=71. SI DARE= _R. J. Tarrant._ The manuscripts' SED DARE is a possible reading; but Professor Tarrant's slight change removes the awkwardness of _nec tamen_ following immediately upon _sed_.

=71. MAVIS= _IF2ul_ MAIVS _BF1_. Either of the two variants could be read from _CMHLT_. The preferable reading is _mauis_, since it links more closely to _potes_ in the pentameter, and would be especially liable to corruption after _maiora_ two lines previous. I have found no good parallel for singular _maius_ 'a more important thing': for the plural _OLD maior_ 5 cites from verse _Fast_ IV 3 'certe maiora canebas' and its model, _Ecl_ IV 1 'paulo maiora canamus'.

=72. NEC TAMEN EX TOTO DESERERE ILLA POTES.= Graecinus was another of Ovid's addressees who, while a soldier, kept up his other pursuits: 'artibus ingenuis [=_lIberAlibus_], quarum tibi maxima cura est, / pectora mollescunt asperitasque fugit. / nec quisquam meliore fide complectitur illas, / qua sinit officium militiaeque labor' (_EP_ I vi 7-10).

=72. EX TOTO.= 'Altogether'. Compare _EP_ I vi 27-28 'spes igitur menti poenae, Graecine, leuandae / non est _ex toto_ nulla relicta meae'. The idiom was probably subliterary: the only instances from the time of Ovid cited by _OLD totum_ 2 are Celsus III 3 71b 'neque _ex toto_ in remissionem desistit' and Columella V 6 17 'antequam _ex toto_ arbor praeualescat'.

=73. NVMERIS ... VERBA COERCES.= 'You arrange words in metrical patterns'.

Similar wording at Cic _Or_ 64 'mollis est enim oratio philosophorum ... nec _uincta numeris_ ['not in rhythmic prose'], sed soluta liberius'.

Professor E. Fantham points out to me that Ovid may also be playing on _numerus_ 'military contingent' (_OLD numerus_ 9): 'you draft words in squads'.

=75-76. NEC AD CITHARAM NEC AD ARCVM SEGNIS APOLLO, / SED VENIT AD SACRAS NERVVS VTERQVE MANVS.= Apollo is similarly described at _Met_ X 107-8 (of Cyparissus) 'nunc arbor, puer ante deo dilectus ab illo / _qui citharam neruis et neruis temperat arcum_'.

=76. VENIT= = _conuenit_. In Latin verse a simple verb can carry the sense of any of its compounds, even when this sense is quite different from the usual meaning of the simple verb. Compare Catullus LXIV 21 'tum Thetidi pater ipse _iugandum_ Pelea _sensit_', "where it is plain that iugandum is for coniugandum, and this leads the reader to the conclusion that sensit is for consensit, where the omission decidedly affects the sense" (Bell 330).

The line should not be taken as an instance of the expression _uenire ad manum_ (_OLD uenio_ 7c), since the idiom's sense 'be convenient' does not fit the context here: for the sense compare Livy XXXVIII 21 6 'quod [_sc_ saxum] cuique temere trepidanti _ad manum uenisset_' and Quintilian II xi 6 'abrupta quaedam, ut forte _ad manum uenere_, iaculantur'. _Venire in manus_ offers a somewhat more satisfactory meaning, almost equivalent to 'have, hold' (compare Cic _Q Fr_ II xv [xiv] i 'quicumque calamus _in manus meas uenerit_' and Persius III 11 '_inque manus_ chartae nodosaque _uenit_ harundo'), but seems to be a separate idiom.

=79. QVAE QVONIAM NEC NOS.= 'Since she continues to give poetic inspiration to myself as well as to you'. _Quae quoniam_ seems very prosaic, but Ovid uses the phrase again at _Tr_ I ix 53-54 '_quae_ [_sc_ coniectura] _quoniam_ uera est ... gratulor ingenium non latuisse tuum'.

=79-80. VNDA ... VNGVLA GORGONEI QUAM CAVA FECIT EQVI.= Hippocrene, the spring of the Muses, said to have been created by the hoof-beat of Pegasus. Similarly described at _Met_ V 264 'factas pedis ictibus undas', _Fast_ V 7-8 'fontes Aganippidos Hippocrenes, / grata Medusaei signa ... equi' and Persius prol 1 'fonte ... caballino'.

=80. VNGVLA ... CAVA.= Professor J. N. Grant points out to me the possible borrowing from Ennius _Ann_ 439 Vahlen3 'it eques et plausu _caua_ concutit _ungula_ terram'.

=80. GORGONEI ... EQVI.= The same phrase in the same metrical position at _Fast_ III 450 'suspice [_sc_ caelum]: _Gorgonei_ colla uidebis _equi_'.

For the birth of Pegasus from the blood of the Gorgon Medusa, see _Met_ IV 784-86,

=81. COMMVNIA SACRA TVERI.= _Sacra_ similarly used of poetry at _Tr_ IV i 87, _Tr_ IV x 19 'at mihi iam puero caelestia _sacra_ placebant', _EP_ II x 17 'sunt tamen inter se _communia sacra_ poetis', and _EP_ III iv 67 'sunt mihi uobiscum _communia sacra_, poetae'. For _tueri_ 'observe, maintain' compare Cic _Tusc_ I 2 'mores et instituta uitae resque domesticas ac familiaris nos profecto et melius _tuemur_ et lautius'.

=82. ISDEM STVDIIS IMPOSVUISSE MANVM.= Similar phrasing at _Tr_ IV i 27-28 'non equidem uellem ... _Pieridum sacris imposuisse manum_'.

=82. IMPOSVISSE= has the sense of the present infinitive, as is shown by _tueri_ in the previous line; compare as well ii 27-28 'uix sumptae Musa tabellae / _imponit_ pigras, paene coacta, _manus_'. For the idiom, see Platnauer 109-12. It is particularly frequent in the latter half of the pentameter, immediately before the disyllable: compare, from many instances, _AA_ III 431-32 '_ire_ solutis / crinibus et fletus non _tenuisse_ decet' and _Tr_ IV viii 5-12 'nunc erat ut posito deberem fine laborum / _uiuere_, me nullo sollicitante metu, / quaeque meae semper placuerunt otia menti / _carpere_ et in studiis molliter _esse_ meis, / et paruam _celebrare_ domum ueteresque Penates ... inque sinu dominae carisque sodalibus inque / securus patria _consenuisse_ mea'.

The idiom, although more common in elegiac verse, is also found in epic: compare _Aen_ X 14 'tum _certare_ odiis, tum res _rapuisse_ licebit'.

=83. LITORA PELLITIS NIMIVM SVBIECTA CORALLIS.= Compare ii 37 'hic mea cui recitem nisi flauis scripta Corallis'. Strabo mentions the Coralli as inhabiting the region near Haemus (VII 5 12); they are rather obscurely described at Val Fl VI 89-94 'densique leuant uexilla Coralli, / barbaricae quis signa rotae, ferrataque dorso / forma suum ['of pigs'], truncaeque Iouis simulacra columnae; / proelia nec rauco curant incendere cornu, / indigenas sed rite duces et prisca suorum / facta canunt ueterumque, uiris hortamina, laudes'.

Nothing else is known of the tribe.

=83. PELLITIS.= Elsewhere in Ovid only at x 2 'pellitos ... Getas'.

=83. NIMIVM SVBIECTA.= Compare vi 45 'nimium nobis conterminus Hister'.

=85. VLLO= _M_ ILLO _BCFHILT_. _Illo_ is not a possible reading, since of course most parts of the empire would have been less isolated than Tomis. Ovid does not specify a preferred place of exile at either _Tr_ IV iv 49 'nunc precor hinc alio iubeat discedere' or _EP_ III i 29-30 'non igitur mirum ... altera si nobis usque rogatur humus', nor in any of the passages listed in the next two notes.

=86. QVI MINVS ... DISTET.= For this constant prayer of the exiled Ovid, see _Tr_ II 575-78 (the concluding lines) 'non ut in Ausoniam redeam, nisi forsitan olim, / cum longo poenae tempore uictus eris; / tutius exilium pauloque quietius oro, / ut par delicto sit mea poena suo', _Ibis_ 28, _EP_ III i 4 & 85, _EP_ III iii 64, _EP_ III vii 30, _EP_ III ix 38, and _EP_ III ix 1-4 'Quod sit in his eadem sententia, Brute, libellis, / carmina nescio quem carpere nostra refers, / _nil nisi me terra fruar ut propiore rogare_, / et quam sim denso cinctus ab hoste loqui'.

=86. DISTET= _FHILM2c_. Lenz and Andre print DISTAT (_BCT_); however, the defining subjunctive seems to be required, and is supported by _EP_ II viii 36 'daque procul Scythico _qui sit_ ab hoste locum'.

=87. LAVDES.= See at 45 _laudum_ (p 268).

=88. MAGNAQVE QVAM MINIMA FACTA REFERRE MORA.= At _EP_ III iv 53-60 Ovid speaks of how a poem of his on a recent triumph has been late in being written, and will be late in reaching Rome: 'cetera certatim de magno scripta triumpho / iam pridem populi suspicor ore legi. / illa bibit sitiens lector, mea pocula plenus; / illa recens pota est, nostra tepebit aqua. / non ego cessaui, nec fecit inertia serum: / ultima me uasti distinet [_scripsi_: sustinet _codd_] ora freti. / dum uenit huc rumor properataque carmina fiunt / factaque eunt ad uos, annus abisse potest'.

=90. SOCERO PAENE ... TVO.= See at 11 _eadem mihi filia paene est_ (p 262).

IX. To Graecinus

C. Pomponius Graecinus (_PIR1_ P 540), suffect consul in 16, was the recipient of _EP_ I vi, an appeal for his assistance, and of _EP_ II vi, a request that he be more lenient towards Ovid's faults and continue to assist him. He must have been an old friend of Ovid, for _Am_ II x is addressed to him ('Tu mihi, tu certe, memini, Graecine, negabas / uno posse aliquem tempore amare duas'), and he was clearly a literary patron (_EP_ I vi 7-8 'artibus ingenuis, _quarum tibi maxima cura est_, / pectora mollescunt asperitasque fugit').

The poem begins with Ovid's wish that his letter might arrive on the day Graecinus becomes consul (1-4). He imagines himself present when Graecinus enters his magistracy; since he will not be there, he will at least in his mind imagine Graecinus carrying out his consular functions (5-56). He then speaks of Graecinus' brother Flaccus, who will succeed him as _consul ordinarius_ for 17: the two brothers will take pleasure in each other's office (57-65). He describes the brothers' devotion to Tiberius, and asks for their assistance in obtaining his removal from Tomis (65-74). The mention of his exile serves as a bridge to the topic of his life in Tomis. Flaccus can attest to the hardships Ovid endures, since he was recently stationed in the area (75-86). Once Graecinus has learned of these hardships from Flaccus, he should ask what Ovid's reputation in Tomis is. He will learn that Ovid is well liked, and has even received public honours (87-104). His loyalty to the imperial family is well known: Flaccus may have heard of this, Tiberius will eventually learn of it, but Augustus has certainly observed it from heaven; Ovid's poems are perhaps inducing Augustus to yield to his prayers (105-34).

The poem is the longest in the book, and combines several almost unrelated sections dealing with a number of subjects. The first section of the poem, the celebration of Graecinus' nomination to the consulship, is very heavily indebted to IV iv, Ovid's first poem on Sextus Pompeius'

election to the consulship. The section detailing Flaccus' presence near Tomis owes something to IV vii, the letter to Vestalis. The description of Ovid's reputation in Tomis is new, and shows a softening of his attitude towards his fellow-townsmen, but the description of his piety to the imperial family owes much to III ii, a letter of thanks to Cotta for the gift of images of the members of the family. The poem's discursiveness and large number of derived elements suggest a hasty composition.

=1. GRAECINE.= Graecinus became a _frater Arualis_ in 21 (_CIL_ VI 2023); the C. Pomponius Graecinus of _CIL_ XI 5809 (Iguvium) seems not to have survived to enter the Senate (Syme _HO_ 74-75). Graecinus is not mentioned in literary sources apart from Ovid, but his brother Flaccus was rather more famous: see at 75 (p 308).

=3. DI FACIANT= looks like a colloquial expression. Other instances at iv 47-48 '_di faciant_ aliquo subeat tibi tempore nostrum / nomen', _Tr_ V xiii 17, and Prop II ix 24.

=3. AVRORAM= here is virtually equivalent to _diem_; it is not found elsewhere in the poetry of exile, but compare _Fast_ I 461 & II 267-68 'tertia post idus nudos aurora Lupercos / aspicit'.

=3. OCCVRRAT.= 'Arrive', as commonly: compare Cic _Phil_ I 9, Livy XXXVII 50 7 '_ad comitiorum tempus occurrere_ non posse', and Pliny _Ep_ VI xxxiv 3 'uellem Africanae [_sc_ pantherae] quas coemeras plurimas _ad praefinitum diem occurrissent_'.

=4. BIS SENOS= = _duodecim_, metrically difficult because of its initial three consecutive short vowels. Roman poets avoid using the usual names for numbers above _nouem_, with the obvious exceptions of _centum_ and _mille_; sometimes, as here, metrical exigencies left them with no alternative. For _bis seni_ (_sex_) Tarrant at Sen _Ag_ 812 _bis seno ... labore_ cites Ennius _Ann_ 323 Vahlen2, _Ecl_ I 43, _Aen_ I 393, Prop II xx 7, _Met_ VIII 243, _Fast_ I 28, Sen _Tro_ 386 & _Oed_ 251, and from Greek Callimachus _Aetia_ I fr. 23 19 Pfeiffer.

=6. TVRBAE.= Compare iv 27 'cernere iam uideor rumpi paene atria _turba_'.

=7. IN DOMINI SVBEAT PARTES.= _Partes_ = 'function'; see at ii 27 _uix uenit ad partes ... Musa_ (p 170). For _subeat_ 'undertake' compare Quintilian X i 71 'declamatoribus ... necesse est secundum condicionem controuersiarum plures _subire personas_' and the passages cited at _OLD subeo_ 7b.

=8. FESTO= _Burman_ IVSSO _BCMFHIL_ IVSTO _T, sicut coni Merkel_. _Iusso_ has been explained since Merula as meaning that Ovid hopes the letter will arrive on the day it is told to; but the word seems rather strange, and lacks the point it has in the passages cited by Ehwald (_KB_ 64), _AA_ II 223-24 'iussus adesse foro, _iussa_ maturius _hora_ / fac semper uenias, nec nisi serus abi' and Prop IV vi 63-64 (of Cleopatra) 'illa petit Nilum cumba male nixa fugaci, / hoc unum, _iusso_ non moritura _die_' (she would commit suicide at a time of her own choosing), or at _Aen_ X 444 (cited by Owen in 1894) 'socii cesserunt _aequore iusso_', where _iusso_ stands by hypallage for _iussi_. The meaning of _iusto_ is inappropriate for the present passage, as will be seen from Suet _Tib_ 4 2 'retentis ultra _iustum tempus_ ['the time allowed'] insignibus'. Burman's conjecture _festo_ was not placed in the text even by its author, but it seems a reasonable solution to the difficulty. For it Burman cited 56 'hic quoque te _festum_ consule _tempus_ agam'; see as well _Fast_ I 79-80 'uestibus intactis Tarpeias itur in arces, / et populus _festo_ concolor ipse suo est'. The corruption of so straightforward an epithet may seem unlikely, but compare Prop IV xi 65-66 'uidimus et fratrem sellam geminasse curulem; / consule quo, _festo_ [_Koppiers_: facto _codd_] _tempore_, rapta soror'.

=9. ATQVI= _unus e duobus Hafniensibus Heinsii_. The ATQVE of _BCMFHILT_ is possibly right. For the adversative sense here required, _OLD_ _atque_ 9 cites Plautus _Aul_ 287-88 '_atque ego_ istuc, Anthrax, aliouorsum dixeram, / non istuc quod tu insimulas', _Mer_ 742, and Ter _Heaut_ 189 (apparently a misprint for 187 'atque etiam nunc tempus est') from comedy, but from the classical period only Cic _Att_ VI i 2 'ac putaram paulo secus' and _Fam_ XIV iv 5 '_atque ego_, qui te confirmo, ipse me non possum', and instances of _ac tamen_ at _Fam_ VII xxiii 1, Caesar _BC_ III 87 4, and Tac _Ann_ III 72. In view of the doubtful status of adversative _atque_ at the time of Ovid and the ease of corruption of _atqui_ to _atque_ I have followed Heinsius in reading _atqui_. Heinsius similarly restored _atqui_ from his _codex Richelianus_ for the other manuscripts' _atque_ at _Tr_ II 121-24 'corruit haec ... sub uno ... crimine lapsa domus. / _atqui_ ea sic lapsa est ut surgere, si modo laesi / ematuruerit Caesaris ira, queat'; and _atque_ is found for the correct _atqui_ in some manuscripts at Hor _Sat_ I ix 52-53 '"magnum narras, uix credibile!" "atqui / sic habet"'

and _EP_ I ii 33-34 '_atqui_ / si noles sanus, curres hydropicus', and in most manuscripts at _Ep_ I vii 1-5 'Quinque dies tibi pollicitus me rure futurum / Sextilem totum mendax desideror. _atqui_, / si me uiuere uis sanum recteque ualentem, / quam mihi das aegro, dabis aegrotare timenti, / Maecenas, ueniam'.

=10. SINCERO.= 'Unbroken'.

=12. SALVTANDI MVNERE ... TVI.= Professor R. J. Tarrant points out to me the notably prosaic use of the defining gerundive.

=13. GRATATVS= has the force of a present participle, as is shown by _cum dulcibus ... uerbis_; Andre mistranslates 'apres t'avoir felicite, je t'embrasserai avec des mots tendres'. The perfect participle of deponent verbs takes past or present meaning indifferently, according to context.

=16. VT CAPERET FASTVS VIX DOMVS VLLA MEOS= seems strange, as does Nemethy's explanation 'poeta elatus superbia tectum uertice tangere sibi uidetur'. Perhaps the distich means something like 'on that day I would be filled with a pride which no ancestry, no matter how illustrious, could justify'.

=16. FASTVS.= 'Haughtiness'--Wheeler. The same sense at _AA_ II 241-42 'exue _fastus_, / curam mansuri quisquis amoris habes' and _Aen_ III 326-27 (Andromache speaking) 'stirpis Achilleae _fastus_ iuuenemque superbum ... tulimus'. Ovid generally uses _fastus_ of the arrogance of women to their suitors (_Am_ II xvii 9, _Met_ XIV 762, _Fast_ I 419); the word is not found elsewhere in the poetry of exile.

=17. DVMQVE LATVS SANCTI CINGIT TIBI TVRBA SENATVS.= Compare iv 41 'inde domum repetes toto comitante senatu'; Ovid is here obviously referring to the earlier procession _from_ the new consul's house.

=20. LATERIS ... LOCVM= is a strange phrase, but is made easier by _latus ... cingit_ in 17. Compare also such passages as _Met_ II 448-49 'nec ... iuncta deae lateri nec toto est agmine prima' and _Aen_ X 160-61 'Pallas ... sinistro / adfixus lateri'. It is possible that _latus_ here means 'companion', as at Martial VI lxviii 4 'Eutychos ille, tuum, Castrice, dulce latus'.