastres, the second at 400, the third at 200, and the fourth, which consists of many sub-divisions, from 100 to 5. A few of the poorest families pay nothing. The whole amount received by the Vezir is 70,000 piastres, or about 4,000£ sterling. The town has now a public debt of 250 purses, the interest of which is to be provided for, in addition to the sums just mentioned; it is lower than in any other place, on account of the better security, and 10 per cent, from the κάσσα, or public chest of Kalaryrtes, is considered by the monied men of Ioannina as preferable to a higher gain anywhere else. The Vezir sometimes however, when he wishes to satisfy a favourite, sends him with a buyurdi, ordering the Kalarytiotes to take the bearer’s money at 12 per cent, whether they want it or not.

§ 4.208 The lands around the village, which formerly supplied the inhabitants with corn for a part of the year, have been neglected with the increase of wealth and population, as it answers better in general to import corn and flour than to cultivate such a wretched soil. In the present year, however, they feel the loss of this culture, in consequence of a dearth of corn in Thessaly, which, it is reckoned, has caused an expenditure of 60,000 piastres for provisions beyond that of last year. A part of the territory is destined to the pasture of oxen, for which 3 piastres a head are paid to the kassa of the town: for a horse or mule fed on another common destined for them, 5 piastres. The remaining pastures are destined to sheep and goats, the charge for which is 500 piastres a year for a mandra of 2000. The Eleftherokhoria of the mountains of Greece in general derive their principal revenue from their pastures, which are common property. This illustrates the ancient ϊπινομία,—a privilege which was often conferred, together with other rights of citizenship, on foreign benefactors, who then had the right of feeding their sheep and cattle on the same terms as citizens. Sometimes the pasturage was in part let for the benefit of the ταμειον, or public chest, or allotted as security for borrowed money. Building is expensive at Kalarytes. A woman who brings a large stone upon her back from the quarry, which is about a mile from the extremity of the town, receives each time 6 paras, and can make ten trips per diem; the expence in quarrying is 2 paras more, so that by the time the stone is put into the wall it costs not less than 10 paras: the smaller stones are brought by mules.

§ 4.209 Plakes, or rough slabs of the same kind of calcareous stone, for roofing, are 10 piastres the hundred, great and small. Timber is cut and brought from Pramanda and Melisurgus, or from a large wood on the eastern side of the mountains in the road to Trikkala, distant about 3 hours. A scantling of fir from the forest of Pramanda, which is generally brought on the shoulder of the person who shapes it there, is sold here from 35 to 40 paras. According to a sumptuary custom of this republic, which has all the force of a law, the head of a family of the first class cannot give his daughter more than 1000 piastres dowry, her wardrobe included: the other classes in proportion. Another confines the dresses of the women to particular kinds of stuff. The Albanian Subashi who is head of the police has only two palikaria to assist him. He decides all trifling differences, receiving a fee from the parties, and even adjudicates in small processes for debt, for which he takes 10 per cent.; but all the more important civil questions are subject to the arbitration of the primates, and are ultimately submitted to the Vezir. Another perquisite of the Subashi is 2 per cent, for collecting the contributions, according to a list furnished by the hodja-bashi and his assessors. The Subashi has the power of punishing in three modes: 1. by imprisonment; 2. by quartering his palikaria upon the house of the offender; 3. by turning out the family and sealing up the door; all which in the greater number of instances may be remitted for money.

§ 4.210 Without quarrels in the village the Subashi would starve, as he receives no pay from his master. But in fact his place is so desirable for a poor Albanian soldier, that he takes care never to exceed his powers, or to give cause for complaints against him by the primates. Matzuki has become a tjiftlik of the Vezir since my last visit to these mountains. Unable to pay the impositions, the poor villagers were obliged to borrow money at Ioannina or elsewhere, at an interest of 20 per cent., or even at 2 per cent, per mensem. Their difficulties having been of course increased by this measure, some of the inhabitants fled to Agrafa, the rest presented themselves to the Vezir with an offer to sell the whole village and its territory. The price demanded was 12 purses and the public debt. His Highness had no difficulty in declaring the place his tjiftlik; but instead of 12 purses he gave only 2, and instead of paying the debts, referred the creditors to the Matzukiotes who had fled to Agrafa. Khaliki, at the sources of the Achelous, once the most important modern village of Pindus, and from whence came many of the chief families of Kalarytes, is on the point of being deserted on account of the excessive burthen of the taxes, and of a debt of 100 purses. When the village was in its prosperity the inhabitants abandoned their corn for sheep, and have now very little of either, their property consisting almost entirely of horses and mules, with which they gain a livelihood as carriers. Yet the annual contribution is still from 400 to 700 piastres from the head of a family.

§ 4.211 Last winter an avalanche buried ten or twelve of the houses in the village, and filled the bed of the river. I before remarked that the name Khaliki, a common Romaic corruption of Chalcis, explains a verse of Dionysius Periegetes, in which he states that the Achelous rises at Chalcis, but I was not then acquainted with a passage of Stephanus, which confirms it. It is in alluding to the involution of the Echinades in the mud of the Achelous that the remark of Dionysius occurs, in reference to which Stephanus observes that Chalcis was a city of Aetolia, from whence the Achelous flows. There may be some inaccuracy in describing this country as Aetolia, but neither Stephanus nor Dionysius could possibly have had in view the Chalcis of Aetolia, as some critics have supposed, for that place, of which the ruins still exist on the coast opposite to Patra, was more than 20 miles distant from the nearest part of the Achelous. The only ancient position besides Chalcis in the mountainous country around Kakardhista that can be stated with any confidence is that of Theudoria, a place mentioned only by Livy, on the authority of Polybius, but from whom it appears to have been one of the chief towns of the Athamanes. The resemblance of name seems to identify it with the modern Thodhoriana, a village situated near Mount Tzumεrka in a pass which leads from the Achelous to the Arachthus.

§ 4.212 Theudoria was recovered from the Macedonians, with the other towns of Athamania, by the expelled king Amynander, with the assistance of the Aetolians, in the year B.C. 189, just before the siege of Ambracia by the consul Fulvius. Argithea is mentioned by the historian on this occasion as the capital of Athamania, and as situated amidst rocky mountains and deep valleys. Tetraphylia was the royal treasury; the other towns were Heracleia and Theium, besides which were Ethiopia, near Argithea, and the fortress Athenaion, apparently not far from Gomphi. Philip, as soon as he heard of the defection of Athamania, marched with all his forces from Gomphi towards Argithea, but having been repulsed at Ethopia, he was obliged to effect a retreat to Gomphi, during which he sustained great loss until he had crossed a certain river, in consequence of the mountainous nature of the country, and the better knowledge of it possessed by his enemies, the Athamanes and Aetolians. As Gomphi was in some part of the plain of Trikkala, it follows, if Theudoria was at Thodhoriana, that Athamania extended from the plain of Trikkala to the crest of the Tzumεrka chain, or perhaps to the river Arta, thus comprehending the modern Aspropοtamo, and a part of Agrafa. That a large portion of the valley of the upper Achelous was included in Athamania, seems evident from the name Paracheloitis, which Livy shows to have been a part of Athamania. The districts of Matzuki, Kalarytes, and Syrako, which are so remarkably separated from the rest of the world by the surrounding ridges, I take to have been the country within the narrow limits of which the once extensive western Perrhaebi were reduced in the time of Strabo, for the geographer describes them as situated to the north of Aetolia, near the Athamanes and Dolopes, and as occupying the western side of the summits of Pindus.

§ 4.213 Aug. 20.—This afternoon having recrossed the bridge at the foot of the mountain of Kalarytes, I leave to the right the ordinary road to Ioannina, which crosses the ridge of St. George, and follow that which leads to Pramanda and Arta along the eastern side of the same ridge. At the end of an hour from the bridge, we arrive at Kiepina, a monastery formed like that of Megaspilio in the Morea, by means of a wall built in front of a cavern, but on a diminutive scale compared with that building, and containing only a small church with two apartments, inhabited by two monks and a young laic. The cavern is very curious, as being the entrance of a horizontal passage into the body of the mountain, of which the monks affirm, that neither they nor any other person have ever yet reached the extremity.

§ 4.214 I followed the passage for 20 minutes by the watch, without any considerable ascent or descent, over a level ground of hard clay, and without meeting any impediment, except occasionally that of stooping under some projections of the roof, or of climbing over some hollows where a single plank would save the trouble. Not having been able to procure a sufficiency of candle, I was obliged to return; the air was cold and loaded with vapour, which increased as I advanced. Near the entrance of the cavern, the sides are a bare calcareous rock; in the farther parts are some large stalagmatic columns. The monastery is situated exactly in the gorge which gives passage to the united stream, formed by the three branches of Matzuki, Kakardhista, and Kalarytes, or Syrako, and not far above the junction of this river with the Arta, or Arachthus. It commands a magnificent view of mountain scenery. Below the monastery, on the side of the hill towards the river, are some gardens watered by springs, which there issue from the mountain. A few years ago Kiepina was a scene of action between the people of Kalarytes and Syrako, upon an occasion on which they ought rather to have united against Aly Pasha, but which exemplifies the characteristic readiness of neighbours in every part of Greece to break out into hostility. A Kalarytiote was on the eve of marriage with a Syrakiote girl, whose family was connected with his own, when the Vezir having received intimation of the beauty of the girl, by a Kalarytiote who is in his service, dispatched some men who took her out of her father’s house at night.

§ 4.215 All Syrako was instantly in arms, and intercepting the road to Ioannina, forced the abducers to retire into Kiepina, where they were blockaded by the Syrakiotes. As the monastery belongs to Kalarytes, the Vezir’s agent had no great difficulty in persuading a body of his countrymen to proceed to the rescue of His Highness's emissaries. Some fighting ensued, when the affair appearing serious, Aly affected to consider the persons who carried away the girl as belonging to a band of robbers, and ordered peace to be made between the two towns, on condition that the girl should be taken out of the monastery to Kalarytes, and from thence be restored to her parents at Syrako; and thus for once the Pasha was obliged to give up his point. Some shepherds who are feeding their flocks around Kiepina confirm the existence of an absurd custom in these mountains, which I had often heard mentioned by the Kalarytiotes. With the view of making their sheep healthy and strong, and the flesh coarse and ill-flavoured, the first for the sake of enabling the sheep to resist the weather, and the latter to render it less tempting to the wolves, they are in the habit of taking a piece of the fibula of a dog, two inches long, and of inserting it into the fleshy part of the thigh of the lamb when it has nearly attained its growth, after which the opening is sewed up. So persuaded are they of the efficacy of this custom, that the shepherd at Kiepina expressed his belief, that lambs born of a ram or ewe so treated have a similar bone. Such a practice could only obtain, where the greater part of the lambs were destined only to be shorn, to breed, and to make cheese.

§ 4.216 The bone, doubtless, is soon carried away by suppuration. As education extends in Greece, this absurd custom, which is already ridiculous among the higher class, will gradually cease, as well as the use of charms and some other superstitious practices which still prevail among the common people, especially among the women. It would be difficult now to meet with an example of the most barbarous of all those superstitions, that of the Vrukolaka. The name being Illyric, seems to acquit the Greeks of the invention, which was probably introduced into the country by the barbarians of Sclavonic race. Tournefort’s description is admitted to be correct. The Devil is supposed to enter the Vrukolaka, who, rising from his grave, torments first his nearest relations, and then others, causing their death or loss of health. The remedy is to dig up the body, and if after it has been exorcised by the priest, the demon still persists in annoying the living, to cut the body into small pieces, or if that be not sufficient, to burn it. The metropolitan bishop of Larissa lately informed me, that when metropolitan of Grevena, he once received advice of a papas having disinterred two bodies, and thrown them into the Haliacmon, on pretence of their being Vrukolakas. Upon being summoned before the bishop, the priest confessed the fact, and asserted in justification, that a report prevailed of a large animal having been seen to issue, accompanied with flames, out of the grave in which the two bodies had been buried. The bishop began by obliging the priest to pay him 250 piastres: (his holiness did not add that he made over the money to the poor). He then sent for scissors to cut off the priest's beard, but was satisfied with frightening him. By then publishing throughout the diocese, that any similar offence would be punished with double the fine and certain loss of station, the bishop effectually quieted all the vampires of his episcopal province.

§ 4.217 Aug. 21.—We pursue the road from Kalarytes to Pramanda to the foot of the mountain, cross and recross the river, and then ascend a beautiful woody slope where the soil appears fertile, to the small ruined village of Mikhalitzi, distant one hour and twenty minutes from Kiepina, and thence, in an hour and a half, through rocky passes over a bare mountain which borders the left bank of the Arachthus to Kuliaradhes, a village in a lofty situation, not far short of which is στο Καστέλι, a place so called from the foundations of a Hellenic building of small dimensions, which were brought to light not many years ago in clearing the ground of wood. Opposite to Kuliaradhes, beyond the precipitous gorges of the Arachthus, is seen the district of the Tomaro-khoria, one of which, Fortosi, stands on the edge of the cliffs immediately opposite to Kuliaradhes. Four or five of these villages are situated on an elevated, fertile, and well-cultivated vale, lying below the northern side of the summit called Xero-vuni, and watered by a tributary of the Arachthus. The remaining villages stand on more distant slopes of the same mountain towards the plain of Ioannina. Lozetzi, the largest of the Tomaro-khoria, lies to the northward of Fortosi, in a lower situation. Farther on the descent occurs Seriana, a small place three hours from Ioannina, where are said to be some remains of antiquity.

§ 4.218 Kuliaradhes is a tjiftlik of Mukhtar Pasha. The inhabitants complain of the expence to which they are subjected in finding fire-wood for his use, and that of the Vezir, their lands not producing any. But none of the villages within a certain distance of Ioannina are exempt from this charge. The supply of snow to the palace is another severe angaria; and the more burthensome to individuals, as it is required only from the places which are near the glaciers. In the beginning of summer the snow is furnished by the villages of Mount Mitzikeli, then by Kotzista and Bozgoli, and towards the end of the season by Syrako. On leaving Kuliaradhes our road lies for an hour along the summits, where the land produces corn or feeds cattle. On the right of us, in a high situation, is Vestavetzi; soon after passing which, we begin a very steep and tedious descent to the Arta, where the road is rendered so difficult by the loose soil and stones, that it is only practicable on foot. We cross the river at the ruins of a bridge where formerly stood some mills. The place is called Tjimovo. In building the bridge advantage was taken of a great rock in the middle of the river which served for a pier. There are some remains of other piers made of bricks and mortar, apparently of the time of the Roman or Byzantine empire. From hence, in three-quarters of an hour, we cross the heights to Prodhovali, a village of eight families, situated at three hours’ distance from Kuliaradhes, on the edge of the plain which separates Mount Dhrysko from the hill of Kastritza.

§ 4.219 Having lodged here under an arbour of vines attached to one of the cottages, I proceed in two hours to Ioannina, on the morning of the 22d of August. Ioannina, September—October, 1809.—On the 1st of September, Omar Vrioni Bey, of Verghiondi, near Berat, entered Ioannina with a suite of led horses and mules, baggage, and attendants, on his return from Egypt, after eight or ten years spent in warring with the Mamluks and plundering the country. His harem and treasures have been sent forward to his native town. Several of the led horses are destined as presents to the Vezir, together with a valuable Damascus blade, riehly mounted, a mule which cost 8,000 piastres, and a hedjin, or saddle-camel, with its furniture.

§ 4.220 The Porte, having little hope of regaining its authority in Egypt while the Albanians remain there, is endeavouring to detach the principal chiefs from Mehmet Aly, in which Aly of Ioannina willingly concurs, having heard of the riches which many of his countrymen have accumulated in that excellent field of plunder, and having a good prospect of obtaining a portion of whatever may reach Albania. A mulatto, brother of Hassan Aga of Margariti, who, before he went to Egypt, was a robber among the Khaitali of Thrace, and who submitted himself to the Vezir some years ago at Monastir, is now one of the richest Albanian Beys in Egypt, and refuses to return, being equally afraid of Aly Pasha, and of his own brother, who is still in Albania, and with whom he had never been on good terms. Τούρκο είδες; ασπρα θελει, a modern proverb, meaning literally that a Greek never sees a Turk approach without knowing that he is coming to demand money, is too applicable to the character of the Turks in general, and especially to those in office, from the highest to the lowest degree. There may be a few examples of moderation in Asia Minor, but among the far greater number of chieftains who have established an independent authority, such as has left them at liberty to pursue their own ideas of governing, the same extortion has prevailed as under the obedient delegates of the Sultan, nor have the countries so ruled experienced any relief from the fatal influence of the Turkish system. It is evident from the account of those who have lately returned from Egypt, that the remark applies not less to Mehmet Aly of Cairo than it does to Aly of Ioannina.

§ 4.221 Aly, since he has become of political importance in Europe, shows some wish that foreigners should have a favourable opinion of him. Nevertheless, he has little scruple in alluding to those actions of his life which are the least likely to obtain such favour, though he generally endeavours to give such a colouring to them as shall make them appear less criminal. Sometimes he manifests a wish to perpetuate his power in his own family, but evidently without great hopes of success, betraying proofs not only that he has little confidence in the ability of his sons to maintain his conquests, but that he even entertains lively fears for his own safety, as he often inquires whether, in case of being driven out of his native country, he should find security for his person and property in the British dominions. Sometimes he listens to counsel for a moment, and endeavours to amend the vices of his government. In August, by advice of the bishop of Larissa, he summoned deputies from all the villages of Trikkala, deposed the hodja-bashis who had acted oppressively, and substituted others. But this is the extent of his reforms. He is perfectly aware that his subjects detest him; and lately in conversation with one of his ministers, remarked, that he should very much prefer the love of his people if it answered equally well to his treasury. He is not incapable of understanding that it might answer better in the end; but his habits are now too inveterate to allow him to act upon such views, and his favourite maxim of να ειμαι καλά εγω, which has actuated many an illustrious despot, not so honest as Aly in declaring it, will doubtless continue to be the rule of his conduct to the end, as well as that sentiment naturally arising from it,

§ 4.222 which Nero is said to have expressed in the words εμου θανοντοc, γαια μιχθητω πυρί. He may perhaps find some excuse for such short-sighted policy in the constant state of anxiety and suspicion in which he is placed by the known treachery of the Supreme Government, by the hatred which the Osmanlis in general entertain against him as an Albanian, by the personal hostility of the Sultan, and by the conduct of his ambitious neighbours in the Islands. It can hardly be doubted that he would better consult his safety by increased efforts to strengthen himself, as well by military power as by cultivating an influence among those who constitute three fourths of his subjects, having already paved the way for the latter by plundering and degrading almost all the Mahometans within his reach, while he favours and employs the Christians, though in neither instance with any longer views than those of immediate advantage. In augmenting his possessions and power in Albania, he seems to proceed upon a more settled principle, and one which offers the better prospects to his ambition, as the Turks have never so completely subdued Albania as to destroy hereditary power and influence, or have been able to keep the country in a state of subjection, but by promoting a balance of power between the principal chieftains,—the best policy, in fact, for the Sultan to adhere to, as he is sure of the mercenary services of the Albanian soldiers whenever they are wanted, and cannot reasonably hope, even if he were to achieve such a conquest of Albania as Mahomet the Second succeeded in effecting, to maintain a permanent and complete authority in the country, which neither that victorious monarch, nor any of his successors, when the military character of the Ottomans was in its meridian, were able to accomplish.

§ 4.223 It must be admitted that the success with which Aly has indulged his ambition in Greece and Albania, not only in defiance of the Porte, but hitherto with a constant increase of influence over the Supreme Government, is a proof of skill, foresight, and constancy of purpose, in which few statesmen or monarchs have ever excelled him, and shows that had he any enlightened and steady views of benefit to his country, he has the talent requisite for pursuing them to completion. He sometimes compares himself to Burros, because Pyrrhus was his predecessor in Epirus, and possibly because Pyrrhus is the only great man of antiquity he ever heard of except Alexander; of Alexander's father at least, whom Aly most resembles in character, I find he has no knowledge. He shows equal art and activity in the various measures of force or fraud by which his advantages are obtained, and exhibits a degree of patience and command of temper, especially when the object is to gain partisans to his cause, which in such an impetuous character is very remarkable. It is surprising to see with what apparent good humour he listens to the interminable discourses of every petty Albanian officer, whose momentary importance may give him the enviable privilege of conversing with the great chieftain. Aly himself also probably takes some pleasure in a mode of transacting business from which he has derived advantage through the whole course of his career, and in exercising upon these men his talents of flattery and deception.

§ 4.224 He professes his determination not to make peace till he has obtained Berat, but admits that he is tired of the contest, which has already cost him 1500 purses, though it began only in May, and that in addition to his expenditure in Albania, he is obliged to meet the loss of credit at Constantinople, which has been the consequence of his having undertaken this contest against the will of the Porte. In fact, it operates as a diversion in favour of the external enemy, by preventing many of the Albanians from recruiting the Grand Vezir's army on the Danube. Though there is great difficulty in ascertaining correctly what passes in the Vezir’s harem, it is known that he never had but one wife, the sister of Khotad and Morteza, Beys of Arghyrokastro, who is still alive; unless the widow of a rich Turk of Ioannina, whom he married for a day in order to obtain her property, may be called another. He asserts that he has 200 women in his harem; for these are subjects on which he has no scruples in conversing. Like most Turks, he is desirous of consulting medical men, the great objects of such inquiries being philters and poisons. His women are all either slaves bought at Constantinople, or presented to him by Turks, or they are Greek women, noticed for their beauty by him or some of his servants, and conveyed by his order to the harem.

§ 4.225 His only favourite at present is a Christian Albanian, from the neighbourhood of Tepeleni, whose father having been ordered to Ioannina, with his whole family, for some real or imputed offence about ten years ago, this girl, then a child, was remarked by Aly, and ordered to be educated in his harem. She is still a Christian, and allowed to have her chapel, and service performed by a papas in the palace. Indeed, he never troubles himself to make religious converts of either sex; on the contrary, it is more common to see the boys who are brought up in the serai in his service, reading and writing with the Greek papas than with the Turkish hodja. Nor has Aly ever deprived any of the higher class of Greeks of their daughters. With the exception of the favourite and two or three of her attendants, whom he often removes in a close carriage at night from one palace to another, at which he intends to pass the ensuing day, not one of the unfortunate inmates of the harem in the castle ever quits her prison, unless, as a rare instance of favour, to be married to one of his servants. There, very indifferently clothed, fed, and lodged, confined to latticed apartments, without amusement or exercise, in a situation where the air in summer and autumn is unhealthy, they cannot but soon lose their health and attractions. Indeed it may be said that when once the palace in the castle becomes their constant residence, they are as much neglected as the building itself. As few women, even of the higher classes in Ioannina, possess either elegance or beauty, it cannot be supposed that these peasant girls can have much to recom mend them after the first glow of health is worn off in their sickly confinement.

§ 4.226 Not many weeks ago a country girl was recommended to the Vezir by his Kalarytiote secretary, who caused her to be brought to the palace. In the course of a few days he resolved to have her married, and fixed upon P. a young man of Kalarytes for her husband, but who having obtained intimation of the honour intended him, immediately made off for Corfu, accompanied by another person of the same town and family. They had nearly reached the coast when the emissaries of Tahir Aga overtook them, and not knowing exactly which was the destined bridegroom, brought them both back to Ioannina. In the mean time, however, the Proestos of Kalarytes, having interceded with the Vezir, P. escaped marriage, and the girl was sent back to her parents. Scarcely any two persons agree as to the number of female children which Aly has had, but it is generally believed that several have been put to death. It is difficult to understand his reasons for this cruelty, as he has made so good a use of female alliances in the furtherance of his political projects, but the practice is supposed to be not uncommon among the great Turks, including the Sultan. They think probably that it enhances the honour of the alliance to have no more than the exact number of daughters required. It has often been remarked, that the life of a man is of no consideration in the East, compared to that which is attached to it among the nations of civilized Europe; but it is difficult for a native of the latter to conceive how much more strongly this remark applies to the female sex.

§ 4.227 The daily rains ceased at Ioannina towards the latter end of June, after which the heat was excessive till about the 10th of July; showery weather then followed, with a cool northerly wind till the 21st; the heat then increased daily till about the 26th, when it arrived at its maximum, the, thermometer at 2 p.m. ranging from 85° to 95° in the coolest parts of the house, and so continued during the first half of August. About the 12th of that month the cold north wind again set in, causing dysentery on board our ships off the coast. These sudden changes of temperature are one of the worst peculiarities of the climate of Ioannina. I have known the north wind blow for several days almost as hot as an Egyptian khamsin, and then suddenly become cold, without any rain having fallen within the visible horizon. At Kalarytes in the middle of August, the thermometer fell below 60°; on my return to Ioannina, on the 22d of August, the weather for about a week was calm, with a light north-west wind, and hot though very moderate compared with the two former periods of heat. About the first of September the southerly winds began, and continued with an accompaniment of violent rain and thunder for a fortnight. A letter which I received from Capt. Brisbane, senior officer of the ships on the coast, dated Sept. 16, stated that he had been prevented for many days from approaching the coast, and that the weather at sea had been severe in the extreme.

§ 4.228 After this little monsoon there was a series of the fairest and calmest days with sometimes very light north-westers in the afternoon, till the 28th September, when the southerly winds again set in, and the rains became almost incessant until I departed for Arta, on the 20th October. The wind seldom blows for twenty-four hours from between the west and south-east without bringing rain. From these and other remarks which I have before incidentally made, it is apparent that, in respect of climate, Ioannina is more northerly than any part of Italy, except the mountains, and may perhaps be ranked with Vienna. It is possible, however, that this year has hardly been an average one, and that there has been more rain during the winter and spring, and less hot and dry weather in the summer than usual. In some years the drought is said to be distressing, and in consequence of the dry gravelly soil is probably much sooner felt here than in Thessaly, for Providence seems to have admirably adapted the soils to the climates on either side of the Pindus, the deep rich mould of Thessaly requiring a much smaller degree of moisture to render it productive than the light stony calcareous soil of the greater part of Epirus, but particularly the plain of Ioannina, which can only be rendered productive by frequent and copious irrigation.

§ 4.229 Arta, October 1809.—The district of Arta contains 170 villages, in eight kolis or subdivisions:—1. Kambo, or the plain; 2. Luro; 3. Lamari; 4. Laka; 5. Kervasara; 6. Vrysis; 7. Radhovizdhi, or Radhovisi; 8. Tzumerka, or Tzumerniko. The chief villages in Kambo are Mehmetjaus and Rakhi, to the westward of Arta—Kostakius, half an hour from Arta in the road to Salaghora, and Neokhori, 2 hours from the sea on the right bank of the river of Arta, which is so far navigable to small vessels. Kambo is divided by the river of St. George from Luro, which contains the valley of Luro as far as the mountains of Suli. In Lamari, are Libokhovo, before described, as well as Kastro-sykia or -skia, Kamarina, Mikhalitzi, and Mytika. In Laka, are Lelovo —Pogortissa, one hour and a half from Tervitziana, which is in the district of Ioannina—Filipiadhes on the river of St. George half an hour from Strivina—Podhogora, in the plain of LOlovo opposite to that village, and Papadhates in a lofty situation an hour and a half from Podhogora and 2 hours from St. George. In Kervasara are Strivina, Kometzadhes, Muliana, Klisura, near Pendepigadhia, and Akoghi. In Vrysis, Peta is the only considerable village. In Rhadhovisi are Komboti, Skulikarya, and Velitziko, the two latter in the mountains. In Tzumerka are Pramanda, Melisurgus, Thodhoriana, Lupsista, and Vurgareii. The villages above named contain about 2000 families;

§ 4.230 all the others in the district of Arta being small, have not more than 3000, so that the whole rural population of Arta is about 30,000, to whom may be added for that of the town, 5000 Greeks, 500 Turks and as many Jews. The plain of Arta and the sub-district of Lamari belong entirely, with the exception of two Turkish tjiftliks, to the Vezir or his sons; and it is computed that the yearly revenue of their lands and flocks amounts on an average to 1400 purses. From the farmers he receives four-tenths of the crop in kind, which includes one-eighth in wheat, barley, and oats, and one-tenth in rokka, due to him as voivoda. Wheat and kalambokki are the principal produce of Arta, then wine, of which there are 20,000 horse-loads, but of indifferent quality, as the vineyards are in the plain; barley, oats, cotton, flax, tobacco, rice, and pulse. The landlord's wheat and maize are partly consumed by the troops and household of the Pasha and his sons: the tobacco and rice are sold at Ioannina. The exports by sea from the district in the commodities abovementioned, to which are to be added the oranges, lemons, and hazel-nuts, of the gardens of Arta, are reckoned to be of the annual value of 1000 purses. The Vezir pays to the Porte for the mukata of the voivodalik 300 purses, and receives more than that sum from the persons alone to whom he underlets the customs, and who collect them at Arta, Mytika, Luro, and Kastroskia.

§ 4.231 For the tithe of the lands not owned by him he receives annually 290 purses. A duty on wine collected at the wine-houses, he lets for 16,000 piastres; the kumerki, or excise on goods entering the town, together with the statiri, or fee on public weighing, for 10,000 piastres; the monopoly of tobacco, for 16,000. About 15,000 piastres are collected by his agent for the kumerki, or toll upon sheep and goats passing through Arta from the mountains of Epirus to feed in the winter in Acarnania. The subashilik and vostina, which are fees paid to the Vezir as possessing the ziamets and timaria of Arta, comprehending about a fourth of the district, amount to 13,000 piastres. Other contributions of various kinds accruing to him as governor are reckoned at 75,000; so that the amount of his revenue from his gain upon the mukata added to the produce of his landed property, including that of his sons, is near 2000 purses, or 60,000£ sterling, from the kaza of Arta alone. The customs of Ioannina, of the ports of the Forty Saints and Vutzintro, are let by the Vezir, together with the voivodalik of Ioannina, and an excise levied upon merchandize passing over the bridge of Raiko, or in entering the gates of Ioannina, for 450 purses. His estates in that district are more extensive than in Arta, but probably not more productive: he possesses, however, various sources of superior profit in the larger population, and the amount of his net revenue from both, may be estimated at 120,000£ per annum. It is supposed by the person who gave me the particulars of his Arteno property, and who is one of the farmers of the revenue, that his whole net income, exclusive of payments and presents at Constantinople, is about five millions of piastres, or 300,000£—a sum capable of effecting twice as much as in England, but for which he is charged, it must be remembered, with all the expence of the civil government and military defence of the country.

§ 4.232 The djezye-guebran (tribute of the infidels) commonly called the kharadj, is farmed by the Vezir from the farmer-general of this well-known capitation tax on male non-Musulmans. It amounts to three, six, or twelve piastres, according to the person's age, and amounts in the kaza of Arta to 35,000 piastres; in that of Ioannina to nearly twice as much. Upon this the Vezir makes some profit. Besides this and the other imposts which have been mentioned, each family is assessed for the local expences by the Proesti; the whole sum levied under this head in the district of Arta is about 100,000 piastres, without taking into account avanias and forced loans, which are seldom or never repaid in toto. To these burthens must be added also the quartering and feeding of soldiers, Turks, and public characters of all descriptions, as well as angaries, or contributions of horses, personal labour, and materials, which the Vezir frequently demands for public works, or more properly speaking, for the execution of his caprices in building palaces and castles. The old inhabitants of Arta speak with great respect of Suleyman, who was cut off by the Porte forty-five years ago, and in whose time those who now pay 700 piastres in khrei were not taxed more than 30, which, however, was equal to at the present day.

§ 4.233 At that time the spahiliks of the district belonged to Turks, and the remaining lands almost entirely to Greeks, subject to a tithe of about an eighth, the khardaj, and a few other general or local taxes. The mode in which Aly has acquired all this property, and the effects of the change upon Arta, is exemplified in the house in which I am lodged: like many others in the town it is large, and shows signs of former comfort and opulence, but belongs to a poor widow who can only afford to occupy a part of it, the Pasha having purchased all her landed property at her husband’s death for his own price. He was much indebted to Kyr P. of Arta for having arranged his finances in this district, and for having in particular made the customs much more productive than before. When at the close of his labours P. made an humble application for some remuneration, the only answer he received was: Turn and sit down prudently; your father died rich, yet I never tormented you. Arta supplies Ioannina with the greater part of its fruits and vegetables, particularly with the orange tribe, which are in a state of maturity here nine months in the year. The blossoms only which expand in September and October, produce fruit that does not ripen. The following method of planting slips of orange-trees is generally practised at Arta; the bark having been taken off round the place where the separation is to be made, a strip of sheep-skin leather is tied tight round the wood.

§ 4.234 A quantity of earth contained in two half pots is then placed at the ligature, and bound so as to be supported by the main body of the tree, in order that the branch may not be injured by the weight. Roots soon strike into the earth from the branch, after which it may be cut off and placed in the ground. In this manner a tree may be planted with fruit upon it, and will bear a good crop in one, two, or three years, according to its strength, instead of ten, which the seedling requires in coming to perfection. It is found that a tree, however good, improves in its fruit by being grafted every three or four years: the graft is taken from a choice tree, and sometimes from the same tree. In the church of the ευαγγελισμός της Παναγίας stands a square stele of marble or hard limestone, three feet high, adorned at the top and down the angles with a plain moulding, but broken and incomplete at the bottom. Whether this monument was not discovered at the time of my former visit, or whether I missed it, trusting too much to the guidance of Bishop Ignatius, I cannot assert. It is a dedication by the community of the deacons in the priesthood of Canopus to Sarapis, Isis, Anubis, and Harpocrates. Both priest and deities seem to have been an importation from Egypt in the time of the Ptolemies, with which date the form of the characters accords.

§ 4.235 Oct. 25.—Peta is a village of sixty houses, distant an hour's ride to the N.E. of Arta, on the heights just above the Arachthus, where it issues into the plain. On the river’s bank, below the village, stands the monastery of Theotoki. Two hours higher, at Sarandaforo, the Arta is joined by a large tributary flowing from Mount Tzumerka, above the left bank of which, on the mountain facing Tzumerka, are Vurgareli and the monastery of Vela. Between these places and Peta is a range of fertile heights, which are now quite uncultivated, the small villages, which formerly occupied them having been totally ruined by robbers. Peta, which has lately become a tjiftlik of Mukhtar Pasha has a small territory producing oil, corn, and kalambokki, all which are of excellent quality: its tobacco is not in such repute. Oct. 26.—From Peta to Komboti: distance I hour and 45 minutes, without baggage. To the right of the road, in the valley which is included between Petro-vuni, as the ancient Perrhanthe or hill of Ambracia is called, and the heights of Peta and Komboti, we pass the ruin of a building apparently of the time of the Lower Empire. The territory of Komboti is a fertile slope at the foot of the range of inferior hills which are backed by the great range, the continuation of Tzumerka. Komboti was once a large Eleftherokhori, but having become a tjiftlik of Mukhtar Pasha, is now in a declining state. It still produces, however, corn, wine, maize, tobacco, and oil, and consists of 120 houses.

§ 4.236 A river here issues from the mountains, and crossing the plain enters a lagoon which stretches along the shore from the north-eastern angle of the gulf to within a short distance of the mouth of the Arta. Formerly there was a saltwork in this lagoon. At the head of the valley, above Komboti, an hour distant from thence, are said to be the remains of a Hellenic town, at a place called Kastri. After dining with the Proestos, I descend to Koprena, the name of an anchorage at the mouth of the river of Komboti, and embark at sunset in a Kefaloniote boat which I had ordered from Salaghora. Koprena is midway between the mouth of the river of Arta and Menidhi, a small bay at the north-eastern angle of the gulf, exactly at the beginning of the pass of Makrinoro. On a height which rises from the bay of Menidhi to the N.E. are the remains of a Hellenic fortress which commanded the northern entrance of the pass. The ruins are called Paleopyrgo. In the night we sail to Vlikha, a distance of about eight miles in a right line, but by land reckoned a march of five hours, of which the pass of Makrinoro is about the half in distance, but greater in time, the road being very bad and impeded by woods. Arapi and Vlikha are small tjiftliks, each consisting of a tower and quadrangle of cottages, situated, the former at the northern, the latter at the southern end of a height called Mavro-vuni, which is covered with wild olives, and projects to the westward of the general line of the coast, thus forming a promontory at the head of the gulf.

§ 4.237 Between Mavrovuni and the great range of mountains which fall on the opposite side to the Achelous is a plain, commencing at the southern extremity of the Makrinoro, and extending to the valley of Xerokambo, which is on a higher level. From Arapi the coast retires eastward, and forms the bay of Kataforno, where a lagoon occupies the lower part of the plain, extending two miles from the foot of the height of Arapi. Here was formerly a valuable fishery, which has been abandoned in consequence of the kleftic wars. Two torrents descend into this lagoon, one from the back of Makrinoro, the other from the mountains on the eastern side of the plain, where are situated the villages of Syndhikno and Dunista. Beyond Vlikha, southward, the coast retires to Armyro, another lagoon, or rather shallow bay; for it has an opening of considerable breadth and depth between a low point and the cape of Spartovuni, which forms, with cape Kendromata on the opposite side, the entrance of the bay of Kervasara. The bay or lagoon of Armyro extends eastward to Xerokambo. On the summit of the cape which terminates the ridge of Spartovuni, are the ruins of a small fortress called Kastriotissa, which consists partly of Hellenic and partly of more modern work. Below it, on the border of the bay, are a few houses named Armyro.The plain of Vlikha, although not less fertile than that of Arta, is now cultivated only about Arapi, Vlikha, and a third hamlet called Neokhori, lately established by the Vezir, and peopled with Prevyzans, whom he has deprived of their property at Prevyza on pretence of their having aided the French against him in the war. The plain in the uncultivated parts is chiefly covered with fern, and there are many large plane-trees on the banks of the torrents.

§ 4.238 Oct. 27.—At Neokhori, distant three-quarters of an hour eastward of Vlikha, on the last fall of the mountain, are the ruins of an ancient city, the general form of which may be understood from the annexed sketch. [sketch] The walls were more than a mile in circuit; and though very little of them remains above ground, they are traceable in every part, except in a marshy level near the village.

§ 4.239 The masonry, like that of Ambracia at Arta, is nearly regular, and is thus unlike that of the Acarnanian ruins, which are generally of the second order. A peaked hill formed the citadel, the wall of which is still traceable. Without the walls, on the southern side, are the foundations of a large quadrangular building, probably a temple; but not a fragment of sculpture is anywhere to be seen, except a fluted cippus in the village church. The city was well protected by the mountain, which rises abruptly to the east, by a deep ravine to the north, and to the south by that of the river of Ariadha, which here enters the plain, and seems recently to have taken a course to the southward of its former direction. The ancient site is in many places overgrown with trees; and not far from it inland begin the impassable woods, consisting of large oaks mixed with underwood, which extend to the Aspro, with the interposition only of a little cultivated land around a very few villages. Ariadha, the largest, is two hours above Neokhori, bearing east from Vlikha. Notwithstanding some objections which may be deduced from Strabo and Thucydides, I believe these to be the ruins of Argos, and that the river of Ariadha, which flows on the southern side of them, is the Inachus—a name derived, as well as that of the city, from the Peloponnesian Argos, from whence a colony founded the Amphilochian Argos about the time of the Trojan war.

§ 4.240 The following are the objections to this opinion. Strabo reports Hecataeus to have asserted that the river of Amphilochia, which took its name from.the Inachus of Peloponnesian Argos, had its rise in the same Mount Lacmus in which were the sources of the Mas or Aous; that the former, flowed to Argos, and discharged itself into the Achelous; the latter into the Adriatic Sea. In some verses of Sophocles cited by the geographer, the same origin is ascribed to the Inachus, and the Perrhaebi are said to have occupied the country at the sources of the river. If this were a correct description of the Inachus, it could not possibly have flowed into the Ambracic Gulf, and Argos of Amphilochia should be sought for to the eastward of the great ridge lying between the gulf and the Achelous.

§ 4.241 But Argos was certainly near or upon the shore of the gulf, as appears from Strabo himself, who, in contradiction to the tenor of his citations from Hecataeus and Sophocles, testifies that the Amphilochi occupied the coast of the Ambracic Gulf, between Ambracia and Acarnania, and that the Inachus flowed into the gulf. And of this there cannot remain any question, upon a reference to the still better authority of the historians, Thucydides, Polybius, and Livy; the first of whom relates some transactions which absolutely require a position for Argos on or very near the shore of the gulf; while Polybius describes Argos as 180 stades from Ambracia, towards Acarnania; and Livy, who copies him, states the same distance at twenty-two Roman miles. From these facts it is evident, that the plain of Vlikha was a principal part of the Amphilochia, and that Argos was in some part of that plain. It would seem, therefore, that Hecataeus was misinformed as to the course of the Inachus and the situation of Argos, and that Strabo had not a knowledge of the country sufficient to correct the historian.

§ 4.242 As to the verses of Sophocles, their weight, as a geographical testimony, is much diminished by their forming part of a passage in which the poet represented the Inachus, after flowing to the Achelous, as then crossing the sea, and re-appearing in Lyrceia of Argolis, an acknowledged fable, justly compared by Strabo to that of the Alpheius flowing to the fountain Arethusa at Syracuse, to that of the Nile flowing to the Inopus of Delos, and to that of the origin of the Sikyonian Asopus in Phrygia.

§ 4.243 The strongest objection to Neokhori as the site of Argos is, that Thucydides describes Argos as a maritime city, which, it must be admitted, better suits the remains at Kervasara, the only place besides Neokhori near the eastern shore of the gulf, where any remains are found deserving the character of a polis,—all the others in this quarter being those of fortresses or of comae. At Kervasara there are not only the fortifications of a large town, but they stand so near the sea as to answer perfectly to the description of επιθαλασσια. Kervasara, however, is considerably more than twenty-two Roman miles from Arta: there is no river corresponding to the Inachus, and the position seems exactly to accord with the description of Limnaea, as lying on the confines of the country of the Agraeae, and as being the nearest harbour to Stratus, or that which afforded the most short and convenient approach to that city from the Ambracic Gulf. It may be thought, perhaps, that assuming the ruins at Kervasara to be those of Argos, Limnaea may be placed at Lutraki, or at Ruga, where the situation of the ruins in a lake would be well adapted to that ancient name; but in this case Argos would have been exactly interposed between Limnaea and the Agraei, which is contrary to Thucydides. Nor would Limnaea in that case have been on the road from Argos to Stratus, as it appears to have been upon two occasions described by the same historian:—1. When Cnemus the Spartan, in the third year of the Peloponnesian war, (B.C. 429,) invaded Acarnania, in conjunction with the Epirotes and Ambraciotae;

§ 4.244 on which occasion he proceeded from Ambracia through the Argeia, and having ravaged Limnaea, marched from thence to Stratus. 2. When Eurylochus in the sixth year was opposed to the Athenian allies in the Amphilochia, to which occurrence I shall presently revert. It is true that Thucydides in the former passage describes Limnaea as a small unfortified town; which is better suited to Lutraki, where no vestiges of Hellenic antiquity are visible, than either to Ruga or Kervasara. We may easily conceive, however, that the importance of the situation of Kervasara may have caused that place to have been augmented and fortified subsequently to the events related by the historian: nor is the name Limnaea unsuitable to Kervasara, there being a marsh near two miles in length, at no great distance inland from the ruins. I am still, therefore, disposed to adhere to the opinion that Kervasara was the ancient Limnaea. It is no slight evidence of Argos having been near Vlikha, that I purchased from the peasants of that village three coins of that city in copper, of great rarity, and which I had never before met with; for coins of cities in that metal, unless where the coinage was very abundant, are seldom found at any distance from the places themselves. As to the adjective επιθαλασσια applied by Thucydides to Argos, it is to be observed, that the inlet of Armyro, although very shallow,

§ 4.245 does not resemble the other lagoons around the gulf of Arta, which, like such lakes in general, are separated from the sea by stripes of very low land, through which there are one or two narrow entrances. Armyro, on the contrary, has a considerable depth of water at the entrance, with a breadth of three or four hundred yards, and is still one of the Skales or harbours of the gulf. It is very possible, therefore, that the part of the inlet nearest to Neokhori, which is now a marsh or a lagoon, according to the season of the year, may have been rendered shallower than it was formerly by the alluvion of the rivers, or by other causes which constantly though variously operate on the coasts of Greece, and that it may once have afforded a commodious harbour to Argos. There may still perhaps be another conjecture as to the site of Argos, namely, that it stood at Vlikha, a word having some appearance of being a corruption of Amphilochia, and that the ruins at Neokhori are not those of Argos, but of some other city,—for example, that of the Agraei. But this would leave an insufficient space at the head of the Gulf for the Amphilochia. Nor will this or any other situation, except that of Neokhori, perfectly accord with the mention of Argos by Thucydides on the occasion already referred to, when in the winter of the sixth year of the Peloponnesian war, the Amphilochi and Acarnanes, headed by Demosthenes the Athenian, gained a complete victory over the Ambraciots and their Peloponnesian allies, under Eurylochus the Spartan.

§ 4.246 Eurylochus, after having failed in an attempt upon Naupactus, had marched into Aetolia, and instead of returning into the Peloponnesus, had remained at Proschium in that province until the winter, with the view of assisting the Ambraciotae against Argos, after which it was intended, in case of success, to proceed against the allies of Athens in Acarnania. When Eurylochus.learned that 3000 Ambraciot hoplitae, advancing from Ambracia, had occupied Olpae, a strong fortress upon a height above the sea, 25 stades from Argos, he advanced from Proschium, through the territories of Phcetiae, Medeon, and Limnaea, into the Agrais, which was friendly to him, and from thence, having crossed Thyamus, an uncultivated mountain, he entered the Argeia in the night, passing unperceived between Crenae, where a body of Acarnanes had been stationed to prevent him, and the city Argos, where the rest of the Acarnanes were assembled with such of the Amphilochi as had not been prevented by the Ambraciotae. Having thus effected a junction with the Ambraciotae at Olpae, he took post with the combined force at Metropolis, soon after which Demosthenes arriving in the Gulf, anchored near Olpae with twenty Athenian ships, on board of which were 200 Messenian hoplitae from Naupactus, and sixty Athenian archers.

§ 4.247 Having disembarked these, and taken the command of the Acarnanes and Amphilochi, he encamped near Olpae, where he was separated only by a great ravine from the army of Eurylochus. It was not until the sixth day that the opponents drew out their troops for battle. Demosthenes, who was on the right with the Messenians and the archers of Athens, opposed to Eurylochus and the Peloponnesians, finding himself in danger of being outflanked by means of the numerical superiority of the enemy, concealed 800 Acarnanians, half hoplitae and half light armed, in a hollow way. The stratagem was successful: Eurylochus in attempting to turn the right of his adversary was attacked in the rear by the Acarnanians, who were in ambush, and was slain, with many of his best men. The Ambraciotae in his right wing, meantime, had so far prevailed over the Acarnanes and Amphilochi opposed to them as to drive them towards Argos, when perceiving the defeat of the other part of their line, they turned, and found some difficulty in making good their retreat into Olpae. The next day Menedaeus, who succeeded to the command, made proposals for permission to retreat, when Demosthenes, with the concurrence of the Acarnanian leaders, and with a view to bring the Peloponnesians into discredit in that part of Greece, agreed to allow the latter to retire separately. This was soon afterwards effected; the Peloponnesians, who went out from Olpae on pretence of gathering herbs and dried bushes for firewood,

§ 4.248 were permitted to move forward and escape, while the Ambraciotae who followed them, ignorant of the secret treaty of the former with Demosthenes, were slain. There fell, however, not more than 200, because having been colonists of Corinth, and hence resembling the Peloponnesians in armour, customs, and language, it was difficult for the enemy to distinguish them. All those who escaped took refuge with Salynthius, king of the Agraei, whose territory confined on the Amphilochia, from whence they proceeded to Oiniadae. In the mean time Demosthenes, who had received advice that all the disposable force of the Ambraciotae was advancing through Amphilochia towards Olpae, from whence their comrades had sent for assistance on the first arrival of the enemy, detached parties to beset the roads and seize the strong posts,

§ 4.249 particularly one of the summits of the mountain Idomenae, the other pinnacle of which, called the lesser Idomene, the Ambraciotae occupied in their advance. In the evening Demosthenes moved forward to the pass with half his army, sending the remainder, which consisted chiefly of Amphilochi, into the Amphilochian mountains. Before daylight the next morning he attacked the Ambraciotae, and the surprise was rendered more complete by the Messenians, who, advancing in front, addressed the enemy in Doric. The consequence was that the Ambraciotae were all either slain on the spot, or that endeavouring to escape into the mountains they fell into the hands of the Amphilochi, while some, to avoid these their most rancorous enemies, (as neighbours often were in Greece) preferred rushing into the sea, in order to swim to the Athenian ships, which happened at that moment to be near the coast. Demosthenes then endeavoured to persuade the Amphilochi and Acarnanes to attack Ambracia, which might easily have been taken, so great had been its loss of men, had not the Acarnanes been afraid of making the Athenians too powerful in this quarter. Soon afterwards the Acarnanes and Amphilochi made a treaty of peace and alliance with the Ambraciotae for a hundred years.

§ 4.250 From this interesting narrative which so well illustrates the military system, the manners and the politics of Greece, we learn that the mountain of which the abrupt termination at the head of the Ambracic gulf, in the great line of communication between the northern and southern provinces of Western Greece, causes the pass to be one of the most important in the whole country, was named Idomene, or Idomenae in the plural, with reference to the two summits, both which were fortified posts, if not in the time of the Peloponnesian war, at least at a subsequent date, as remains of them still exist. Of that at the northern end there are considerable ruins, now called, as I before remarked, Paleopyrgo. The εσβολη, through which Demosthenes advanced on the eve of his second victory, seems clearly to have been the pass of Makrinoro itself, especially from the circumstance of the routed Ambraciotae having endeavoured to swim to the Athenian ships. It follows that the southern extremity of the mountain above Kataforno was that which Demosthenes occupied, that the northern at Menidhi was the position in which the Ambraciotae were attacked and defeated, and that Paleopyrgo was the lesser Idomene. By taking possession of the southern summit, Demosthenes obtained a post which both protected the advance and secured the retreat in case of ill success, as well of the division which he led through Makrinoro as of that which marched through the mountains to the right.

§ 4.251 The anchorage of the fleet of Demosthenes seems to have been in the bay of Kataforno. Arapi, or more probably a position on the adjacent part of Mavrovuni, where some Hellenic remains are still said to exist, exactly at the distance of 25 stades from Neokhori, which the historian gives as the interval between Olpae and Argos, I conceive to have been the site of Olpae, of which name Arapi is a very natural corruption. The torrent which separated the combatants seems to have been the northern of those which enter the lagoon of Arapi, and Metropolis to have been a place on its right bank, at the southern extremity of Makrinoro. The hollow way where Eurylochus fell was probably a higher part of the same ravine which separated the two armies. As changes have occurred on all the alluvial coasts of Greece since the time of the Peloponnesian war, in some instances by the filling up of harbours, as I conceive to have happened at Argos, in others by the extension of the low coast, and the formation of lagoons within the beach, which is likely to have been the case at the bay of Kataforno, the lagoon near Arapi may not have existed, or may not have been of such extent as it now is, in the time of the Peloponnesian war. Armyro I conceive to have been the position of Crevice, where the Acarnanes were stationed.to intercept the enemy; for that place lying on the route from the southward into the Amphilochian plain, was exactly suited to that purpose. But Eurylochus suspecting, or having intimation of their design when he arrived in the vale of Limnaea, crossed Mount Thyamus, which is thus identified with Spartovuni, and descended into the plain of Vlikha, between Armyro and Neokhori.

§ 4.252 The same transactions in the fourth year of the Peloponnesian war leave no doubt of the situation of the Agrais, or country of the Agraei, which appears to have been separated from the district of Limnaea in Acarnania by Spartovuni, and farther inland from the Medionia by a continuation of the same ridge, thus comprehending the vale of the Kekhriniatza up to the ruined fortress of that name, which was probably on the frontier of the Agrais and Stratice. The Agraei comprehended therefore the modern villages of Varetadha, Serdhiniana, and Ariadha, and separating the Aperanti from the Amphilochi, touched the southern borders of the Oreita, and the north-western frontier of Aetolia. At Xerokambo and cape Kastriotissa they extended to the shore of the Gulf. As I learn from Captain Mitjo Kondoianni, who commands 200 armatoli in Valto, and is therefore well acquainted with that and the neighbouring districts, that the only Hellenic ruins which can be compared with those at Neokhori, Surovigli, and Preventza, are at Serdhiniana, this was probably the position of the chief town of the Agraei; if therefore the name Ariadha has been formed from Αγραιδα, it is in a situation different from that of the ancient city, in the same manner as Vlikha, if it be a corruption of Amphilochia.

§ 4.253 Captain Kondoianni describes another great ruin, called Syvisti, near the left bank of the Achelous, about 4 hours above the monastery of Tetarna, in a part of Agrafa named Velaghora, but it seems rather to be the remains of a town of the lower Empire, though possibly occupying the site of a city of the Eurytanes. The Achelous, according to the same informant, often contains very little water in summer above Tetarna, where it receives a large subterraneous stream; and a little lower down its principal tributary, the Megdhova, or Migdova, which is composed of three rivers. Of these, the western rises near the village Agrafa, from a range of heights, beyond which are situated Leontito, Petrilo, and several other large Agrafiote villages. The middle or main stream of the Megdhova, rises on the eastern side of Mount Karava, flows first through an extensive valley named Nevropoli, and then traverses a more confined country between the Dolopian ridges bordering upon Thessaly, and a secondary parallel range, until it receives the Aguliano or river of Karpenisi; after which, turning to the north-west, it joins the Agrafiotiko; and at no great distance below that junction falls into the Aspro, at a spot to which the union of a third stream from the mountain of Syndekno to the westward gives the name of Tripotamo. Not far below Tripotamo, the river is said to flow between precipices so closely approaching, as to be crossed by a bridge of ropes, whence the place is called Sta Kremasta. If the Megdhova be the Campylus, as I before suggested, the name may have been derived from its reflex course, caused by the southerly projection of the mountain of Kerassovo, near the southern extremity of Agrafa. This remarkable peak is visible from Prevyza, a few degrees to the right of the still more striking Kalana, which latter lies from Saint George at Prevyza, exactly in a line with the two capes forming the entrance from the gulf of Prevyza, into that of Arta. Viena, or Panaetolicum, opposite to Vlokho, is also visible from the same station, as well as the northern summit of the same great ridge near Arakhova, in Suvalako, in the district of Karpenisi, the latter subtending 15°1/4 with the centre of Kalana and the two capes, the former 28°.

§ 4.254 November 9.—From Prevyza to Luro in 3 ½ hours, the loaded horses in 5 hours. The road passes through the ruins of Nicopolis, and over the height of Mikhalitzi into the plain of Lamari, which has the same rich kind of soil as that of Prevyza, but is overgrown with small oaks and brambles, and supplies only pasture to the Vezir’s flocks. Near Prevyza the hawthorn was in blossom, and there were lambs of a month old. As we advance towards the interior the season is less forward. The castle of Luro, built only a few years ago, is already falling to ruin, having been constructed like the dwelling houses of Epirus, of loose stones and mud interposed between strata of wood. In the opening of the vale above Luro, Suli presents itself to view in a very imposing manner. The vale and the slopes of the adjacent mountains, as far up as Suli and Tervitziana, are covered with oaks. These are for the most part small and of the velani kind; towards the river of St. George are some of larger girth.

§ 4.255 In the afternoon I proceed in 3 hours to Memetjaus, or Mamutjaus. At Kanza we leave the pass which leads into the vale of Lelovo to the left, traverse soon afterwards a marsh where the water comes up to the horses' knees, and arrive at a wooden bridge across the river of St. George, from whence a winding muddy path leads through a plain of the most fertile soil, but producing only a small quantity of maize, to Riba. This is a small village on the left bank of the river of St. George, opposite to the ruins called Rogus, which occupy an extremity of the hills. The remains are those of a fortified town of the Byzantine empire, built upon Hellenic foundations, and composed in part of materials of ancient times. Although the place has probably for ages been ruined and deserted, the name still gives title to a suffragan bishop of the metropolitan See of Arta. It is found among the bishoprics of the metropolis of Naupactus, in the tenth century, when Arta was not yet an episcopal See, and from Cantacuzenus it appears to have been one of the chief towns or fortresses of Epirus in the fourteenth century. The Hellenic remains seem from Polybius to have belonged to a town named Charadra, for he relates that Philip passed by Charadra in his march from Ambracus to the strait of Actium, that is to say, from Fidho-kastro in the marshes of Arta to Prevyza, from which line the marshes towards the sea would have caused a divergence to the northward.

§ 4.256 When Fulvius the Roman consul was preparing to besiege Ambracia, in the year B.C. 189, some Aetolian deputies, who had been intercepted by the Epirotes on their way to Rome, were sent to Charadra as prisoners, and from thence transferred by the Epirotes to Buchaetium. The name occurs also in a fragment of Ennius, which seems to refer to the good quality of a particular kind of shell-fish in the adjoining river or marshes. The town evidently took its name from the river which forming a continued cataract from one of its chief sources at St. George to the plain of Lelovo, well merited the appellation of Charadra. This stream, after emerging from the gorge between Strivina and Rogus, follows the foot of the woody heights to Rogus, and then turns towards Luro, receiving that river and joining the gulf not above two or three miles to the northward of the mazoma of Nicopolis. The marshes which intersect the plains below Lamari and Luro, and around Rogus, aided by the wild vegetation which surround them, render the maritime Molossis very unhealthy in summer, and scarcely any persons then remain in the villages except those engaged in the harvests. Nov. 10. — The road from Mamutjaus to Ioannina takes the direction of Arta, but winds very much in order to avoid the marshes, which seem to be chiefly formed by the river of Strivina and by sources between that place and Rogus, added to those of Khanopulo. In the latter marsh there are two streams, one of which joins the St. George, the other finds its way separately to the sea. There is no plain in Turkey where drainage would be attended with greater profit.

§ 4.257 Having arrived in two hours at Khalikiadhes, which is about one hour west of Marati, the suburb of Arta, we join, in another half hour, the paved derveni, at a point one hour distant from Arta, and proceeding from thence by the beylik, or high road, through Pendipigadhia, reach Ioannina, at the end of twelve hours and twenty minutes from Mamutjaus, having halted one hour and twenty minutes on the road. This route, the only one practicable at all seasons from the coast to the plain of Ioannina, is now paved in all the difficult places, and may be travelled in a four-wheeled carriage.

§ 4.258 CHAPTER 39 EPIRUS. THESSALIA. Nov. 19.—From Ioannina to the three Khans [Τά τρία χάνια] seven hours. From October the 20th to this day the weather has been equally divided between fair and showery. At Prevyza there was a heavy rain for five days, with very little intermission, succeeded by a north-east wind for seven days, with an atmosphere perfectly serene. For the last four days there has been rain, with a south-west wind. This day, the wind having moved more directly westward, has brought only light showers. Mounts Kakardhista and Tzumerka were permanently tipped with snow on the first of this month.

§ 4.259 On the 10th, Kahardhista, the ridge of Metzovo, and the summits of Zagori, were entirely covered with snow, the first more extensively than the others, while Nemertzika and Olytzika had not any. All the summits around Kalarytes had been sprinkled two or three times during the rains of October; but the snow had all disappeared, even from Kahardhista, on the 20th of October. Our route, although it is the high road to Constantinople, and the only frequented communication between Epirus and Thessaly, is in the most neglected condition. The late autumnal rains have left it only just passable. The ascent and descent of the hill of Dhrysko, which divides the plain of Ioannina from the valley of the Arachthus, is in the worst possible state. After a halt at the khan of Dhrysko, we descend from thence to the Zagori branch of the Arta, cross it by the bridge της Κυράς, follow up the bed of the Metzovo branch for near an hour; and then, after passing for a short distance along the right bank, cross to the opposite side by a bridge of one arch, follow the heights on that side, again descend into the bed, and then follow the right bank to the Three Khans. Nov. 20.—From Tria Khania to Metzovo two hours and a quarter. We continue to follow the river along its bed, or over the heights on the left bank, as far as the bridge, which is a little below the junction of the tributary descending from the mountain of Khaliki; then quit the river, and ascend to Prosilion, or the northern Makhala of Metzovo, where I am lodged in the house of Kyr S. N.,

§ 4.260 a merchant who usually resides at Ioannina — receive visits from the primates, the stipendiary physician, who is a Neapolitan; the bolu-bashi a Tepeleniote, and Captain Dehli-Ianni a native of Metzovo, chief of the armatoli who defend the pass, and the terror of the robbers. The fields of the Metzovites producing only a small quantity of corn, they are now paying ten paras an oke for barley at Grevena or Trikkala, chiefly for the use of the horses of travellers. Though complaining of the Vezir’s extortions, they admit that he is not quite forgetful of the expences to which their situation in this great derveni renders them liable; and, like most of his subjects, they allow him the merit of defending them from inferior agents and highway robbers, though he seldom suffers any good opportunity to pass of plundering them himself: for instance, my host was three months in prison this year, under the pretext of being concerned in a correspondence with the Russians, and was obliged to pay the Vezir 1000 sequins to obtain his liberty; nor would it have been effected at this price without the intercession of Omer Bey Vrioni, whose assistance was wanted by Aly against Berat. The affair of Berat has already cost His Highness so much that he is collecting money from all quarters. A tatar overtook us to-day, going to collect 80 purses from fifteen villages of Aspropotamo; only in three or four of which, as Captain Ianni informs me, are there any inhabitants left, the rest having fled from the robbers into the plain of Trikkala. The most conspicuous building in Anilio, the quarter of Metzovo on the opposite side of the ravine, is a house which has been lately built by Mukhtar Pasha for a favourite youth of that place.

§ 4.261 Nov. 21.—A southerly gale, succeeding a single serene day, set in last midnight with great violence, and continues all this day, with torrents of rain. Nov. 22.—The southerly wind of yesterday had melted the snow upon the ridge, but last night it fell again and covered all the woods on the summit; but by no means to such a degree as when I crossed the Zygos a week earlier, in the year 1805, which was noted for the early severity of the weather. During the last two years there has been no permanent snow before January. Our passage today is not difficult; especially as I have the assistance of ten or twelve khamalidhes, or porters of MOtzovo, who were directed by the Vezir’s commandant to accompany me, together with nine of his palikaria. We arrive at the khan of Malakassi in three hours and a half. A very useful khan has been established since my last visit, just under the eastern side of the summit. Here the Metzovites are responsible for having hamals constantly in attendance. Nov. 23.—A fall of snow during the night renders the road extremely bad; and, together with the effect of the late rains in swelling the Salamvria, and obliging us to follow the akres on the left bank, lengthens the time to Kalabaka an hour and a half. In descending the heights to the river of Kratzova or Miritza, which we cross by a high bridge,

§ 4.262 the view opens of the entrance into Thessaly between the superb rocks of the Meteora on the north, and the woody mountain opposite to them on the south, together with a part of the plains of Upper Thessaly, not far from Trikkala; with the exception of this part of the landscape the whole is now covered with snow. The river of Kratzova in the lower part of its course flows through a thick wood of large planes. Farther up, its valley is well cultivated, and on the heights still farther, at the distance of six or eight miles to our left, is seen the large village of Miritza. On the northern side of this valley the heights are covered with oaks, which extend also over the mountains, inclosing the vale of the Salamvria, but where the forest is not so thick, the trees being in general intermixed with vineyards around several villages: few of the oaks are of any considerable size. Nov. 24.—Kalabaka has suffered extremely of late from the vexations of the last hodja-bashi Ιanaki, who built a superb house with the produce of his plunder, and ended his days in the prison at Ioannina. But it is injured more permanently by the expence of konaks, to which it is continually subject, in consequence of its lying at the exit of the most frequented pass in Greece. The master of the house in which I lodge, who among his other misfortunes has left an eye with the thieves, had the honour not long since of having a Bey with a party of Albanians quartered upon him for ten or twelve days: they burnt his furniture and his silk frames, and finished by borrowing a valuable mule, which he saw no more.