=15. AD SVMMAM= means 'in short' or 'to sum up', and is used to introduce a recapitulation of what has just been expressed or concluded. The line should therefore be taken as the end of a debate which Ovid has had with himself. For the idiom, Ehwald (_KB_ 45) cites Cic _Att_ VII vii 7, XIV i 1, Hor _Ep_ I i 106 'ad summam, sapiens uno minor est Ioue, Petronius _Sat_ 37 5 'ad summam, mero meridie si dixerit illi tenebras esse, credet', 37 10, 57 3 & 9, 58 8 (in all these passages the narrator's neighbour at table is the speaker) and 71 1 (Trimalchio speaking).

Professor R. J. Tarrant cites Sen _Apoc_ 11 3 'ad summam, tria uerba cito dicat et seruum me ducat'.

AD SVMMVM is the reading of _L_ and _T_ and is printed by Burman (who punctuates _uiderit ad summum_) and Merkel (_ad summum dixi_). _OLD summus_ 8b gives only one instance of _ad summum_, where it means 'at most' (Scribonius Largus 122). The phrase does not seem appropriate to the present context.

=15. IPSE= (_FTP_) is so much better in sense ('although _he_ may object') than the ILLE of most manuscripts that I have followed all previous editors in accepting it.

=16. HANC.= This, the reading of _H_ and _I_ (perhaps recovered by conjecture), must be preferred to HA (AH, A), the reading of the other manuscripts, since without it _licet ipse queratur_ would have to be linked to _uiderit_, which seems awkward. The corruption of _hAc_ to _ha_ is not difficult, especially in view of the following _pudet_; compare _Met_ IX 531 'pudet, a pudet edere nomen'.

=17. SI QVID EA EST.= 'If it really exists'. The affirmation would be 'est aliquid Lethe'; compare Prop IV vii 1 'Sunt aliquid Manes: letum non omnia finit'.

=17. HEBETANTEM PECTORA.= I have found no other instance in Ovid of this transferred sense of _hebetare_, but compare _Aen_ II 604-6 'omnem quae nunc obducta tuenti / mortalis hebetat uisus tibi ... nubem eripiam' and _Aen_ VI 731-32. The transferred sense is found at Celsus II i 11 'Auster aures hebetat ... omnis calor ... mentem hebetat'; compare as well Pliny _NH_ XVIII 118 '[faba ...] hebetare sensus existimata' and Suet _Cl_ 2 'animo simul et corpore hebetato'.

_Oblitus_ in 18 indicates that _pectus_ is virtually equivalent to 'mind' or even 'memory'. In Ovid it often has the sense 'poetic feeling', as at xii 16 'pectus habere neger'.

=17. LETHEN.= Compare _Tr_ IV i 47-48 'utque soporiferae biberem si pocula Lethes, / temporis aduersi sic mihi sensus abest'.

=21. ET= can be construed, as connecting with the preceding _nec_; compare _Fast_ VI 325 '_nec_ licet _et_ longum est epulas narrare deorum'. SED should however possibly be read, the word contrasting with the preceding _nec_ as at ii 15-16 'nec tamen ingenium nobis respondet ut ante, / _sed_ siccum sterili uomere litus aro'. The error could easily be induced by the final _s_ of the preceding _putes_; compare _Med_ 55-56 'par erui mensura decem madefiat ab _ouis_ / (_sed_ [_uar_ et] cumulent libras hordea nuda duas)'.

=21. LEVIS HAEC ... GRATIA.= 'This unimportant expression of gratitude'.

The same use of _leuis_ at _EP_ II v 35-36 'hoc tibi facturo, uel si non ipse rogarem, / accedat cumulus gratia nostra leuis'.

=21. HAEC MERITIS REFERATVR GRATIA.= Similar phrasing at _Met_ V 14-15 'meritisne haec gratia tantis / redditur?', _Tr_ V iv 47 'plena tot ac tantis referetur gratia factis', _EP_ I vii 61 'emeritis referenda est gratia semper', and _EP_ III i 79-80 'nec ... debetur meritis gratia nulla meis'.

=23. NVMQVAM PIGRA FVIT NOSTRIS TVA GRATIA REBVS.= Wheeler rightly points out Ovid's play in 21-23 on the varying senses of _gratia_ (thanks), _gratus_ (grateful), and _gratia_ (favour, kindness).

=26. FERETQVE= is Heinsius' correction for the REFERTQVE of the manuscripts (REFERT _B1_, REFERTA _C_); it is made necessary by the following _fiducia tanta futuri_. Owen, Lenz, and Andre report _feretque_ as the reading of the thirteenth-century _Canonicianus lat 1_, but Professor R. J. Tarrant, who has examined the manuscript, informs me that it in fact reads _refertque_.

For the pattern compare _Tr_ III viii 12 'quae non ulla tibi _fertque feretque_ dies' and _Tr_ II 155-56 'per superos ... qui _dant_ tibi longa _dabuntque_ / tempora'.

The corruption was natural enough, particularly in view of such passages as _Fast_ VI 334 'errantes _fertque refertque_ pedes', _Tr_ I vii 5-6 (to a friend who owned a ring with Ovid's portrait) 'hoc tibi ... senti ... dici, / in digito qui me _fersque refersque_ [_codd_: ferasque _Heinsius_] tuo', and _Tr_ V xiii 29 'sic _ferat ac referat_ tacitas nunc littera uoces'.

=28. QVOD FECIT QVISQVE TVETVR OPVS.= 'Everyone protects the work he has created'. This is hardly a commonplace of ancient poetry, and the catalogue which follows of famous works of art does not serve to illustrate it.

=29-34.= Ovid's description of the works of Apelles, Phidias, Calamis, and Myron was influenced by Propertius' catalogue of artists at III ix 9-16; in particular, he imitates 10-12 'exactis Calamis se mihi iactat equis; / in Veneris tabula summam sibi poscit Apelles; / Parrhasius parua uindicat arte locum', and 15 'Phidiacus signo se Iuppiter ornat eburno'.

Professor E. Fantham points out to me the inclusion of Apelles, Calamis, and Myron as canonical figures in a catalogue of artists at Cic _Brut_ 70 and of all four in a similar catalogue at Quint XII x 6-9.

=29. VENVS.= Ovid is speaking of the famous Aphrodite Anadyomene painted by Apelles (fourth century BC) in Cos; hence the epithet _Coi_ later in the line--Apelles was in fact from Colophon. Ovid had probably seen the picture in Rome, for Augustus brought it there from Cos (Strabo XIV 2 19; Pliny _NH_ XXXV 91).

Ovid refers to the painting at _Am_ I xiv 33-34 and _Tr_ II 527-28. At _AA_ III 223-24 (quoted in the next note) Ovid seems to be describing a cut gem copied from the painting.

=30. AEQVOREO MADIDAS QVAE PREMIT IMBRE COMAS.= _Imbre_ depends on _madidas_. _Premit_ is equivalent to _exprimit_, as is shown by _AA_ III 224 'nuda Venus madidas _exprimit_ imbre comas'. For _exprimere_ taking as object that out of which something is pressed or squeezed see Celsus IV 24 and Pliny _NH_ XXIX 31.

The Romans would not have found _aequoreo ... imbre_ strange. Although the primary transferred sense of _imber_ would be rain-water, it is used of sea-water as early as Ennius _Ann_ 497-98 Vahlen 'ratibusque fremebat / imber Neptuni', and without defining qualifier at _Aen_ I 123.

=31. ACTAEAE= = the metrically difficult _Atheniensis_. The word is generally confined to high poetry (_Ecl_ II 24, _Met_ II 554 & 720, VI 711, VII 681 & VIII 170), but its first occurrence is in prose, at Nepos _Thras_ 2 1 'hoc initium fuit salutis Actaeorum'; some manuscripts read _Atticorum_, which may be right.

=31. VEL EBVRNA VEL AEREA CVSTOS.= There were at Athens two famous statues of Athena sculpted by Phidias: 'Phidias ... fecit ex _ebore auroque_ [_Mayhoff_: aeque _codd_] Mineruam Athenis quae est in Parthenone stans, ex _aere_ uero ... Mineruam tam eximiae pulchritudinis ut formae cognomen acceperit ['was named the Minerva Formosa']' (Pliny _NH_ XXXIV 54); the second, less famous statue is described at Pausanias I 28 2.

Heinsius' note is something of an oddity. He begins by reading AVREA for the AENEA of most manuscripts, taking _uel eburna uel aurea custos_ to refer to the chryselephantine statue in the Parthenon, 'sed altius consideranti locum apparet de duplici statua Mineruae agi, altera eburnea, altera aenea'. _Aenea_ therefore continued to be the accepted reading until 1873, when Haupt (_Opuscula_ 584) pointed out that it was unmetrical, and restored _aerea_, found in some manuscripts.

The inverse error occurs at _Her_ VI 32, where most manuscripts have the unmetrical _aeripedes_ for _aenipedes_. But Merkel, followed by Palmer, considered 31-38 an interpolation; and _aeripedes_ may have been what the interpolator wrote.

=32. PHIDIACA ... MANV.= Ovid is recalling Prop III ix 15 'Phidiacus ... Iuppiter'. For the Latin poets' use of a personal adjective for the genitive of the noun, see Austin's interesting note on _Aen_ II 543 _Hectoreum_.

=33. VINDICAT VT CALAMIS LAVDEM QVOS FECIT EQVORVM.= 'As Calamis lays claim to the praise given his horses'. Calamis, a sculptor of the fifth century BC, was particularly famous for his statues of horses; see Pliny _NH_ XXXIV 71 'habet simulacrum et benignitas eius ['Praxiteles'

generosity is seen in one of his statues']; Calamidis enim quadrigae aurigam suum imposuit, ne melior in equorum effigie defecisse in homine crederetur. ipse Calamis et alias quadrigas bigasque fecit equis sine aemulo expressis'.

=33. QVOS FECIT EQVORVM.= Similar instances of hyperbaton at 28 'quod fecit quisque tuetur opus', _Met_ IV 803 'pectore in aduerso quos fecit sustinet angues', and _Fast_ VI 20 'tum dea quos fecit sustulit ipsa metus'.

=34. VT SIMILIS VERAE VACCA MYRONIS OPVS.= The _Cow_ of Myron (late fifth century BC) was his most famous work. Praise of the statue's lifelike appearance was a stock theme of Hellenistic writers of epigram; it appears from Pliny _NH_ XXXIV 57 that the poetry written about the statue was as notable as the statue itself. Thirty-six poems of the Palatine Anthology deal with the theme (IX 713-42 & 793-98). Ausonius wrote eight epigrams on the same subject (_Ep_ LXVIII-LXXV), of which I quote LXVIII as a typical example of what both the Greek and Latin epigrams are like:

Bucula sum, caelo ['chisel'] genitoris facta Myronis aerea: nec factam me puto, sed genitam, sic me taurus init, sic proxima bucula mugit, sic uitulus sitiens ubera nostra petit.

miraris quod fallo gregem? gregis ipse magister inter pascentes me numerare solet.

The statue was in Athens during Cicero's lifetime (_II Verr_ IV 135); Ovid is likely to have seen it during his visit to the city (_Tr_ I ii 77). He would certainly have seen the four statues of cattle sculpted by Myron which Augustus placed in his temple of Apollo, and which Propertius described: 'atque aram circum steterant armenta Myronis, / quattuor artificis, uiuida signa, boues' (II xxxi 7-8).

=35. VLTIMA.= 'Smallest, least important'. For this rare sense compare Hor _Ep_ I xvii 35 'principibus placuisse uiris non ultima laus est', _Cons ad Liuiam_ 44 'ultima sit laudes inter ut illa tuas', Vell Pat I 11 1, and the other instances cited by _OLD ultimus_ 9.

=35. SVM= ('I am not the least of your possessions') seems unobjectionable enough; most editors have, however, accepted PARS from the _excerpta Politiani_.

=36. MVNVS OPVSQVE= is a Latin phrase with the general meaning of 'creation'. It is used in this sense at Cic _Tusc_ I 70 'haec igitur et alia innumerabilia cum cernimus, possumusne dubitare quin iis praesit aliquis uel effector ... uel ... moderator tanti _operis et muneris_?', _ND_ II 90, _Off_ III 4 'nulla enim eius ingenii [_sc_ Africani]

monumenta mandata litteris, nullum _opus_ otii, nullum solitudinis _munus_ extat', and _Met_ VII 435-36 (to Theseus) 'quodque suis securus arat Cromyona colonus, / _munus opusque_ tuum est'.

II. To Cornelius Severus

Cornelius Severus (Schanz-Hosius 268-69 [-- 317]) was one of the most famous poets contemporary with Ovid; of him Quintilian said 'etiam si uersificator quam poeta melior ['even if his facility outruns his inspiration'], si tamen (ut est dictum) ad exemplar primi libri bellum Siculum perscripsisset, uindicaret sibi iure secundum locum [_sc_ after Virgil]' (X i 89). The elder Seneca quoted with approval Severus' lines on the death of Cicero, as the finest lament produced on the subject (_Suas_ VI 26: Winterbottom _ad loc_ refers to a commentary by H.

Homeyer, _Annales univ. Saraviensis [phil. Fak.]_ 10 [1961], 327-34).

_EP_ I viii was addressed to a different Severus: in the third and fourth lines of the present poem, Ovid expresses his embarrassment at having addressed no poem to Severus previously, and in the earlier poem no mention is made of the addressee's poetry.

The poem is an apology to Severus for Ovid's not having sent a poem to him before; he offers two excuses for the omission. In the first fourteen lines, he flatters Severus by saying that so good a poet hardly needs to receive verse from someone else; in the twenty-four lines that follow he describes how his poetry, because of the conditions at Tomis, is now less abundant and of poorer quality than before. The subject is one Ovid had employed before: _Tr_ III xiv, a request for indulgence to Ovid's verse, and _Tr_ V xii, a reply to a friend who had urged him to write more poetry, treat the same topic in much the same way. The theme is similar to that of Catullus LXVIII 1-40, where the poet explains that his brother's death has caused his lack of interest in poetry.

In 39-46 Ovid moves to the somewhat discordant topic (which serves however to re-emphasize his misery at Tomis) of how he continues to write poetry to take his mind off present evils, a theme he had used several times before, most notably in _EP_ I v. He ends the poem with a request that Severus send him some of his recent work (47-50).

=1. QVOD LEGIS.= Similar beginnings to verse-epistles at _Her_ III 1 '_Quam legis_ a rapta Briseide littera uenit', _Tr_ V vii 1, _EP_ I vii 1-2 'Littera pro uerbis tibi, Messaline, salutem / _quam legis_ a saeuis attulit usque Getis', and _EP_ III v 1 '_Quam legis_ unde tibi mittatur epistula quaeris?'.

Compare as well _Her_ X 3-4 '_Quae legis_ ex illo, Theseu, tibi litore mitto / unde tuam sine me uela tulere ratem'. This poem has suffered from two separate interpolations at its beginning. Certain manuscripts start the poem with the distich 'Illa relicta feris etiam nunc, improbe Theseu, / uiuit et haec aequa mente tulisse uelis', which is universally condemned; but the formulaic nature of 3-4 suggests that 1-2 'Mitius inueni quam te genus omne ferarum, / credita non ulli quam tibi peius eram', found in all manuscripts, is a second interpolation. Micyllus was the first to see this; a recent discussion at Kirfel 69-70.

=1. VATES MAGNORVM MAXIME REGVM.= Severus apparently wrote a poem dealing with pre-Republican Rome, to judge from xvi 9 his most famous work: 'quique dedit Latio carmen regale, Seuerus'. Heinsius took the two passages as meaning that Severus was a writer of tragedy, citing _Tr_ II 553 'et dedimus tragicis scriptum regale cothurnis'; compare as well Hor _Sat_ I x 42-43 'Pollio regum / facta canit pede ter percusso ['in iambic trimeter']'. Heinsius' suggestion is possible enough, but since Seneca and Quintilian speak of Severus as an epic poet and there is no mention of the stage in this poem, it should be rejected.

Similar language is used of epic poetry at _Ecl_ VI 3 'cum canerem _reges_ et proelia' and Prop III iii 1-4 'Visus eram ... reges, Alba, tuos et _regum facta_ tuorum, / tantum operis, neruis hiscere posse meis'.

=1. REGVM.= VATVM (_M1FIL_) is a conscious or unconscious attempt to extend the etymological figure seen in _magnorum maxime_.

=5-6. ORBA TAMEN NVMERIS CESSAVIT EPISTVLA NVMQVAM / IRE PER ALTERNAS OFFICIOSA VICES.= Other mentions of what was clearly an extensive prose correspondence between Ovid and his friends at _Tr_ V xii 1-2 and _EP_ I ix 1-2.