Skip to main content

Full text of "On the Culex and Other Poems of the Appendix Vergiliana"

See other formats


STOP 



Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world byJSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 
purposes. 

Read more about Early Journal Content at http://about.istor.org/participate-istor/individuals/early- 
journal-content . 



JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 
contact support@jstor.org. 



AMERICAN 
JOURNAL OF PHILOLOGY. 

Vol. III. No. ii. 



I.— ON THE CULEX AND OTHER POEMS OF THE 
APPENDIX VERGILIANA. 

The Culex, with the Ciris Dirae Moretum Copa and epigrams 
generally known as Catalecta, as well as the Aetna now usually 
ascribed to Lucilius, have recently been re-edited by Bahrens in 
the second volume of his Poetae Latini Minores. This work 
marks a great advance on the Appendix Vergiliana of Ribbeck, 
published in 1867, and has suggested to me many new views on 
these poems, which, from their peculiarly intimate relation to those 
of Catullus, have at all times had an attraction for me much beyond 
their intrinsic merit. None of them have come down to us in a 
more corrupt state than the ' Gnat,' and it is therefore of some 
importance to record from time to time the readings of a Bodleian 
MS, Auct. F I, 17, which here, as also in the Dirae (see Cam- 
bridge Journal of Philology, VIII 72), may be reckoned among 
the uninterpolated class, generally exhibiting a close resemblance 
to Bahrens' B. 

24-27. 

Et tu cui meritis oritur fiducia cartis, 
Octaui uenerande, meis adlabere coeptis 
Sancte puer, tibi namque canit non pagina bellum 
Triste lovis ponitque canit non pagina bellum 
Phlegra giganteo sparsa est quo sanguine tellus. 

It is obvious that in 26 canit non pagina bellum has been erro- 
neously repeated from the following verse. Ribbeck completed 
the lacuna by reading tibi namque humilis conamine prima, 
Bahrens tibi namque sonant mea carmina, quamquam. Retaining 



272 AMERICAN JOURNAL OF PHILOLOGY. 

this in outline, I would change sonant to merenf, which would 
repeat with emphasis the expression of 24 cui meritis, unneces- 
sarily altered by B. to cuius monitis. The metaphor is natural 
enough to a Roman : the poem takes service under the banners of 
Octavius : ponitque, which has been altered to Rhoetique, Rhoeci- 
que, Coeique, Cottique, Phorcique, may after all be Pontique, for 
Poseidon took his part in the war with the Giants (Apollod. I 6, 
2). I cannot agree with Ribbeck and Bahrens in supposing the 
Octavius to whom the poem is addressed to be any one but the 
youth who became later Octavianus and Augustus ; only so can 
the strong expressions Octaui tienerande, Sancte puer, which last 
occurs twice (26, 37), be adequately explained ; this too gives a 
meaning to the elaborate invocation to Apollo, a god especially 
associated with the history of Augustus. It is not necessary to 
suppose the poem actually written to the young Octavius ; for my 
own part I have never been able to regard it as anything but the 
composition of a later, but still early period, when the tradition 
that Virgil had written a Culex prompted some versifier to supply 
the required poem. Such a forger would naturally inscribe his 
Culex to Augustus, and as Virgil was supposed to have written it 
in boyhood, to Augustus still a boy. No one can, I suppose, read 
the verses eulogizing a country life (58 sqq.) and not feel certain 
that they are an imitation of the famous passage in Georg. II, 
O fortunaios nimium. 

37> 8. 

Haec tibi, sancte puer, memorabilis et tibi certet 

Gloria perpetuum lucens mansura per aeuum. 

Et tibi sede pia maneat locus, et tibi sospes 

Debita felicis memoretur vita per annos 

Grata bonis lucens. 

It is surprising that Bahr. retains et in 37. It is not only weak, 
but cacophonous in view of the double et in 39. I would read : 

Haec tibi, sancte puer, memorabimtcs : haec tibi restet 
Gloria, etc. 

And what can B. find so absurd in the words Et tibi s. p. maneat 
locus ? which he alters to Serum s. p. m. locus. Surely the poet, 
whose gnat finally rests in Elysium, might reasonably enough wish 
his patron the same good fortune ; tibi with maneat as in Cat. 
VIII 15; Phil. II 5, II, 



POEMS OF THE APPENDIX VERGILIANA. 273 

55- 

O bona pastons si quis non pauperis usum 

Mente prius docta fastidiat et probet illis 

Omnia luxuriae spretis incognita curis. 

In my Catullus of 1867 I conjectured thdX illisomnia had been 
corrupted into illis omnia, and suggested that the error arose from 
the two verses having at some time been written continuously. 
Subsequently I found that Haupt had conjectured somnia, leaving 
illis unchanged. I still prefer my original explanation of the cor- 
ruption and venture to think that most critics will consider illi 
more elegant, as it is certainly on other grounds more probable than 
illis. ' Happy the shepherd's lot, should there be any who scorns 
not the employment of the poor and commends the dreams that 
proud life of luxury never knew, despising the cares that torture 
the covetous.' 

57- 

Haec teneras fruticum sentes rimatur, at ilia 

Inminet in riui praestantis imaginis undam. 

Here imaginis is usually supposed to be a corruption of mar- 
ginis, wrongly, I fancy ; at least none of the emendations which it 
necessitates in the rest of the verse can be considered very prob- 
able. On the other hand the goat might well be described as 
hanging over the water to look at her own reflected image, like 
the horse in a well-known fragment of Sophocles (593). Hence I 
would read praesentis imaginis undam, an image-presenting 
stream. The double genitive, the latter of quality, is not harsher 
than the double abl. in 153. So Spenser in his translation, ' The 
whiles another high doth overlooke Her owne like image in a 
christall brooke.' 

89, 90. 

Illi dulcis adest requies et pura uoluptas 
Libera simplicibus curis. 

Rather duplicibus. 

lOI. 

Tendit ineuectus radios Hyperionis ardor, 
Lucidaque aethereo ponit discrimina mundo. 

This Strange word ineuectus, which is supposed to recur in 342 : 

Ne quisquam propriae fortunae munere diues 
Iret ineuectus caelum super 



274 AMERICAN JOURNAL OF PHILOLOGY. 

seems to raise no doubts in lexicographers, who explain it as 
' mounted upon.' We must then suppose that in the first passage 
the sun's heat is described as mounted on its rays, tetidit radios 
quibus inevedus est, and in the second that the rich man mounts 
on the chariot of his wealth above the sky. The first of these is 
undeniably harsh, and the word itself is spelt in Bahrens' MS 
V in eicectus ; while in 342 the Bodl. MS above alluded to (Mr. 
Macray, one of our best experts, dates it about 1230) gives euec- 
tus. In this passage indeed there are other signs of the traditional 
reading being wrong, for though the Bodl. MS (which I shall call 
F) like the others collated by Bahr. gives Iret, a Paris Anthology 
has Tendit, and it seems more than probable that Tenderet euectus 
is the right reading. But may not ineuectus in loi be, as we should 
at first sight more readily believe (cf inexcitus, inexhaustus, etc.), 
a negatival adj., ' not yet borne aloft,' i. e. to the highest part of 
the sky ? We must then suppose the poet to mean that up to that 
time of the day the sun had not reached the zenith, and now begins 
to approach it and disperse his rays equally to both sides of the 
sky. This quite agrees with what immediately follows, 107 lam 
m-edias operum partis euectus erat sol, which is a further step 
onwards, that part of the day when the sun had got beyond the 
zenith, at the hottest part of the afternoon. Still as V gives in 
eicectus and no authority is quoted but the Culex for ineuectus, it 
is possible that the right reading is in erecttim, ' the sun stretches 
his rays in an upright line,' not slanting at an angle as at an earlier 
or later period of the day. 

109. 

Vt procul aspexit luco residere uirenti, 
Delia diua, tuo, quo quondam uicta furore 
Venit Nyctelium fugiens Cadmeis Agaue, 
Infandas scelerata manus et caede cruenta. 

Pastor ut ad fontem densa requiescit in umbra, 
Mitem concepit proiectus membra soporem. 

Such I believe to be the real apodosis of Vt procul. With quo 
quondam victa furore begins a description of the grove, which 
continues for more than 50 lines, and thus disguises the fact that 
the sentence began with a protasis and remains incomplete. The 
nominative is accordingly resumed in Pastor ut. This will enable 
us to dispense with the changes suggested by Ribbeck, Bahrens 
and earlier editors. Nor can I think that any alteration is required 



157- 



POEMS OF THE APPENDIX VERGILIANA. 275 

in 112, either Bern bo's ^ or Ribbeck's very problematical ec : as 
abl. cruenta would be tautologous, as nom. ' gory with a deed of 
blood ' it is Virgilian (Aen. I 475) and animated. 

117. 

Tantum non horridus Hebrum 
Kestantem tenuit ripis siluasque canendo 
Quantum te per nigre morantem diua chorea 
Multa tuo laetae fundentes gaudia uultu. 

For horridus or ori'idus in 117 is ordinarily printed (as in Pithou's 
Collection, p. 6, ed. 1590, and even by Ribbeck and Bahrens) 
Orpheus, which is found only in one of B.'s MSS (V) as a first-hand 
reading, and can scarcely be right, whether on metrical or palaeo- 
graphical grounds. It is quoted indeed by L. Miiller (de r. m. 
p. 268) as a trisyllabic nominative with Orphet/s in Cul. 269 ; but 
this was before the MSS had been accurately collated ; and in 
269 Orpheos as a genitive is rightly restored by Ribbeck. The 
nearest approach I can find to orridus is odrisis, and we might 
then suppose the Odrysian region to be substituted for the Odrisian 
bard, which is a mild, almost tame license in the poet, if coni- 
pared with the • parallel description in Seneca's Hercules Oetaeus, 
1043 sqq., where Athos breaks part of its crags away with the 
Centaurs on them to come and stand near Rhodope, while Orpheus 
sings. If this should seem too bold, I would suggest Non tantum 
Oeagrius. V. 1 19 was emended by Haupt Quantum te, pernix, 
remorantur, diua, chorea, andpernix is actually written in V. But 
here again, as in 117, I hold the truer reading to be that of the 
other MSS, including F, and would read Quantum te pernice 
morantur, diva, chorea, by which the awkwardness of chorea as 
nom. followed by the plural laetae fundentes is obviated. 

123. 4- 

Nam primum prona surgebant valle patentes 
Aeriae tplatanos, inter quas impia lotos. 

So F ; B has platane with us written over ^ in a more modern 
hand ; V and several other MSS give platani. I should here, 
against Ribbeck, incline to regard platanos as the less corrupted 
reading, and, with Bembo, restore the rare but not incredible form 
platanus, which Neue seems to accept, Formenlehre I 536. 

127. 

At quibus tinsigni curru proiectus equorum 
Ambustus Phaethon luctu mutauerat artus 
Heliades. 



276 AMERICAN JOURNAL OF PHILOLOGY. 

This is no place for styling Phaethon's chariot splendid; an 
obvious correction is indigne. So Ovid speaking of the Sun's 
anger at Phaethon's death says, M. II 400 Saeuit enim natumque 
obiectat et inputat illis. 

131- 

Posterius cui Demophoon aeterna reliquit 

Perfidiam tlamentandi mala perfide multis 

Perfide Demophoon et nunc tdefende puellis. 

Bahrens is, I believe, right in reading lamentanti, as certainly 
wrong in his i nunc defendeque vela. It would be difficult to im- 
prove on Scaliger's deflende, which Ribbeck retains. ' Thou faith- 
less Demophoon, to many a maiden faithless, aye still a memory 
to rouse their tears,' a pleasing and natural apostrophe to the oft- 
repeated story of Phyllis' betrayal. 

137-9- 

Hie magnum Argoae naui decus edita (so F with most MSS, adita V) pinus 
Proceros (Proceras, MSS) tdecorat (decoras, F) siluas hirsuta per artus. 
Ac petit aeriis tcontingere Imontibus astra. 

I can hardly think decorat right. Possibly superat. Montibus 
was corrected by Scaliger to motibus, a very weak word ; by Hein- 
sius to frondibus. Audacious as to some it will seem, I believe 
the right word is morsibus ; for the successive growths by which 
the fir and pine are continually rising, a new apex marking the new 
growth, might not inaptly be described as so many bites in the air. 

153- 

Argutis et cuncta fremunt ardore cicadis. 

Bahrens, ingeniously, a rore. I doubt, however, whether the 
fact is so, and suggest stridore, the regular word for the peculiar 
sound of the cicada, Plin. XI 266 alia murmur edere, ut apis, alia 
cum tractu stridorem, ut cicadas, receptum enim duobus sub pectore 
cauis spiritum, mobili occursante membrana intus, attritu eius sonore. 

166-8. 

Obuia uibranti carpens grauis ore trilingui 
Squamosos late torquebat motibus orbes. 
tTollebant aurae uenientis ad omnia uisust. 

In ad omnia I think abdomina probably lurks. For aurae V 
has arte. Bahrens reads Tendebant acres venientis ad omnia 



POEMS OF THE APPENDIX VERGILIANA. 277 

uisus, which certainly gives a clearly defined picture of the restless 
eyes of the advancing snake ; but seems to me, as Latin, a little 
strained ; tendebant especially is hardly the right word, to say 
nothing of the fact that omnia several times marks a corruption, 
as in 217, 233, 242. Accepting nisus for uisus from Ribbeck 
I would read Tollebant acres {f) venienii abdomina nisus, the 
contortions of the snake in its progress cause the belly to be 
constantly lifted from the ground and exposed to view. Silius 
has nisu se concitat acri of a warrior, v. 235. Or can aurae con- 
ceal caudae ? Haupt's Pallebant aura uementis gramina uiri is 
inexpressibly violent, and will, I should fancy, convince no one, a 
remark which extends to many of his alterations of the Culex, 
especially in reference to his introduction of elisions against the 
MSS and in violation of the laws observed by the poet. See Birt's 
careful examination, Halieut. p. 50. 

In 177 Saepius arripiens should be retained, as a repeated dart- 
ing at objects in the way would be natural in an enraged serpent; 
similarly spiritibus rumpit fauces is not to be changed into spiritus 
erumpit f. (Heinsius), the plural expresses the convulsive and 
continual motion of the hissing throat. 

185, 6. 

Qua diducta genas pandebant lumina gemmis 
Hac senioris erat nature pupula telo 
Icta leui. 

Forbiger explains ' where the unclosed eyes laid open the lids to 
the pupil,' i. e. for the eye-ball to exert its function of seeing, sup- 
posing gemma to be another word for pupula. But no instance 
of such a meaning is quoted, and the resemblance of sound in 
genas gemmis, as well as the iteration /w/w/a in 186 (Bahrens alters 
this to palpebra), is suspicious. Possibly pennis ' to the gnat's 
wings,' i. e. to the approach of the whirring gnat. Nature is, of 
course, as Bothe saw, a mistake for mature, ' in time' to avoid 
the serpent's bite. I do not think palpebra is right ; (i) it is not 
the MS reading ; (2) the word seems only to occur in the plural 
and with the e long, Lucr. IV 952 ; (3) if the eyes are stated to 
have been unclosed, it was because the eye-ball, not the eye-lid, 
was stung by the gnat. 

193-5- 

Quam casus sociarit opem numenue deorum 

Prodere sit dubium, ualuit sed uincere tali 

Horrida squamosi uoluentia membra draconis. 



278 AMERICAN JOURNAL OF PHILOLOGY. 

It is not necessary to change tali (V) into talis. Here tale is 
' such a thing,' ' so slight a thing,' as omne is used for ' everything.' 
F with two of B.'s MSS has talis, which perhaps points to tale 
(nom.) as what the poet wrote. 

198-201. 

Et quod erat tardus TOmni languore remote 
tNescius aspiciens timor obcaecaverat artus 
Hoc minus implicuit dira formidine mentem 
Quem postquam uidit caesum languescere sedit. 

Bahrens is perhaps right in transposing 201 before 198, for 198- 
200 seem to explain sedit ; the shepherd having killed the snake, 
instead of moving away at once from the scene of danger, sat 
down with less appearance of dismay than might have been 
expected, (1) et quod, because he was still drowsy from the 
sleep from which he had been suddenly awoke (remoto) ; 
(2) because the sudden alarm of the sight of the serpent 
had for a while paralyzed his limbs and made him unwilling 
to move. Hence for Nescitis I would read Nee secus. Bahrens' 
Quo plus seems to me too remote for the MSS, nor can 
I think his astringens for aspiciens probable. F has tonor for 
timor ; but though Quintilian (I 5, 23) says tonor was an old 
form of tenor in the sense of accent, it can hardly mean anything 
like rigor or tension of the limbs, and must therefore, I think, be 
dismissed. There is, however, some weakness in timor , formidine 
in two consecutive lines. If aspiciens timor is thought, as perhaps 
it may be, too harsh, ' and similarly fear at the sight of the snake,' 
it would be easy to read ad speciem. 

225-7. 

Praemia sunt pietatis ubi, pietatis honores r 
In uanas abiere uices tet iure recessit 
lustitiae prior ilia fides. 

For et iure, the reading of F and most MSS, V has uita, whence 
Bahrens reads et uicta recessit lustitia et (Schrader) prior ilia fides. 
Is not et here somewhat weak ? If V represents the true tradition, 
I should prefer euicta, ' driven out of its holdings, dispossest '; if 
the other MSS, perhaps abiere, a repetition corresponding to that 
of pietatis in 225. 

239 sqq. 

Terreor a tantis insistere, terreor, umbris. 
Ad Stygias reuocatus aquas uix ultimus amnl 
Kestat nectareas diuum qui prodidit escas 



POEMS OF THE APPENDIX VERGILIANA. 279 

Gutturis arenti reuolutus in omnia sensu. 
Qui saxum procul aduerso qui monte reuoluit 
Contempsisse dolor quem numina uincit acerbas 
Otia querentem frustra siblite puelle 
Ite quibus tedas accendit tristis Erinis 
Sicut himen prelata dedit conubia mortis. 

In this difficult passage the poet recalls himself to the descrip- 
tion of the infernal world : ' I shudder to dwell on such grim 
shadows, to return to the waters of Styx.' Hence Ad St. reuocatus 
aquas should be constructed with terreor, not with extat. At uix 
begins the description of Tantalus' punishment. Extat for Restat 
(Heinsius) is certain, which cannot be said of any emendation yet 
proposed for revolutus in omnia. We saw above that omnia is a 
frequent residue of error ; in 217 it seems to represent moenia 
(Sillig) ; in 233 Quem circa tristes densentur in omnia (in oninua 
F) Poenae, it is, I believe, a mistake for ostia, as the Poenae would 
naturally gather at the door of Hell ; in the line before us Ribbeck 
may be right in conjecturing inania, and if so, reuolutus (which can 
hardly stand with reuoluit in the next line) may be a mistake for 
releuatus, a word peculiarly appropriate to relief of hunger or 
thirst. Or is it possible that in omnia is here for insomnia f then 
resolutus may represent some active participle, reparans, renouans 
or the like. The next five verses I would write sis follows : 

Quid saxum procul adverse qui monte reuoluit, 
Contempsisse dolor quem numina vincit acerba«s, 
Otia quaerentem frustrarfb«j- ? Ite puellae, 
Ite quibus taedas accend^KJ tristis Erinys, 
Sicut Hymen, praefata dedit conubia mortis. 

The reference is to Sisyphus and the Danaides. Acerbans is, I 
imagine, better than acerbus or acerba, and here again I find the 
Bodl. MS a reliable guide ; acerbas is another instance of the 
suppressed n of the nomin. participle of which Corssen collects so 
many instances. Frustratibus is rare, but occurs in Plautus ; it 
might aptly enough express the baffled attempts of Sisyphus to 
roll the stone to the top of the mountain. The allusion in the last 
two verses is to the deadly bridal of the Danaides, ' to whom the 
Fury, speaking the words of prelude, as it were Hymen (Cat. 
LXIV 382), assigned a bridal that was death.' 

265, 6. 

Ecce Ithaci coniunx semper decus Icariotis 
Femineum concepta decus manet. 



28o AMERICAN JOURNAL OF PHILOLOGY. 

Decus in 265 is generally altered to ducts, in consequence of 
decus in 266. But it is not certain that this is the right word there, 
for F has what looks like recus. May not this represent secus, 
sex ? With this Bahrens' consaepta would well agree. 

274. 5- 

Ecfossasque (Necfossasque MSS) domos ac tartara nocte cruenta 
Obsita, nee faciles ditis sine iudice sedes. 

Ecfossas, not Defossas, is what MSS point to, ' homes dug out 
of the earth,' i. e. subterranean and dark. The form ecfodere is 
indubitable in Tacitus and Cicero as well as Plautus, as Lewis and 
Short show from Neue Formenl. II 767. Didaeo (Bahrens) is 
very plausible, yet sine must, I think, be genuine ; perhaps, there- 
fore, Dictae sine is what the poet wrote. There is too strong a 
tendency in editors to eliminate difficult negatives or words imply- 
ing a negative. Thus in Heroid. XII 169, 170, Medea says Non 
mihi grata dies, nodes uigilantur amarae, Nee tener a misero pec- 
tore somnus abit, for so I would modify A. Palmer's conjecture, fol- 
lowing the MSS, which would hardly have changed Nee into Et. 
Nee qualifies tener, ' and sleep, not the soft sleep of a happy lover, 
flies from me.' So in the passage of the Culex before us, nee 
extends both to faciles and sine iudice, ' and the abodes that smile 
not with Dicte's judge away,' i. e. ' the abodes where Dicte's judge 
is ever present to make them forbidding.' Cf the remarks of Birt, 
Halieut. p. 49. 

286-288. 

Haec eadem potuit Ditis te uincere coniunx 
Eurudicenque ultro ducendam reddere : non fas 
Non eiat inuitam dire exorabile mortis. 

' This same lyre had power to persuade thee, consort of Pluto, 
and to restore Eurydice unasked to be led away. But it might 
not be ; to traverse the path of dreadful death was not to be won 
by entreaty.' I read then ^ ire uiam. F, both here and in 268, has 
Erudice, in which I trace a vestige of the old spelling Eurudice. 

294. 

Dignus amor venia tgratiam si Tartara nossent. 

So F, gratum most MSS, as" I incline to think, rightly. ' Grati- 
tude,' viz. for Orpheus' devotion. Birt reads gratam, explaining 
of Proserpine, Halieut. p. 53. 

^ Birt, Non fas, Non erat : Inuictae diuae exorabile numen. Halieut. p. 53. 



POEMS OF THE APPENDIX VERGILIANA. 281 

295-303- 

Peccatum meminisse graves tuos sede piorum 

Vos manet heroum contra manus, hie et uterque 

Aeacides, Peleus namque et Telamonia virtus 

Per secura patris laetantur numina, quorum 

Conubiis uenus et uirtus iniunxit honorem. 

Hunc rapuit ferit ast ilium nereis amavit. 

Adsidet hac iuvenis sociat de gloria sortis 

Alter in excissum referens a navibus ignis 

Argolicis Phrygios turba feritate repulsos. 

No passage of the Culex is more corrupt than this. I will give 
what appears to me the connexion of thought. ' Yet it were shame 
to remember Orpheus' sin : ye are both (Orpheus and Eurydice) 
destined to rest in Elysium with the heroes of old time. In Ely- 
sium are both the Aeacids Peleus and Telamon, rejoicing in the 
tranquil assurance of their father's divine power (ApoUod. Ill 12, 

15 Tt^OTat hi K.di irapa TiKovraivi reXeuT^cray AlaKos koI ras kKcis tov "Atdov 

(^uXaTTct), and in life raised by their prowess and the love they in- 
spired to marriages of high consideration. Seated near is Ajax, 
associated with them by the allotment of destiny — Ajax of boldness 
unapproachable, telling how the Trojans were beaten back in con- 
fusion from the Greek ships which they would fain have set on fire.' 
The whole passage I would write thus : 

Peccatum meminisse gravej'^ (Bahr.) : uos sede piorum 
Vos manet heroum contra manus. Hie et uterque 
Aeacides : Peleus namque et Telamonia virtus 
Per secura patris laetantur numina, quorum 
Conubiis uenus et uirtus iniunxit honorem. 
Hunc rapit Hesiona, ast ilium Nereis amauit. 
Adsidet huic invenis, sociat ^uem (Bahr.) gloria sortis, 
Acer (Bemb.) inaccwjum, referens a nauibus ignis 
Argolicis Phrygios turba trepidante repulsos. 

The most doubtful point in these verses is the obviously corrupt 
feritast {feritas V) and again feritate (303). It is remarkable 
thdX feritatis recurs in 311 where it is undoubtedly right ; but it 
cannot but be wrong I think in each of the former places. Bembo 
conjectured in 300 serva ast, Schrader Periboea, which Ribbeck and 
Bahrens adopt. I greatly doubt the possibility of peribea becom- 
ingferiias; ast is thoroughly in its place in a contrast of this kind; 
in some forms of writing, Hesionast might easily be misread feri- 
tast; while to supply an exact parallel might be quoted Ovid M. 



282 AMERICAN JOURNAL OF PHILOLOGY. 

215 sqq. Nee pars militiae Telamon sine honore recessit, Hesione 
que data potitur. Nam coniuge Peleus Clarus erat diua. In the 
next verse Bahrens seems right in recaUing quern of H for de of F 
and most MSS, but I see no reason for changing sortis to sorti. 
For in excissum [excidium H, excelsum V) which Bahr. alters to 
in excessum, with very dubious meaning, I would write inaccessum, 
a rare word which easily became obscured; iurba seems to be 
right, as Homer speaks of the confused scene which ensued when 
the Trojans were driven back from the attack on the ships, II. XVI 
294: 

'H/iuSaijf 8' apa vrjvs XiVer' avroBi ' rot S' ecpo^ridev 
Tpaes 6c<r7re(ri<a 6fjui8<f. Aavaol 8' eTrexovro 
N^af ava y\a<jivpds ' ofiaSos S' aXiaaros iTv^Bi). 

And again, 367 : 

Qy t5>v ck vrjwv yevero laxr) re (fio^os re, 
OvSc KUTa jiotpav mpaov iraXtp. 

Besides, torua feritate is feeble, and everything points to the cor- 
ruption lying not in iurba, but feritate, for which V has feritare, 
H fremitante. What word these variants conceal is of course 
doubtful ; trepidante is tolerably near and gives excellent sense. 

304- 

O quis non referat talis diuortia belli ? 

Diuortia is perhaps a translation of the Homeric n-oXtp^io y€<t>vpas. 
3". 312. 

Ipsa uagis namque Ida potens feritatis et ipsa 
Ida faces altrix cupidis praebebat alumnis, 
Omnis ut in cineres Rhoetei litoris ora 
Classibus ambustis flamma lacrimante daretur. 

Bembo wrote iugis for uagis,'which Heinsius completed by writing 
frondentibus iox feritatis et. But (i) the repetition of the two words 
ipsa Ida might well be accompanied by a connecting et; (2) potens 
ox patens is an obvious corruption oi parens which, retaining yi?rz- 
tatis, will then be a translation of the Homeric p-ipirip 6ripS>v. Hence 
uagis {uatis H) must conceal some accusative, possibly trabes, the 
material of spears. \{Jtamma lacrimante is right, it can only mean 
an oozy flame such as is produced by pitch and similar resinous sub- 
stances. My friend Mr. Shadworth Hodgson suggests lambente. 
In the difficult passage which follows this the word Tegminibus 



POEMS OF THE APPENDIX VERGILIANA. 283 

can, I think, hardly represent Ignibus hie, but either Fragminibus 
or perhaps Hie manibus ; for this last cf. II. XV 716 'EKzap Be n-p^/x- 
prj6ev cVei Xd^ev, ovxt fi(6iei ' h(f)Ka<TTov /lera )(ep(riv c;fa>i'. 

325. 6. 

Rursus acerba fremunt Paris hunc quod letat et huius 
Arma dolis Ithaci virtus quod concidit icta. 

Arma surely cannot be Alma, for who would think of applying 
such an epithet to the valor of Ajax? Bahrens suggests that 
Arma is a relic of two lost verses, in which the adjudication of the 
arms of Achilles to Ulysses and the subsequent death of Ajax were 
narrated. F omits the words after uirtus. 

327-330. 

Huic gerit auersos proles Laertia uultus 

Et iam Strymonii Rhesi victorque Dolonis 

Pallade tiam laetatur ouans, rursusque tremiscit 

lam Ciconas iamque horret [atrox Laestrygonas ipse]. 

It is inconceivable that iam should be repeated four times so 
meaninglessly. Read Pallade laetabatur ouans, and cf. 50 sqq. 
tondebant, earpuntur, petuniur. The words after horret are omitted 
in F ; so in 334 it omits Atrides after gener amplis {sie') ; in 340 
it has only one word, Neque ; in 362 it omits moriiura metelli. 

363. 4- 

Curtius et mediis quem quondam sedibus urbis 
Deuotum jbellis consumpsit gurgitis unda. 

For bellis a not improbable emendation is uiolens. 
370. I- 

Scipiadaeque duces, quorum deuota triumphis 
Moenia frapidis Libycae Carthagiuis horrent. 

H gives iapidis. This suggests lappis, the burs or weeds 
which spring up on neglected sites, Virg. G. I 152. Haupt's 
uepretis conveys the same idea, but is farther from the MSS. Sub 
seems to have fallen out. 

374. 5- 

Et uastum Phlegethonta pati, quo, maxime Minos, 
Conscelerata pia discernis uincula sede. 

' Phlegethon by which Minos separates the prison of the guilty 
from the abode of the blest.' I cannot see that vincula requires 
any change against all MSS. 



284 AMERICAN JOURNAL OF PHILOLOGY. 

399- 

Et rosa purpureum crescent rubibunda terrorem. 

So B ; F has quiescant rubicunda ; V pudibunda ; H gives 
tenorem ; C per orbem. The old reading crescens is to my mind 
made probable by the peculiar form it assumes in F ; the whole 
line I would read 

Et rosa purpureum crescens pudibunda per orbem, 

' growing in the folds of a crimson disk.' Per denotes the gradual 
accretion of the petals into the full flower. 

Dirae 83. 

Tuque inimica ttui semper discordia eiuis. 

Bahrens rightly calls tui meaningless ; but boni is not so prob- 
able a restoration as pii. In Prop. Ill 13, 56 hospitio non, Poly- 
dore,pio most of the MSS have tuo ; and in II 25, 31 Namque 
in amore stio semper sua maxima cuique Nescio qtw pacto uerba 
nocere solent, the meaning is in favor of pio, a faithful love, as 
opposed to a wandering and shifting passion. 

91. 3- 

Tardius a miserae descendite monte capellae. 

Mollia non iterum carpetis pabula nota. 

Tuque resiste pater, tea prima nouissima uobist. 

The general sense is clear ; the she-goats and their male leader 
are leaving forever their browsing-ground. They are therefore 
told to linger and crop their last meal. Possibly then we should 
read ea thym-bra nouissima uobis, ' that is the last meal of savory 
you will ever see,' or cyma, ' the last sprout.' 

Lydia 14. 

Membra reclinarit tteneremque illiserit herbam. 

The Bodleian MS like most of Bahrens' has fenerem (not tene- 
ram) ; H veneri. Hence I would read temere atque. 

R. Ellis.