¹ Herodotus used conversation passim; Thucydides' conversations are often extensions of the antithetical debate: 2. 71–4, 5. 87–111, but cp. 3. 113; for Ctesias, see Demetrius, , On Style 216Google Scholar; there are no extant passages of conversation from the H.O. or Ephorus: see Bruce, I. A. F., An Historical Commentary on the Hellenica Oxyrhynchia (Cambridge, 1967), p. 6Google Scholar on direct speech in the H.O.; for Theopompus see Jacoby, , FGrHist 115F 21, 75Google Scholar. Conversation is found in the Hellenica of Xenophon himself at 3. 1. 22–8, 3. 3. 1–3, 3. 3. 4–7, 3. 4. 5, 3. 4. 9, 3. 4. 25–6, 4. 1. 6–14, 4. 1. 29–38, 4. 3. 1–2, 5. 4. 25–33. Briefer exchanges also occur, like those at 1. 5. 6, 1. 6. 2, 1. 6. 32, 2. 1. 25–6, 2. 3. 15–16, 2. 3. 56, 3. 2. 6–7, 5. 3. 15. All the historians mentioned above, with the possible exception of the H.O., made use of speeches.

² Sordi, M., ‘I Caratteri del'Opera Storiografica di Senofonte nell'Elleniche’, Athenaeum 28 (1950), 15–16Google Scholar. Cp. Jacoby, FGrHist 115F 21Google Scholar.

³ Breitenbach, H. R., Die Historiographische Anschauungsformen Xenophons (Basel, 1951), pp. 101–4Google Scholar.

⁴ Henry, W. P., Greek Historical Writing (Chicago, 1967), pp. 156–60Google Scholar.

⁵ Bruce, op. cit., n. 1, p. 134.

⁶ See Henry, op. cit., n. 4, pp. 155–61 for a criticism of the theory, which Sordi develops on pp. 10–16, 24–5 and 35 n. 1. See also infra pp. 330–1. Briefly, Sordi's theory is that Xenophon composed the middle section of the Hellenica, which she defines as 3. 1. 1–2, 20 and 3. 4. 1–29 and 4. 1. 1–2. 2 and 4. 3. 1–4. 7. 7, as a memoir of his adventures with Agesilaus in Asia and Greece, and that conversation is characteristic of this genre. Later he turned it into a history of Greece, adding sections to this middle section, and this is marked by avoidance of conversation and preference for the set speech. My principal objection to this is that conversation occurs outside the alleged memoir section at 2. 3. 56 for instance and particularly at 5. 4. 25–33. Sordi notices this discrepancy but does not even try to explain it. Henry's criticisms are equally telling.

⁷ Xenophon, , Ages. 3Google Scholar. 3 indicates that Spithridates had been insulted by Pharnabazus in the matter of his daughter's marrige and Agesilaus would bear this in mind in proposing this better match for her. However, Xenophon does not mention the insult here. In view of the wealth of detail in the passage we would expect him to do so if he had thought it important for the reader to know. I conclude that it can have little direct bearing on our interpretation of the passage.

⁸ See Bruce, op. cit., no. 1, p. 144. For the end of the alliance Hell. 4. 1. 27–8.

⁹ Bruce, op. cit., n. 1, p. 134.

¹⁰ Indeed it is treated in narrative form in Xenophon, 's Agesilaus 4Google Scholar, under the heading of justice.

¹¹ See Gould, J., ‘Dramatic Character and Human Intelligibility in Greek Tragedy’, PCPS n.s. 24 (1978), 43–67 and particularly p. 55Google Scholar.

¹² The sentences are long and balanced and note the indignant alliteration as well in τ⋯ μ⋯ν κατακεκομμ⋯να, τ⋯ δ⋯ κατακεκαυμ⋯να.

¹³ Plutarch, , Life of Agesilaus, 12. 3–5Google Scholar.

¹⁴ Breitenbach, op. cit., n. 3.

¹⁵ In its symbolism this scene resembles many from the dialogues of Plato, on which Friedländer, P. (Plato, vol. i, New York, 1958, pp. 158–61)Google Scholar says ‘Plato could not tolerate any accidental element in his work. He was compelled to select the participants…to attune the surroundings to the inner content, to strip the natural setting of accidental factors so that it could become an effective agent in the total work.’ He concludes ‘Thus, for an understanding of the dialogues it may be necessary to inquire more deeply…into the symbolic meaning of the spatial setting and the physical happenings…For the frame action is not constructed accidentally…’ The same could apply to Xenophon's dramatic setting here.

¹⁶ Ages. 25. 5.

¹⁷ op. cit., n. 2.

¹⁸ See Demetrius, , On Style 128–89Google Scholar for the characteristics of this style.

¹⁹ See Jacoby, , FGrHist 115F 21Google Scholar.

²⁰ For the nature of Theopompus' style, see Dionysius of Halicarnassus, Ep. ad Pomp. c. 6.

²¹ See Friedländer, op. cit., n. 15.

²² Xenophon's attachment to Socrates is mentioned by Laertius, Diogenes, Lives of the Philosophers 2. 48Google Scholar. His Oeconomicus is a Socratic dialogue and his Memorabilia feature Socrates the conversationalist. His Hiero is a proper philosophical dialogue in the Platonic style.

²³ Demetrius, , On Style 298Google Scholar.

²⁴ Ibid. 216 for an extract from his Persica.

²⁵ For the remaining fragments, Jacoby, FGrHist. For his tragic flair see above extract.

²⁶ Xenophon refers to his work at Anabasis 1. 8. 26–7.

²⁷ Demetrius, , On Slyle 137Google Scholar.

²⁸ See examples at Herodotus 7. 101. 3 and Xenophon, , Cyropaidia 8. 4. 13Google Scholar.

²⁹ Hatzfeld, J., ‘Notes sur la chronologie des Hélléniques’, REA 35 (1933), 387–409Google Scholar, had Leotychidas born in 412 b.c. and Agesilaus succeed in 398 b.c., so that Leotychidas would be a mere boy when the quarrel took place. But Littman, R. J., ‘A New Date for Leotychidas,’ Phoenix 23 (1969), 269–77Google Scholar argued most plausibly that Agis would not have waited so long (or been allowed to wait so long) to produce an heir to the Spartan throne. He identifies the earthquake that Xenophon mentions with regard to Leotychidas' conception (3. 3. 2) with that mentioned by Thucydides at 3. 89 and so has Leotychidas a man of 25 years when he contests the throne with his uncle Agesilaus.

³⁰ H.O. 22 (17). I refers to a meeting between Agesilaus and the man Xenophon calls Otys, in editions subsequent to the original one. The reading is a conjecture based on Xenophon's text. It does not refer to the betrothal, however.