BMCR 2009.07.23

Sexti Properti Elegi; Cynthia

, Sexti Properti Elegi. Oxford Classical Texts. Oxford: Oxford University Press, 2007. lxxxi, 217 pages. ISBN 9780198146742.
, Cynthia. A Companion to the Text of Propertius. . Oxford: Oxford University Press, 2007. xvi, 647 pages. ISBN 9780199228706.

‘There exists no good text of Propertius,’ Goold rightly stated some years ago (HSCPh 71, 1967, 59). Until now we have been able to read the text of Propertius in the editions of Barber (OCT, 1960, 2nd. ed.), Hanslik (Teubner-Leipzig), Fedeli (Teubner-Stuttgart, 1994, 2nd. ed.), Luck (Zürich, 1996, 2nd. ed.) and Goold (LCL, 1999, 2nd ed.). Other texts to have appeared recently are those of the radical Giardina (Roma 2005; cf. BMCR 2006.03.20) and the conservative Moya del Baño (with Ruiz de Elvira, Madrid, 2001) and Viarre (coll. Budé, 2005). In addition, a balanced text and worthwhile commentary have been published by G. Hutchinson for book 4 (Cambridge 2006, cf. Dimundo, Exemplaria Classica 12, 2008, 375-82).1 We also have at our disposal the extraordinary textual commentary by Shackleton Bailey (Propertiana, 1956) and the very useful list of variants by Smyth (Thesaurus criticus ad Sexti Propertii textum, Leiden 1970), together with the very good monographs of Butrica (The Manuscript Tradition of Propertius, Toronto, 1984), and Günther (Quaestiones Propertianae, Leiden, 1997), who is also the editor of the excellent Brill’s Companion to Propertius (Leiden, 2006), which contains obligatory reading in the contributions by P. Fedeli ‘The History of Propertian Scholarship’), Butrica (‘The Transmission of the Text of Propertius’) and Tarrant (‘Propertian Textual Criticism and Editing’).

The edition by Heyworth (hereinafter, H.), accompanied by a very useful and thought-provoking textual commentary (hereinafter, Hc.), now replaces the above-mentioned edition by Barber in the prestigious Oxford catalogue. The introduction, written not in Latin but in English, as were the Sophocles of Lloyd-Jones and Wilson (1990), the Ausonius of Green (1999) and the Vegetius of Reeve (2004), consists of a Preface (vii-lxvii), Bibliography (lxviii-lxxvii), Stemma (lxxviii), and Sigla (lxxix-lxxxi). It is followed by the Latin text (1-190), an Appendix with III vii as found in cod. N (193-5), an Index orthographicus (196-204), and Index nominum (205-17). The commentary (Cynthia), for its part, is divided into a brief Preface (vii-xiii), the Stemma and the Sigla of the edition (xiv-xvi), the rich and extensive commentary itself (1-514), a translation of the elegies (515-607), a comprehensive bibliography (608-615), Indexes of passages cited and of Latin words, and a general Index including proper names and an extremely useful list of subject-matter and topics covered (617-47). Both volumes have been excellently produced by the publisher, with clear print and high-quality paper. Over so many pages I have found extremely few errata.2

The preface deals clearly and precisely with the manuscripts and the bases for this new edition of Propertius:

a) The archetype, N, and the medieval tradition (vii-xi) with an analysis of the Neapolitanus (nunc Guelferbytanus Gudianus 224, ca. 1200) and the florilegium Vat. Reg. Lat. 2120 ( Flor. 1);

b) A and its descendants (π, xi-xxviii) with a study of A (Leidensis Vossianus lat. O.38, ca. 1240, which goes as far as II 1 63), the florilegium Par. Lat. 16708 (lor. 2) and the descendants of A stemming from a lost manuscript of Petrarch (π): F (important for what is missing from ἀ λπβθ;

c) Poggio’s manuscript (λ) and the 15th-century tradition (xxviii-xlix), which for H. and Butrica constitutes a third branch in the tradition of the Propertian text, independent of the previous two; H. argues that the manuscript Poggio sent to Niccoli is not the Neapolitanus, but another, now lost one which he calls λ, from which descends a series of manuscripts which must be taken very much into account when establishing the text and critical apparatus of Propertius: T (1427), S (ca. 1460-70), J (ca. 1430-45), K (1469), W (1450-75), M (1465), U (1465-70), R (1466), C (ca. 1470-1);

d) The Stemma (xlix-li, lxxviii), on which the conclusion reached is that ‘the vital point is to cite the new group of manuscripts accurately’ (li);

e) Indirect transmission (li-lii) through the ancient grammarians, Donatus, the codex Salmasianus, Lactantius, Isidore and the Pompeian graffiti;

f) The constitution of the apparatus criticus (lii-liii), which is positive, as is only to be expected from an editor who has carefully collated all the manuscripts used in the edition, for which service we will forever be in his debt;3

g) Punctuation and spelling (liii-lvi), where H. defends modern punctuation conventions for ‘a writer of English’, just as he also uses modern spelling conventions in his edition (e.g., -es for the 3rd decl. acc. pl.), rightly ignoring medieval orthography;

h) The editions and textual criticism of Propertius (lvi-lxi), in which H. justifies the use of the catch-all symbol ς for conjectures proceeding from 15th-c. MSS. (‘unauthoritative readings found in fifteenth-century μσσ’, η. CQ 36, 1986, 199). However, I would have preferred to see all the readings accompanied by at least the sigla of a manuscript4 and by the name of the first editor clearly known to have accepted this or that conjecture (cf. the Index below). Citing a manuscript and, if possible, an editor offers rather more than the conformism of a ‘nescioquis per coniecturam ante annum 1600’ (lxxx). From p. lix on there is an overview of the best contributions made to the text of Propertius from the edition of Scaliger (1577) down to the above-mentioned Thesaurus by Smyth. H. says much, and says it well, in very little space of Scaliger, Heinsius (Adversariorum liber IV, 1742), Passerat (1608), Burman (1680), Lachmann (1816), Baehrens (1880), Housman (CP 29-54, 232-378), Goold (1990 and 1999), Butrica (1984; cf. CQ 47, 1997, 176-208) and Günther (1997). Nor does he forget to mention Luck (1964, 1996), Morgan ( CQ 36, 1986, 182-98), Liberman (RPh 66, 1992, 337-44; MEFRA 107, 1995, 315-44; RPh 76, 2002, 49-100) or Giardina (2005), but he does ignore Shackleton Bailey (1956) and Fedeli (1994, 2nd. ed., 1965, 1980, 1985, 2005). These two distinguished scholars deserve a place, it seems to me, alongside those mentioned above, as well as in the bibliography;

i) Divisions, numeration, omissions, fragments, transpositions (lxii-lxiv). Given that antiquity has only handed down to us the division of the books and not of the poems, or their numbering (cf. H. ‘Dividing Poems’, in Pecere-Reeve, Spoleto 1995, 117-48) — since the text of Propertius presents numerous omissions, transpositions and fragments — and also in view of the fact that the state of book II is defective, perhaps due to the loss of some pages of an ancient manuscript, H. presents us with a very different (and very reader-unfriendly) look from that of other editions, the details of which are noted in the Appendix below. Perhaps the numbering of the lines ought to have been changed once and for all (read H., PLLS 8, 1995, 171-2). I, for one, would certainly have been less put out by this editorial approach (Hc. x, n. 3) than I am by the head-swimming experience of reading book II with so many numbers jumping around, so many lacunae, and so many distichs relegated to the end of the poems;

j) The text. In contrast to the editions by Housman that were intended to be ‘editorum in usum‘, the text presented here by H. is ‘rather lectorum in usum‘ (p. lxv) and on the same page H. clarifies what type of edition lies in the reader’s hands: ‘This may be thought a radical edition; it is certainly not an edition that tries to plot a middle course. But I suspect that there are more couplets that should be deleted, more lacunae marked, and far more places where we need rather to deviate from the MS tradition than to return to it. Without an extraordinary papyrus find, or time travel cheap enough to be exploited by the editors of classical texts, future investigation of the MSS is likely to increase our knowledge of the activities of the humanists while barely affecting our investigation of the text of Propertius’.

In his edition (1-190) H. works, as stated above, on the basis of the existence of a third branch of the tradition of Propertius, made up of a group of 15th-c. manuscripts (TSJKWMURC) descended from a lost manuscript which Poggio Bracciolini took to Florence in 1423 (X for Butrica and λ for H.), and which deserves to be cited alongside the two branches that descend from the archetype, N (ca. 1200) and A (ca. 1240) together with their descendents (φλπ [βθ]). At the same time he discards manuscripts DVVo (Δ), which have been used in previous editions (e.g., Barber 1960; Hanslik 1979; Fedeli 1994); since they are not independent of the archetype, their readings contribute nothing new, and they lack independent authority. The result is an edition that is completely different from all that precede it, including the novel edition by Goold of 1990 and 1999. The Appendix below lists scores of proposals by H., not counting the ones featured in the critical apparatus. There are almost 200 such proposals. Hardly anything has come through the process without having been weighed, analysed, and either emended or subjected to conjecture. It will take a great deal of time and effort to digest such a vast number of novelties in the text of Propertius.

In his monumental commentary, which throughout denotes expertise and professional dedication, H. analyses poems, elegies, lines, words, and punctuation with the skill of a scholar consistent in his own ideas and ruthless with the scalpel of reason and common sense (Housman CP 1058), making sure all the while that everything fits grammar, metre and sense (Luck AJP 102, 1981, 186). We have here 514 pages of textual commentary unparalleled since the ‘editio Burmanniana’ of 1780 or Shackleton Bailey’s ‘Propertiana’ (a slightly more conservative work) of 1956 (the closest thing to it today being the new Catullus by Trappes-Lomax, 2007; cf. BMCR 2008.09.32). My fullest recognition and admiration therefore for this Herculean task of clearing away the dead wood and shedding new light on the work of the great poet from Assisi. I shall limit myself, therefore, to making a few observations on certain passages.

BOOK I

i 5 puellas I do not believe the reference here is to the Muses (an old interpretation due to Vulpius, cf. also Burman, 1680, 3), but to women, matrons or hetairai who, for whatever reason, turn down offers of love; cf. Luck (AJPh 100, 1979, 73-5) and Fedeli (1980, 67-8).

ii 13 In the text H. accepts Hertzberg’s praelucent (1845, 11) and is inclined to opt for gaudent depicta in the commentary (Hc. 13), but perlucent (recc., Volscus) is supported by Catullus (69.4 aut perlucidi delicii lapidis) and Varro ( ling. 5.31 (140) quod perlucet, ut lapides); cf. Manil. 5.531 et perlucentes cupiens prensare lapillos, with Housman’s commentary (1920, 51); Mart. 9.2.9 splendet Erythraeis perlucida moecha lapillis.

v 15 ‘maestus tremulis nescioquis‘ says H. on p. 8; in the edition by Barth (1777, 17) we read ‘et moestus tremulis Colb. alt.’ (cf. Paldamus 1827, 9), the reading of the Parisinus BN lat. 8459 (Butrica, 1984, 340 and 283-5); 20 H. incorporates domo into his text (dub. Heinsius, 1742, 659, Lachmann, 1822, 22; not so in his edition of 1829, 6), but abire domum is ‘a colloquial dismissal’, as Mankin notes on Hor. epod. 11.20 (1995, 203; cf. Fedeli 1980, 163).

vii 26 H. is right in proposing to read honos (honor Rossberg), accepted by Giardina (2005, 50) and Luck (Exemplaria Classica 12, 2008, 52).

viii A 21 taedae (recc., Ayrmann) is unnecessary (cf. Housman, JPh 22:43, 1893, 117), since the complaints at the door are directed at the beloved (de te), as Goold understands (1999, 61); 22 H. accepts fata (Henry) in place of verba, but the expression ‘verba queri’ is the one used by the elegiac poets to refer to words of lamentation at the door of the beloved (1.5.17; Ov. Rem. 509; Fast. 3.507); the passages cited by H. (Tib. 1.5.51 2.6.34; Ov. Met. 5.298) have nothing to do with the elegiac context of the Propertian ‘exclusus amator’.

viii B 43 H. rightly accepts palmis (Scaliger, Luck) in place of plantis, since it is said that the stars can be touched with the hand or the head, but not with the feet; to the examples presented add Ov. Fast. 3.34 contigeratque sua sidera summa coma.

ix 4 The app. crit. would be clearer thus: empta] illa Fontein, fort. ista; read also solos] sola H. (p. 66 on 2.19.7).

x 13 H.’s recitare for reticere was accepted by Luck (1996, 28 and 368), but reticere dolores from Ω or calores (Dousa f. and Guyetus) can be kept, since the poet is telling Gallus that he has learned to be discreet about the passionate love between Gallus and his beloved.

xii 20 I prefer a comma, not a semi-colon, after fuit : Cynthia prima fuit, Cynthia finis erit.

xiii 8 There is no need to change primo to primum; cf. the same expression in Ov. Met. 4.59 primosque gradus vicinia fecit with Bömer’s pertinent commentary, 1976, 38-9 (‘primosque gradus, sc. amoris: ein seltener und vor Ovid nicht bekannter Sprachgebrauch); the translation is not ‘and, having slipped for the first time, to go astray’ (Goold, or ‘and, having . . . to abandon your stance’, Hc. 525), but ‘to give up on slipping at the first step (sc. of love)’; see also Fedeli 1980, 304-5.

xiii 13 in the app. crit. H. does not cite vago (Francius, Luck 1996, 34 and 368 : malo codd.).

xv 29 Neither H.’s nam prius e vasto nor other emendations improve the proposal ( alta prius retro) by Burman (1780, 151; cf. Ov. trist. 1.8.1-2 in caput alta suum labentur ab aequore retro / flumina; see Hc. 68-9), accepted by Luck and Goold; 39 quid (recc., DVVo), on which H. expresses his doubts in the app. crit., was previously incorporated into the text by Luck (1996, 40; quod on p. 368 is an error).

xvii 11 The proposal by Paldamus (revolvere in his notes; reponere in the text on p. 32) is defended in ch. III of his Observationes criticae on pp. 257-8 of his 1827 edition, which is not cited by H. on p. lxviii.

xviii 11 H.’s proposal, femina, for altera, which is surely a gloss, as he himself points out (Hc. 80), is convincing and deserves to be incorporated into the text in view of the parallel presented (3.15.9-10 nec femina post te / ulla dedit collo dulcia vincla meo).

xix 12 On the meaning of magnus see Weber (CPh 103, 2, 2008, 184-8); 15 the easy emendation harum of Francius, accepted by H., seems unnecessary to me, as veniant would have a concessive-optative force and quarum a demonstrative force (= earum, cf. Szantyr, 1965, 569-72; Bassols, 1967, 252-5).

xx 25-30 I understand these lines as follows: from the air the brothers Zetes and Calais were harassing the fair Hylas (vv. 25-6), intent on snatching kisses from him in suspended flight (v. 27: suspensis … plumis, cod. Guelf. Helmstadensis 373, Livineius, Luck) and on taking these kisses from him, one and then the other, as he was looking up (28), but Hylas, still in suspense, took refuge at the end of one wing, protecting himself from the trickery of the two birds with a branch (vv. 29-30). If this is a valid interpretation — and emphatic repetition is prominent in the text — I fail to see why so many changes have to be made, as the text could be taken as follows:

hunc duo sectati fratres, Aquilonia proles,
hunc super et Zetes, hunc super et Calais;
oscula suspensis instabant carpere plumis,
oscula et alterna ferre supina fuga;
ille sub extrema pendens secluditur ala
et volucres ramo submovet insidias.

The text of H., it seems to me, does not improve upon that transmitted by the manuscripts;

xx 32 Why can the repetition ibat … ibat? not be maintained? According to H., who changes the first ibat into unus, because ‘Propertius uses repetition for antithesis or addition, not for plangency’ (Hc. 90). This seems a very weak argument for changing a text; 49 H. accepts Fontein’s brilliant proposal (cui procul Alcides ter ‘Hyla’ respondet, at illi) for cui procul Alcides iterat responsa, sed illi. Is the change necessary? It depends on whether we tend towards the direct style of Virgil (ecl. 6.43-4) or the narrative style of Val. Flaccus, 3.596-7, since iterat and responsa are very Latin terms.

BOOK II

i 48 solus is preferable to semper (Hc. 112), as it picks up on the uno (sc. amore) of the previous line. Propertius pins his good fortune on love, but his happiness will be more complete if he enjoys just one love on his own, without having to share it with anyone else (‘sine rivali’, as pointed out by Passerat [1608, 250], cf. 2.7.19 tu mihi sola places; placeam tibi, Cynthia, solus).

ii 3 In the app. crit. H. is tempted by the diuina of a codex of A. Perreius (Burman, 1780, 234). I believe that this reading, accepted by Dousa and Burman in their notes, corresponds better to the sense of her role as ‘puella divina’: ‘Why does this divine face dally on earth?’ (from Lyne, CQ 48, 1998, 540).

iii 1-4 It is better to take these lines as the literal words of an interlocutor (cf. Fedeli 2005, 123) and a colon should be placed after haesisti, because cecidit . . . tuus is like an epiphonema of what has gone before.

v 3 hoc merui spectare does not improve the text of the manuscripts, haec merui sperare, a variation of Catull. 64.140 (non haec miseram sperare iubebas); 4 the proposal aliquo is not from Bosscha (1801) but is noted by Burman before he presents his proposal alio (1780, 244, where he devotes three pages to clarifying the passage!); 28 Peiper’s verna (also accepted by Goold 1999, 118) for verba destroys the sense Propertius wished to transmit, which is summarised by Lachmann (1816, 120-1) as follows: ‘Certe Graecismus iste (referring to verba, an accusative of respect) poetis nimis familiaris est, quo levem Cynthiam Propertius indicat, cuius verba nihil ponderis habeant, certi nihil’; see also Burman (1780, 247: ‘sed verba levis Graeco more pro in verbis levis et variabilis’) and Fedeli (2005, 190).

vi 5 deiectas is Gebhard’s reading, explained paleographically by Hc. (134) from delectas as the regular use of l for consonantal i (here l for i longa).

vii 8 I do not think that it is more that is the corrupt term but perdere (H. prints perdere iure from Allen*; cf. Hc. 140-1), since the verb to be expected would be tollere faces nuptae, as in Catull. 61.121 (tollite, o pueri, faces); cf. Ov. Her. 11.101.

xi 2 H. makes an attractive proposal in the app. crit., suggesting ponat instead of ponit.

xiii 1 In view of the incomprehensible combination armatur etrusca, the option armantur Susa (recc., Beroaldus, Luck, Goold, alii) is preferable to the rare armatur Itura (Pontanus, Hanslik), accepted by H., who in Hc. (162) also proposes armata est Creta; 11 me iuvet is preferable to me iuvat, cf. 2.34.59, 3.5.19 and 21; 19, 21 on tunc/tum (Hc. 164-5) see also Gaertner, ‘Tum und tunc in der augusteischen Dichtersprache’, RhM 150, 2007, 211-24; 58 Hc. (171) clearly shows how both here and at 4.1.86 we ought to read in N quid and not qui, because the scribe abbreviates quid consistently in this manner.

xv 1 Propertius opens a makarismos, addressed to himself, with triple anaphora of the interjection o, as in 1.10.1-4. The text of the manuscripts can be preserved if a hiatus is accepted after the second interjection or if we follow the correction by Puccius (ex codd.), as Passerat does (1608, 302; cf. Luck 1996, 86). Housman’s proposal (io . . . io . . . io), accepted by H. and Goold, is characteristic of an epithalamium, not a makarismos; 22 I would not question pudet (Hc. 175-6) and replace it with piget, as Giardina does (2005, 146), since Propertius is referring to women who feel embarrassment, not ‘regret or irritation’, when displaying their breasts after having given birth once or twice, and not to Cynthia, who is one of those in whom carent quoque pectora menda (Ov. Ars 3.781).

xvi B 27 I lean not towards exutis . . . lumbis (Sandbach, PCPhS 5, 1958-59, 3-5), accepted by H. and others, but to the reading excussis . . . lumbis (recc., and, e.g., Beroaldus, Passerat [1608, 311], Broukhusius [1702, 139], Burman [1680, 329], Sh. Bailey [1956, 95-6], Hanslik [1979, 65], Günther [1997, 37, n. 153]); cf. 2.29.35, 2.9.45; Catull. 16.11 qui duros nequeunt movere lumbos. The best explanation of v. 27 I have seen is in the edition by Broukhusius (cf. also ‘Notae variorum’ [1822, 810])): ‘Ego expono: Barbarus iste, me excluso, agitat et calcat mea vestigia, quae in tuo lecto restant, inque iis volutat sese interea, dum lumbos excutit et clunem movet’; 49 tonitrus (Francius), accepted by H., is unnecessary and no reason put forward in Hc. (183) can persuade me to break the poetic zeugma of vidisti from both sonitus percurrere and fulmina desiluisse, as in Aesch. Prom. 115, Sept. 103, Verg. Aen. 4.490 (with Pease’s comm. on pp. 405-6), 6.256-7 (with Norden’s comm. on p. 205), Ov. Trist. 3.8.37-8 (with Luck’s comm. on p. 207), and Augustine civ. 10.54; see also Munro on Lucr. 4.598, Fedeli 2005, 506. Propertius makes use here of synaesthesia to combine two images or sensations proceeding from different sensory domains: seeing and hearing the lightning crossing the sky and leaping from Olympus.

xviii 31 H. accepts Goold’s vultus but does not even refer to Watt’s cultus (MH 49, 1992, 234), accepted by Fedeli (2005, 551-2).

xix 24 Watt’s fallere (ap. Goold) has better support than H.’s tangere; cf. esp. Mart. 14.216.1 (non tantum calamis, sed cantu fallitur ales / callida dum tacita crescit harundo manu); see also Passerat 1608, 324; Fedeli 2005, 575. Hc. himself (192) admits that ‘I chose tangere to give each a presence in published texts’; 31 I take the last two lines (cf. Passerat 1608, 324 and Kuinoel 1805, 140), to mean that neither the solitary woods nor the rivers flowing down from the mossy heights will prevent me from having you continually on my lips, changing your name for other fictitious ones so that you can always be present, since everyone seeks to harm those who are absent. Attempts to modify the text (e. g., metuam Jacob for mutem) and try to find out-of-the-way explanations (see Fedeli 2005, 579-583) seem pointless to me; see also Passerat 1608, 325; Enk 1962, 269-70.

xx 31 H.’s suggestion of sitque inter Tityi volucres mea poena iacere (or ligari) is attractive, ingenious and clear, but is it what Propertius actually wrote? For mea poena is a use of abstract for concrete (see Bömer on Ov. Met. 5.373 patientia nostra) and our poet has constructed a powerful image of the poet who, as punishment, vagetur, ‘wanders about’ among the birds that are devouring a body of nine hectares, as described by Homer ( Od. 11.576-9) and Virgil ( Aen. 6.596).

xxiv 51 in me (attributed to ς by H.) for ut me seems to come from Lachmann (1816, 189).

xxv 1 I do not think that ‘the vulnerability of initial letters’ (Hc. 218) can justify such a drastic change as Cynthia, defended by Phillimore, Goold and H., for the Vnica of the manuscripts; see also Passerat (1608, 345), Hertzberg (1702, 173) and Fedeli (2005, 707).

xxvii 1 H., like Richmond and Richardson, considers there to be a lacuna before v. 1, since elegiac poems do not start on a regular basis with at, except Ov. Her. 12.1 (about which H. expresses doubts in note 62 because Ovid is not following the model of Euripides, Ennius and Catullus) and am. 3.7.1; why should Propertius not be able to begin an elegiac poem abruptly (‘initium abruptum’, Enk, 1962, 345) with a iunctura characteristic of an apostrophe (Passerat 1608, 359) or with an ‘ at indignantis’? Cf. Hertzberg (1845, 183); Bömer (1976, 14) on Ov. Met. 4.1; Bessone (1997, 62); Fedeli (2005, 768).

xxix B 41 In view of the stated opinions of Broukhusius (1702, 186), Heinsius (1742, 706) and Burman (1780, 430), the lemma of the app. crit. should include the following information: excludor Guyetus, def. Broukhusius : secludor Passerat in notis : eludor dub. Burman in notis : deludor Palmer; cf. Palmer, 1880: sanctae casto deludor amore; Otto, Hermes 23, 1888, 38: cultor deludor amoris; Housman, JPh 22, 1893, 87 (= CP I, 316): custos deludor amoris.

xxxii 23 Surprisingly, not even in the app. crit. does H. include the reading pervenit (recc.), defended by Ellis (JPh 15:29, 1886, 17) and accepted by Luck (1996, 130) and Fedeli (2005, 885; cf. also Lachmann, 1816, 216), given that the expression appears in Virgil (Aen. 2.81) and also several times in Ovid (cf., e. g., Met. 5.256; see also Fedeli 2005, 903); 47 see the extensive note by Lachmann (1816, 218) on Tatios veteres; H.’s singular (Tatium veterem) is unnecessary; cf. 3.1.14, Ov. Ars 1.179; Fast. 2.135, Hor. carm. 1.12.37; 58 the reading corripuit, found in the recc., was proposed independently by Luck ( AJPh 100, 1979, 88-9) and in verse 33 correpta (Fontein) should also be read.

xxxiii A 21 H. accepts Housman’s proposal (pia causa), as does Goold (1999, 206), but I fail to understand the sense of the hexameter in relation to the pentameter (‘But you, who through excess of piety are the cause of my pain, once we are free of these nights, let us thrice make <love’s> journey’, trans. on p. 561). In the hexameter Propertius is referring to Cynthia, who has spent 10 days of secubatio in honour of the goddess Isis and is therefore now appeased (placata) or tranquil (pacata, Oxon. bibl. Bodlei. Canon. class. lat. 31 and Markland) at the cost of the suffering of Propertius (who has had to bear those ten days without her); and with Cynthia in this state, once the ten nights have passed, he invites her to make love three times. The difficulty, then, would seem to lie in nimium, covering a iam nunc (rather than the nunc iam of Hertzberg (p. 222, cf. 2.16.13), since in poetry the phrase used is iam nunc and not nunc iam (cf. 2.19.17, 4.11.93; Tib. 1.3.53,1.5.71, 2.3.3, 4.5.11). The manuscript abbreviations for iam and nunc may have contributed to the corruption. The editor himself recognizes (Hc. 260) that ‘The absence of nunc (or something similar) in the couplet also helps show that we are not to read this as a reaction to the actual end of the period of Isiac worship’, but with iam nunc Propertius is certainly referring to the ending of the ten days of forced separation.

xxxiv 12 Markland’s proposal (tu in place of in) was included in the edition by Luck (1996, 136); 20 nullo in place of stulto should, I think, be attributed in the app. crit. to Heinsius (1640-1680; cf. 1742, 711), as Goold and Luck do, and not to Fontein (1708-1788); 46 if recta is to be changed, because ‘the usage is not idiomatic’ (Hc. 273), the epithet dura (cf. 1.1.10, 1.17.16, 2.1.78, 4.2.23) would suit the situation described better than the possessive adjective nostra, proposed by H.; 64 it is surprising that in the quotation from Virgil’s Aeneid (1.2-3) the editor should write (Hc. 275) Laviniaque venit / litora rather than Lavinaque venit/litora, as defended, among others, by Hertzberg (1845, 236-7), Courtney (BICS 28, 1981, 27) and Goold (‘The Voice of Virgil’, in Woodman and Powell, eds., Author and Audience in Latin Literature, Cambridge, 1992, 115; and in his edition of Virgil, 1999, 262).

BOOK III

iv 11 It is unnecessary, pace Hertzberg (1845, 244 on 3.1.1: ‘sacer de rebus diis sacratis, numquam de ipsis diis, qui sancti sunt, dicitur’, although on p. 268 he defends sacrae . . . Vestae with a clear ‘recte et docte’), to adopt Postgate’s proposal of sanctae for sacrae fatalia lumina Vestae, since Vesta here refers metonymically to fire, as in Ov., Fast. 6.291: nec tu aliud Vestam quam vivam intellege flammam (cf. Bömer 1958, 359); see Sh. Bailey (1956, 141-2), Fedeli (1985, 166), Camps (2001, 71); 14 ducere (recc.) . . . licet (Livineius, Guyetus) seems to me an option not to be rejected outright; cf. 3.8.30, Ov. Rem. 522 with Lucke’s comm. (1982, 159); Prud. c. Symm. 1.239-40 gaudia ducunt / festa kalendarum (example quoted by Heinsius, 1742, 658).

vi 31 Instead of the phrase vana canunt, Passerat, Gebhardus, Schoppius (cf. Heinsius, 1742, 719; Burman, 1780, 529; Hertzberg, 1708, 237) and H. prefer vana cadunt, but see Stat. Theb. 3.646 sed quid vana cano; Claud. 15.353-4 (In Gildonem): per somnos mihi, sancte pater, iam saepe futura / panduntur multaque canunt praesagia noctes (example quoted by Herzberg); see, however, Burman (1680, 529-30): ‘quod [ cadunt ] non facile repudiandum . . . eleganter enim cadere pro evenire’; cf. Passerat 1608, 437: ‘ cadunt, id est, accident, evadunt, eveniunt. Sic cadere pro evenire ubique M. Tullius’.

viii 31 Dardanus, as proposed by Heinsius 1742, 722 (Francius had proposed Dardana for barbara in Ov. Her. 8.12; cf. Burman 1727, I, 108), is unnecessary for barbarus; cf. Fedeli (1985, 296) and Nisbet-Hubbard (1978, 71-2 on Hor. Carm. 2.4.9).

x 21-24 I see nothing wrong in taking it that Cynthia and Propertius, having fulfilled the religious rites of her dies natalis (vv. 19-20), should wish to dine together and spend the night drinking (v. 21: noxque inter pocula currat [ surgat is Cornelissen’s emendation, accepted by H.], cf. OLD s.v. ‘curro’, 6), in a very pleasant atmosphere (v. 22) and to the accompaniment of soft background music (tibia . . . rauca) and dancing suited to the night (v. 23: nocturnis [not continuis from Housman] succumbat . . . choreis). Nor do I consider it indispensable to break the parallelism et . . . et (vv. 22 and 24) by changing et sint to adsint (H.); 25 dulcia . . . convivia, not dulcia . . . convicia (Broukhusius ap. Barth. 1777, 221-2: ‘ convicia i.e. cantus amoebaei iuvenum ac puellarum vult Broukh.’; cf. Burman, 1780, 582 and 700 on 3.25.1) is the appropriate term for the context (cf. Ov. Fast. 1.401, pointed out by Hc., 330; Passerat [1608, 465] quite rightly points to 3.7.43 dulcis conviva and Hor. carm. 3.8.6 dulces epulas).

xi 29 nexerit from ς (= cod. Vat. Barberinianus VIII 23 [Hanslik, xiv, xxii; Fedeli 1984, 175]) was proposed independently by Sh. Bailey in the app. crit. to Barber’s Oxford edition (1960, 107).

xiii 59 The criticism levelled at Cairns seems somewhat excessive in Hc. 355, since the reading verus haruspex can be interpreted in the way Barth does (1777, 236): ‘verus cui creditur, cum Casandrae non fuisset creditum. Nam sic etiam Roma posset a sua pernicie revocari’, or as Camps interprets it (2001, 120): ‘The poet means ‘may I be a true prophet in the eyes of my countrymen’, i.e. ‘may they believe me (and be warned in time)”, Lee (Propertius, The Poems, Oxford 1994, 90) adroitly translates ‘would my country thought me a true prophet!’ And the words of v. 61 (certa loquor, nulla fides) ratify, in my view, the verus of this line.

xv 33 sub (Livineius) tacito, close to the sic tacito of ω and accepted, for example, by Camps, Fedeli and Luck, should, I think, appear in the app. crit.

xvii 3 flatus (anon. ap. Camps, 2001, 40) Veneris is an unattested combination and insanae Veneris stands in place of insani amoris (cf., e. g., 2.14.18 with Enk’s comm. [1962, 207-8] and Hor. Carm. 3.21.2 with Nisbet-Rudd’s comm. [2004, 248]); fastus is a standard amatory term in Propertius (cf., e. g., 1.18.5); the poet is beseeching Bacchus to free him from the pride he is enslaved to in the form of the mad sexual attraction (insanae Veneris) he feels towards Cynthia.

xviii 14 Hertzberg ends his commentary on this line as follows (1708, 355): ‘quid referam Marcelli ipsius gesta, quid praeterea omnia illa, quae eius nomine mater gesserit?’ and Goold translates (1999, 285) thus: ‘and all that his mother’s influence had procured?’; cf. Lee 1994, 176. None of this is even cited before the unnecessary decision to accept the adjective maturas from Barber in place of maternas (Hc. 384); 29-30 see also Belfiore, Emerita 31, 1963, 4-5.

xxi 7 vix makes perfect sense (‘she however [in spite of my having tried everything] hardly ever receives me or does so just once’ / ‘But she receives me hardly ever or just once after many snubs’ [Goold 1999, 292]) and there is no need to replace it with Cornelissen’s symmetrical bis.

xxiv 2 I follow the simplest explanation and understand oculis . . . meis as ‘too proud in my eyes’, that is, ‘dall’ammirazione con cui ti guardavo’ (Fedeli 1985, 676) or ‘oculorum meorum iudicio, quo tuam olim formam probavi’ (Kuinoel, 1805, 302; cf. Ov. Her. 17.125), with no need to fall back on the proposed elegis (Schrader) . . . meis.

BOOK IV

ii 57 The six remaining lines (sex superant versus) are the six lines of the inscription below (59-64), without counting vv. 57-8. In passages of Ovid ( Trist. 1.7.33-40) and Martial (11.20.1-8) not cited by Hc. the counting of the six lines begins with the direct quotation; cf. Sh. Bailey (1956, 229), Fedeli (1965, 117). In addition, the present tense of the verb ( superant) seems to me to be right, not wrong (Hc. 444) if taken as a general present (Bassols 1948, 198-9; Kühner-Stegmann 1912, 114). H.’s proposal of suberunt for superant is therefore unnecessary.

iii 11 Faced with this ‘crux Propertiana’, I am inclined, pace Morgan (haecine pacta fides? haec noctis iura maritae at CQ 36, 1986, 193-4) and H. (et pacta haec foedera nobis) to opt for a proposal closer to the Neapolitanus (et parce avia noctes) and to the context of a night of love and sex in the pentameter: haecne marita fides et pactae gaudia noctis, / cum rudis urgenti bracchia victa dedi?; cf. 2.14.9, Cat. 61.116-7, Tib. 2.1.12, Ov. Her. 18.108, 19.68-9, Ars 2.308, 3.462 (et vos gaudia pacta date with Gibson’s comm.. [2003, 287]: ‘ pangere is common in the context of marital undertakings’), rem. 107, 400-1, 727-8, Met. 11.306-10. Nor should we reject outright the option of Shilleto’s et pactae tum mihi noctes, accepted by Luck (1996, 226), although the position of mihi would be abnormal (cf. Hc. 446); see also Hutchinson (2006, 105).

iv 68 Luck (AJPh 100, 1979, 92) supports, rightly, in my view, Heinsius’ succubuisse (Burman, 1780, 778: ‘N. Heinsius ad marginem ed. Ald. coniecerat’); 94 H. on p. 164 fails to do justice to the manuscripts and to other scholars in his brief comment ‘sunt qui iniuste legant’; Hutchinson (2006, 39) also offers a dismissive ‘ iniuste sunt qui scribant’); The variant iniuste is found in FLPDVo (Fedeli 1984, 241), and was proposed by Lachmann (1816, 362) and accepted by Jacob (1827, 112, 220: ‘Gron. Rheg. Quod iam Lachmannus scribendum viderat’), Tovar (1963, 209), Fedeli (1965, 28 and 153, but cf. 1984, 241) and Hanslik (1979, 167).

v 55-6 I am not as convinced as H. and other editors since Scaliger that this distich is spurious. Without regarding it as highly as Hertzberg does (p. 455 ‘hos versus Scaliger excidit, nervos scilicet totius elegiae’), I feel, like Lee (1994, 190), that ‘Acanthis scornfully quotes i.2.1-2’; Hanslik (1979, 170) and Luck (1996, 240) also keep the distich; see also Sh. Bailey (1956, 242 and PCPhS 1952-3, 16-7).

vi 72 blanditae . . . rosae should be read here rather than the emendation by Lachmann blandae utrimque (1816, 374); this iunctura from Par. Lat. 8458 (cf. Butrica, 1984, 283-4) was an independent emendation by Scaliger (cf., among others, Passerat [1608, 636], Broukhusius [1702, 376], Heinsius [1742, 750], Barth [1777, 312 in the critical notes], Burman [1780, 823], Camps, Hanslik and Hutchinson), and is supported by Plin. Nat. 25.17 (ut radicem silvestris rosae, quam cynorrhodon vocant, blanditam sibi aspecto); see Hutchinson (2006, 167) and Bömer on Ov. Met. 10.555 (1980, 186).

vii 69 Against Housman (CR 48, 1934, 139 = CP III 1235) Sh. Bailey (1956, 252-3) rightly defended sanamus, for which there is no need to find a substitute (Rossberg’s sancimus); cf. 1.10.17. The variant solamur, cited by H., is interesting; it is cited by Passerat in his manuscript notes (1608, 651 ‘alii lib. Solamur, id est, levamus, minuimus’).

viii 4 I note the absence in the app. crit. of huc (recc., Paley, Luck, Hanslik), and tibi (recc., Goold); 22 iocos from λ was conjectured by Markland and revived by A. Tovar from two Salmanticenses (S and Se) in CPh 59, 1964, 34-5; 36 Palmer’s discubitus is unnecessary, since the question quaeris concubitus? would appear perfectly relevant in the context (‘Though sexual intercourse is intended’, Hc. 478 recognizes).

ix 29 ‘glaucis is one of Housman’s happiest suggestions (CP I 39) in this author’ (Sh. Bailey, 259), and it is accepted by Goold and H., in place of longis, but see Passerat (1608, 676), Fedeli (1965, 227) or, even better, Hutchinson (2006, 212): ‘P. is probably rendering Gk. τανύφυλλος / τανίφυλλος’; 71-2 after 74 Passerat (1608, 683 ‘mutanda disticha, ut sequens praecedat’), as H. says (Smyth and other editors mistakenly attribute the modification to Schneidewin and Dilthey).

xi 64 The correction by Scaliger (manu for sinu), followed by many editors (H. included), is a very easy one, but the image Propertius wishes to get across is that Cornelia’s eyes were closed in the bosom of her children, in an added detail of filial love (Passerat 1608, 705 ‘compositi et clausi mihi oculi a vobis, cum in sinu et amplexu vextro exspiravi’; 706 ‘in complexu carorum emori solent et libenter homines’); cf. 1.17.12 ossaque nulla tuo nostra tenere sinu?; cf. Verg. Aen. 4.686, Ov. Met. 14.743-4 with Bömer’s comm., Epiced. Drusi 95-6, Call. Epigr. 48 (Page = AP 7.728.5-6). See also Burman (1780, 911); Luck reflects the intentions of Propertius well in his translation (1996, 275): ‘in euren Armen schloss ich meine Augen’, as does Tovar in his edition (1963, 240; cf. Ramírez de Verger 1989, 266) ‘en vuestro regazo se cerraron mis ojos’.

And now, some final considerations. First of all, if H. and Butrica are right in establishing a third branch in the textual tradition of Propertius (recensio), as reflected in a group of 15th-century manuscripts, we should accordingly and unreservedly accept the establishment of a different apparatus criticus along the lines suggested by Butrica (1984, 170-201) and H.’s edition. However, this theory has been rejected by La Penna (Gnomon 61, 1989, 120-3), Murgia ( MD 45, 2000, 195-222), Giardina (2005, 14-7) and Fedeli (2005, 35). Goold (1999, 13) and Hutchinson (2006, 22) are sceptical. Whatever the case, Goold’s words are incontestable: ‘Whether or not we accept X (Heyworth’s Λ) as independent, it does not seem to make any practical contribution to the textual criticism of our author’.

Secondly, it is true that the text of Propertius has come down to us in a deplorable condition and that it is necessary to intervene whenever there are doubts regarding the reliability of the paradosis (Butrica 1997, 183), since the question is not whether to opt for a conservative or a radical approach but to decide between a search for the truth and the upholding of a comfortable credulousness (cf. Butterfield, Exemplaria Classica 12, 2008, 23, n. 67; Kenney 1974, 113-4). However, there is a limit to all things and in the noble task of conjecture hypercriticism is one such limit, and I would point to the clarifying contribution of Kenney on this danger in textual criticism (‘Ut erat novator: Anomaly, Innovation and Genre in Ovid, Heroides 16-21′, Adams and Mayer, Aspects of the Language in Latin Poetry, Oxford 1999, 399-414). The proposals H. makes in the text (cf. the Index below) number almost two hundred (more if we add those confined to the app. crit.), but many of them are put forward in his Cynthia in hesitant terms. I believe that one must be very sure of a conjecture, transposition, lacuna or deletion ‘to be embodied in the text’, to use the now-famous phrase of Butler-Barber (1933, lxx). Thus, there are doubts in Hc. which are converted into certainties in H.: e.g., on pp. 7, 34 ‘ et would be an improvement’, 41 ‘I follow Postgate (quidvis), partly because this is a possible solution that has been neglected’, 61 ‘if any change is to be made, we should remove both infelicities; hence my embellishment of Baehrens’s conjecture: quae cano non‘, 79 ‘much the most plausible is revolvere‘, 83 ‘I therefore prefer to postulate a lacuna between verses 23 and 24’, 87 ‘and I print it (durus ut) ‘faute de mieux”, 89 ‘Hesitantly I print raram‘, 115 ‘The solution, I suggest, is to suppose that et is an error for it’, etc., etc., etc.

And thirdly, I do not agree with those who believe that ‘Propertius’ mental processes were as orderly as his critic’s’ (Sh. Bailey, 1956, 253) and I do believe that we should be most attentive, not only to the principles of ‘reason and common sense’, but to Propertius’s own sensitivity towards the poetic text (cf. Burman sec. ad Lotichii Secundi Poëmata, Amstelaedami 1754, I, 267: ‘sed haec sufficiant pauca, ut pateat quam periculosum facinus admittant illi, qui ad suam trutinam pensant et castigant veterum artificium, verissimumque est, notante Broukhusio, illud Jul. Caes. Scaligeri dictum, paucis mortalibus datas esse aures poëticas’). I certainly do not believe that Propertius could have written lines like Cynthia forma potens; Cynthia verna levis (also Goold 1999, 118) instead of Cynthia forma potens, Cynthia verba levis (2.5.28) or that there is any need to intervene, for example, in 1.13.8, 20.32; 2.1.48, 5.3, 15.1, 16 A.27, 19.24, 25.1, 27.1, 32.47, 33A.21; 3.4.11, 6.31, 8.31, 10.21-4, 13.59, 17.3, 21.7, 21.28; 4.2.57, 7.79, 9.29, 11.64.

Great scholars, such as Housman, Sh. Bailey or Butrica, for different reasons were not able to edit the text of Propertius. Nor should it be forgotten that Goold in his second edition of 1999 changed his mind in 68 passages or that Fedeli introduced 150 changes in his 2005 commentary on book II with respect to his Teubner edition. H. has had the courage and conviction to offer us, in spite of everything, not only a reader-friendly edition, but also a Herculean commentary (a paperback edition is announced for April 2009 at £40), which represent a great step forward in clarifying both the Propertian recensio and emendatio and we must therefore be profoundly grateful to him for having devoted so many years of study to the great poet of Assisi. We will be forever in his debt for his work, which deserves my fullest praise and respect.

APPENDIX: ηεψωορτη’ς ρεαδινγς

Compared with previous editions, H.’s presents many substantial differences. I believe it is worth pointing out the most important differences, and apologize for exceeding the space normally granted for reviews (Trappes-Lomax, for example, includes this useful feature for his readers in his recent textual commentary on Catullus (Swansea, 2007, 21-32). I take as my reference Goold’s second edition of 1999 in the Loeb Classical Library, the most innovative to date (I cite in the first place the readings of H., indicating the proposer of each reading, and secondly Goold’s). For greater clarity I begin each elegy with the numbering of each editor and go on to give the different readings.

BOOK I

i 1-11, lacuna of two lines (Housman), 12-38 / 1-38

2 Cupidinibus / cu-; 5 Puellas / pu-; 12 ferire (Heinsius on Fast. 1.287 [1661, III 17 (not.); 1742, 248, 655]) / videre; 23 vobis et sidera et umbras (Jeverus) / Manes et sidera vobis; 25 et / aut; 33 nam me (H., Goold); 36 locum / torum.

ii 1-32 / 1-32

8 formae / formam; 13 praelucent (Hertzberg) / praegaudent; 23 vulgo / fuco; 24 nimis (H.) / satis; 25 ergo (H.) / non; sim/sis.

iii 1-46 / 1-46

8 non certis / consertis; 16 et arma / tarda; 26 malaque (H.) / munera; 39 producas (recc., edd. Gryph., Heinsius, Luck) / perducas; 45 lapsam / lassam.

iv 1-28, v 1-2 (Enk) / 1-28, v 1-2

14 discere (Heinsius, Goold, Luck) / ducere.

v 3-32 / 3-32

3 meos / meae; 8 solet / sciet (not reflected in the app. crit. of H., cf. Sh. Bailey [1956, 18-9]); 20 domo (Heinsius) / domum.

vi 1-36 / 1-36

26 hanc . . . aeternae (H.) / huic . . . extremam; 27 longaevo (Fontein) / longinquo.

vii 1-10, 13-4, 11-12 (H.), 15-26 / 1-14, 23-4, 15-22, 25-6

11 laudet (H.) / laudent; solitum (Markland) / solum; 16 (quam (Heinsius) nolim nostros te violasse deos!) / quo nollem nostros me violasse deos; 26 honos (H. from honor (Rossberg) / Amor.

viii A 1-12, 15-18, 13-14 (Carutti), 19-26 / 1-12, 15-18, 13-14, 19-26

7 calcare (Passerat) / fulcire; 13 faciles (Heinsius) / talis; 19 et (Livineius) / ut; victa (Heinsius) / lecta; 21 taedae (recc., Ayrmann) / de te; 22 fata (Henry) / verba.

viii B 27-46 / 27-46

29 Livor (Luck) / livor; 43 palmis (Scaliger, Luck) / plantis.

ix 1-30, 33-4, 31-2 (deleted and relegated to the end of the poem (H.) / 1-34

1 Amores / amores; 3 taces (H.) / iaces; 4 quidvis (dub. Postgate) / quaevis.

x 1-30 / 1-30 8 luna / Luna; 13 recitare (H. 1984) calores (Guyetus, Dousa f.) / reticere dolores; 16 surdas (Heinsius) / tardas; quaecumque / quae cuique; 27 Amori / amori.

xi 1-3, lacuna (H.), 4-8, 15-16 (Housman), 9-14, 17-30 / 1-8, 15-16, 9-14, 17-30

1 tepidis (Heinsius) / mediis; 6 pectore restat amor (Heimreich) / restat amore locus; 18 veretur (Lachmann) / timetur; 21 ei (H.) / ah (Lachmann); 24 omni . . . tempore (Fontein) / omnia . . . tempora; deliciae (Fontein) / laetitiae; 30 Amoris / amoris.

xii 1-20 / 1-20

2 Cynthia (recc., Passerat) / Pontice.

xiii 1-36 / 1-36

8 primum (H.) / primo; 12 inultus (Watt 1992) / amicus; 13 quae cano non (H.) / haec non sum; 25 amores (Lachmann) / amantes; 29 digna et / dignae; 30 partu, gratior (dist. Scaliger; see also Booth CQ 56, 2006, 528-37) / partu gratior.

xiv 1-24 / 1-24

5 altas (Barber 1953) tibi tendat (Barber 1960) / satas intendat; 23 Lyda (Markland) / ulla.

xv 1-14, 17-42, 15-16 (secluded by Pescani and relegated to the end by H.) / 1-14, 17-22, 15-16, 23-42

13-14 visura dolebat, illa tamen longae / visura, dolebat illa tamen, longae; 19 illum (Burman) / illos; 21 ablata (Lachmann) / elata; 29 nam prius e (H.) / alta prius.

xvi 1-10, 13-48, 11-12 (secluded by Lachmann and relegated to the end by H.) / 1-24, 27-36, 25-6, 37-48

13 gravibus . . . querelis / gravius . . . querela; 38 ingratos . . . iocos (Enk) / irato . . . ioco; pota (Heinsius) / tanta; 40 moras, / moras?; 47 nunc . . . nunc (Francius) / nunc . . . et.

xvii 1-28 / 1-28

3 solvit (Madvig) conversa (or resupina) (H.) / salvo visura; 11 revolvere (Paldamus) / reposcere.

xviii 1-23, lacuna (H.), 24-32 / 1-32

16 turgida (Heinsius, Luck) / turpia; 21 tenera . . . umbra (Hall) / vestras . . . umbras; 22 vestris / teneris; 23 an tua quod / ah tua quot; 27 dumosi (Heinsius) / continui.

xix 1-26 / 1-26

1 manes / Manis; 5 Puer / puer; 10 Thessalus / Thessalis; 15 harum (Francius) / quarum.

xx 1-51, lacuna (H.), 52 / 1-52

1 haec / hoc; 2 quod (H.) / id; 4 durus ut (dub. Butrica 1984, 176, 187) / sic erat; 7 hunc / huic; Vmbrae sacra (Hoeufft) / Umbrae rate; 11 nympharum / Nympharum; cupida . . . rapina (edd. vett., Passerat) / cupidas . . . rapinas; 32 unus Hylas ibat (H.) / ibat Hylas, ibat; 33-35 fons . . . quem (H.) / hic . . . quam; 33 Pegae (Scaliger) / Pege; 34 nymphis / Nymphis; 52 nymphis credere / ni vis perdere. xxi 1-10 / 1-10

6 me (La Penna 1952) soror Acca (Scaliger, Luck) / haec soror acta.

xxii 1-10 / 1-10

6 sed (Palmer, Luck) / sic.

BOOK II

i 1-36, 39-48, 51-78, 37-8 (deleted by Fontein and relegated to the end by H.), 49-50 (deleted by Carutti and relegated to the end by H.) / 1-38, III ix 33-4, 39-48, 49-50 (secluded), 51-78

5 cerno (Leo, dub. Fedeli 2005) / vidi; 11 cum poscentes / compescentes; 31 tractus / attractus; 33 et (Lachmann) / aut; 40 intonat (‘primus Leidensis’ [Leiden BPL 133A, Butrica 1984, 340], cf. Burman 1680, 209; H., CQ 34, 1984, 399) / intonet; 48 semper (H.) / solus; 51 Thesei (dub. Burman in notis) / Phaedrae; 58 habet (Flor. i) / amat; 71 Fata / fata.

ii 1-2, lacuna (Scaliger), 3-8, lacuna (Housman), 13-16, 9-10 (extraneous to this elegy [Housman] and relegated to the end (H.), 11-2 (extraneous to this elegy [Housman] and relegated to the end (H.) / 1-8, (9-12 after xxix 28), 8a-b, 13-16

4 tibi (Heinsius) / tua; 5 it (H.) / et; 6 ut (ed. Gryphiana 1548) / et; 7 ut (H.) / ceu; 11 Mercurio sacris / Mercurio aut qualis; 13 iam, divae quas / iam, divae, quas.

iii 1-10, 13-44, 11-12 (secluded and relegated to the end by H.), 45-6 and 47-54 (fragments, cf. Lemaire, 1832, 162) / 1-10, 13-16, 11-12, 17-29, 32, 31, 30, 33-44 (45-54 ante 3.4.1)

1-4 without inverted commas / with inverted commas; 15 papilla (Dorvillius) / puella; 23 nam / non; 45 saltem (H.) / saltem ut; 47 detrectat / detractat; 50 faciles (Günther 1997, 111) / post haec.

iv fragmentary (H.): 1-4, 5-6, 7-16, 17-22 (H.) / iii 45-54, 1-8, 15-16, 11-14, 9-10, 17-22 13 comitantur (Giardina 1977) / mirantur.

v 1-30 / 1-30

3 hoc / haec; spectare (H.) / sperare; 4 alio (Burman, Luck) / aliquo; 5 tandem (Burman, Luck) / tamen; 18 mihi (H.) / tibi.

vi 1-14, lacuna (Ribbeck), 15-22, 25-6 (Enk), 23-4, 27-40, 41-2 after vii 6 (Sandbach*) / 1-16, III xviii 29-30, 17-40, 41-2 before 7.1 (Luck)

5 deiectas (codd., coni. Gebhard) / delectas; 13 laedunt / laedent; 26 quolibet ire (Schrader) / quidlibet esse; 36 male (recc., Heinsius, Luck) / mala; 39 sed (H.) / nam.

vii 1-6, vi 41-2 (Sandbach*), 7-20 / vi 41-2, 1-20

8 iure (Allen) / more; 11 ei mihi (H.), tum quales caneret tibi, Cynthia (P in ras., recc.) / ah mea tum qualis caneret tibi tibia.

viii 1-6, 9-10, lacuna (Baehrens), 17-40, 7-8 (deleted and relegated to the end (H.), 11-6 extraneous (Dietrich) and relegated to the end (H.) / 1-40

17 Properti; / Properti?; 19 manes / Manes; 31 fugam et (H. 1984) / fuga; 38 idem (Müller, Luck) / ille.

ix 1-14, 17-20, 29-30 (Housman), 21-8, 31-6, 41-8, 15-16 (deleted by Carutti and relegated to the end by H.), 37-40 (extraneous and relegated to the end by H.), 49-52 (marked as extraneous by Walker and relegated to the end by H.) / ix A 1-20, 29-30, 31-48 and ix B 49-52 (fragment)

13 adusti (H.) / tanti; 15 cum tibi / cui tum (Housman).

x lacuna (Lachmann), 1-26 / lacuna, 1-26

11 ex humili iam carmine; (H.) / ex humili; iam, carmina; 23 culmen (Passerat ex codice quodam Memmii hodie deperdito) / currum (Markland); 25 montes (Butrica 1996) / fontes.

xi lacuna (Lachmann), 1-6 / lacuna, 1-6.

xii 1-24 / 1-24

11 tutos (H.) / tuti.

xiii 1-16, lacuna (Schrader), 17-58 / 1-58

1 armatur Itura (recc., Pontanus, Passerat, Hanslik) / armantur Susa; 11 iuuat (PBT, Burman, Luck) / iuuet; 32 manes / Manes; 37 hinc (Burman in notis) / haec; 40 huc iterum (Kiessling, Fedeli 2005, 379 and 400) / hoc iter; 44 saevis (Heinsius, cf. Burman 1680, 302) de tribus una soror / quaevis de Tribus una Soror!; 45 servatur (recc., Puccius) / servetur; 47 cui si (recc., Hanslik) / qui si; 48 Iliacus Grais (Bergk) / saucius Iliacis; 53 qui / cui; 57 manes / Manes.

xiv 1-32 / 1-10, 13-14, 11-12, 15-32

26 munere (recc., Scaliger, Luck) / nomine.

xv 1-2, 7-10, 3-6 (Carutti), 11-38, 41-54, 39-40 (deleted by Guenther and relegated to the end by H.) / 1-54

1 io . . . io nox . . . io (Housman) / o . . . nox o . . . et o; 10 ista . . . meis (H.) / nostra . . . tuis; 27 vinctae / iunctae; 35 calores (Vat. Chig. H IV 123, Volscus, Beroaldus, Luck) / dolores; 47 me (recc., Heinsius) / haec.

xvi A 1-6, lacuna (Enk), 7-10.

xvi B (Havet) 15-6, 11-2 (Fontein), 17-24, 27-8 (Keil), 25-6, 31-40, 43-6, 29-30 (Carutti), 47-56, 41-2 (marked as extraneous by Fontein and relegated to the end by H.) / xvi 1-12, 17-8, 13-28, 31-46, 29-30, 47-56

34 Musa (PTS, Rothstein) / mensa; 39 Amor / amor; 49 tonitrus (Francius) / sonitus.

xvii fragments: 1-2, lacuna (Tremenheere), 3-10, 11-12 and 15-18, 13-4 (deleted by Günther [1997, 20 and 127] and relegated to the end by H. / after xxii 50: xxii 43-50, xvii 1-2, 13-4, 3-18, 18.1-4

14 nostras . . . manus / nostram necem; 15 libet (Guyetus) / licet.

xviii fragments: 1-4, 5-8+13-14+9-12+15-20 (end of the elegy), 21-2, 23-6+31-2 (Lachmann)+27-30, lacuna+35-6, 33-4, 37-8 / xviii A (1-4 after 17.18) 5-8, 13-4, 9-22 + xviii B 23-6, 31-2, 27-30, 33-8

5 iam actis aetas canesceret (H. 1984) / canis aetas candesceret; 9 undis (codd., Pinotti, cf. Hc. 187) / ulnis; 29 deme mihi / desine! mi.

xix 1-32 / 1-32

10 fanave (Guyetus) / fanaque; 18 Veneri / Veneris; 20 movere / monere; 24 tangere (H. 1984) / fallere (Watt); 25 sua formoso (Postgate) / formosa suo; 29 sic / hic.

xx 1-34 (end of the elegy), 35-6 (omitted by cod. Gron., deleted by Jacob and relegated to the end by H.) / 1-36

7 Niobae . . . superbae (codd., Luck) / Niobe . . . superba; 8 lacrima sollicito (Phillimore) defluit / sollicito lacrimas depluit; 9 me / mi; vincula (Jeverus) / bracchia; 12 insiliamque (dub. Burman in notis) / transiliamque; 23 numquam non (Keil) / non numquam; 24 numquam nonnon numquam; 28 tum me, vel tragicae, vexetis, Erinyes, et me / tum me vel tragicae vexetis Erinyes, et me; 31 sitque . . . iacere (or ligari, H.) / atque . . . vagetur; 33 nec / ne (Camps); 35 hoc . . . ius / haec . . . laus.

xxi 1-20 / 1-20 17 nunc quoque quid restat? (dub. Burman in notis) / huic quoque, qui restet.

xxii A 1-10, 13-20 and xxii B (Hall*) 35-40, lacuna (H.), 41-42, lacuna (H.), 21-34 (end of the elegy), fragments (43-50 [recc.], 11-2) / xxii A 1-10, 13-24, 11-12, 25-42 (43-50 constitute the beginning of xxvii, placed after xxii)

4 et nimis (H.) / o nimis; 39 irata . . . ministro / ingrata . . . cubili; 43 sive es (H., Goold); 48 quae (codd.) non venerit (recc.)

ipsa (Guyetus) / quasi non noverit, illa non noverit ille; 50 dicere plura / promere furta.

xxiii = xxiii 1-24, xxiv 1-16, / xxiii 1-24, xxiv 1-10, xxiv A 11-16

1 fuit indocti fugienda / fugienda fuit indocti; haec semita (recc., Burman, Hertzberg) / semita; 4 promissa / commissa; 7 pertulerit (Ayrmann) / pertuleris; 8 haec (Guyetus) scribet (H.) / ut scribat; 10 immundae . . . casae (Heinsius) / immunda . . . casa; 11 venerit (H.) / vertitur; 12 quas (recc., Wassenbergh) / quos; 14 custodum (Hemsterhusius) / custodum et; 21 immisit / mihi misit; 22 capiant ( ΠΛ, Luck) / iuverint; 23 clam (Scheidweiler) / iam; 24 liber erit, viles (or multas, H.) / nullus liber erit; xxiv 1 sic (Baehrens) / tu; 4 ingenuo est (Munro) / ingenuis; 5 tam / iam; 8 darem / daret (Barber); 12 duram . . . pilam (Luck 1979, 82) / durae . . . pilae; 13 subit iratam (Heinsius) / cupit interdum.

xxiv B 17-22, lacuna (Mueller), 23-46, 49-52, 47-8 (secluded by Postgate and relegated to the end by H.) / xxiv B17-22, 47-8, 23-46, 49-52

27 libet taetra venena (Hendry 1996) / taetra venena libens; 35 tu mea / tum me; 46 viro est (H.) / viro; 49 offerre (Heinsius) / consuesse; 51 hic / hi; in (Lachmann) / ut.

xxv 1-30, 33-44, lacuna (H.), 45-8, 31-2 (marked as suspect by Baehrens and relegated to the end by H., Hc. 219) / 1-48

2 line in parentheses (Luck) / no parentheses; fuit (Watt 1992) / vehit; 9 deducet / diducet; 14 aves? / aves.; 17 amator (Heinsius) / amor qui; 35 antiqua his grata (Dousa f.) / antiquis grata; 39 renovatis (Barber) / revocatis; 41 vidistis / vidisti; 43 patria (H.) Argivas (Barber) / quandam Argiva; 47 suis (Foster) / tuis; 48 semicolon after cuivis / unpunctuated; malum (Carutti) / mala.

xxvi A 1-20, 21-6 (fragment [van Lennep 1747, 125]), 27-8 (deleted by Jeverus and relegated to the end by H.) / 1-20, xxvi B 21-8

1 vidi ego te ( γ, Burman) / vidi te; 23 gaza Midae (Palmer) / iam Gygae; munera (Camps) / flumina.

xxvi B 29-58 / xxvi C 29-58

39 montes duo (Palmer) / duo litora; 47 quaerit (H.) / ferret; 49 nam (Vahlen) deus amplexae (Postgate) / quam deus amplexus; 51 negabit (Livineius) / negavit; 57 pro (Havet) / sit.

xxvii lacuna (Richmond), 1-9, lacuna (H.), 10-6 / 1-16

6 terrae / terrent; 7 flemus / fles tu; 8 dubius (H.) / dubias; 9 domibusque / metuisque.

xxviii 1-20, 23-32, 35-46, 33-4 (H.), 47-56, 59-62, 21-2 (deleted by Heimreich and relegated to the end by H.), 57-8 (excluded by Fontein and relegated to the end by H.) / 1-32, 35-46, 33-4, 47-56 (57-8 after 3.18.24), 59-62

5 neque (recc., Guyetus) / non; 10 semper (Markland) / prae se; 11 templa / planta; 12 Pallados / Palladis; 15 per / post; 16 hora / aura; 37 luna / Luna; 47 et (H. 1984, Goold); 51 Antiope / Antiope est.

xxix A 1-22 / xxix A 1-22.

xxix B 23-8, lacuna (Richardson 1976), 29-42 / 23-8, ii 9-12, 29-42

27 intactae (Postgate) / hinc castae; 37 e (Liberman 1992) / in; 41 excludor (Guyetus, Broukhusius, Heinsius) / deludor.

xxx fragments: 1-10, lacuna (Sh. Bailey), 11-12, 19-22, 13-18, lacuna, 23-40 / xxx A 19-22, 1-12 and xxx B 13-18, 23-40

7 semper Amor supra caput improbus instat (H.) / instat semper Amor supra caput, instat; 8 ipse (codd., Luck) / ipsa; 18 Pallados / Palladis; 20 sola (Knauth) / rauca; 33 virginibus / Virginibus.

xxxi + xxxii: 1-4, lacuna (H.), 5-16, xxxii 7-10, 1-6, 11-62 / xxxi, xxxii 1-16, xxxii 7-10, 1-6, 11-62

8 quattuor, artificis vivida signa, boves / quattuor artifices, vivida signa, boves; 10 vel (Bergk) / et; xxxii 3 Preaeneste in (Lachmann) / Praenesti; 13 creber platanis / platanis creber; 15 sonitus (H. 1994) nymphis tota crepitantibus urbe / sonitus lymphis toto crepitantibus orbe; 23 manavit (Giardina 1977) / maledixit (Schneidewin); 25 attendere (H.) / credere; 28 manere (Liberman 1992) / videre; 34 non (recc., Hertzberg) / nec; 35 deam (Clausen 2000) / illam (Barber); 44 facit? / facit!; 45 iam ante illam impune (H.) / ante illam iam impune; 47 Tatium veterem (H., Goold); 53 et / at; dum (H.) / cum; 54 at (recc., Vulpius) / et; 61 si tu es (H. from seu tu es Luck 1964).

xxxiii A 1-22 / 1-22

5-6 amantes? quaecumque, illa suis (Housman*) semper amara fuit / amantes, quaecumque illa, semper amara fuit.

xxxiii B 25-6 (Reedy*), 23-4, 27-40 (end of the elegy, H.), 41-2 (fragment, H.), 43-4 (fragment, H.) / 23-40, xxxiii C 43-44, 41-2 23 me (recc.) ecfundere (Housman*, Watt 1992) / mea ludere; 36 i (Birt 1911) / iam.

xxxiv 1-66, 77-80 (Ribbeck), 67-76, 81-2, 85-94, 83-4 (deleted and relegated to the end by H.) / 1-46, 51-4, 47-50, 55-66, 77-80, 67-76, 81-94

8 virum est / virum; 12 tu (Markland, Luck) / in; 15 socium (Cornelissen) / dominum; 16 dominum (Cornelissen) / socium; 20 quod falso (H. 1984) / et in nullo (Heinsius); 31 potius (Schrader) memorem Musis / satius Musam leviorem; Philitan / Philitae (Santenius); 39 Amphiaraëae (recc., Unger) nil (Muretus, Luck) prosint (recc.) / non Amphiareae prosint tibi; 45 aut (Stroh 1971) / tu; 46 nostra (H.) / recta; 52 frenatis (Allen 1975) luna / fraternis Luna; 59 me iuvat (PJC, Luck) / me iuvet; 61 Vergilio est (Baehrens) / Vergilio; custodis / cordi sit (Butrica 1997); 85, 87, 89 sic . . . sic . . .. sic (H.) / haec . . . haec . . . haec; 83 aut sim <. . .. . .. . .. . .> / aut sit minor ore, canorus.

BOOK III

i 1-38 / 1-38

1 manes / Manes; 7 in armis. / in armis!; 22 honos / Honos; 29 Pulydamantos (Postgate) / Pulydamantis.

ii 1-26 / 1-26

3 detinuisse / delenisse; 5 Thebanam (Heinsius) / Phoebeam; 16 et / nec.

iii 1-52 / 1-51

6-12 no parentheses / in parentheses; 41 ne (Passerat) / nil; 48 morae (H.) / fugae.

iv 1-13, lacuna, 17-8, lacuna (H.), 14-16, 19-22 / 1-13, 18, 17, 14-16, 19-22

4 tua / nova; 8 ad (Eldikius) / et; equos (Baehrens; armigeros . . . equos, already in Hertzberg 1702, 266) / equi; 11 sanctae (Postgate) / sacrae.

v 1-39, 42, 41, 40 (Housman), 43-8 / 1-2, 5, 4, 3, 6-39, 42, 41, 40, 43-8

2 stant / sat; 9 in arto (Housman) / in arte; 11 incauti (Alton) / in tantum; 14 at (Schrader, Luck) / in; 27 decidit (Hutchinson 1986) / defici; 31 si (recc., Volscus 1482, alii) / sit; arcem / arces; 39 si (recc., Lachmann) / sint; 47 superest / superet.

vi 1-2, 5-12, 3-4, lacuna (H.), 13-42 / 1-2, 5-8, 3-4, 9-11, 14, 13, 12, 15-42

9 sic illam (Havet) / sicin eram; 13 et (Keil) / ac; lacertis, / lacertis?; 19 sono. / sono?; 28 exsucis (Burman) / exsuctis; 29 recentia (Heinsius) / iacentia; 30 raptaque (H.) / cinctaque; 40 ipse (Housman) / esse; 41 tanto / e tanto.

vii 1-8, 29-36, 19-20 (Housman), 37-42, 47-53, 52, 51, 54-64, 17-18, 65-66, 43-46 (H.), 67-70, 13-16 (Postgate), lacuna, 11-12, 9-10 (H.), 25-28 (Carutti), 71-72, 21-24 (deleted by Willymot and relegated to the end by H.); / 1-8, 43-50, 53, 52, 51, 54-66, 17-18, 9-16, 67-70, 25-36, 19-24, 37-42, 71-72

1 vitae es / vitae; 33 Penates / penates; 47 nunc tulit (Barber) et (H.) / noluit hoc; 50 effultum est (H.) / effultum; 57 Aegaei et (Voss) / Aegaei; aequora venti / aequora, venti,; 66 ultimaque haec (Beroaldus) / ultima quae; 43 patrio / patrios.

viii 1-2, 5-8, 3-4 (H.), 9-24, 27-34, 37-40, 25-6 (secluded by Dorvillius and relegated to the end by H.), 35-6 (deleted by Burman and relegated to the end by H.) / 1-2, 5-8, 3-4, 9-40

13 grege si (Butrica 1984, 85) / grege seu; 14 Maenas ut icta vias / maenas ut icta, vias; 31 Dardanus (dub. Heinsius 1742, 722) / barbarus.

ix 1-20, 23-34, 21-2 (H.), 35-46, 57-8 (H.), 47-56, 59-60 / 1-32, 33-4 after II i 38, 35-48, 51, 50, 49, 52-60

7 nervis (Palmer) / rerum; 8 haec ex quo (Sandstroem) / ex aequo; illa (recc., Reid) / ulla; 11 ponit / poscit; 12 iocum (Lachmann) / locum; 17 contingit (Vannini) / concurrit; 23 tu cum (H.) / cum tibi; 25 arcus (Helvetius) / hastas; 37 tepentes (H. in Morgan, CQ 36, 1986, 188-9, n. 28) / paternos; 38 septem (Lipsius) / semper; 48 Oromedonta / Eurymedonta; 59 nunc (H.) / hoc; laudes (H.) / laudis.

x 1-32 / 1-32

1 risissent (Passerat) / visissent; 12 praesentes (‘quidam libri teste Passerat’ [Barth 1777, 221], Luck) / poscentes; 13 ac / at; 21 surgat (Cornelissen) / currat; 24 adsint (H.) / et sint; 25 convicia (Broukhusius ut vid., cf. Burman 1780, 582) / convivia.

xi 1-4, 7-57, lacuna, 58 (omitted as in ν 65-8 (Housman), 59-64, 69-72, 5-6 (deleted by Georg* and relegated to the end by H.) / 1-57, 58 ( stat non humana deicienda manu), 65-8, 59-64, 69-72

10 armifera (Livineius) / armigera; 17 quin etiam (Heinsius) / Omphale; 22 et . . . sustulit (Giardina 2005, 272) / ut . . . tolleret; 23 ut (Puccius [attributed by Hanslik and Giardina to ν] et; 29 qui (Baehrens) / quae; 32 patres. / Patres?; 35 tua (Hoeufft) / ubi; 37 quam ibi (H.) / tibi; 38 sua . . . erat (Butrica 1993) / tua . . . eras; 40 Philippei sanguinis usta (H.) / Philippeo sanguine adusta; 41 nostro est (H.) / nostro; Anubin / Anubim; 52 accepere tuae / nec cepere tuae; 56 dixerat (Housman) / dixit et; 58 omitted by ν stat non humana deicienda manu; 65 condiderant / condiderunt; 59 [fera tela] (H.) / monumenta; 70 tantum operis belli / tanti operis bellum; 5 venturam / ventorum; mortem / morem.

xii 1-38 / 1-38

7 immunda (H.) / iniecta; 17 marito (Heinsius) / timore; 28 alterna . . . aqua (dub. Camps 2001, 114) / alternas . . . aquas; saeva (Faltin) / scissa; 34 lyras (H.) / lacus.

xiii 1-37, lacuna, 39-40, lacuna (H.), 38, 41-66 / 1-66

8 culti messor (Fontein) / multi pistor; 9 puellas (Markland) / pudicas; 21 gaudent (Stephanus, Enk) / ardent; 35 iunctos (Sh. Bailey) / stratos; 37 lentis (Baehrens) / laetas; 39 Arcadii (Hertzberg) / atque adeo (Watt); 42 dextris (H.) . . . focis / vestri . . . foci; 53 mons / mox; 59 falsus (recc.) / verus; 61 nempe (Housman*) Ilia / neque vilia; 62 Pergamei . . . mali (Housman*, Luck 1964); Maenas / maenas.

xiv 1-16, lacuna (H.), 17-34 / 1-10, 15-6, 11-4, 17-34

5 flectit (H.) / fallit; 14 turma (Gulielmius, Heinsius, Ramírez de Verger 1989, 209, Luck) / turba; vagatur agris (Heinsius) / lavatur aquis; 23 est ulli (dub. Broukhusius 1702, 289) / aut ulla est; 31 sint / sit; dent verba (Enk) / sit acerba (Burman sen.).

xv 3-10, 45-6 (Fisher), 1-2 (H.), 43-4 (Vulpius), 11-42 / 1-46

4Amoris / amoris; 7 iam (Postgate) / cum; 11 vero / vano (Franz); 33 subtractae (Richardson 1976) / sollicito (Nairn); 35 sera tamen, pietas / sera, tamen pietas; 39 tibi gloria; Dirce / tibi gloria Dirce; 40 obitura (Heinsius) / habitura.

xvi 1-14, 19-20 (Struchtmeyer), 15-18, 21-30 / 1-14, 19-20, 15-18, 21-30

7 haec distulero / distulero haec; 12 media . . . via / medias . . . vias; quis (Watt 1992) / his; 20 ecce suis (Fisher) it (Dorvillius) / et cuius sit; 16 concutit (Francius) / praecutit; 21 casus / cursus; 22 talis / tali.

xvii 1-2, 7-8 (H.), 3-6, 9-42 / 1-42

7 amoris (Burman) / in astris; 12 animum (recc., Puccius) / animos; 15-16 colles . . . vites, quas (Guyet) / vites . . . colles, quos; 24 et triplici funera grata gregi (H.) / in triplices funera rapta (Scaliger) greges; 29 onerabo (recc., Passerat) / onerato; 38 libabit (Foster) / libatum.

xviii 1-8, lacuna (Guyetus), 9-28, 31-4, 29-30 (extraneous here, according to Scaliger, and relegated to the end by H.) / 1-24, II xxviii 57-8, 25-28, (29-30 after II vi 16), 31-3

1 ludit / tundit; 2 fumida[que exundant] (H.) / fumida Baiarum; 3 Euboica tubicen Troianus (dub. Heinsius, 1742, 733) / et Troiae tubicen Misenus; 5 olim, [Hesperias] (H.) / ubi, mortales; 9 his pressus / Marcellus; 14 maturas (Barber) / maternas; 19 omnia conchis / ostra smaragdis; 22 via (recc., Burman) / via est; 31 qui / quo (Jacob); 32 hac (Guyetus) / huc.

xix 1-14, 17-20, 15-6 15-6 (Postgate), 21-8 / 1-28

19 [tuum facinus] (Housman) / Clytaemestrae; 22 tondens purpuream (Markland, Luck), regna paterna, comam (Markland, Luck) / tondes purpurea regna paterna coma; 24 Amor (Luck) / amor.

xx 1-14, 19-20 (Lachmann), 15-8, 21-30 / 1-10, 13-14, 11-12, 19-20, 15-18, 21-30 5 stulta, deos/stulta adeo es; 22 vigilata (Palmer) / vigilanda; 25 tacta sic . . . ara (Heinsius) / tactis haec . . . aris.

xxi 1-34 / 1-34

4 Amor / amor; 6 possit / posset; exsomnis (Barber) / ex omni; usque (Heinsius) / ipse; 7 bis (Cornelissen) / vix; 10 Amor / amor; 19 Lechaei (Guyetus, Luck) / Lachaeo; 22 isthmos / Isthmos; 25 in (H.) spatiis (Broukhusius) / vel stadiis; 28 munde (Kuinoel) / culte; 31 aut / et.

xxii 1-5, lacuna, 15-16, lacuna, 6 (H.), 7-14, 17-36, 39-42, 37-8 (deleted by Knoche and relegated to the end by H.) / 1-6, 15-16, 7-14, 17-36, lacuna, 37-42

3 e vite (Haupt, Luck) / in vite; 15 sive et (H.) / si tibi; 6 et / nec; 9 Geryonae (recc., Volscus) / Geryonis; 11 Phasin / Phasim; 13 immissa (H. 1986) / Argoa; 38 curvatas / curtatas; sua / fera.

xxiii 1-24 / 1-24

6 atque eaedem (Heinsius) / et quaedam; 14 bene (recc., Guyetus, Luck) / bona; 16 paravit / parabit (Heinsius); 18 ducitur / dicitur; iocis (recc., Heinsius) / dolis.

xxiv 1-8, 11-14, 9-10 (Tremenheere), 15-20, xxv 1-12, 15-18, 13-14 (interpolated from Tib. 1.8.45 and relegated to the end by H.) / xxiv 1-20, xxv 1-18

2 elegis (Schrader) . . . meis / oculis . . . tuis (Burman); 6 cum (Dousa f.) / ut; Amor / amor; 11 haec / hoc; nunc . . . nunc (H.) / non. . . non; 12 verba loquebar (H.) / vera fatebor; xxv 6 ad insidias (Baehrens) / ab insidiis.

BOOK IV

i A 1-18, lacuna (H.), 21-38, 55-6 (Lange), 39-52, 87-8 (Mueller), 53-4, 61-2 (H.), 57-60, 63-70, 19-20 (deleted and relegated to the end by H.) / 1-33, 36, 35, 34, 37-40, 47-8, 41-6, 49-52, 87-8, 53-70

4 concubuere / procubuere; 7 pater / Pater; 8 murus (H. 1986) / bubus; 9 nunc (H.) / qua; se sustulit; olim / se sustulit olim; 11 curia / Curia; 12 patres / Patres; 23 pauca (Lachmann) / parva; 33 Vrbe / urbe; 38 putat / putet; 45 tunc / hinc; 46 venit et ipsa, sui Caesaris arma Venus (Hollis) / vexit et ipsa sui Caesaris arma Venus; 52 lata (H.) / vera; 88 candida regna (Murgia 1989) canam / regna superba cano; 65 si quis / quisquis; cernet (F, Hutchinson 2006) / cernit; 19 celebrare / celebrante.

i B 71-81, 82 Iuppiter <. . .> (Sandbach*) obliquae signa iterate rotae, 83-6, 89-124, 127-50, 125-6 (interpolated [Richmond] and relegated to the end [H.]) / 71-6, 83-4, 77-82, 89-102, 85-6, 103-40, 143-4, 141-50

77 Orops / Horops; 81 fallitur / fallimus; 94 bene . . . sibi (H.) / sibi . . . bene; equo / equum; 103 haec / hoc; 112 Atrides / Atridis; 113 tu diruta fletum / tu, diruta, fletum; 119 devehor (recc., edd. vett., Hutchinson 2006) / devehar; 128 Lares / lares; 135 fallax / pellax; 139 pararis (Murgia 1989) / parasti; 141 decusserit (H., ut vid.: Hc. 432) / decusseris (Broukhusius); 142 ipsa (H.) suo / ansa tuo.

ii 1-12, 19-44, 13-18 (H.), 45-7, lac., 51-4, lac., 48-50 (H.), 55-64 / 1-4, 51-4, 49-50, 55-56, 5-12, 19-32, 35-6, 33-4, 37-42, 13-18, 43-8, 57-64

2 paterna dei / fatente deo; 4 proelia / proelis (typographical error, ut vid.); 6 forum / Forum; 9 stagnum (Housman*) / spatium; 10 Vertumnus / Vertamnus; 12 Vertumni / Vertanni; 19 falsa es (Lachmann) / alius; 34 Faunus (recc., Heinsius) / fautor; 35 Vertumnus / cum verbere; 40 scirpiculis (recc., ed. Ven.) / sirpiculis (recc., Burman); 52 atque / quoque; 57 suberunt (H.) / superant; 60 pauper pauper (H.) / grata pauper.

iii 1-36, 39-40, lacuna (H.), 51-6 (H.), 41-50, 57-72, 37-8 (deleted and relegated to the end, H.) / 1-72

8 ferreus (Postgate) / Persicus; 10 ustus / tu[n]sus; 11 foedera nobis (Watt 1992) / munera nuptae; 34 lecta (Heinsius) / secta; tuas (Lee 1994, 110 ‘for your parade cloak’) / suo; 53 limina (Heinsius) / omnia; Kalendis / kalendis; 56 tori . . . tuam (H.) / tui . . . toro; 46 issem (Heinsius) / essem; 49 rapto (Hoeufft) / adempto; 72 salva (Burman ad v. 61 on p. 763) / grata.

iv 1-2, 9-14 (Sh. Bailey), 7-8 (Postgate), 3-6 (Baehrens), 15-16, 19-86, 17-8 (Housman), 87-94 / 1-2, 9-10, 13-14, 11-2, 3-8, 15-16, 19-82, 85-6, 83-4, 87-92, 17-8, 93-4

2 nemus / scelus; 11 namque (H.) / atque; 12 foro / Foro; 14 e vivo (Waardenburgh) / ex illo; 7 fontem / contra; 15 hunc (Canter) / hinc; fontem / laticem; libarat (Fontein) / libavit; 27 cumque (H. 1999) / dumque; 29 sua Tarpeia / Tarpeia sua; 30 vicinae . . . Iovis (H. 1999) / vicino Iovi; 47 potabitur (Palmer, Luck) / pigrabitur; 48 tu / tum; 50 semper / caespes; 52 hanc (Baehrens) / haec; 55 dic (Passerat) / sic; spatierne / spatiorne; 73 urbi / Urvi; 77 sacros (Passerat) / raros; 83 ascensu / ascensus (Jacob); dubius / dapibus (Jacob); 88 ipsa / ipse; 94 iniustae . . . sortis / iniuste . . . mortis (dub. Lachmann).

v 1-19, lacuna (Richardson 1976), 20-54, 55-6 (omitted), 57-78 / 1-28, 45-6, 31-44, 29-30, 47-54, 55-6 (omitted), 57-78

2 perpetuam (H. 1999) / quod non vis; 3 manes / Manes; 7 quae (Guyetus) / quoque; 19 verbis / tenebris; tu (Barber) blanda peruris (H. 1999) / ceu blatta papyron; 20 saxosamque / suffosamque; terat (recc., Beroaldus) / forat; turba (Vat. Barberinianus VIII 23) / talpa; 21 si te Eoa lecta lapis / chrysolithus si te Eoa; 29 moram (Lütjohann) / virum; 61 odoratum . . . Paestum (Schippers, Luck) / odorati . . . Paesti; 71 fuerint (Graevius, Luck) / fuerunt (Passerat); 74 claustra (recc., Beroaldus, Rothstein) / clatra.

vi 1-86 / 1-86

17 celebrant (H.) / Leucas; 18 nunc onerata via est (H.) / non operosa via; 20 pinea; / pinea,; 28 nam / non; 45 audent prope / audent: pro; 49 quodque / quotque; 55 hostes (H.) / arcus; 72 blandae utrimque (Lachmann) / blanditiaeque; 75 irritet / irritat; 79 confectum (Livineius) / confessum; 80 reddit (recc., Ayrmann) / reddat; 81 aequus (Housman) / aliquid.

vii 1-96 / 1-40, 47-8, 41-6, 49-96

1 manes / Manes; 2 exstinctos (Passerat) / evictos; 11-2 (at illi . . . manus) / at illi . . . manus; 19 trivio est (H.) / trivia; 23 euntes / eunti; 33 hocne (Hall*) / hoc; 57-8 cumba (Rossberg) Clytaemestram (Lütjohann) stuprumve in Tartara (Weidgen) Cressae / portat mentitae lignea monstra bovem (Postgate) / unda Clytaemestrae stuprum vehit altera, Cressam / portans mentitam lignea monstra bovis; 59 vecta (recc., Hanslik) / rapta; 64 foedera (Heinsius) / pectora; 79 tumulo mihi, quae / tumulo, mihi quae; ramosis . . . pomifer / pomosis . . . spumifer.

viii 1-18, 21-88, 19-20 (deleted by Butler and relegated to the end by H.) / 1-2, 19-20, 3-8, 11-2, 9-10, 13-18, 21-88.

4 ubi / tibi; 5 nam (H.) / qua; 6 qua / hac; virgo (tale iter omen habet) Housman*) / (virgo, tale iter omne cave!); 10 tenera (recc., Scal.) / temere; raditur (Cornelissen) / creditor; 17 dic, quaeso / dic quaeso; 23 sed vaga iam (Bonazzi) / serica nam; 31 inter stat (H.) / est inter; 39 Nile, tuus tibicen erat / Miletus tibicen erat; Orontes (Morgan*) / Byblis; 41 contractus / concretus; 72 cum / cui; 76 forum / Forum; 78 nudet (Koch) / se det; operta (recc., Koch) / aperta; 82 risit et (H. from risit at Burman) / riserat); 84 suffiit et (recc., Beroaldus) / suffiit ac; 88 res pacta (Mueller) / despondi (Puccius); toto / noto (Heinsius).

ix 1-41, lacuna (Scal.), 65, 66 (= 42) [65-6 transp. Jacob], 43-64, 67-70, 73-4, 71-2 (Passerat) / 1-41, 42 ( et gemere abstractum Dite vetante canem?), 65-6, 43-70, 73-4, 71-2

5 quoque / quaque; 11 manifestaque (Luck 1962) / manifestae; 13 deo furtum est (H. 1986) / deo : furem; 18 quaesiti (H., Goold); 20 forum / Forum; 24 murus (Fontein) / lucus; saepserat (Fontein) / fecerat; 36 suscepto / succepto; 40 notas (recc., Broukhusius) / vastas; 43 etsi (H.) / quodsi; 45 autem (H.) / aliquam.

x 1-48 / 1-22, 25-6, 23-4, 27-48

5 primus (Jeverus) / primae; 6 exuviis (recc.) / exuvio; 15 hic / haec; 18 Lare / lare; 19 equus (Guyetus) / eques; 23 inficitur (Phillimore 1916) / insequitur; 32 ab hoste (H.) petit (Heinsius) / ab urbe dedit; 34 vineaque inductum (recc., Volscus, Beroaldus) / vinea qua ductum; 39 Eridano (Passerat) / at Rheno; 42 nobilis (recc., Schuster) / mobilis; evectis (Rothstein) / e rectis; 46 ipse (Damsté) / ense; 47 huc (Broukhusius) / haec.

xi 1-32, 43-4 (Peerlkamp), 33-6, lacuna (H.), 65-6 (H.), 37-42, 45-62, 97-8 (Peerlkamp), 63-4, 67-8, 71-2 (Baehrens), 69-70, 73-96, 99-102 / 1, 6-7, 4-5, 2-3, 8-60, 65-6, 61-2, 97-8, 63-4, 67-8, 71-96, 69-70, 99-102

3 leges / sedes; 8 umbrosos . . . locos (recc., Markland) / eversos . . . rogos; invida (Boot) Parca (Markland) / lurida porta; 9 sat (Dousa pater) / sic; 13 Parcas? / Parcas:; 15 nocti (Sandbach) / tenebris; sedes (Butrica 1984, 199-200) / et vos; Acherontis (Butrica 1984, 200) / paludes; 16 aut (H.) / et; unda / ulva; 27 loquor / loquar; 31 materni hos (Heyne) / maternos; 66 consule quo, fausto (Peerlkamp) tempore, / consul quo factus tempore; 39 et <. . .. . .. . .. . .., et illum> (H.) / et, Persen proavi stimulat dum pectus Achilli,; 41 nec / neque; 53 cuius / cui; sacros (Rothstein) / sacra suos; 61 generosos / generosae; 64 manu (Scaliger) / sinu; 73 natos / Paulle; 82 reddita (Graevius) / credita; 89 laudate: nimis collate / laudate nimis: collate; 103 aquis / avis.5

Notes

1. And if we prefer to go back to older editions, ‘Google books’ will give us access to a number of editions and commentaries (e. g., Broukhusius, 1702; Barth, 1777; the ‘Notae variorum ex editione Graevii at Propertii opera omnia ex editione Kuinoelis. . . in usum Delphini’, 1822; Jacob, 1827; Paldamus, 1827; Lemaire, 1832; Hertzberg, 1845).

2. I ix 26 (H. 15) es must be a typographical error, as Hc. (523) translates ‘once a woman is yours’; xiii 8 (H. 19) read -o, not -us in the app. crit.; xv 29-32 (H. 23) read alienos for alicnos; II xxii B 35 (H. 71) read ministret, not minstret; IV vi, not v (H. 168); straightforward, not staight- (Hc. 84); at Hc. 158 there should, I believe, be a comma, not a full stop, after ‘others’ at the end of l. 20.

3. The apparatus criticus does not include the manuscripts’ later hands, and the numerous conjectures presented in the apparatus come from the indispensable Thesaurus by Smyth (1970), from later scholars cited in the bibliography, or from unpublished conjectures by scholars, which are marked by an asterisk. To these must be added the conjectures that Housman noted down in his copy of the 1880 edition by Baehrens (Hc. xi-xii) but never got round to publishing (made public here for the first time).

4. Hanslik’s edition of 1979 (cf. Kenney, CR 31, 1981, 201; La Penna, Gnomon 54, 1982, 516) is very useful in this regard, since the chronological system of Greek lettering as used by Mynors for Catullus and by Butrica in some of the elegies of Propertius seems to me just as imprecise as that used by H. and other editors.

5.This review has been translated from the Spanish by J. J. Zoltowski. Thanks are due to the Spanish MEC (FFI2008-01843) and the Junta de Andalucía (HUM2005-04375) for their financial support. I am deeply grateful to Professors G. Luck and L. Rivero García for their many valuable suggestions and corrections. The remaining errors are all my own.