^{page 93 note 1} Berve, , Alexanderreich, ii. 17–19 (nr. 37).Google Scholar

^{page 93 note 2} Arr. 1. 25. 2; Curt. 7. 1. 6–7; Just. 11. 2. 2.

^{page 93 note 3} Badian, E. {Phoenix, xvii (1963), 244 ff.)Google Scholar poses the question cui bono? and delineates convincingly the insecurity of Alexander's position in 337. C. B. Welles (Loeb edition of Diodorus, vol. viii, p. 101 n. 2) makes the stimulating suggestion that Pausanias, Philip's murderer, was killed to ensure his silence. The two hypotheses are amalgamated by Hamilton, J. R. (Greece &Rome, xii (1965), 120 ff.).Google Scholar

^{page 93 note 4} Diod. 17. 2. 1; Just. 11. 2. 1; Arr. 1. 25. 1.

^{page 93 note 5} Diod. 17. 51. 2; Curt. 4. 7. 27; Plut. Al. 27 3–4; Just. 11. 11. 9.

^{page 93 note 6} Diod. 16. 93. 3–94. Aristotle {Pol. 5. 131 ib2) explicitly places the murder in the category of personal, not political, attacks.

^{page 93 note 7} Berve, ii. 309 (nr. 614).

^{page 93 note 8} P.Oxy. 1798 ═ FGrH 148; Wilcken, U. cf., SB. Berlin, 1923, 151–7Google Scholar. Papyro-logists are undecided whether the script is Hellenistic or post-Augustan (cf. Jacoby, FGrH ii D. 533). The work may have been fairly extensive. On the left margin of the second surviving column (1. 17, F 5–6) there is a stichometric sign Ψ indicating 2,300 preceding lines. This comes in the context of the Granicus, and so it is inferred that the previous narrative was of considerable length and detail (so Wilcken, op. cit. p. 151 n. 2). One cannot, however, assume that the whole of the lost narrative was devoted to the youth of Alexander and his early regnal years, or even that the history covered the whole of the recto of the papyrus. The brevity of the Issus narrative suggests that the treatment of events elsewhere was not particularly discursive.

^{page 94 note 1} Much obscurity arises from the word . Wilcken (op. cit., p. 152) interpreted the expression as crucifixion, and found corroboration in an anecdote of Justin (9. 7. 10), in which Pausanias is described as crucified by the time of Philip's funeral. Opinions have since changed, and LSJ Suppl., ad loc, translate ‘cudgel to death’. This seems to have been the standard mode of execution for heinous crimes at Athens (Lys. 13. 56, Arist. Ath. Pol. 45. 1, cf. E. Ar. Plut. 476. Photius s.v. , but the word seems to have been used in some contexts for any violent mode of execution (Plut. Suil. 6, Mor. 1049 D, Euseb. HE 5. 1. 47; Barnes, T. D. cf., JRS lviii (1968), 40, n. 91).Google Scholar The word is at least consistent with stoning to death by the Macedonian army assembly, which seems to have been the regular method of execution (Curt. 6. 11. 10 and 38 —more patrio). Crucifixion, however, is not impossible; the consensus of evidence is that Callisthenes died in this way (Plut. Al. 55. 9; Arr. 4. 14. 3 (Ptolemy)).

^{page 94 note 2} This detail is unique; the subsequent reference to a night sacrifice is paralleled in Curtius (3. 8. 22). It is interesting that Alexander had offered sacrifices to Poseidon and the Nereids at the Hellespont crossing (Arr. 1. 11. 6), and there is no reason why he should not have invoked them before Issus, close at hand as they were (so Wilcken, op. cit., p. 155). However, the absence of corroboration for such a striking detail tells strongly against its authenticity.

^{page 94 note 3} The phrasing indicates that the subject of the anecdote is Darius rather than Alexander . It makes little difference, however, whether one refers the anecdote to Darius or Alexander (so Grenfell, and Hunt, , P.Oxy. xv. 134)Google Scholar; it is still unattested elsewhere.

^{page 94 note 4} The figure for the Persian mercenary losses is missing in the papyrus, but even allowing a fairly substantial total the Persian casualties are by far the lowest in any source other than Justin, who gives 61,000 infantry and 10,000 cavalry (11. 9. 10). The papyrus, it may be added, is the only source which volunteers a total for the mercenaries' losses.

^{page 95 note 1} The first detailed analysis was byVolquardsen, J.,Untersuchungen über die Quellen der griechischen und sicilischen Geschichte bei Diodorus XI-XVI (Kiel, 1868)Google Scholar, summarized by Schwartz, RE v. 679–80 ═ GG, pp. 59–60. There has been little doubt since that Diodorus’ narrative of Greek affairs in this period is taken directly or indirectly from Ephorus.

^{page 95 note 2} Diod. 16.84–85. 1 ═ Dem. 18. 169–178. Note also Diodorus’ citation of Lycurgus’ speech against Lysicles (88. 2).

^{page 95 note 3} Diod. 16. 85. 4, citing verbatim Dem. 18. 136; for the dating to 344/3 b.c. see Dem. 7.20.

^{page 95 note 4} Schwartz, RE v. 682, adduces Aelius Aristeides, 38. 483. 5 and 39. 503 f., where Python is made to lead the embassy to Thebes. But it seems impossible that Python represented Philip on that occasion; Mar-syas, cited by Plutarch {Demosth. 18 ═ FGrH 35/6 F 20) stated that Philip's representatives to Thebes were Amyntas and Clearchus of Macedon with Daeochus and Thrasydaeus of Thessaly. There is, moreover, no mention of a Byzantine in Philochorus‘ list of nations comprising the embassy (FGrH 328 F 56).

^{page 95 note 5} Cf. 16. 93.

^{page 95 note 6} 16. 91. 2; 92. 2–3. A similarly ambiguous response from the Sibyl of Erythrae is cited by the late Paroemiographers (Zeno-bius, 4. 78, Diogenianus, 5. 75, Apostolius, 9 83).

^{page 96 note 1} So Justin, 11. 2. 1; cf. Diod. 17. 2. 1; Plut. Al. io. 8.

^{page 96 note 2} Cf. Beloch, iii². 77; Berve, ii. 17, 80, 169.

^{page 96 note 3} This is implied by Badian, (Phoenix xvii (1963), 248)Google Scholar and stated as fact by Welles (op. cit., p. 350 n. 1).

^{page 96 note 4} We hear of another Arrhabaeus, the father of Neoptolemus and Amyntas (Berve, nrs. 59, 547); there is no reason other than the name to identify him with the son of Aeropus.

^{page 96 note 5} Diod. 17. 32. 1; 80. 2; Curt. 7. 1. 5; 8. 8. 6; Just. 11. 7. 1. A parallel might be the non-Argead usurper, Ptolemy, son of Amyn-tas, who is referred to as ό Άλωpίτη (Diod. 5. 71. 1; 77. 5)

^{page 96 note 6} For Perdiccas see Curt. 10. 7. 8; Arr. 6. 28. 4; Ind. 18. 5. For the sons of Machatas see Berve, ii. 75–6 (nr. 143); the evidence linking them with the royal house of Elimi-otis is no stronger or weaker than that linking the sons of Aeropus with the royal house of Lyncestis.

^{page 97 note 1} Plutarch (Al. 10. 6–7) claims that Olympias bore the brunt of the slander but Alexander was not free from suspicion. Justin goes further and claims as fact that Olympias had horses ready for Pausanias’ escape and surreptitiously placed a crown on his crucified body, quod nemo alius audere nisi haec superstite Philippi filio potuisset (9. 7. 9–10).

^{page 97 note 2} Plut. Al. io. 6 f.; Just. 9. 6. 4 ff.; Diod. 16. 93. 3–9.

^{page 97 note 3} Cf. Badian, art. cit., p. 247.

^{page 97 note 4} Pausanias came from Orestis—Diod. 16. 93. 3

^{page 97 note 5} The best impression of the geographical position is given by Hammond, , Epirus (Oxford, 1967)Google Scholar, Maps 14 and 15. There is a useful summary of evidence by F. Geyer, , Makedonien bis zu Thronbesteigung Philipps II (Historische Zeitschrift, Beiheft 19 [1930]), 11–15.Google Scholar

^{page 98 note 1} Str. 7. 7. 8 (326); cf. Hecataeus FGrH 1 F 107.

^{page 98 note 2} Str. 9. 5. 11 (434).

^{page 98 note 3} Str. 7. 7. 8 (326–7) states that Pelagonia was part of the old area of Upper Macedonia, in the upper reaches of the Erigon by Lyncestis. (Ps.-Scymnus 621 also juxtaposes the areas.) The route taken by Philip V in 211 also implies that Pelagonia was to the north (Liv. 26. 25. 4). cf. Geyer, op. cit., p. 74.

^{page 98 note 4} Tod, GHI ii. 143, 1. 7 ═ IG ii², no; cf. Tod, 148.

^{page 98 note 5} Berve, ii. nr. 654. Note also his list of attested domiciles (ii. 445 b).

^{page 98 note 6} Diod. 17. 57. 2 (cf. Curt. 4. 13. 28). Thucydides has no hesitation in calling theLyncestians, Elimiotians, and other mountain peoples Macedonian (2. 99. 2).

^{page 98 note 7} Ait. i. 2. 5.

^{page 98 note 8} Hammond, , Epirus, pp. 444 ff.Google Scholar, argues that Strabo's ultimate source was Hecataeus of Miletus. Whether or not this is so, the passages refer to a period before Philip–s incorporation of the upper kingdoms into the Macedonian state.

^{page 98 note 9} Thuc. 2. 80. 6.

^{page 98 note 10} First published by Evangelides, D., Hel-lenica xv (1957), 247 fr.Google Scholar, and revised by Hammond, , Epirus, pp. 527 ff.Google Scholar

^{page 98 note 11} Evangelides merely dated the inscription before 330 b.c.; Hammond's restoration of the name of King Neoptolemus (p. 529) is avowedly speculative.

^{page 99 note 1} Apart from Strabo–s claim that they were Epirote by origin we only hear of communications for trade (Diod. n. 56. 3).

^{page 99 note 2} Thuc. 4. 79. 2; 83; 124–6. Compare his brief earlier statement (2. 99. 2):

^{page 99 note 3} Thuc. 4. 125.

^{page 99 note 4} Arist. Pol. 5. 1311^b12–14.

^{page 99 note 5} Str. 7. 7. 8 (326)

^{page 99 note 6} Geyer, op. cit., pp. 79–80 argues that Irrhas was king of Elimiotis; Beloch, iii². 2, 79 claims Orestis as his domicile. Both hypotheses rest on a priori arguments, not positive evidence.

^{page 99 note 7} Plut, de lib. educand. 20 (14 b), Libanius, Vit. Demosth. 9, Suda s.v. Kάρανοςpavos; cf.Momigliano, , Filippo il Macedone, p. 30 n. 2.Google Scholar There seems to me no force in Beloch's counter-argument–'aber die makedonischen Herrscher haben sich wohl Nebenfrauen aus Illyrien geholt, aber nicht ihre Koniginnen'.

^{page 99 note 8} Tod, ii. 148 = SIG³ 188.

^{page 99 note 9} There is no reason except the patronymic Άρρβαίου to suppose that Menelaus was ruler of Lyncestis. If IG ii², i, 190 is correctly assigned to the archonship of Chion (so Ferguson, , Klio, xiv (1915), 393 n 5)Google Scholar. Menelaus cannot have been king in 365/4; in that year proxenia and euergesia are awarded to II[…τόνΠε]λαγόνων βα[σλα.

^{page 100 note 1} Satyrus ap. Athen. 13. 557

^{page 100 note 2} Elimiotis seems to have supplied an entire phalanx battalion (Diod. 17. 57. 2).

^{page 100 note 3} Scholiast to Thuc. 1. 57. 3; it is invariably (and correctly) assumed that the Derdas in Thucydides is of the same family as the Derdas who is named in Xenophon as (Hell. 5. 2. 38). The treaty of alliance between Athens and Per-diccas II, almost certainly datable to 423/2 (SEG x. 86 — Bengtson, Staatsverträge, nr. 186), mentions a of the signatories(1.61).

^{page 100 note 4} Arist. Pol. 5. 131 1^b1 1 ff.

^{page 100 note 5} Xen. Hell. 5. 2. 38; cf. 5. 2. 12; Diod. 15. 19. 2.

^{page 100 note 6} Str. 7. 7. 8 (326), Just. 7. 4. 4–5; for the date see Beloch iii². 2, 66.

^{page 100 note 7} On the vexed question of chronology see Costanzi, V., Klio vi (1906), 297 fr.Google Scholar Beloch, iii². 2, 57 f., Geyer, pp. 112 ff., Momigliano, p. 31 n. 1. Diodorus dates Amyntas' expulsion by Illyrian raiders to 393/2, the second year of his reign (14. 92. 3). He adds that Amyntas was restored with Thessalian support, having given his border lands to the Olynthians for defence. As Geyer observes, this is a note inserted from Diodorus' main historical source, which certainly dated the expulsion within a few years of Amyntas' accession. The subsequent difficulties raised over the ‘doublet’ in Diod. 15. 19. 2 seem to me unreal. Diodorus is explaining the reasons for Amyntas' appeal to Sparta in 383/2, and he merely gives a resumptive note, explaining that the border lands had been surrendered in the crisis of the Illyrian invasion and that Amyntas was now attempting to recover them. There is no reason to posit a second Illyrian invasion; the emphasis in this second passage is on the feud with Olynthus, and the dispute is explained as having originated ten years before in the Illyrian crisis. Xenophon (Hell. 5. 2. 12) says nothing about the Illyrians, his concern being to present Olynthian aggression in its most lurid colours so as to justify the Spartan expedition of 382; the original gift of the border lands was best omitted. The insoluble problem is how long Amyntas was in exile after the Illyrian raid of 393/2. At that time he was replaced by a pretender named Argaeus (pace Geyer, the usurpation must be dated immediately after the Illyrian invasion; there is no evidence for a second expulsion). Diodorus says that there was a tradition of an usurpation of two years (14. 92. 3, repeated by the chronographers— cf. Beloch iii². 2, 51); this is at variance with the statement in his main historical source that the exile was short-lived. However, even allowing a two-year exile, Amyntas would have been back in Macedon by 391/0 in ample time to marry Eurydice. Isocrates, as so often, adds nothing but confusion with his rhetorical statement in the Archidamus (6. 46) to the effect that Amyntas was deprived of virtually his whole kingdom by the barbarians, and after sending for Spartan help recovered his entire realm . There is nothing here about the dispute with Olynthus, only a vague statement about a barbarian invasion and a quick recovery with Spartan help (perhaps confused with the Thessalian restoration). A vague historical parallel put in Archi-darrius' mouth to encourage the Spartans in their resistance to Thebes cannot be pressed against the largely coherent narrative of Diodorus, based as it is on the work of Ephorus. I conclude that there was only one major Illyrian invasion, which occurred shortly after Amyntas' accession.

^{page 101 note 1} Geyer, p. 111, argues on a priori grounds that the Lyncestians were overrun by the Illyrians. There is no evidence for this.

^{page 101 note 2} Cf. Arr. Succ. F 1, 22–3, Berve, ii, nr. 456 (Kυννάνη).

^{page 101 note 3} Alexander was born in 356, about the time of an Olympic festival (Plut. Al. 3. 8). Olympias then married Philip before autumn 357 (so Beloch, iii². 2, 68).

^{page 102 note 1} Plut. Al. 9. ii. Badian, art. cit., p. 244 n. 8, suggests that Alexander's host in Illyria was Langarus, king of the Agrianians.

^{page 102 note 2} Athen. 13. 557 d; Plut. Al. 9. 7; Just. 9. 7. 3.

^{page 102 note 3} Cf. Badian, art. cit., pp. 245 f.

^{page 103 note 1} Berve, ii. 46 (nr. 94), cites the Alexander Romance for the story of Antipater hailing Alexander immediately after his father's murder, and Badian (p. 248) accepts the detail. However, the account of Philip's murder in the Alexander Romance (Ps.-Call. 1. 24 fr.) is pure fantasy, and it is arbitrary to select one detail out of the farrago as authentic.

^{page 103 note 2} Curt. 6. 9. 17: Hie Amyntae, qui mihi consobrinus fuit et in Macedonia capiti meo impias comparavit insidias socium et conscium adiunxit (cf. 6. 10. 24).

^{page 103 note 3} Curt. 6. Io. 24; cf. Badian, p. 245.

^{page 103 note 4} Cynnane, the ex-wife of Amyntas, was betrothed to Langarus, king of the Agri-anians, in 335 shortly before the invasion of the Thebaid (Arr. 1. 5. 4). Amyntas must have been already dead, and the most probable assumption is that he was liqui- dated immediately after Alexander's accession.

^{page 103 note 5} Plutarch (de Al. fort. i. 3270) lists Alexander's difficulties before the Asiatic campaign and states: . This is the sole evidence that Alexander's accession was a stormy period at home, but it seems a roughly accurate description of the chaos which must have ensued after Philip's murder.

^{page 103 note 6} The details of their arrest are unknown, but there is certainly no evidence for Badian's dramatic statement (p. 248)—‘The two brothers who were supposed to have procured Philip's assassination were taken entirely by surprise and presented themselves as meek and helpless victims to his successor.’

^{page 104 note 1} Berve, ii. nr. 182 (Άτταλος). N.B.Justin, 11. 5. i; proficiscens ad Persicum bellum omnes novercae suae cognatos, quos Philippus in excel-siorem dignitatis locum provehens imperilspraefecit, interfecit. The purge must have been extensive.

^{page 104 note 2} Diod. 17. 16. 2. The incident is unattested elsewhere, but Aristobulus claimed that Parmenion advised the king to form a liaison with the captive Barsine after Issus (Plut. Al. ai. 7 ═ FGrH 139 F 11).

^{page 105 note 1} Polyb. 18. 47. 6; cf. Liv. 39. 23. 6. Walbank in his commentary ad loc. suggests that the Orestian defection came in the truce of 198/7 (so Hammond, , Epirus, p. 620).Google Scholar