¹ Hammond, N. G. L., A History of Macedonia I (Oxford, 1972) 100–2Google Scholar. There are unfortunately three errors in these pages, ‘Kastoria’ being printed instead of ‘Koritsa’ on the middle of p. 100 and twice on the upper half of p. 102. I refer to this book hereafter as Macedonia.

² I am grateful to Mr G. T. Griffith, Mr E. I. McQueen and Professor P. A. Brunt for commenting on the first draft of this article. Professor Brunt helped me particularly in the chronology of the campaign. The maps which accompany this article are based on the British Staff Maps of 1945 entitled ‘Albania: 1: 5,000, Sheets Bilishti and Poyani’. In giving the detail of the Gryke e Ujkut I have used the notes which I made on the spot in 1972 and a sketchmap in Iliria I (1971) 48, Tab. 1.

³ For a tumulus near here see p. 76 below.

⁴ See Macedonia I. 117 for this pass.

⁵ For this cave see Studia Albanica 1967. I. 142–5 and 1971. I. 134 f.; Iliria I (1971) 31–47 and Plates i–xvii; cf. Macedonia I. 242 and index.

⁶ This wall has been reported in Studia Albanica 1971, I. 135, and it was remarked there that the bluff is very steep and rocky so that there was no need of fortification elsewhere. Pieces of bronze pendants were found, as at Kuç i Zi (see n. 3 above); also flat-topped thumb-rests springing from the wall of a pot, as in Macedonia Fig. 16 f and s, and sherds from imported Greek Geometric pottery, one piece each ‘being from Argos, Corinth and Delphi’. There was also some Late Bronze Age pottery.

⁷ I was told that this is the only agger yet known in Albania but such aggeres are known in Bosnia at this period. The tumuli indicate that there was an important route here.

⁸ See Macedonia I. 230.

⁹ We visited the headsprings at Izvor on the Albanian side of the border. I was told by Professor M. D. Petruševski that these headsprings are the sources of the ancient Drilon (Byzantine Drymon, modern Drin i Zi); its waters can be seen running through Lake Ochrid to Strouga, if one flies above the lake.

¹⁰ See Macedonia I. 52 f. and Map 7.

¹¹ For the two lakes see Megale Hellenike Encyclopaedeia s.v. Prespa. Lake Prespa is the largest lake in the Balkans, but its greatest depth is only 50 m. Lake Little Prespa (Lake Ventrok) is shallower. Both are very rich in fish.

¹² Leake, W. M., Travels in Northern Greece I (1835) 332 f.Google Scholar

¹³ Arr. An. 1.5.6 and 12 describes as ‘wooded’ two areas, which, if I am correct in my identifications, are now barren or have only sparse bushes of garigue. The blocking of the river-outlet was probably due to deforestation, erosion and perhaps an avalanche or cliff-fall.

¹⁴ For this place on the Via Egnatia route see Macedonia I. 32 f.

¹⁵ Anonymi Gesta Francorum (ed. Hagenmeyer, ) 4. 3–5Google Scholar and Sivracensis, TudebodusHistoria de Hierosolymitano itinere, 16.Google Scholar

¹⁶ The Suggestion of Hagenmeyer that the Crusaders committed this atrocity because they confused the place-name ‘Palagonia’ with the fifth-century theologian Pelagius seems to be far-fetched; surely ‘Diabolis’ should have been plundered even more brutally! We need in the vicinity a wealthy and renowned community, which was a centre of Christianity but not of Catholic Christianity. Hagenmeyer suggests that the Crusaders went from Kastoria not northwards but eastwards to Florina, Ostrovo (where he assumes that there was then but not now a considerable island in the lake), and Thessalonica. But Anna Comnena 5.5.I mentions expeditions by the Crusaders to Achris (Ochrid), Scopia (Skopje) which is on the Vardar, and ‘the two Poloboi’ (perhaps the two Prespas); it is only after these expeditions that the Crusaders marched from Ochrid to ‘Ostrobos’ (Ostrovo) and thence by ‘Soskos’ to Servia and Beroea (Verria).

¹⁷ The technique of the painting is reminiscent of the chalk figures on the Wiltshire Downs; for the silhouetted figures are filled with chalky lime (‘de la chaux qui remplie toute la silhouette’, Studia Albanica 1971. I. 136).

¹⁸ See Macedonia I. 96.

¹⁹ For this pass see Macedonia I. 42.

²⁰ Tarn, W. W., Alexander the Great (Cambridge, 1948)Google Scholar I. 6 is mistaken in putting Pelion inside Macedonia. The fact that Cleitus ‘occupied’ Pelion (Arr. An. I.5.5), means that the town was not Dardanian; but it does not follow that it was Macedonian. Diodorus Siculus 17.8.1 ‘Having subjugated the Thracian tribes Alexander went on to Paeonia and Illyris and the lands co-terminous with these, reduced many of the barbarian inhabitants who had revolted, made all the neighbouring barbarians subject to his rule … and returned to Macedonia’, the jibe of Demosthenes when Alexander ‘was among Illyrians and Triballians’ (Plu., Alex. II.3Google Scholar), and Athens' congratulations to Alexander on his safe return ‘from the Illyrians and Triballians’ (Arr. I.10.3) also indicate that the campaign was in Illyris and not in Macedonia. So too Curt. 3.10.6 and 5.1.1.

²¹ For the occurrence of this name in various forms see Macedonia I. 415, where it is noted that ‘a common feature of the districts in which the name survived is the proximity of a lake’.

²² A preliminary report is in Buletin Arkeologjik 1971. 31f. I saw some of the material in the exhibition cases at Tirana in 1972.

²³ See Buletin Arkeologjik 1969. 27–36; Studia Albanica 1971. 1.151 f.; and Iliria I (1971) 350. The excavator, Zhaneta Andrea, showed me some of the material which was exhibited at Tirana and in the museum at Korcë in 1972. There were bronze pendants and beads as in Macedonia I. Fig. 19b, Fig. 17m, y and cc and Fig. 18k, two gold-leaf mouthpieces with their strings intact as ibid. 1.352, and painted pottery of the ‘north-western Geometric style’ as ibid. I. 281 f. but with plastic nipples.

²⁴ For this diary (the Ephemerides) see Wilcken, U., ‘Hypomnematismoi’, Philologus 53 (1894) 112 f.Google Scholar; he infers that the diary included the campaigns in Europe.

²⁵ See Strabo 7. 301 and 305 for the position of the island of Peuce in the Danube.

²⁶ For the location of these peoples see Macedonia I. 78 and 202.

²⁷ Their territory at this time was probably in the Metohija, a canton of Yugoslavia, and farther west and north. See BSA LXI (1966) 249 f. and Macedonia I. 81 f.

²⁸ See Macedonia I Map 17.

²⁹ Str. 7. 327. Alexander marched ‘near the Erigon’ (Arr. 1.5.5), not ‘up the Erigon’ in the words of Fuller, J. F. C., The Generalship of Alexander (London, 1958) 224Google Scholar, since the river is impassable in the Mori hovo.

³⁰ See BSA LXI (1966) 243 f.; Cleitus was not the son of Bardylis I, as Fuller op. cit. 223 n. 2, has suggested.

³¹ It seems likely that Alexander kept his main camp as at Arr. 1.5.5. ‘by the river Eordaicus’ and moved only his assault force into a position near Pelion at Arr. 1.5.8.

³² The Loeb translation of Arrian 1.5 and 6 contains many errors.

³³ As we have seen, Alexander's army did not use during the operation the narrowest part of the entry, namely that lying between the river and Mt Trajan. It was to this part that Arrian referred in saying that there would not have been passage for the army even at four abreast ( I.5.12). The fact that Arrian, perhaps copying his source verbatim, says not that there was not passage but that there would not have been passage, seems to me to imply that the army did not use that particular passage when the action developed. In other words someone who knew the ground made this comment after the event.

³⁴ If we take the pursuit to have gone ‘as far as the mountains of the Taulantians’, which were north of Elbasan (see BSA LXI (1966) 247), Alexander and his companions rode for 90 or 100 kilometres and passed through some mountainous country in the latter part of the pursuit. The route is not in doubt: Goricë—Tsangon Pass—Korcë plain—Malik —Gramsh—Shtërmen—Elbasan (see Macedonia I. 98, map 10 and 236, map 20). The total distance is about 100 km and includes rough going and scrub-covered country in Gramsh. An alternative has been suggested to me by G. T. Griffith. This is that the words τῶν Ταυλαντίων should be taken not with τὰ ὄρη but with ἡ δίωξις. If this is done, then ‘the pursuit of the Taulantians’ went only as far as the mountains on the southern side of the Korcë plain, a distance of some 25 kilometres, and one would assume that the name of these mountains was in the original source but was omitted by Arrian. This suggestion reduces Alexander's pursuit to reasonable proportions. But it seems doubtful whether Arrian would have written μέχρι πρὸς τὰ ὄρη ρῶν Ταυλαντίων if he had wished to avoid the obvious misunderstanding inherent in his word-order. Moreover ἡ δίωξις used absolutely is a regular feature in Arrian's descriptions of Alexander's battles (see Arr. 1.16.2; 2.11.7; 3.15.5–6), and the pursuit was sometimes carried to a great distance, e.g. at 3.15.5 for 600 stades = 120 km. I have therefore adopted the former interpretation in my text. I am grateful to Mr B. Bosworth of the University of Western Australia for his advice on this matter.

³⁵ It seems likely that Alexander had already been through the Gryke e Ujkut, so that he knew the ground well enough to make his plan in advance. If so he had either come that way to find Cleitus at Pelion, or he had been there either with his father on a campaign against the Illyrians or during the period of estrangement, when he was ‘among the Illyrians’.

³⁶ Arrian's concentration on Alexander leaves some points obscure. For example, we are not told what the cavalry as a whole did during the withdrawal into the Gryke e Ujkut. We may surmise that they formed the rear-guard and so prevented the enemy in Pelion from sallying out across the plain to attack the Macedonians or join their friends. Again the Somatophylakes were named only once, at 1.6.5, where they were mounted and acted together with Alexander's personal Companion cavalry, τοῑς ἀμφ’ αὐτὸν ἑταίροις; the Somatophylakes and these Companions were resumed in the single phrase σὺν τοῖς ἑταίροις at 1.6.6. The Hypaspists were different; being infantrymen, they led the way across the river at the head of the column and when it moved to the left, they came to face the Kalaja e Ventrokut. The phrases τοῑς μετ’ Ἀλεξάνδρου and τοῖς ἀμφὶ Ἀλέξανδρον seem to be variant phrases for ‘the troops with Alexander’ at 1.6.7 and 1.6.11; so too we have τοῖς ἀμφὶ Φιλώταν and οἱ ξὺν Φιλώτᾳ at 1.5.9, and οἱ ἀμφὶ Γλαυκίαν and τοῖς ἀμφὶ Κλεῖτον καὶ Γλαυκίαν vel sim. in a general sense at 1.5.6, 1.5.10, 1.5.11 and 1.6.9. When Alexander planned his surprise attack at night, he did not use cavalry, because it was difficult to move horses silently. He took the Hypaspists, the Agrianians, the Archers and ‘the regiment of Perdiccas and Coenus’, and in the event the Agrianians and the Archers led the attack. It is generally held that Arrian meant the regiments commanded respectively by Perdiccas and Coenus.

³⁷ He left Glaucias on the throne (Glaucias had dealings later with Cassander and he adopted Pyrrhus; see BSA LXI (1966) 246), and Illyrian soldiers fought under Alexander's command in Asia.

³⁸ Additional Note. After writing this article I read in the report of excavations on the island of Saint Achilleiou that the destruction of the great basilica has been dated by coins and other evidence to “before the beginning of the twelfth century, most probably within the last quarter of the eleventh century” (PAE 1967.68 and 1969.39; Ergon 1967.40). This is when the Franks were in the vicinity.