3. nititur; Virg. Aen. 4. 252 'paribus nitens Cyllenius alis.'

4. nomina, for plural cp. 3. 27. 76. Compare with the stanza the way in which Horace speaks of the enterprise when it is undertaken by a friend, not offered to himself, Epp. 1. 3. 10 [Titius] 'Pindarici fontis qui non expalluit haustus Fastidire lacus et rivos ausus apertos.'

6. quem. . aluere. This seems to be the certain reading, though the old Bland. is among a few MSS. which have cum.. saliere.' The vulg. was the reading interpreted by Acr. and the Comm. Cruq.

7, 8. Boils and rushes in a fathomless flood of words.' As so often in Horace, the interpretation of the simile is clothed still in language almost wholly metaphorical and borrowed from the simile itself, see on Od. 1. 35. 19, 2. 2. 1, 4. 4. 59. Ore' belongs more to the poet than to the river, 'profundo ore' being the analogue of 'ore rotundo,' A. P. 323, magno ore,' Virg. G. 3. 294, of varieties of poetical style. The epithet, on the contrary, belongs primarily to the river, and even 'ore' is a word which was probably felt to be capable of an interpretation in the same connection, though neither fountain-head' (Virg. Aen. 1. 245) nor 'mouth' (Virg. G. 4. 292) is a meaning which would bear pressing here.

10. audaces covers, probably, boldness of treatment and of tropes, besides the two points afterwards named-vocabulary and rhythm.

nova verba, novel words, long compounds. Tŵv dovoμáтav Tà μὲν διπλᾶ μάλιστα ἁρμόττει τοῖς διθυράμβοις, Arist. Poet. 22. 14. 11. devolvit, as the torrent rolls boulders down its bed.

12. lege solutis. Two technical expressions possibly contribute to the full force of this phrase: (1) legibus solutus,' said of any one exempted from the operation of any law (Cic. Phil. 2. 13), in later times of the emperor as above the laws (Merivale, vol. iii. p. 466); (2) ‘soluta oratio,' the common designation of prose, as exempt from strict laws of prosody; so that the words of the text form a sort of oxymoron, 'verse which is as free of law as if it were not verse.'

13. regesque. The kings obviously of mythology-Pirithous, Theseus, Bellerophon, not the kings of Pindar's day. He is speaking of Hymns and Paeans. Que' is the reading of all the best MSS., as against the vulg. 've,' and the change would hardly be necessary, even if it were clear that Pindar's Odes on the mythical exploits of demigods were classed under a different name from those addressed to gods. Horace might rank them together.

17. Elea palma. The Olympian Odes are taken as representatives of the ἐπινίκια.

18. caelestes, 'very denizens of heaven'; see on 1. 1. 6 'terrarum dominos evehit ad deos.'

equum, as Pindar makes mention of Pherenicus, Hiero's horse,

Ol. 1. 18, Pyth. 3. 74; but singing of horses may well mean of their riders.

19. potiore signis. Pindar makes the comparison himself, Nem. 5. 1 οὐκ ἀνδριαντοποιός εἰμι . . ἀλλὰ γλυκεῖ ἀοιδὰ στεῖχ ̓ ἀπ ̓ Αἰγίνας, κ.τ.λ. ; so Horace, Od. 3. 30. 1, 4. 8. 1-12.

21. flebili, tearful,' as 'flebilis Ino,' A. P. 123.

ve varies the 'sive,' 'sive,' of the last two stanzas, although he is adding a fresh department of poetry, viz. the opĥvo. Cp. 1. 22. 5-7 'sive. . sive . . vel.'

23. aureos is predicative, as all golden,' 'painted as golden'; cp. 1. 5. 9. The double contrast of the bright stars above, the gloomy shades (all colours are alike in the dark') below, is indicated by the verb 'educit,' on one side, by the adjective 'nigro' on the other.

educit in astra, as 'Musa vetat mori: Caelo Musa beat,' 4. 8. 27, of immortality in fame.

25. multa aura; there is no fear of his falling; the free and buoyant winds of heaven are beneath him as he soars into the upper air. Horace contrasts Pindar's higher flights and original inspiration with his own humble aims and laborious imitative method, gathering honey from flower to flower of Greek lyrics, and enshrining it in the "curiosa felicitas" of his own poetical style'; in v. 29, 'circa nemus.. Tiburis,' we get a slightly different idea, from wood to wood of his Sabine neighbourhood.' The simile was introduced to express the difference in genius and literary habits between Pindar and Horace, but it suggests to the poet his own wanderings in the valley of the Anio. The Matinian bee (save in its antitype, the poet of Venusia) has no business at Tibur.

Dircaeum; Virg. E. 2. 24 Amphion Dircaeus'='Theban'; but a 'spring' will be specially germane to a 'swan.'

26. Antoni. As Orelli points out, the name is repeated to mark the summing up and the practical conclusion. 'You know, Antonius, the difference between Pindar's powers and mine. You will treat this high theme yourself better than I can.'

27. Matinae; see Introd. to 1. 28.

28. more modoque, a common formula. Cicero's fragment, Timaeus, seu de Univers. 1 Carneadeo more et modo.'

30. plurimum, probably with laborem,' and with a slight relation of antithesis, as its position shows, to 'grata.' Dill., after Bentley, joins 'plurimum nemus'; but the epithet would not be very forcible (Bentley wished to escape from it to floreum'), and 'nemus ripasque uvidi Tiburis' is a thoroughly Horatian arrangement.

uvidi; 1. 7. 14, 3. 29. 6.

31. operosa parvus . . fingo, probably not without some reference

back to the first description of Pindar's poetry, its vast spontaneous rush. 'immensus ruit profundo ore.'

33. concines. We should rather expect an emphatic 'tu.' Cp. 2. 17. 30, 3. 17. 5. The 'ego. parvus' of the last stanzas contrasts not only with Pindar before, but with Antonius maiore plectro,' who follows. We may note, also, that part of the contrast lies in the occasion. Not now, neither you nor I. By and by, whenever it is that Caesar comes home in triumph, a greater poet than I shall sing his praises, and even I (v. 45) shall find a voice in my delight.' 'Maiore plectro' occupies a double relation, as instrumental ablative with concines,' and descriptive ablative with 'poëta'; see on 1. 3. 6. With the expression itself cp. 'leviore plectro,' 2. 1. 40, aureo plectro,' 2. 13. 26. 34. quandoque='quandocunque'; see 4. I. 17, A. P. 359. 35. per sacrum clivum; see on Epod. 7. 7 Britannus. . descenderet sacra catenatus via.' Cp. Mart. 1. 71. 5. The name was given to the slope by which the 'sacra via' descended, from the spot where its pavement is still visible under the Arch of Titus, into the Forum, a fall of 53 feet. Burn's Rome and Campagna, p. 78.

36. Sygambros; see Introd. to the Book.

37. Cp. Epp. 2. I. 17 (of Augustus) Nil oriturum alias, nil ortum tale fatentes.'

42. publicum ludum. The technical and usual phrase would be 'ludos,' but Horace varies it slightly, after his custom; 2. 4. 24, 3. 5. 42, 4. 14. I, 4. 15. 9.

super impetrato.. reditu, in discharge of the vows for his return, which have won their purpose. Orelli mentions the fact that coins have been found of the year B. C. 16 (A. U. C. 738) with the inscription S.P.Q.R.V.S. (vota suscepta) PRO S. (salute) et red. aug. The ludi votivi' here anticipated were really celebrated in B. C. 13; Dio C. 54. 27

45. audiendum, 'that merits hearing.' Horace is speaking, at the moment, not of shouting with the shouting mob, but of singing, i.e. writing poetry; but there is probably a metaphorical play in the expression, if I can make my voice heard in the din.'

46. bona pars; Sat. 1. 1. 61, A. P. 297; so ‘bona copia,' Epp. I. 18. 109. 'A large part,' i. e. I will raise my voice ungrudgingly. my delight.'

48. felix, 'in 49 foll. The difficulties of this stanza are well known, and no solution of them is thoroughly satisfactory. The reading of the text is that of the great preponderance of good authorities. Tuque' is found in some MSS. of secondary value, and adopted by Ritter; procedit' is found in the Berne MS., having been proposed, in ignorance of that fact by Heinsius, and supported by Bentley. To whom does the pronoun refer? Only two answers worthy of consideration have been given.

(1) To Antonius. In favour of this is the fact that 'te' at the beginning of the next stanza undoubtedly belongs to him, and it is unlike Horace's finished workmanship to put the same pronoun in an emphatic place in two consecutive stanzas, when the subjects to which it refers are wholly different. Against it must be counted the necessity which it involves of accepting, on very slender authority, the reading 'tuque,' and the difficulty of putting any satisfactory meaning on 'dum procedis.' Newman thinks that Antonius would ride in Augustus' chariot as a member of the imperial family, by marriage with Marcella. Ritter, that he would take part in the procession as praetor (this would oblige us to postpone the date of the Ode to B. C. 13). Bentley seems really to dispose in advance of both suggestions. Antonius' place in the procession, if he had one, could hardly be important enough to bear the weight of this stanza. 'Dum procedis,' with no qualification or hint that the cries of triumph were not for him, could hardly be said of any person but the triumphing general. (2) To 'Triumphus,' the triumph personified. 'Thy name will we pronounce as thy procession passes by, Ho Triumph! again and again, Ho Triumph!' This is the interpretation of Acr. and Porph. ( ad ipsum triumphum conversus haec dicit'), and of Orelli, save that he accepts 'procedit' from the Berne MS., and takes it of Caesar, as he passes along.' The only serious objection to it is the separation before noticed of the 'te' of this stanza from the 'te' of the next. Bentley objects, also, that we are taking ‘io Triumphe' in two different senses; in the first line as in the poet's mouth, in the second materialiter,' as the cry to be uttered by the people. But this is hardly true. The construction in both cases is 'te dicemus, io Triumphe.' 'Io Triumphe' is an invocation of the personified Triumph-god; it is also the actual cry uttered. Bentley ought equally to object to Virgil's 'Evoe Bacche fremens, solum te virgine dignum Vociferans,' Aen. 7. 389, where 'Evoe Bacche' is at once the actual cry of Amata and the vocative case which explains the 'te dignum' (not 'tu dignus es'), in which the poet, not Amata, is the direct speaker. For 'io Triumphe' was the cry raised as the procession passed along (it was an old cry, and not originally limited to triumphant processions, for it occurs in the Hymn of the Fratres Arvales), cp. inter al. Ov. Trist. 4. 2. 51 Tempora Phoebea lauro cingentur, Ioque, Miles, Io, magna voce, Triumphe, canent.' That it was held to involve a personification of Triumphus appears from Livy's expression, 45. 38 [Milites] triumphum nomine cient, suasque et imperatoris laudem canentes per urbem incedunt.' But compare especially Hor. Epod. 9. 21 foll. 'Io Triumphe, tu moraris aureos Currus et intactas boves? Io Triumphe, nec Iugurthino parem Bello reportasti ducem,' &c.; see note there. Orelli's 'procedit' rather improves the stanza, by giving it a more con

6

tinuous connection with the last. 'It is Caesar's triumph that will unlock my voice, as it will of that of all Rome'; and there is more ob vious motive in the 'te,' helped by the corruption to 'tu,' for altering the third person to the second, and vice versa. Bentley, dissatisfied (and it must be admitted, after all, with some reason) with both these inter pretations, cuts the knot by reading 'Isque dum procedit,' of Caesar.

51. civitas omnis seems to add the climax to vv. 33-44 of Antonius' share in the rejoicings, and vv. 45-48 of Horace's own, and to prepare the way for the division again of their duties in the sacrifices. 'We will shout together in the shouting town, but when we come to offer our thank-offerings, we shall be as unequal as we were in our poetical offer. ing, ten bulls and cows for you, and a little calf for me.'

dabimus thura, of altars by the wayside, on which incense was offered as the procession passed.

53. te decem tauri; cp. 2. 17. 30' reddere victimas Aedemque votivam memento: Nos humilem feriemus agnam,' and 3. 23.

54. solvet, sc. a voto.

55. iuvenescit, is growing to a 'iuvencus'; the common use of the verb is 'to grow young,' when one has been old.

56. in mea vota, 'to pay my vow.'

57. Whose horns are like the moon three days old.' Others, offended at the exaggeration, for the horns of a calf recently weaned are only just budding, take the whole description of a crescent-shaped mark on the forehead. But the stanza reads like a reminiscence of the young bull in Mosch. 2. 84 foll., of which tò μèv äλλo déμas ¿avlóxpoov ÉTKEY,| κύκλος δ ̓ ἀργύφεος μέσσῳ μάρμαίρε μετώπῳ, | ισά τ ̓ ἐπ ̓ ἀλλήλοισι κέρα ἀνέτελλε καρήνου | ἄντυγος ἡμιτόμου κεραῆς ἅτε κύκλα σελήνης. The homely ending of the Ode has been severely criticised: Desinit in vitalum. . formosa superne.' It is at least intentional on Horace's part; see Introd. to Odes i-iii, § 11. 3, and on 3. 5. 55. On the immediate topic of the last six lines see on 3. 13. 4. In this place the detailed account of the intended offering increases the contrast of vv. 53, 54, and balances to some extent the inequality. You will sacrifice animals brought by the score from your "latifundia," I the calf which I bred myself, whose birthday and marks I know by heart.' It is like his offer to Maecenas in I. 20. 1-4 q. v. 'common wine; but home-made, carefully stored, and with pleasant memories about it.'

59. niveus videri, see App. 2, § 2.

ODE III.

The poet is one set apart from his birth by the Muse's favour; neither his mode of life nor his objects of ambition are the same as