^{page 245 note 2} Cf. Berve, H., Das Alexanderreich, i. 113Google Scholar; Tarn, W. W., Alexander the Great, ii. 142Google Scholar: Momigliano, A., Filippo it Macedon, pp. 8–10Google Scholar

^{page 245 note 3} Demosth. 2. 17.

^{page 245 note 4} Harpocration and the ‘Suda’ s.v. (Jacoby F.Gr.H. 72 F 4); cf. Etym Magn., P. 699.

^{page 245 note 5} An author unknown, cited by Magn, Etym. p. 699, 47–8, implies that the were a select group: This conflicts totally with all other evidence, which implies that was the name for the totality of the phalanx, and we must assume that the quotation has been garbled in transmission. Presumably the author of the fragment (Anaximenes?) stated something to the effect that a select group of the so-called were used against the Illyrians, and in the transmission of the text an artificial antithesis has crept in.Google Scholar

^{page 245 note 6} Arr. I. 24. 3. This corps was detached from Alexander all through the winter of 334/3 and only rejoined the main force at Gordium (I. 25. 3; 29. 3).

^{page 245 note 7} See Roos's prolegomena to his Teubner text of Arrian, pp. v-xi.

^{page 246 note 1} I Arr. 4. 23. I; 5. 22. 6; 6. 6.; 6. 21. 3.

^{page 246 note 2} The only exception I can find is E. huff Robson, who in his Loeb text of Arrian (i. 208 n. 1) suggests that is the correct reading at 2. 23. 2 and elsewhere. But in his text he leaves Blancardi's emendation, and his explanation of the term (‘civilian volunteers’) is wholly unacceptable.

^{page 246 note 3} The manuscripts of Hesychius s.v. unanimously read and some of the glosses in the Etymologicum Magnum are obviously meant to explain p. 658, 41 ff.). The manuscripts of the ‘Suda’ have yet another corruption

^{page 246 note 4} The force of in this passage is ‘other than Macedonian’; in exactly the same sense Arrian refers in the next line, in the context of cavalry, to (i.e. ‘another (non-Macedonian) royal agema’). With his usual passion for variatio Arrian uses and to convey exactly the same meaning. The sense is definitely not ‘Persian and other i.e. ‘other who were not

^{page 246 note 5} He was none the less followed unquestioningly by Gronovius, by Dübner, and by Robson in the Loeb edition.

^{page 246 note 6} The crucial passages are Diodorus 17. 57. 2 and Curtius 4. 13. 27. Both name a force of argyraspids in Alexander's battle line at Gaugamela, occupying exactly the position where Arrian places the hypaspists (Arr. 3. 11. g), and both the argyraspids of the vulgate and the hypaspists of Arrian are commanded by Nicanor, son of Parmenion. The two units must be identical (cf. Tarn ii. 116–18).

^{page 246 note 7} Schmieder was followed by Abicht (1871), who does not condescend to note that he has excised the phrase from his text, and most recently by Roos in the standard Teubner edition.

^{page 247 note 1} 4. 22. 7. There is no doubt that the three named officers with Hephaestion were phalanx commanders. Meleager served as a battalion commander throughout the reign (cf. Berve ii Nr. 494), Cleitus' is attested as a phalanx battalion at the Hydaspes (5. 12. 2), and there is a very high probability that Gorgias had taken over Craterus' battalion in Bactria (cf. Tarn ii. 145).

^{page 247 note 2} It has been argued that in the latter years of the reign there were seven phalanx battalions (so Berve i 115–16; Tarn ii. 190–1; Milns, R. D., G.R.B.S. vii [1966], 159 ff.). This is nowhere explicitly stated, and the hypothesis depends on complex manipulations of the data in Arrian which I do not find wholly convincing. In particular it is not true that Arrian names seven battalions participating at the battle of the Hydaspes. At 5. I I. 3 he states that Craterus was left at the base camp with the battalions of Polyperchon and Alcetas, and he also claims that Alexander's crossing force contained the battalions of Coenus and Cleitus (5. 12. 2). Now at 5. 12.1 Arrian says that Meleager, Attalus, and Gorgias were strung along the river bank Arrian says nothing about their battalions. He only mentions that they commanded the mercenary forces. We know that battalion commanders could be detached from their regular units to perform other duties. Coenus seems to have commanded a hipparchy at the Hydaspes (5. 16. 3), and both he and Attalus were apparently left behind with Craterus during the invasion of India while Alexander took their battalions ahead (4. 23. 5–24. 1). At the Hydaspes Meleager, Attalus, and Gorgias are not attested with battalions and cannot have commanded battalions; otherwise we should be faced with the absurdity of Alexander going into battle with two battalions and leaving five unengaged on the far bank of the Hydaspes (5. r8. 1 is conclusive that the forces left across the river only crossed over when the victory was won). So we have no explicit evidence that seven battalions were present at the Hydaspes, and the other evidence hitherto cited is quite inconclusive. It does seem, however, that Alexander's phalanx troops comprised seven battalions at the time of the invasion of India (the evidence is given on p. 249, n. 4).Google Scholar

^{page 247 note 3}

^{page 248 note 1} See the observations of Brunt, P. A. and Griffith, G. T. in J.H.S. lxxxiii (1963), 28 f.; 70 f. It is not explicitly stated that a hipparchy contained two ilai and no more, but it is very unlikely that Alexander would have given Hephaestion more cavalry than he himself retained.Google Scholar

^{page 248 note 2} Arr. 6. 17. 3. Antigenes is attested as battalion commander on this occasion alone, and there is no way of identifying his predecessor in the command. Previously he had figured once in Arrian's narrative in some sort of phalanx command at the Hydaspes (5. 16. 3). Then he could have deputized for Coenus, who had temporarily been transferred to a cavalry hipparchy, but equally possibly he could have stood in for one of the three phalanx commanders separated from their units to command the diversionary mercenary cavalry (5. 12. 1; cf. p. 247, n. 2). Subsequently he succeeded to a phalanx battalion; it may have been that of Cleitus or even of Gorgias, who is not explicitly mentioned as a phalanx commander after the march on India. Tarn's suggestion (ii. 146) that Antigenes succeeded to Coenus' battalion is wholly speculative; most probably Coenus was followed in his command by Peithon (see below, p. 249).

^{page 248 note 3} Arr. 5. 2 I. 5

^{page 248 note 4} 5.20. I: Craterus is left at the Hydaspes 5. 21. I: Coenus is left with his battalion at the Acesines; 5. 21. 4: guard-posts are established in the area up to the Hydraotes to enable the forces of Coenus and Craterus to command the intervening territory.

^{page 248 note 5} Arr. 5. 27. I; Curt. 9. 3. 3. Tarn, who believed Coenus' speech in Arrian to be a forgery, argued that Coenus could not have been present at the Hyphasis, since he had been left behind at the Acesines (ii. 287). Badian, E., J.H.S. lxxxi (1961), 20 n. 24, points out that Coenus could easily have rejoined the main force before the arrival at the Hyphasis.Google Scholar

^{page 249 note 1} 6. 6. I; cf. 7. 2; 8. 2–3.

^{page 249 note 2} 6. 2. I; cf. Curt. 9. 3. 20.

^{page 249 note 3} So Beloch iii². 2. 329 and, more tenta tively, Berve ii 312 n. I.

^{page 249 note 4} The battalion of Polyperchon is attestec in action with that of Coenus during the march on India (4. 25. 6). This attack on the Assaceni must have taken place during the absence of Hephaestion and Perdiccas who two chapters later are specifically state to be still absent from the main force (4. 27. 5 cf. 4. 30. 9). Also operating with Alexande was the battalion of Alcetas (4. 27.1 and 5) which is later attested as part of the phalam infantry (5. 11. 3). Alcetas was the brother of Perdiccas (Berve ii. Nr. 45), and in all probability he commanded the battalion lef vacant by Perdiccas' elevation to the body guard (so Berve ii. 22; Tarn ii. 144–5) Perdiccas' battalion had been recruited fron Lyncestis and Orestis (Diod. 17. 57. 2), and if it is correct that Alcetas succeeded hi brother, Alexander's column in the march on India included all the Upper Macedonian battalions. In Alexander's main force there seems to have been another battalion, that of Attalus (4. 24.1 and 10). At 4. 24. I Attalus' is named in a context which implies overwhelmingly that it belonged to the phalanx battalions. In other words Alexander had four phalanx battalions and Hephaestion and Perdiccas three, making a grand total of seven. See also the discussion in the Appendix of the complex problems raised by Arrian 4. 24. 10.

^{page 249 note 5} Diod. 17. 57. 2; Curt. 4. 13. 28.

^{page 249 note 6} Cf. Geyer, F., Makedonien bis zur Thronbesteigung Philipps II. HZ Beih. xix [1930] PP. 77–83Google Scholar

^{page 249 note 7} Arr. I. 2. 5 mentions cavalry from Upper Macedonia, which was used in 335 against the Triballi. The unit is not mentioned subsequently, and presumably Alexander left it in the home army under the command of Antipater.

^{page 249 note 8} Diodorus and Curtius, locc. citt. (n. 5).

^{page 250 note 1} For a good survey of rival theories se( Momigliano, Filippo it Macedon, pp. 8–10 Momigliano himself opts for Alexander the Philhellene as the author of the reform, but that is surely excluded by Thucydides' conotemptuous description of the Macedoniar foot: That refers to 427, almost fifty year: after the acme of Alexander I. If we accept Anaximenes' attribution of the reform to an. Alexander, the only choice is Alexander II.

^{page 250 note 2} Cf. Beloch, iii². I. 227; 473.Google Scholar

^{page 250 note 3} For a convenient list of Macedoniar words which apparently bear no relatior to any known Greek dialect forms see Hoffmann, O., R.E. xiv. 694–5.Google Scholar

^{page 250 note 4} Xen. Hell. 5. 2. 13.

^{page 250 note 5} Arr. Anab. 7. 9. 2.

^{page 250 note 6} Diod. 16. 71. 2; cf. Theopompus F.Gr.H. 115 F 110; Plut. Mor. 520B; Plin. N.H. 4. 41 (Philippopolis); Strab. 7. 6. 2 (320); F.Gr.H. 115 F 220 (Calybe); Etym. Magn. s.v. For discussion of these settlements see Momigliano, Filippo it Macedone, pp. 138, 146; F. R. Wüst, Philipp II. von Makedonien and Griechenland (5938), pp. 103–4.

^{page 250 note 7} Justin 8. 5. 7–6. 2. Philip's eponym and imitator, Philip V, organized a similar shift of population from the coastal cities of Thrace to Paeonia (Polyb. 23. 50. 4–7; Livy 40. 3. 3–6; cf. Walbank, F. W., hilip V of Macedon [1940] p. 243).Google Scholar

^{page 251 note 1} (Hesychius s.v.—Aeschylus F 66 Nauck).

^{page 251 note 2} Schwyzer 54 B = Buck 70 (Tegea) Schwyzer 424 = Buck 65 (Olympia)

^{page 251 note 3} e. g., For further examples of haplology see Schwyzer, E., Griechische Grammatik i. 263. The most interesting examples of syllabic losses come from Thessaly. There is attested an alternative form, Lasa, of the city Larisa (Hesych. s.v. I.G. ix. 2, 517, 19). More significantly, Thessalian proper names beginning tend to be abbreviated into I.G. ix. 2, 414 b 3; 536, 10; 506, 5; 580, 6; 234; 281, 3). It would be tempting to derive from and except that there is no reason why Philip should have used this peculiarly Thessalian contraction. An enigma would also be presented by the term ‘best companions’. The superlative would have been provocative to the other troops, and there seems no reason for the Upper Macedonians to have been singled out in this way.Google Scholar

^{page 251 note 4} Thuc. 2. 99. 2; Strab. 7. 7. 8 (326) erroneously states that the Upper Macedonians were of Epirote extraction. For further detail see my discussion in C.Q. xxi (1971), 97–102.

^{page 251 note 5} At 7. II. 3 Arrian mentions side by side with argyraspids. This is the sole intrusion in Arrian of this term for the hypaspists, although it occurs in the vulgate description of the battle line at Gaugamela (cf. p. 246 n. 6). Tarn ii. 117–18 argues that the term argyraspids slipped in through Arrian's reading of Hieronymus of Cardia. More probably Arrian took the passage from some source other than Ptolemy.

^{page 251 note 6} Arr. I. 28. 3; 7. II. 3; Plut. Flamin. 17 = Apophth. Flamin. 4 (1970).

^{page 252 note 1} F.Gr.H. 138 F 18; cf. Strasburger, H.Ptolemaios and Alexander (1934), PP. 41–2;Google ScholarSeibert, J., Untersuchungen zur Geschichte Ptolemaios' I. (1969), pp. 19–20.Google Scholar

^{page 252 note 2} This expression must mean ‘the taxis of Philippus and that of Philotas’, and not ‘the taxis jointly commanded by Philippu: and Philotas’, which is linguistically the more natural interpretation. There arc numerous parallels in Arrian (e.g. I. 6. 9: I. 22. 4: and in no instance does the expression denote a unit commanded by two or more officers. The most conclusive example is 1. 20. 5, when Alexander is described attacking Myndus with various army units, which included In 334 Amyntas, Perdiccas, and Meleager each commanded a phalanx battalion (cf. Berve i. 114), and it is to the last degree unlikely that they were amalgamated in the command of a single battalion for the attack on Myndus, and never again. For other examples of this peculiar usage, see Tarn ii. 144.

^{page 252 note 3} Compare I. 22. 4, where Arrian uses to describe the hypaspist companies of Adaeus and Timander.

^{page 253 note 1} Hamilton, J. R., C.Q. v (1955), 220. accepts Tarn's general reconstruction, but balks at emending Alcetas out of the text at 4. 22. 1. He therefore suggests, unconvincingly, that Alcetas was only a deputy commander.Google Scholar

^{page 253 note 2} This was observed by Milns, R. D.G.R.B.S. vii (1966), 160.Google Scholar

^{page 253 note 3} Tarn ii. 144 n. 1, amplifying Berve ii 397 Nr. 803. His arguments have strangel) been accepted without comment (e.g. C. B Welles, Miscellanea Rostagni [Turin 5963], p. 107 0. 40).

^{page 253 note 5} This force is suspiciously large for its task, and it might be that the detail is apocryphal (so Welles, Miscellanea Rostagni, pp. 109–10, and, more tentatively, Seibert, op. cit., pp. 10–16). But even if Ptolemy concocted a bogus expeditionary force, he presumably named authentic units of the Macedonian army.