=41. DOMESTICVS.= The word is rare in verse; Ovid uses it as a substantive at iii 15 'ille ego conuictor densoque _domesticus_ usu'. Here Ovid may be recalling the language of _Met_ VI 685-86 (of Boreas) 'ira, / quae solita est illi nimiumque _domestica_ uento'.

=42. VIRES.= Merkel proposed MORES, citing Virgil _G_ I 50-52 'at prius ignotum ferro quam scindimus aequor, / uentos et uarium caeli praediscere _morem_ / cura sit' and Statius _Sil_ III ii 87 'quos tibi currenti praeceps gerat Hadria _mores_'. The second passage is not to the point, since it means 'what sort of obedience to your wishes do you expect from the Adriatic as you make your voyage'. In any case, Professor R. J. Tarrant points out to me the poor logic of Merkel's proposed text: Ovid is deriving the _natura loci_ from its surroundings; he should not now be saying that Boreas gets his _mores_ from the area.

The reading of the manuscripts seems acceptable enough if one accepts Meynke's _polo_ for _loco_ ('he gathers strength from the nearby North Pole'). For _sumit uires_ compare _Met_ VIII 882 (Achelous speaking) 'armenti modo dux _uires_ in cornua _sumo_', _Met_ XI 510-11 'ut ... solent _sumptis_ incursu _uiribus_ ire ... feri ... leones' and Hor _Ep_ I xviii 85 'neglecta solent incendia _sumere uires_'. Professor R.

J. Tarrant compares such phrases as _sumere iras_ (_Met_ II 175), _animos_ (_Met_ III 544-45), and _cornua_ (_AA_ I 239, _Tr_ IV ix 27).

=42. POLO= _Meynke_ LOCO _codd_. The pointlessness of _loco_ is made clear enough by Wheeler's 'and he takes on strength from a place nearer to him'. Meynke's _polo_ removes the difficulty, answers well to the following 'at Notus, _aduerso_ tepidum qui spirat ab _axe_', and is supported by the language of _Met_ II 173 'quaeque _polo_ posita est glaciali _proxima_ Serpens', and _Fast_ IV 575-76 (of Ceres) 'errat et in caelo liquidique immunia ponti / adloquitur gelido _proxima_ signa _polo_'. For the corruption, compare the common misreading of _locum_ for _solum_.

=43. ADVERSO ... AB AXE.= Ovid here seeks a contrast with _polo_ in the previous line; but clearly he means only that the south wind comes from the opposite direction, not that it originates at the South Pole.

Bentley conjectured AVERSO for _aduerso_, and the two words are obviously prone to interchange: compare _Tr_ I iii 45 (of Ovid's wife, after his departure) 'multaque in auersos [_Heinsius_: aduersos _codd_]

effudit uerba Penates' and the variations among the manuscripts at Virgil _G_ I 218 'auerso ... astro', _Aen_ XII 647 'auersa uoluntas', and Sen _Tr_ 1123 'auersa cingit campus' (on which see Housman 1076). But _aduerso_ 'opposite' seems to have the sense required here.

=43. TEPIDVM QVI SPIRAT.= For the construction compare _Met_ IX 661 'sub aduentu _spirantis lene_ Fauoni' and Avienus _Descr Orb_ 847 'uel qua _lene_ Notus _spirat_'. The trivialized TEPIDVS QVI SPIRAT is found in _MH2c_. _Tepidus Notus_ occurs four times in Ovid (_Am_ I iv 12, I vii 56 & II viii 20, and _Tr_ III xii [xiii] 42).

=44. LANGVIDIORQVE VENIT.= Compare _EP_ II i 1-2 'Huc quoque Caesarei peruenit fama triumphi, / _languida_ quo fessi uix uenit _aura Noti_'.

=46. AB AMNE.= Similar instrumental uses of _ab_ at _Her_ X 138 'tunicas lacrimis sicut _ab imbre_ graues', _AA_ III 545 'ingenium placida mollitur _ab arte_', _Met_ I 65-66 'contraria tellus / nubibus assiduis pluuiaque madescit ab Austro', _Met_ IV 162-63 'pectus ... adhuc _a caede_ tepebat', and _Fast_ V 323 'caelum nigrescit _ab Austris_'.

=47-58.= For the lengthy catalogue, typical of Ovid, compare the listing of Actaeon's dogs at _Met_ III 206-25 (in particular at 217 'et Dromas et Canache Sticteque et Tigris et Alce') and the catalogue of trees that came to listen to Orpheus sing (_Met_ X 90-107).

=47. LYCVS.= A number of rivers had this name in the ancient world. Ovid presumably means the Paphlagonian Lycus referred to by Virgil at _G_ IV 366-67 'omnia sub magna labentia flumina terra / spectabat diuersa locis, Phasimque Lycumque ...'.

=47. SAGARIS.= The modern Sakarya; it flows into the Black Sea about 125 kilometres east of Istanbul. It is mentioned at Pliny _NH_ VI 1 4 'Sangaris fluuius ex inclutis. oritur in Phrygia, accipit uastos amnes ... idem Sagiarius plerisque dictus'.

=47. PENIVSQVE.= The 'flumen et oppidum Penius' are mentioned at Pliny _NH_ VI 14 as being in the region of the Caucasus on the Euxine coast; nearby were 'multis nominibus Heniochorum gentes'. The river seems not to be mentioned elsewhere in ancient literature.

=47. HYPANISQVE.= The modern Bug empties into the Black Sea about 50 kilometres east of Odessa. It is mentioned again by Ovid at _Met_ XV 285-86 'quid? non et Scythicis Hypanis de montibus ortus, / qui fuerat dulcis, salibus uitiatur amaris?' and Virgil _G_ IV 370 'saxosumque sonans Hypanis'.

=47. CALESQVE.= Isaac Vossius made this correction for the manuscripts'

CATESQVE (_I_ has CHARESQVE) on the basis of 'Eustathio Scholiis in Periegeten'. Heinsius aptly cited a description of the occasionally violent flow of the river at Thucydides IV 75 2.

As indicated by this passage, the modern Alapli flows into the Black Sea near Eregli, about 200 kilometres east of Istanbul.

=48. CREBRO VERTICE TORTVS HALYS.= An imitation of _Aen_ VII 566-67 'fragosus / dat sonitum saxis et _torto uertice_ torrens'. _Tortus_ when used of water generally refers to the disturbance caused by rowing (_Fast_ V 644; Catullus LXIV 13; _Aen_ III 208).

=48. HALYS.= The modern Kizil Irmak flows into the Black Sea about 600 kilometres east of Istanbul. Andre compares Apollonius' description of the river (II 366-67) '[Greek: rhoai Halyos potamoio / deinon ereugontai]'.

=49-50.= The three rivers mentioned in these lines are all named for their swiftness.

=49. PARTHENIVSQVE RAPAX.= The modern Bartin flows into the Black Sea about 280 kilometres east of Istanbul and about 240 kilometres west of Sinop. It is in fact a very calm river: this information was available to Ovid from Apollonius II 936-37 '[Greek: Parthenioio rhoas halimureentos, / preutatou potamou]' (cited by Andre).

=49. VOLVENS SAXA.= Similar phrasing at _Met_ VIII 552-53 '[undae ...]

ferre trabes solidas obliquaque _uoluere_ magno / murmure _saxa_ solent'.

=49. CINAPSES= _BC_ CINAPSIS _L_ TYNAPSES _H_ CINASPES _FIT_ NIPHATES _M_.

Editors read CYNAPSES; but since the river is not otherwise known, restoration is dangerous. _M_'s reading looks like an interpolation from Lucan III 245 'Armeniusque tenens _uoluentem saxa_ Niphaten' (cited by Micyllus).

=50. NVLLO TARDIOR= = _uelocior omni_; Andre mistranslates 'le plus lent des fleuves'. Compare _Tr_ I v 1 'O mihi post nullos umquam [_uar_ ullos numquam] memorande sodales' and _EP_ I iii 65-66 'Zmyrna uirum tenuit, non Pontus et hostica tellus, / paene _minus nullo_ Zmyrna petenda loco'.

=50. TYRAS.= The modern Dnestr flows into the Black Sea about fifty miles south of Odessa; near its mouth is the city of Ovidiopol. The river is briefly mentioned at Pliny _NH_ IV 82 & 93, and at Mela II 7, where it is called the 'Tyra'; this however seems to be a scribal error induced by the following _separat_.

=51. THERMODON.= The modern Terme flows into the Black Sea about 100 kilometres southeast of the mouth of the Kizil Irmak (Halys). It was conventional to mention the Amazons in connection with the river (_Met_ XII 611, _Aen_ XI 659-60, Prop III xiv 13-14, Ammianus Marcellinus XXII 8 17). Professor E. Fantham suggests to me that Ovid may here be providing Albinovanus with material for the part of his _Theseid_ dealing with Theseus' expedition against the Amazons.

Ovid also mentions the Thermodon at _Met_ I 248-49 (the story of Phaethon) 'arsit et Euphrates Babylonius, arsit Orontes / Thermodonque citus Gangesque et Phasis et Hister'. As in the present distich, the Thermodon and Phasis, both prominent in mythology, are mentioned together.

=51. TVRMAE= _BCM_ TVRBAE _FHILT_. There is a similar variation among the manuscripts at _AA_ III l-2 'Arma dedi Danais in Amazonas; arma supersunt / quae tibi dem et _turmae_, Penthesilea, tuae'. From other descriptions of the Amazons, the Auctor Electorum Etonensium aptly compares Val Fl IV 603 (_cateruas_) and 607 (_turma_); compare as well Statius _Sil_ I vi 56 (_turmas_). It is possible that _turma_ should be read at Prop III xiv 13-14 'qualis Amazonidum nudatis bellica mammis / Thermodontiacis _turba_ lauatur aquis'; but this would make _bellica_ redundant.

=53. BORYSTHENIO ... AMNE= = _BorYsthenE_. The river is the modern Dnepr, which flows into the Black Sea about 120 kilometres east of Odessa, about 50 kilometres east of the mouth of the Bug (Hypanis). For the metrical device here employed, compare Prop II vii 17-18 'hinc etenim tantum meruit mea gloria nomen, / gloria ad hibernos lata _Borysthenidas_', Avienus _Descr Orb_ 448 'inde _Borysthenii_ uis sese _fluminis_ effert' & 721 'ora _Borysthenii_ qua _fluminis_ in mare uergunt'.

=53. LIQVIDISSIMVS= is not found elsewhere in Ovid.

=53. DIRAPSES.= The river is not mentioned elsewhere.

=54. MELANTHVS.= The modern Melet Irmak flows into the Black Sea about 25 kilometres west of Trabzon (Trapezus). It is mentioned in passing at Pliny _NH_ VI 11.

=55-56. QVIQVE DVAS TERRAS, ASIAM CADMIQVE SOROREM, / SEPARAT ET CVRSVS INTER VTRAMQVE FACIT.= The Tanais (Don) is named as the border between Europe and Asia by Pliny (_NH_ IV 78) and Avienus (_Descr Orb_ 28 & 861). Compare as well Lucan III 272-76 'qua uertice lapsus / Riphaeo Tanais diuersi nomina mundi / imposuit ripis Asiaeque et terminus idem / Europae, mediae dirimens confinia terrae, / nunc hunc, nunc illum, qua flectitur, ampliat orbem'.

Vibius Sequester (_Geog Lat min_ [Riese] p. 212) has an entry 'Hypanis Scythiae qui, ut ait Gallus "uno tellures diuidit amne duas": Asiam enim ab Europa separat'. The Hypanis cannot be the river Ovid is here referring to, for it has already been mentioned in 47; but, as Lenz saw, the line from Gallus could well have been in Ovid's mind as he wrote this passage. Professor R. J. Tarrant notes that the extraordinary _Cadmique sororem_ could well be a borrowing from the earlier poet.

=57-58. INTER MAXIMVS OMNES / CEDERE DANVVIVS SE TIBI, NILE, NEGAT.= A similar conjunction at _Tr_ III x 27-28 'ipse, papyrifero qui non angustior amne, / miscetur uasto multa per ora freto'. Herodotus compares the courses of the Nile and the Danube, concluding '[Greek: houto ton Neilon dokeo dia pases tes Libyes diexionta exisousthai toi Istroi]' (II 34), referring to the length of the rivers, however, rather than their volume of discharge. At _NQ_ III 22 Seneca mentions the belief of some that because of their large size and the fact that their sources were both unknown the Nile and the Danube must both have been formed at the creation of the world, unlike other rivers. At IV 1 1-2 he argues against those who equated the two rivers, pointing out that the source of the Danube was known to be in Germany, and that the two rivers flood at different times of the year.

=59. COPIA TOT LATICVM QVAS AVGET ADVLTERAT AQVAS.= The comparative freshness of the waters of the Black Sea was well known in antiquity.

Besides the passages cited at 37-38, see Polybius IV 42 3 and Philostratus _Imag_ I 13 7.

=61-62. QVIN ETIAM, STAGNO SIMILIS PIGRAEQVE PALVDI, / CAERVLEVS VIX EST DILVITVRQVE COLOR.= Ovid's drinking water was, on the other hand, rather brackish: 'est in aqua dulci non inuidiosa uoluptas: / aequoreo bibitur cum sale mixta palus' (_EP_ II vii 73-74).

=63. INNATAT VNDA FRETO DVLCIS.= Similar wording at Macrobius _Sat_ VII 12 32 'superficies maris, cui dulces aquae _innatant_, congelascit'.

=64. PONDVS= _B1CMFHT_ NOMEN _ILB2_. Wakefield conjectured MOMEN on the basis of Lucretius VI 473-74 'quo magis ad nubis augendas multa uidentur / posse quoque e salso consurgere momine ponti'. But _pondus_ seems appropriate to the context in a way that _momen_ 'heaving' does not.

_Nomen habe(n)t_ is a frequent line-ending in Ovid, occurring some twenty-five times (once in _Her_ XVI). _Proprium nomen_ occurs in Ovid at _Fast_ V 191-92 (Ovid is addressing Flora) 'ipsa doce quae sis.

hominum sententia fallax: / optima tu _proprii nominis_ auctor eris'

and _EP_ I viii 13-14 'Caspius Aegissos, de se si credimus ipsis, / condidit et _proprio nomine_ dixit opus'. The phrase would have been very familiar to the scribes from grammatical treatises ('proper noun').

A combination of these circumstances no doubt induced the error.

Professor A. Dalzell suggests to me that _momen_ is perhaps correct, the notion being that the salt water keeps moving, and so does not freeze.

_Pondus_ would then be a (mistaken) gloss that has displaced _momen_ from the text; _nomen_ would be a simple misreading of _momen_.

=66. CERTIS ... MODIS.= 'Metre'; compare _Fast_ III 388 'ad _certos_ uerba canenda _modos_', Tib II i 51-52 'agricola ... primum ... cantauit _certo_ rustica uerba _pede_' and Manilius III 35 '_pedibus_ ... iungere _certis_'.

=67. DETINVI ... TEMPVS, CVRASQVE FEFELLI= _excerpta Politiani_ DETINVI ... TEMPVS CVRAMQVE FEFELLI _LT_ DETINVI ... CVRAS TEMPVSQVE FEFELLI _BCMFHI_. _Tempus fallere_ 'make time pass unnoticed' is perfectly acceptable Latin; compare _Tr_ III iii 11-12 'non qui labentia tarde / _tempora_ narrando _fallat_ amicus adest', _Her_ I 9-10 'nec mihi quaerenti spatiosam _fallere noctem_ / lassaret uiduas pendula tela manus', _Met_ VIII 651 'interea medias _fallunt_ sermonibus _horas_', _Tr_ IV x 112-14 'tristia ... carmine fata leuo. / quod quamuis nemo est cuius referatur ad aures, / sic tamen absumo _decipioque diem_', and _Her_ XIX 37-38 'tortaque uersato ducentes stamina fuso / feminea tardas _fallimus_ arte _moras_'. The difficulty with the manuscript reading in the present passage is that _detinui curas_ is without parallel.

Heinsius therefore accepted Politian's reading, citing in its support _Met_ I 682-83 'sedit Atlantiades et euntem multa loquendo / _detinuit_ sermone _diem_'. The Auctor Electorum Etonensium objected that _detinui tempus_ was inappropriate: 'poeta tempus detinere noluit, quod scilicet per se morari atque haerere uidebatur inuisum'. He conjectured DISTINVI CVRAS and Burman DIMINVI CVRAS, which he later found in one of his manuscripts. But _detinere_ here can have the same meaning 'occupy, keep busy' as it has at the _Metamorphoses_ passage, where A. G. Lee cites the present passage (with Politian's reading) and _Tr_ V vii 39 '_detineo studiis animum_ falloque dolores'.

The interchange of adjoining metrically and grammatically equivalent substantives is very common.

=67-68. "DETINVI" DICAM "TEMPVS, CVRASQVE FEFELLI; / HVNC FRVCTVM PRAESENS ATTVLIT HORA MIHI".= The thought of the passage also at ii 39-40 & 45 'quid nisi Pierides, solacia frigida, restant', _Tr_ V i 33-34 'tot mala pertulimus, quorum medicina quiesque / nulla nisi in studio est Pieridumque mora', and _EP_ I v 53-55 'magis utile nil est / artibus his, quae nil utilitatis habent. / consequor ex illis casus obliuia nostri'.

=69. ABFVIMVS SOLITO ... DOLORE.= Compare Cic _Fam_ IV iii 2 'a multis et magnis molestiis abes'; I have found no parallel from verse.