PA 
 
 6519 
 
 A73 
 
 E44 
 
 1912 
 
 MAIN 
 
 B M D3b 307 
 

 The Amores of Ovid 
 
 A LECTURE 
 
 DELIVERED IN THE HALL OF CORPUS CHRISTI COLLEGE 
 ON TUESDAY, JUNE ii, 1912 
 
 BY 
 
 ROBINSON ELLIS, M.A. 
 
 CORPUS PROFESSOR OF LATIN I.ITF.RATl'RE 
 
 LONDON 
 
 HENRY FROWDE 
 
 OXFORD UNIVERSITY PRESS, AMEN CORNER 
 
 OXFORD: 116 High Street 
 
 Price One Ski/line net 
 
The Amores of Ovid 
 
 A LECTURE 
 
 DELIVERED IN THE HALL OF CORPUS CHRISTI COLLEGE 
 ON TUESDAY, JUNE ii, 1912 
 
 BY 
 
 ROBINSON ELLIS, M.A. 
 
 CORPUS PROFESSOR OF LATIN LITERATURE 
 
 LONDON 
 
 HENRY FROWDE 
 
 OXFORD UNIVERSITY PRESS, AMEN CORNER 
 
 OXFORD: 116 High Street 
 
OXFORD : HORACE HART 
 PRINTER TO THE UNIVERSITY 
 

 THE AMORES OF OVID 
 
 Brandt's recent edition of Ovid's Arnores forms a fittinfj 
 sequel to his edition of the Ars Amatoria. Critically it 
 follows closely in the steps of Ehwald and Martinon, 
 acceptinf; the readings of P (the Puteaneus, Paris, H242), 
 and *S' (Sangallensis, 824). This, considering the immense 
 number of MSS. of the Amures which have never been 
 examined, and the boundless field for new discovery which 
 lies before the competent explorer, is not as satisfactory to 
 the researcher of the XXth century as might be wished. 
 Modern research has made no little advance in the 
 valuation of MSS., in assigning dates to them, in discrimina- 
 ting the valuable and uncorrupted from the worthless and 
 interpolated, as may be seen by comparing the edition of 
 Nicolas Heinsius with those of the present day. 
 
 Yet Heinsius had great advantages ; if a rare or early 
 codex existed in any great collector's library, the intelligence 
 was quickly conveyed to the ears of a savant who knew not 
 only the determinants of age and goodness, but was likely 
 to use every effort to secure it for the library of the Queen, ^ 
 Now the case is altered materially : kings care less for 
 accumulating MSS., while on the other hand the world in 
 general is more and more interested in them. 
 
 The arrangements for the convenience of students of MSS. 
 are infinitely more commodious in our time than in the 
 XVIIth century ; the function of the collator is respected 
 and everything is done to make his task lighter, whether 
 by larger and better lighted apartments (as in the Vatican 
 under the judicious rule of Pope Leo XIII), or by 
 increasing the number of learned referees who may be 
 consulted on points of difficulty, or by the greatly increased 
 number of printed catalogues giving a detailed list of the 
 
 ' Clm^tina o^ Svvexlen, 
 
works contained in each MS. with the page where each work 
 begins and ends. Such are the Catalogues of the MSS. in 
 the Bodleian, prepared by Mr. H. O. Coxe and his assistants. 
 
 The trouble, I repeat, of the researcher in MSS. is now 
 so much reduced as to make it, if the codex is not exorbi- 
 tantly difficult or obscure, a pleasant and even delightful 
 occupation. 
 
 Why is it that in constituting a text of the two elegiac 
 poems of Ovid, which, if any, deserved to give him, and did 
 give him, immortality, so little advance has been made 
 since the time of Heinsius ? Of the MSS. of the finest of 
 his works, the ATYiores, and the most famous, the Ars 
 Amatoria, we have a thoroughly uncorrupted copy in P ; 
 Heinsius called it his sacra ancora, and all subsequent 
 students of the poet's amatory works, Keil, Merkel, 
 Palmer, Ehwald, Martinon, agree with his verdict. Yet 
 even now we have no perfect conspectus of its readings in 
 either Am. or A. A., although Palmer's diligent collation 
 of the Heroides seems to leave little for future critics to 
 explore. I suppose this most unfortunate accident is due 
 to the more licentious character of the Arnores and Ars, 
 for surely no sane critic could rank the Heroides with 
 them as poetry. 
 
 But this is the very reason why a new and complete 
 collation of P in these two works is imperatively required. 
 The Aviores is, as poetry, by far the finest effort of Ovid's 
 genius. In technic (it is true) the Ars is its equal, but not 
 in the quality which is most required in an erotic poet — 
 tender feeling combined with lofty imagination. 
 
 This emotional and imaginative quality seems to exist in 
 the Amores in a higher degree than in any other work of 
 Ovid's; witness the elegy on the death of Tibullus, the 
 Pari'ot, the Complaint to Aurora. And in such proportion 
 as its imaginative quality is finer, we have reason to 
 lament that our information as to its condition in MSS. is 
 still very imperfect. There is indeed no absolute consensus 
 as to what MS. is of sufficiently commanding goodness 
 to be ranked with or next to 1'. It is not often that a 
 
problem of a palaeo^raphical kind is at once so interesting 
 and so repellent ; interesting, because the three books 
 (originally five) of the Amorea must have been written 
 while the poet was still at the height of his genius ; re- 
 pellent, because the critical insight required for eliminating 
 the larger number of the MSS. and confining the attention 
 to the really important ones is a rare gift, and only comes 
 after long practice, and handling of many codices. Some 
 years ago, in reviewing Brandt's Ars Amatoria, I complained 
 that he had made little or no use of the one earliest copy 
 of any of the elegiac poems of Ovid, the first book of the 
 Ars Amatoria (Bodl. Auct. F. iv, 32), of which I have given 
 some photographed facsimiles in the various books of speci- 
 mens which have been executed b}' the Oxford Press for the 
 use of my palaeographical pupils. The date of this MS. is not 
 certainly known,but one of the most skilful judi^es, especially 
 in the case of early Wehh script, Mr. Bradshaw, formerly 
 librarian of the Public Library at Cambridge, assigned it to 
 the latter half of the IXth century (Hermes, xv, pp. 427- 
 430). And no MS. of any complete portion of Ovid's works 
 is ascertained to be earlier than this. 
 
 Yet even of this MS., which was known to Heinsius, 
 I know no entire collation except my own, published in 
 Hermes, xv. Of the Amores no codex of equal antiquity 
 is known to exist ; but the critical problems which it 
 presents are even more perplexing than most of those in 
 the Heroides or Ai's Amatoria. I will give one signal 
 case, about which a far fuller examination of the MSS. of 
 the Amores can alone guide us to a reasonable conclusion. 
 
 The Xlllth elegy of Book I introduces the poet com- 
 plaining of the cruelty of Aurora in parting lovers from each 
 other. The first two lines are written in F (the Puteaneus) 
 and S (Sangallensis) : — 
 
 lam super oceanum uenit seniore marito 
 Flaua pruinoso quae uehit axe diem. 
 
 Heinsius, retaining neait as perfect, propo.sed to write 
 relicto for marito, rejecting the ordinary view that the 
 
preposition a has fallen out before seniore. Though 
 found both in P and S, the MS. reading cannot be right, 
 albeit accepted by Ehwald, Martinon, and Brandt. For, 
 surely the present, not the perfect, is required. Morn is 
 approaching, and its arrival deprecated on behalf of 
 travellers, soldiers,ploughmen,school-boys,lawyers, pleaders, 
 spinning-maidens, none of whom rise to their daily tasks 
 with pleasure. This section of the elegy occupies (5-26) 
 H2 lines ; then follows the poet's own praj^er, that Night 
 might overpower Day, the Stars outshine the Dawn, or 
 violent blast wreck the chariot of the Goddess. Then, 
 the inevitable mythological allusions, Memnon, Tithonus, 
 Cephalus, the former Aurora's husband, the latter her 
 paramour, and Endymion the favourite of the Moon (27- 
 48). In this way a total of 44 lines intervenes between the 
 beginning and end of the elegy — 
 
 lurgia finieram, scires audisse, rubebat 
 Nee tamen adsueto tardius orta dies, 
 
 and in this space of time the full da}^ we may suppose, 
 has risen, and the coming in of Morn is completed. Not 
 less in support of the pi'esent uenit not the perfect lie nit 
 is the emphatic mane in v. 3, 'Quo properas Aurora? 
 'mane ' : for Aurora would not be hurrying on if she was 
 already risen, and the poet would not conjure her to sta}^ 
 unless some time elapsed before she might be expected 
 to arise. 
 
 I must not leave this elegy without calling attention to 
 the extraordinary interpolation after 30 — 
 
 Quid, si Cephalic numquam flagraret amore ? 
 An putat ignotam nequitiam esse suam 1 
 
 This distich is not found in S, or, except by the second 
 hand, in P ; it would be worth while to ascertain by 
 a large examination of ]\rSS. in how many it occurs, and at 
 wiiat period it may reasonably bo thought to have been 
 introduced. The distich, like otliers in the works of Ovid, 
 and in particular those in tlu^ //>/s, seems to have been 
 
suggested by the two which the poet hunself wrote in their 
 proper place in the elegy (vv. 39, 40) — 
 
 At si, quem mauis, Cephalum complexa tencres, 
 Clamares : ' lente currite, Noctis equi,' 
 
 and may have been prompted by some unfortunate prudery 
 to which any such open expression of the Goddess's felicity 
 in the embrace of her mortal lover might appear shocking. 
 
 Of quite another stamp of interpolation is the 5th elegy 
 of Book III, Xox erat et i>omiiuti lassos submisit ocellos. The 
 poem contains 46 verses, and describes a nocturnal vision 
 or dream in which the poet is forewarned of an approach- 
 ing betrayal by Corinna. This elegy was thought spurious 
 by Heinsius and in our own day by Lucian Miiller. It is 
 also rejected by Merkel and Martinon, the latter of whom 
 calls it ' cette mauvaise ^^gie '. It must however have 
 got into some copy of the Aviores very early, since it is 
 quoted as Ovid's in Servius' (or perhaps pseudo-Servius') 
 commentary on Eel. vi. 54 ' Pallentes euminat herbas 
 reuomit et denuo consumit. Atque iterum pasto pascitur 
 ante cibo, sic Ouidius.' In this respect it stands on the 
 same footing with the verse Sisque viiser sernper nee sis 
 ni'uerahil'is uUi (Ibis 117) which is cited as Ovidian by the 
 grammarian Eut^'ches and is imitated by the Christian 
 poet Orientius Ille rtiiser uere nee erit miserabilis uUi, 
 and was obviously considered to belong to a genuine work 
 of Ovid. 
 
 I agree with almost all the modern critics of the 
 Amoves in believing the elegy (iii. 5) to be spurious. So 
 refined a writer would hardly have omitted the personal 
 pronoun which defined Ocellos as the eyes of the poet. 
 Returning to v. 18 ' Atque iterum pasto pascitur ante cibo,' 
 we are on safer ground. The verse is not without clever- 
 ness, but is not up to the mark of Ovid. The writer is 
 defining the meaning of reuomit, ' and feeds a second time 
 on food fed on before '. Here we may feel sure that our 
 inimitable poet would not have constituted his verse in 
 such a way as to suggest at the first hearing that iterur,i 
 
8 
 
 is to be connected with pasto which really belongs to ante. 
 Look again at this distich (13, 14), in which the white cow 
 is compared with milk just fresh from the sheep's udder : 
 
 Candidior quod adhuc spumis stridentibus albet 
 Et modo siccatam lacte reliquit ovem. 
 
 Allowing, what it is difficult to admit, that Ovid would 
 have made the rather grotesque comparison of the white 
 skin of a heifer with the white of a sheep's milk, who can 
 believe that he would have used lacte either as archaic 
 nominative or as ablative strangely interjected in a clause 
 to which it does not belong, and in which it produces 
 a bizarre effect ? 
 
 On the whole, this Elegy presents none of those peculiar 
 graces which mark the rest of the Amoves, and are equally 
 conspicuous in the Eeroides and Ars Amatoria. 
 
 It might therefore be expected that it would be omitted 
 by some at least of the better MSS. It is found, however, 
 both in P and >S' ; and, which is more to the purpose, as 
 was noticed above, it is quoted as by Ovid in Servius' 
 commentary on Vergil. It follows that it must have 
 formed part of the tradition of the Amoves at a very early 
 time. It is not impossible that some future collator of the 
 MSS. of the poem may find a codex equal in antiquity or 
 even superior to the Puteaneus, and which may either omit 
 the Elegy entirely or offer some note by which its presence 
 may be explained. I have above stated that the Bodleian 
 Library possesses a Welsh MS. of Avs Amatovia i, which 
 Henry Bradshaw, a great authority on everything Welsh, 
 ' believed to be as early as century ix.' 
 
 This MS., of which I have published a collation, contains 
 glosses explaining many of the Latin words in early Welsh, 
 others in Latin. 
 
 Some such MS. of the Amoves may yet await a discoverer 
 such as Mr. Winstedt, to whom we are indebted for a frag- 
 ment of the original draft of Juvenal's Sixth Satire, not 
 known to exist in any other Codex. Or some verses of the 
 Elegy may exist as an isolated fragment, just as 18 lines oT 
 
9 
 
 the Epistles from Pontus, in a script of the 6th or 7th 
 century, survive in an old edition of the Aloralia of Nicolas 
 de Lyra at Wolfenbiittel, or as our own Bodleian possesses 
 fragments of early and good codices of Cicero's Tunculan 
 Disjmtations, Sallust's lugurtha, Statius' Thehais. 
 
 It will be seen by any one who takes the trouble to 
 analyse the three books of Amoret< that by far the larger 
 portion of the poem is occupied with the subject of love 
 exclusively. Even those which (like iii. 6) are only remotely 
 amatory, and where little or no mention is made of Corinna, 
 are more or less closely woven into the general tissue. In 
 the Elegy (iii. 6), which begins with this distich 
 
 Amnis harundinibus limosas obsite ripas, 
 Ad dominam proper© , siste parumper aquas, 
 
 our poet on the way to Corinna encounters a turbid 
 river which obstructs his progress. He remonstrates with 
 the stream as playing a role which ill suits its calling. 
 Rivers, says the poet, from the earliest times have always 
 been true to love and lovers. A long list of river names 
 follows, more conspicuous for the dexterity shown in 
 narrating their amours than for any special attractiveness 
 in the legends, often remote, by which the poet has chosen 
 to illustrate his theme. 
 
 Yet from a purely literary point of view^ they are inter- 
 esting as showing how early Ovid's Muse had entered on 
 the line which he afterw^ards pursued more freely in the 
 Metamorphoses and (if it is by him) the Ibis. 
 
 No less interesting is the Elegy (i. 6) addressed to the 
 gate-keeper, who refuses to give the poet admittance to his 
 love. This poem is marked ofi" into sections of 8 lines each 
 by the refrain — 
 
 Tempora noctis eunt, excute poste seram, 
 
 and appears to be unique in Roman Elegy, if we may 
 except the epistle Deianira Herculi. It must not be 
 supposed that all the amatory poems in the volume of 
 Aviores are equally good or at least equally pleasing to 
 a refined taste. This is no doubt true of the larger pro- 
 
10 
 
 portion ; they belong to the period of Ovid's highest 
 inspiration. But there are some where the subject is 
 either offensive, e g. the short but indignant elegy denounc- 
 ing the employment of eunuchs to guard women (ii. 3) : 
 
 Ei mihi quod dominam nee uir nee femina seruas, 
 or grotesque, e.g. the protest against Corinna's dyeing 
 her hair to a Sygambrian red, when its natural black was 
 so much more beautiful. Or lastly, surprising in a work 
 so little serious as the Amoves, e.g. the cai-efully worded 
 and difficult verses in which objection is taken to the practice 
 of abortion. Poems sucli as these, it cannot be denied, are 
 displeasing, or even repulsive to a modern ear, but there can 
 be little doubt that they added to the poet's reputation at 
 the time, and, with two or three sufficiently notorious ex- 
 ceptions, they do not offend on the score of prurient or over- 
 sensual description. 
 
 Sometimes we may even admit they heighten by contrast 
 the beauty of the surrounding elegies. Probably to most 
 readers the poems on the Parrot and on The Death of 
 Tlbvllus will seem the real gems of the collection. It is 
 in these, at any rate, that Ovid has shown most of the 
 pathetic quality which is inseparable from the highest poetry. 
 The Parrot in particular stands out in the list of bird- 
 poems as unrivalled in its combination of ingenious fancy 
 with genuine unaffected grief. Ovid's verses on the Parrot s 
 death equal or perhaps surpass Catullus' hendecasyllables 
 on the Sparrow of Lesbia, and far outshine the hexameters 
 of Statins on a similar subject. 
 
 In the course of the ATUores, Ovid takes occasion to pre- 
 dict his own immortality (i. 15. 41, 2) — 
 
 Ergo etiam cum me supremus adederit ignis, 
 Viuam parsque mei multa superstes erit. 
 and again (iii. 15. 19, 20) — 
 
 Imbelles elegi, genialis Musa, ualete 
 
 Post mea mansurum fata superstes opUvS. 
 And his prediction has proved true, though some of tlie 
 poets for whom he prophesied a similar immortality have 
 
11 
 
 been wholly or pitrtiully lost — most of Menaiider, the hulk 
 of Ciilliniachus' Airia, Eiuiius, Accius, Varro of Atax, and 
 Gallus. Nor is it at all difficult to discern the reason of 
 Ovid's self-confidence. The Amores is not like the Heroi(lef<, 
 a clever presentment of an impossible situation, in wliich 
 mythical heroines write love-letters to equally mythical 
 heroes ; nor is it, like the Ars Amatoria, a manual of love- 
 making in verse — exquisite verse, it is true— but pursuing 
 a more or less avowedly licentious object. As its name 
 implies, the ^7)io?'e6' exhibits a series of episodes or tableaux 
 in the life of the poet himself; that is to say, actual scenes 
 presenting the various phases of love in its many moods. 
 Corinna is a real woman, and this was essential to the 
 success of the poem ; this, however, makes only a small 
 part of its success. It is in the perfection of its form, both 
 in diction and metre, in the illustration of the main subject 
 by learned legends generally drawn from Greek mythology, 
 sometimes from the annals of ancient Rome, in a hundred 
 felicitous nuances, epigrammatic, rhetorical or even gram- 
 matical, which are purely Ovidian and are found in none 
 of the elegiac poets who preceded him — it is in these that 
 the abiding charm of the Amoves is mainly centred. I select 
 the following as among the more favourable examples : — 
 
 Qualis Amymone siccis errauit in agris 
 Cum premeret summi uerticis urna comas, 
 
 Talis eras, aquilamque in te taurumque timebam. 
 
 In the same elegy (i. lo) : 
 
 Nunc timor omnis abest, animique resanuit error. 
 Again : 
 
 Nee Venus apta feris, Veneris nee filius arniis. 
 Or: 
 
 Inque suos fontes uersa recurrit aqua. (ii. i. 26.) 
 The last three words with recurret for recurrit are found 
 also in Nero ides v. 30. 
 
 Again : 
 
 Odi nee possum cupiens non esse, quod odi. (ii. 4. 5.) 
 
12 
 
 Haec tibi sunt mecum, mihi sunt communia tecum. 
 
 (ii- 5-3^') 
 Quid iuuat in nudis hamata retundere tela 
 Ossibus ? ossa mihi nuda relinquit amor. 
 
 (ii. 9. 13, 14.) 
 Indeserta meo pectore regna gere. (ii. 9. 52.) 
 Nauita sollicitus cum uentos horret iniquos, 
 Et prope tam letum quam prope cernit aquam. 
 
 (ii. II. 25, 26.) 
 An admirable specimen of Ovid's superlative art. 
 
 O utinam fieri subito mea munera possem. (ii. 15. 9.) 
 Quod sequitur fugio; quod fugit ipse sequor. 
 
 (ii. 19. s6.) 
 Victima deceptus decipientis ero? (iii. 3. 22.) 
 Cedere iussit aquam : iussa recessit aqua. (iii. 6. 44.) 
 Ipse tuus custos, ipse uir, ipse comes, (iii. 11. 18.) 
 Dicta erat aegra mihi : praeceps amensque cucurri ; 
 Veni, et riuali non erat aegra meo. (iii. 11. 25, 26.) 
 No one, I imagine, can read even one of the Elegies 
 without realizing the extraordinary beauty of the language, 
 so lucid, so tender, so inexhaustible, so Roman, without 
 acknowledging in its author a poet of unique genius, born 
 to arrest the attention of his own time and to command 
 the admiration of all future time. 
 
 Much indeed of his fame as an amatory poet belongs to 
 periods of which even the most accomplished scholars 
 know little, and which require the learning of a Gaston 
 Paris or a Paul Meyer for their adequate illustration. 
 Eternal is the debt which the Middle Age, its Minnesingers 
 and Courts of Love owe to Ovid, and almost to Ovid alone. 
 Catullus escaped oblivion by the accidental preservation of 
 a single MS. ; Propertius and Tibullus were copied by few ; 
 but the amatory works of Ovid were preserved in multiplied 
 npographa and were studied or imitated by an innumerable 
 company of French and Proven(;al singers. In connexion 
 with this voluminous topic I may mention Schrotter, Ovid 
 und die Troubadours^ Halle, 1908. 
 
NOTES 
 
 (to p. I I) 
 
 Amonnr the specialities of metre which Ovid introduces 
 in tlie u47HO;'e,s are (i) the invariable disyllable at the end 
 of the pentameter, ajjainst the use of Propertius and (in 
 a smaller degree) of Tibullus and Lygdanius ; (2) the much 
 more frequent use of short monosyllables like dat, uir, 
 nee in the composition of the dactyl, e.or. Spemque dat, ii, 
 4. 14 ; Vnde uir, ii. 2. 48 ; Dure uir, iii. 4. i ; Qui quod 
 amat, ii. 5. 9 ; Quique tot errmdo, ii. i. 3 1 ; (3) the occasional 
 admission of a word at the end of a pentameter in which 
 the second syllable is a short vowel, e.g. aqua, ii. i. 26; 
 (4) immense care in avoiding any but the lightest elision. 
 So scrupulous is Ovid in this paiiicular as almost to make 
 it questionable whether the well-known narrabam jiumi- 
 iiiiin amoves, iii. 6. 101, is genuine. 
 
 Two Adjkctives of colour in the Amores. 
 
 There are in the Amores no less than two words of 
 colour about which doubts have been raised which cannot 
 yet be thought to be settled finally. 
 
 One of them is purpureas as an epithet of Amor. 
 Carmina, purpureus quae mihi dictat Amor. 
 
 (ii. I. 38.) 
 Notaque purpureus tela resumit Amor. (ii. 9. 34.) 
 
 Brandt adds another passage from A. A. — 
 
 Purpureus Bacchi cornua pressit Amor (i. 232.) 
 
 and compares Phrynichus' Tropcpvperj^^ napfja-L 0co9 epcoTo?. 
 
 The scholia of Acron on Horace, C. iv. i. 10, where the 
 swans of Venus are called purple — purpureis nitidis aut 
 pulchris aut reginae Veneri dicatis ut pro regno pur- 
 2)ureos dixerit — show how uncertain were the early 
 expositors as to the meaning of the word. 
 
 The last of these explanations, viz. that the swans are 
 called purple, as connoting Venus the supreme arbiter of 
 Love, is worth something as an idea, but can hardly be right. 
 
 There must be a definite notion of colour in the word, as 
 distinctly as in Phrynichus' purple or purpureal cheeks, and 
 Gray's ' The bloom of young desire and purple light of love.' 
 
 The precise idea which Horace meant to convey is perhaps 
 not recoverable, but some form of colour we may suppose 
 to be signified, probably in the wings, similar to the tinted 
 
34 
 
 wings of Amor in Catalept. xiv. 9, if the right reading 
 there, which however is quite uncertain, is 
 
 ignicolorius alas 
 In morem picta stabit Amor pharetra. 
 
 That Ovid had in his mind some such conception of 
 Amor, i. e. tinted with a suffusion of pink or roseate 
 colour, seems more than probable. At any rate the thrice- 
 recurring epithet piirpureus, and the fact that the Amoves 
 may almost be marked off from the other elegiac works of 
 Ovid by this combination of ptt^'/^urewy with Amor, points, 
 1 think, to some such conclusion. If the earlier distribu- 
 tion of the Amoves into 5 instead of 3 books had been 
 preserved in any copy, we should be better able to speak 
 with confidence about this; we might then decide whether, 
 as is my suspicion, the verse in which Ovid speaks of 
 purpureal Love as dictating his verses, was the original 
 upon which the other two — 
 
 Notaque purpureus tela resumit Amor, 
 and 
 
 Purpureus Bacchi cornua pressit Amor — 
 
 were framed. As it is, such a guess is little more than 
 a possible hariolation, although I might urge in support of 
 such a view that in the second and third of the above- 
 quoted instances 'purpureus has no special force; in the 
 first of the three it has : Love's roseate gleam lights up the 
 verses it dictates, and Love becomes henceforward pur- 
 pureal throughout the poem. 
 
 Some such theory seems tenable, at least tentatively. 
 It has the merit of not sacrificing the idea of colour which 
 I believe to be inseparable from the word. The only other 
 view which appears to me possible, is that Ovid has adopted 
 the word as a literary epithet from Greek, but this requires 
 a larger exploration. 
 
 Ovid has another adjective of colour in the combination 
 uirldesque Britannos ii. 16. 39. Brandt explains this 
 from Strabo's description of our island as KardSpv/jLO?, full 
 of trees. This can hardly be right. Surely the reference is 
 to the custom of the ancient Britons to dye their bodies and 
 tlius give them a more alarming look. Caesar B. G. v. 14 
 Omnes uero se Brltaniil uUro infuhiiit, quod caeruleum 
 ejjlcit colorem, (d(]v.e hoc horridiore suiU lit pugna afpectu, 
 
 Caesar's caerideus color nui}^ fairly be taken to express 
 or include the greenish tint. 
 
15 
 
 ^ Amor. i. 8. 65, 66 : 
 
 Nee te decipiant ueteres quinquatria cerae. 
 Tolle tuos tecum pauper ainator auos. 
 
 So the Codex Putcaneus : otlier MSS. have ueteris. 
 
 Naugerius and Marcilius both thouglit that quinque 
 atria was to be read here. ' Fortasse,' wrote the former, 
 ' ut in quibusdam, quinque atria legendum, ut maxime 
 nobilis significetur is, cuius maiorum imagines uel quinque 
 compleant atria.' Caspar Barth [Adiicrt^ar. x. 27) cites 
 a marginal scholion from a MS. of the Aniorea which gives 
 a colour to this view : ' Explodit nobiliorem qui totis atriis 
 imagines habebat dispositas, et se hominem nobilem magni 
 faciebat. Ad talem dicit luuenalis cum tota Carthagine 
 migra.' I see no ground for doubting Earth's truthfulness 
 as to this excerpt : doubtless similar or identical scholia 
 may be found on the passage in some of the innumerable 
 MSS. of the Amores. 
 
 Barth, however, while agreeing wnth the interpretation 
 above given as a whole, maintained that quinquatria should 
 be written as one w^ord. He quotes a gloss -, ' Quinquatria 
 porticus est quinque ambulacrorum,' which is repeated by 
 Papias in this form, ' Quinquatria quinque porticorum am- 
 bitus. ' Many years before my attention was called to the 
 difficulty in Ovid, I had found this gloss in the XlVth cent. 
 Balliol Glossary, wath portlcuum for 2iortlcorum. It can 
 be traced back as far as the Xlth century. A Vatican 
 codex of this date (Mai, Auct. Class. VII) gives it thus : 
 'Quinquatria, quinque porticorum ambitus puto.' It is also 
 in the so-called Glossae Isidori, Quinquatria ambitus quin- 
 que porticorum ; but I am aware that Lowe has discredited 
 the value and antiquity of this collection. Yet Arevalo 
 seems right in tracing a connexion between the gloss and the 
 scholion on toils qulnquatrlhus optat, Jnv. x. 11,5 'diebus 
 festis quibus Minerua colitur, aut quod intra quinque atria 
 fit, aut atria almndat, nam dies quinque in uno [contlnui 
 Schurzfleisch) festi ' ; and this carries back the gloss to 
 a comparatively early period. 
 
 Barth's belief that quinquatria was a singular noun (in 
 accordance with which he changed decipiant into deciplat) 
 in no way follows from the gloss. It may quite as well be 
 a neuter plural with a collective meaning ; and at least as 
 regards the MSS. of Ovid there seems to be no support for 
 the singular. 
 
 But was Barth right in accusing Naugerius and Marcilius 
 
 ' Reprinted from the American Journal of Philology, vol. xiii, 1892, p. 347. 
 2 Can this gloss bo ultimatelv biblical? See S. John v. 2. 
 
16 
 
 of great and dangerous error for preferring to write 
 quinque atria in two words rather than quinquatria in 
 one ? Surely not, on the evidence of the gloss. For if the 
 Juvenal scholion is the source of it, it is a mere etymology, 
 in its original form ; a tentative explanation of a word, not 
 an ascertained or really existing meaning. If, on the 
 other hand, it has nothing to do with the Juvenal scholion, 
 it may be, as Heinsius suggested, a gloss on this actual 
 passage of Ovid, but written after quinquatria had occupied 
 the MSS., ousting the more genuine quinque atria. 
 
 It is, I think, highly improbable that in the Augustan 
 age a poet of Ovid's eminence should have written quin- 
 quatria (which, so far as our extant authorities go, is 
 invariably applied to the festival of Minerva in the third 
 week of March) in the sense of quinque atria 'five halls ' 
 or ' hallfuls '. But I see no strong objection to his writing 
 quinque atria with this meaning. Vergil, speaking of the 
 wealthy Galaesus, Aen. vii. 538, describes him as lord of 
 Jive flocks of sheep, ^iw herds : Quinque greges illi balantum, 
 quina redibant Armenta, and the recurrence of this number 
 in familiar combinations such as quinque zonae, planetae, 
 sensus, &c., to say nothing of the nepTuOXou, might be 
 enough to account for its being used by Ovid in the vague 
 and large way more usuall}^ associated with tres, ter. At 
 any rate, the numerical difficulty which Burmann finds in 
 quinque, as if five atria were an impossibility, is not a very 
 insurmountable one. It would not, I suppose, arrest any 
 scholar who found it so written in a MS. otherwise authori- 
 tative. The real difficulties that quinquat^na Bhonld here, 
 and here alone, have a meaning quite distinct from its 
 ordinary sense. But this might well have its origin in 
 palaeography. How little separates QVINQ. ATRIA from 
 QVINQV ATRIA! Let the point which followed the Q 
 once fall out, how likely that it would continue to be omitted, 
 that the MSS. which preserved it would become fewer and 
 rarer ! 
 
 I need not dwell on the natural connexion between the 
 waxen imaofes of ancestors and the atrium. Juvenal's 
 well-known line, viii. 19 ' Tota licet ueteres exornont atria 
 cerae,' is copiously illustrated by Mayor mi loc. One parallel 
 I may be permitted here to cite. Mart. ii. 90. 5, 6 Ditt'orat 
 hoc patrios optat qui uincere census Atriaque inmodicis 
 avtat iinaginibus. Tlie plural here is like the ([uiaque atria 
 of Ovid. It is vague and meant to give an idea of pro- 
 fusion — he crowds his 'halls' with busts, just as Ovid's 
 lover tills 'five ' whole ' halls ' with tlieni. 
 
UNIVERSITY OP CALIFORNIA LIBRARY, 
 BERKELEY 
 
 THIS BOOK IS DUE ON THE LAST DATE 
 STAMPED BELOW 
 
 Books not returned on time are subject to a fine of 
 50c per volume after the third day overdue, increasing 
 to $1 00 per volume after the sixth day. Books not in 
 demand may be renewed if application is made before 
 expiration of loan period. =^=^^=== 
 
 MAR 30 19£5 
 
 JON 9 I960 
 
 *uZC D LD 
 
 ^UN 6 1960 
 
 fe,......«.- ..... 
 
 REC J^ 
 
 MAY 22 1962 
 
 APR 4- 1968 5 
 
 j£C 1 8 1358 6 3 
 
 s 
 
 RECEIVED 
 
 Dtiu 6*68-1 f'W 
 
 LOAN DEPT. 
 
 iJ 
 
 ]5m-12,'24 
 
'VI, 
 
 
 "'"^ J!rr 
 'I. tono 
 
 U C BFRKELEY LIBRARIE! 
 
 Ca5Efl33^TfcD 
 
 33VXoy 
 
 UNIVERSITY OF CALIFORNIA LIBRARY